Bookshelf NBK133363 PDF

Evidence Report/Technology Assessment
Number 211
Making Health Care

Safer II: An Updated
Critical Analysis of the
Evidence for Patient
Safety Practices
Evidence-Based
Practice
Patient Safety

Number 211
Making Health Care Safer II: An Updated Critical

Analysis of the Evidence for Patient Safety Practices
Prepared for:
Agency for Healthcare Research and Quality
U.S. Department of Health and Human Services
540 Gaither Road
Rockville, MD 20850
www.ahrq.gov
Contract No. 290-2007-10062-I
Prepared by:
RAND Corporation, Santa Monica, CA
University of California, San Francisco/Stanford, San Francisco, CA
Johns Hopkins University, Baltimore, MD
ECRI Institute, Plymouth Meeting, PA
AHRQ Publication No. 13-E001-EF

March 2013
This report is based on research conducted by the Southern California-RAND Evidence-based

Practice Center (EPC) under contract to the Agency for Healthcare Research and Quality
(AHRQ), Rockville, MD (Contract No. 290-2007-10062-I). The findings and conclusions in this
document are those of the authors, who are responsible for its contents; the findings and
conclusions do not necessarily represent the views of AHRQ. Therefore, no statement in this
report should be construed as an official position of AHRQ or of the U.S. Department of Health
and Human Services.
The information in this report is intended to help health care decisionmakerspatients and
clinicians, health system leaders, and policymakers, among othersmake well-informed
decisions and thereby improve the quality of health care services. This report is not intended to
be a substitute for the application of clinical judgment. Anyone who makes decisions concerning
the provision of clinical care should consider this report in the same way as any medical
reference and in conjunction with all other pertinent information, i.e., in the context of available
resources and circumstances presented by individual patients.
This report may be used, in whole or in part, as the basis for development of clinical practice
guidelines and other quality enhancement tools, or as a basis for reimbursement and coverage
policies. AHRQ or U.S. Department of Health and Human Services endorsement of such
derivative products may not be stated or implied.
Persons using assistive technology may not be able to fully access information in this report. For
assistance contact [email protected].
None of the investigators have any affiliations or financial involvement that conflicts with the
material presented in this report.
Suggested citation: Shekelle PG, Wachter RM, Pronovost PJ, Schoelles K, McDonald KM, Dy
SM, Shojania K, Reston J, Berger Z, Johnsen B, Larkin JW, Lucas S, Martinez K, Motala A,
Newberry SJ, Noble M, Pfoh E, Ranji SR, Rennke S, Schmidt E, Shanman R, Sullivan N, Sun F,
Tipton K, Treadwell JR, Tsou A, Vaiana ME, Weaver SJ, Wilson R, Winters BD. Making Health
Care Safer II: An Updated Critical Analysis of the Evidence for Patient Safety Practices.
Comparative Effectiveness Review No. 211. (Prepared by the Southern California-RAND
Evidence-based Practice Center under Contract No. 290-2007-10062-I.) AHRQ Publication No.
13-E001-EF. Rockville, MD: Agency for Healthcare Research and Quality. March 2013.
www.ahrq.gov/research/findings/evidence-based-reports/ptsafetyuptp.html.
ii
Preface
The Agency for Healthcare Research and Quality (AHRQ), through its Evidence-based
Practice Centers (EPCs), sponsors the development of systematic reviews to assist public- and
private-sector organizations in their efforts to improve the quality of health care in the United
States. These reviews provide comprehensive, science-based information on common, costly
medical conditions, and new health care technologies and strategies.
Systematic reviews are the building blocks underlying evidence-based practice; they focus
attention on the strength and limits of evidence from research studies about the effectiveness and
safety of a clinical intervention. In the context of developing recommendations for practice,
systematic reviews can help clarify whether assertions about the value of the intervention are
based on strong evidence from clinical studies. For more information about AHRQ EPC
systematic reviews, see www.effectivehealthcare.ahrq.gov/reference/purpose.cfm.
AHRQ expects that these systematic reviews will be helpful to health plans, providers,
purchasers, government programs, and the health care system as a whole. Transparency and
stakeholder input are essential to the Effective Health Care Program. Please visit the Web site
(www.effectivehealthcare.ahrq.gov) to see draft research questions and reports or to join an
email list to learn about new program products and opportunities for input.
We welcome comments on this systematic review. They may be sent by mail to the Task
Order Officer named below at: Agency for Healthcare Research and Quality, 540 Gaither Road,
Rockville, MD 20850, or by email to [email protected].
Carolyn M. Clancy, M.D.
Director
Jean Slutsky, P.A., M.S.P.H.

Director, Center for Outcomes and Evidence
Stephanie Chang, M.D., M.P.H.

Director
Evidence-based Practice Program
Center for Outcomes and Evidence
James Battles, Ph.D.

Task Order Officer
Center for Quality Improvement and
Patient Safety
iii
Authors and Affiliations

Co-Principal Investigators:
Paul G. Shekelle, M.D., Ph.D., RAND Corporation Evidence-based Practice Center
Robert M. Wachter, M.D., University of California, San Francisco
Peter J. Pronovost, M.D., Ph.D., Johns Hopkins University
ECRI Institute:
Scott Lucas, Ph.D., P.E.
Meredith Noble, M.S.
James Reston, Ph.D., M.P.H.
Karen Schoelles, M.D., S.M., FACP
Nancy Sullivan, B.A.
Fang Sun, M.D., Ph.D.

Kelley Tipton, M.P.H.
Jonathan R. Treadwell, Ph.D.
Amy Tsou, M.D. M.S.
Johns Hopkins University:

Sallie J. Weaver, Ph.D.
Bradford D. Winters, M.D., Ph.D.
Elizabeth Pfoh, M.P.H.
Renee Wilson, M.S.
Kathryn Martinez, M.P.H.; Ph.D.

Sydney M. Dy, M.D., M.Sc.
Zack Berger, M.D., Ph.D.
RAND Corporation:
Breanne Johnsen, B.S.
Jody Wozar Larkin, M.L.I.S.
Aneesa Motala, B.A.

Roberta Shanman, M.L.S.
University of California, San Francisco/Stanford:

Kathryn M. McDonald, M.M.
Sumant R. Ranji, M.D.
Stephanie Rennke, M.D.
Eric Schmidt, B.A.
University of Toronto:
Kaveh Shojania, M.D.
Communications Analysts:
Sydne J. Newberry, Ph.D., RAND Corporation
Mary E. Vaiana, Ph.D., RAND Corporation
Acknowledgments
The authors gratefully acknowledge the following individuals for their contributions to this
project.
Technical Expert Panel
Alyce Adams, Ph.D.
Kaiser Permanente
Oakland, CA
iv
Peter B. Angood, M.D., FACS, FCCM

American College of Physician Executives
Tampa, FL
David Bates, M.D., M.Sc.
Harvard University, Brigham and Womens Hospital
Boston, MA
Leonard Bickman, Ph.D.
Vanderbilt University Peabody College
Nashville, TN
Pascale Carayon, Ph.D.
University of WisconsinMadison
Madison, WI
Sir Liam Donaldson, M.B.Ch.B., M.Sc., FRCS(Ed), MFPHM, M.D., FFPHM
Imperial College London
London, United Kingdom
Naihua Duan, Ph.D.
New York State Psychiatric Institute
New York, NY
Donna Farley, Ph.D., M.P.H.
RAND Corporation
Pittsburgh, PA
Trisha Greenhalgh, B.M.B.C.H., M.R.C.P., M.R.C.G.P., M.D. thesis, FRCP, FRCGP, FHEA,
FFPH
Global Health, Policy and Innovation Unit
Centre for Primary Care and Public Health
Blizard Institute
Barts and The London School of Medicine and Dentistry
London, UK
John L. Haughom, M.D.
PeaceHealth
Eugene, OR
Eileen Lake, Ph.D., R.N., FAAN
University of Pennsylvania
Wynnewood, PA
Richard Lilford Ph.D., FRCP, FRCOG, FFPHM

University of Birmingham
Edgbaston, Birmingham, UK
Kathleen Lohr, Ph.D., M.Phil.
RTI International-UNC Evidence-based Practice Center
Research Triangle Park, NC
Gregg S. Meyer, M.D., M.Sc.
Dartmouth Institute for Health Policy and Clinical Practice
Lebanon, NH
Marlene R. Miller, M.D., M.Sc.
Johns Hopkins Childrens Center
Baltimore, MD
Duncan Neuhauser, Ph.D., M.B.A., M.H.A.
Case Western Reserve University
Cleveland, OH
Gery Ryan, Ph.D.
RAND Corporation
Santa Monica, CA
Sanjay Saint, M.D., M.P.H.
Veterans Affairs Ann Arbor Healthcare System
University of Michigan
Ann Arbor, MI
Steve Shortell, Ph.D., M.P.H., M.B.A.
University of California, Berkeley
Berkeley, CA
David Stevens, M.D.
Dartmouth Institute for Health Policy and Clinical Practice
Lebanon, NH
Kieran Walshe, Ph.D.
Manchester Business School, University of Manchester
Manchester, UK
Peer Reviewers
Scott Flanders, M.D.
University of Michigan
Ann Arbor, MI
vi
Rainu Kaushal, M.D., M.P.H.

Harvard Medical School, Division of General Internal Medicine, Brigham and Womens
Hospital
Boston, MA
Harvey J. Murff, M.D., M.P.H.
Vanderbilt Institute for Medicine and Public Health
Nashville, TN
Jeffrey M. Rothschild, M.D., M.P.H.
Harvard Medical School, Division of General Medicine, Brigham and Womens Hospital
Boston, MA
Hardeep Singh, M.D., M.P.H.
Houston VA Health Services Research & Development Center of Excellence
Houston, TX
Donna Woods, Ph.D.
Northwestern University Feinberg School of Medicine
Chicago, IL
Additional Contributors
General Contributors
ECRI Institute
Michele Datko, M.S.
Lydia Dharia
Kitty Donahue
Eileen Erinoff, M.S.L.I.S.
Gina Giradi, M.S.
Allison Gross, M.S., M.L.S.

Janice Kaczmarek, M.S.
Laura Koepfler, M.L.S.
Kristy McShea, M.S.L.I.S.
Kristi Yingling, M.A., M.S.L.S.
Additional Chapter Contributors

Steven C. Bagley, M.D.
Veterans Affairs Palo Alto Healthcare
System
Palo Alto, CA
Chapter 26
Despina Contopoulos-Ioannidis, M.D.

Stanford University
Stanford, CA
Chapter 35
Sean M. Berenholtz, M.D., M.H.S.

Johns Hopkins University
Baltimore, MD
Chapter 11
Kristina M. Cordasco, M.D., M.P.H.,

M.S.H.S.
Veterans Affairs Greater Los Angeles
Healthcare System
Los Angeles, CA
Chapter 39
Vineet Chopra, M.D., M.Sc.

University of Michigan, School of Medicine
Ann Arbor, MI
Chapter 10
vii
Paul Dallas, M.D.

Carilion Clinics CME Program
Roanoke, VA
Chapter 18
Elliott R. Haut, M.D., FACS

Baltimore, MD
Chapter 28
Cyrus Engineer, Dr.P.H.

Baltimore, MD
Chapter 8
Susanne Hempel, Ph.D.

RAND Corporation
Santa Monica, CA
Chapter 19
Mohamad G. Fakih, M.D., Ph.D.

St. John Hospital and Medical Center
Detroit, MI
Chapter 9
Lawrence A. Ho, M.D.

Stanford University, Stanford University
Medical Center
Stanford, CA
Chapter 38
Olavo A. Fernandes, B.Sc.Phm., Pharm.D.

University of Toronto
Toronto, Ontario, Canada
Chapter 25
John Ioannidis, M.D.

Stanford University
Stanford, CA
Chapter 35
Tabor Flickinger, M.D., M.S.

Baltimore, MD
Chapter 32
Devan Kansagara, M.D., M.C.R., FACP

Portland Veterans Affairs Medical Center
Portland, OR
Chapter 22
David A. Ganz, M.D., Ph.D.

Healthcare System
Los Angeles, CA
Chapter 19
Sarah Kianfar, M.S.

Center for Quality and Productivity
Improvement
Department of Industrial and Systems
Engineering
University of Wisconsin-Madison
Madison, WI
Chapter 31
Melinda Maggard-Gibbons, M.D., M.S.H.S.

University of California, Los Angeles
Los Angeles, CA
Chapter 14
Sarah L. Krein, Ph.D., R.N.

Ann Arbor Veterans Affairs Medical Center
Ann Arbor, MI
Chapters 9 and 10
Peter Glassman, M.B.B.S., M.Sc.

Healthcare System
Los Angeles, CA
Chapters 4 and 5
Janice Kwan, M.D.

Chapter 25
Sara N. Goldhaber-Fiebert, M.D.

Stanford University, School of Medicine
Stanford, CA
Chapter 38
viii
Brandyn D. Lau, M.P.H.

Baltimore, MD
Chapter 28
Oanh K. Nguyen, M.D.

University of California, San Francisco
San Francisco, CA
Chapter 37
Lisha Lo, M.Sc.

Chapter 25
Russell N. Olmsted, M.P.H., C.I.C.

St. Joseph Mercy Health System
Ann Arbor, MI
Chapters 9 and 10
Julia Lonhart, B.S., B.A.

Stanford University
Stanford, CA
Chapter 35
Noelle Pineda, B.A.

Stanford University
Stanford, CA
Chapter 35
Lisa Lubomski, Ph.D.

Baltimore, MD
Chapter 33
Michael A. Rosen, Ph.D.

Baltimore, MD
Chapter 40
Yimdriuska Magan, B.S.

San Francisco, CA
Chapters 23 and 37
Nasia Safdar, M.D., Ph.D.

University of WisconsinMadison, School
of Medicine and Public Health
Madison, WI
Chapter 10
Brian Matesic, B.S.

Stanford University
Stanford, CA
Chapter 35
Marin Schweizer, Ph.D.

University of Iowa Health Care
Iowa City, IA
Chapter 7
Jennifer Meddings, M.D., MS.c.

Ann Arbor, MI
Chapter 9
Marwa H. Shoeb, M.D., M.S.

San Francisco, CA
Chapter 37
Isomi M. Miake-Lye, B.A.

Healthcare System
Los Angeles, CA
Chapter 19
David Thompson, D.N.Sc., M.S., R.N.

Baltimore, MD
Chapter 3
Erika Moseson, M.D.

San Francisco, CA
Chapter 23
Anping Xie, M.S.

Center for Quality and Productivity
Improvement
Department of Industrial and Systems
Engineering
University of WisconsinMadison
Madison, WI
Chapter 31
ix
Making Health Care Safer II: An Updated Critical

Analysis of the Evidence for Patient Safety Practices
Structured Abstract
Objectives. To review important patient safety practices for evidence of effectiveness,
implementation, and adoption.
Data sources. Searches of multiple computerized databases, gray literature, and the judgments
of a 20-member panel of patient safety stakeholders.
Review methods. The judgments of the stakeholders were used to prioritize patient safety
practices for review, and to select which practices received in-depth reviews and which received
brief reviews. In-depth reviews consisted of a formal literature search, usually of multiple
databases, and included gray literature, where applicable. In-depth reviews assessed practices on
the following domains:
How important is the problem?
What is the patient safety practice?
Why should this practice work?
What are the beneficial effects of the practice?
What are the harms of the practice?
How has the practice been implemented, and in what contexts?
Are there any data about costs?
Are there data about the effect of context on effectiveness?
We assessed individual studies for risk of bias using tools appropriate to specific study
designs. We assessed the strength of evidence of effectiveness using a system developed for this
project. Brief reviews had focused literature searches for focused questions. All practices were
then summarized on the following domains: scope of the problem, strength of evidence for
effectiveness, evidence on potential for harmful unintended consequences, estimate of costs, how
much is known about implementation and how difficult the practice is to implement. Stakeholder
judgment was then used to identify practices that were strongly encouraged for adoption, and
those practices that were encouraged for adoption.
Results. From an initial list of over 100 patient safety practices, the stakeholders identified 41
practices as a priority for this review: 18 in-depth reviews and 23 brief reviews. Of these, 20
practices had their strength of evidence of effectiveness rated as at least moderate, and 25
practices had at least moderate evidence of how to implement them. Ten practices were
classified by the stakeholders as having sufficient evidence of effectiveness and implementation
and should be strongly encouraged for adoption, and an additional 12 practices were classified
as those that should be encouraged for adoption.
Conclusions. The evidence supporting the effectiveness of many patient safety practices has
improved substantially over the past decade. Evidence about implementation and context has
also improved, but continues to lag behind evidence of effectiveness. Twenty-two patient safety
practices are sufficiently well understood, and health care providers can consider adopting them
now.
x
Contents
Executive Summary .................................................................................................................. ES-1
Part 1. Overview ............................................................................................................................. 1
Chapter 1. Introduction ............................................................................................................. 1
Who Will Use This Report, and for What Purpose?........................................................... 2
References ........................................................................................................................... 3
Chapter 2. Methods ................................................................................................................... 5
Topic Development ............................................................................................................. 5
Project Overview ................................................................................................................ 6
Topic Refinement................................................................................................................ 6
Evidence Assessment Framework .................................................................................... 11
Evidence Review Process ................................................................................................. 13
Assessing Quality of Individual Studies ........................................................................... 15
Assessing Strength of Evidence for a Patient Safety Practice .......................................... 16
Summarizing the Evidence ............................................................................................... 16
Setting Priorities for Adoption of Patient Safety Practices ............................................... 18
Future Research Needs ..................................................................................................... 19
Peer and Public Review Process ....................................................................................... 20
References ......................................................................................................................... 20
Part 2. Evidence Reviews of Patient Safety Practices .................................................................. 21
Part 2a. Practices Designed for a Specific Patient Safety Target ................................................. 23
Section A. Adverse Drug Events ............................................................................................ 23
Chapter 3. High-Alert Drugs: Patient Safety Practices for Intravenous Anticoagulants .. 23
Chapter 4. Clinical Pharmacists Role in Preventing Adverse Drug Events: Brief Update
Review .............................................................................................................................. 31
Chapter 5. The Joint Commissions Do Not Use List: Brief Review (NEW*) ............ 41
Chapter 6. Smart Pumps and Other Protocols for Infusion Pumps: Brief Review
(NEW*) ............................................................................................................................. 48
Section B. Infection Control ................................................................................................... 55
Chapter 7. Barrier Precautions, Patient Isolation, and Routine Surveillance for Prevention
of Healthcare-Associated Infections: Brief Update Review ........................................... 55
Chapter 8. Interventions To Improve Hand Hygiene Compliance: Brief Update
Review .............................................................................................................................. 67
Chapter 9. Reducing Unnecessary Urinary Catheter Use and Other Strategies To Prevent
Catheter-Associated Urinary Tract Infections: Brief Update Review............................. 73
Chapter 10. Prevention of Central Line-Associated Bloodstream Infections: Brief Update
Review .............................................................................................................................. 88
Chapter 11. Ventilator-Associated Pneumonia: Brief Update Review ......................... 110
Chapter 12. Interventions To Allow the Reuse of Single-Use Devices: Brief Review
(NEW*) ........................................................................................................................... 117
Section C. Surgery, Anesthesia, and Perioperative Medicine .............................................. 122
Chapter 13. Preoperative and Anesthesia Checklists ...................................................... 122
Chapter 14. Use of Report Cards and Outcome Measurements To Improve Safety of
Surgical Care: American College of Surgeons National Surgical Quality Improvement
Program (NEW*) ............................................................................................................ 140
xi
Chapter 15. Prevention of Surgical Items Being Left Inside Patient: Brief Update
Review ............................................................................................................................ 158
Chapter 16. Operating Room Integration and Display Systems: Brief Review
(NEW*) ........................................................................................................................... 163
Chapter 17. Use of Beta Blockers To Prevent Perioperative Cardiac Events: Brief
Update Review ............................................................................................................... 169
Chapter 18. Use of Real-Time Ultrasound Guidance During Central Line Insertion: Brief
Update Review ............................................................................................................... 172
Section D. Safety Practices Aimed Primarily at Hospitalized Elders ................................... 178
Chapter 19. Preventing In-Facility Falls ......................................................................... 178
Chapter 20. Preventing In-Facility Delirium .................................................................. 201
Section E. General Clinical Topics ....................................................................................... 212
Chapter 21. Preventing In-Facility Pressure Ulcers ........................................................ 212
Chapter 22. Inpatient Intensive Glucose Control Strategies To Reduce Death and
Infection (NEW*) ........................................................................................................... 233
Chapter 23. Interventions To Prevent Contrast-Induced Acute Kidney Injury .............. 248
Chapter 24. Rapid Response Systems (NEW*) .............................................................. 257
Chapter 25. Medication Reconciliation Supported by Clinical Pharmacists (NEW*) ... 270
Chapter 26. Identifying Patients at Risk for Suicide: Brief Review (NEW*)................ 287
Chapter 27. Strategies To Prevent Stress-Related Gastrointestinal Bleeding (Stress Ulcer
Prophylaxis): Brief Update Review .............................................................................. 297
Chapter 28. Prevention of Venous Thromboembolism: Brief Update Review............. 303
Chapter 29. Preventing Patient Death or Serious Injury Associated With Radiation
Exposure From Fluoroscopy and Computed Tomography: Brief Review (NEW*) ...... 310
Chapter 30. Ensuring Documentation of Patients Preferences for Life-Sustaining
Treatment: Brief Update Review .................................................................................. 320
Part 2b. Practices Designed To Improve Overall System/Multiple Targets ............................... 325
Chapter 31. Human Factors and Ergonomics ................................................................. 325
Chapter 32. Promoting Engagement by Patients and Families To Reduce Adverse Events
(NEW*) ........................................................................................................................... 351
Chapter 33. Promoting Culture of Safety ....................................................................... 362
Chapter 34. Effect of Nurse-to-Patient Staffing Ratios on Patient Morbidity
and Mortality................................................................................................................... 372
Chapter 35. Patient Safety Practices Targeted at Diagnostic Errors (NEW*) ................ 385
Chapter 36. Monitoring Patient Safety Problems (NEW*) ............................................ 405
Chapter 37. Interventions To Improve Care Transitions at Hospital Discharge
(NEW*) ........................................................................................................................... 425
Chapter 38. Use of Simulation Exercises in Patient Safety Efforts ................................ 439
Chapter 39. Obtaining Informed Consent From Patients: Brief Update Review .......... 461
Chapter 40. Team-Training in Health Care: Brief Update Review ............................... 472
Chapter 41. Computerized Provider Order Entry With Clinical Decision Support
Systems: Brief Update Review...................................................................................... 480
Chapter 42. Tubing Misconnections: Brief Review (NEW*) ........................................ 487
Chapter 43. Limiting Individual Providers Hours of Service: Brief Update
Review ............................................................................................................................ 493
Part 3. Discussion........................................................................................................................ 499
xii
Chapter 44. Discussion ......................................................................................................... 499

Introduction ..................................................................................................................... 499
Limitations ...................................................................................................................... 505
Conclusions ..................................................................................................................... 506
Future Research Needs ................................................................................................... 506
References ....................................................................................................................... 512
Abbreviations/Acronyms ............................................................................................................ 514
Tables
Table A. Format for in-depth reviews....................................................................................... ES-6
Table B. Summary table ........................................................................................................... ES-8
Table C. Strongly encouraged patient safety practices ........................................................... ES-12
Table D. Encouraged patient safety practices ......................................................................... ES-13
Table 1, Chapter 2. Initially excluded topics .................................................................................. 8
Table 2, Chapter 2. Proportion of technical expert panelists expressing a preference
for the level of evidence review for each PSP ................................................................................ 9
Table 3, Chapter 2. Format for in-depth reviews .......................................................................... 13
Table 4, Chapter 2. Criteria for assigning strength of evidence for effectiveness/harms
questions ....................................................................................................................................... 16
Table 1, Chapter 3. Summary tableheparin effectiveness studies ............................................ 26
Table 2, Chapter 3. Summary table .............................................................................................. 28
Table 1, Chapter 4. Summary of studies ....................................................................................... 33
Table 1, Chapter 9. Description of outcomes evaluated (adapted from the prior meta-analysis). 75
Table 2, Chapter 9. Indications for indwelling urethral catheter use (from the 2009 CDCs
guideline) ...................................................................................................................................... 76
Table 3, Chapter 9. Number of avoided CAUTI episodes per 1,000 catheter-days ..................... 77
Table 1, Chapter 10. Categories and recommendations for CLABSI reduction practices
from the Healthcare Infection Control Practices Advisory Committee of the Centers for Disease
Control and Prevention ................................................................................................................. 90
Table 2, Chapter 10. Summary table .......................................................................................... 100
Table 1, Chapter 13. Governmental and nongovernmental organizations adopting
or recommending adoption of the WHO checklist ..................................................................... 134
Table 1, Chapter 14. Example of interventions and associated impact on outcomes in American
College of Surgeons national surgical quality improvement program
for hospitals/collaboratives ......................................................................................................... 144
Table 2, Chapter 14. Comparison of American College of Surgeons national surgical quality
improvement program use options.............................................................................................. 147
xiii
Table 3, Chapter 14. Percent of Medicare surgical cases covered by the national surgical
quality improvement program..................................................................................................... 148
Table 4, Chapter 14. List of American College of Surgeons national surgical quality
improvement program collaboratives including type, number of sites, and payor ..................... 149
Table 5, Chapter 14. Example of reductions in complications and associated costs .................. 152
Table 1, Chapter 19. Components of multi-factorial falls prevention trials in hospitals,
1999 to 2009 ............................................................................................................................... 180
Table 2, Chapter 19. Meta-analytic estimate of the effect of multicomponent fall intervention
programs on inpatient fall rates................................................................................................... 184
Table 3, Chapter 19. Abridged evidence tables, adapted from Oliver and colleagues ............... 186
Table 4, Chapter 19. Implementation themes highlighted in implementation studies................ 191
Table 1, Chapter 21. Components of pressure ulcer prevention trials in U.S. hospitals,
2000 to 2011 ............................................................................................................................... 214
Table 2, Chapter 21. Components of pressure ulcer prevention trials in long term care,
2000 to 2011 ............................................................................................................................... 215
Table 1, Chapter 22. Large trials (n > 500) evaluating the health outcome effects of intensive
insulin therapy............................................................................................................................. 234
Table 1, Chapter 24. RRS Summary table: effectiveness ........................................................... 261
Table 2, Chapter 24. RRS Summary table: implementation studies........................................... 264
Table 1, Chapter 25. Studies of medication reconciliation that include assessment of clinically
significant unintended discrepancies and emergency department visits and hospitalizations
within 30 days of discharge ........................................................................................................ 272
Table 2, Chapter 25. Key features of the 12 included medication reconciliation interventions . 276
Table 3, Chapter 25. Medication reconciliation in varying levels of intensity as seen
in published studies ..................................................................................................................... 282
Table 1, Chapter 31. Key characteristics of HFE and its application to patient safety .............. 327
Table 2, Chapter 31. HFE mechanisms between system design and patient safety ................... 331
Table 3, Chapter 31. Examples of HFE design principles .......................................................... 332
Table 4, Chapter 31. HFE issues in selected care settings .......................................................... 339
xiv

Table 1, Chapter 32. Patient engagement in safety: effectiveness studies.................................. 353
Table 2, Chapter 32. Hand-hygiene intervention studies ............................................................ 354
Table 3, Chapter 32. Rapid response team intervention studies ................................................. 356
Table 4, Chapter 32. Falls prevention intervention studies ........................................................ 357
Table 5, Chapter 32. Surgical checklist intervention studies ...................................................... 358
Table 1, Chapter 33. Results of included studies on patient safety culture ................................ 367
Table 1, Chapter 34. Pooled odds ratios of patient outcomes corresponding to an increase of one
registered nurse full-time equivalent per patient day .................................................................. 376
Table 2, Chapter 34. Odds ratios indicating the effect of nurse staffing on 30-day inpatient
mortality and failure to rescue, in California, New Jersey, and Pennsylvania ........................... 378
Table 3, Chapter 34. Risk of death associated with exposure to a shift with an actual RN staffing
level 8 hours or more below target, high patient turnover, and other variables ......................... 380
Table 4, Chapter 34. Summary table for increasing nurse-to-patient staffing ratios to prevent
death ............................................................................................................................................ 382
Table 5, Chapter 34. Summary table for increasing nurse-to-patient staffing ratios to prevent
falls, pressure ulcers, and other nursing sensitive outcomes (other than mortality) ................... 382
Table 1, Chapter 35. Study design distribution........................................................................... 390
Table 1, Chapter 36. Overview of the purposes of different methods for detecting patient safety
problems ...................................................................................................................................... 407
Table 2, Chapter 36. Advantages and disadvantages of different methods used to measure errors
and adverse events in health care (from the Thomas and Petersen study).................................. 408
Table 3, Chapter 36. Advantages and disadvantages of different methods for hospitals to monitor
for internal patient safety problems (from the Shojania study) .................................................. 409
Table 4, Chapter 36. Evidence-based rating of the main methods used in developed countries for
estimating hazards in health care systems .................................................................................. 417
Table 5, Chapter 36. Subjective rating, where there was no evidence-based data, of the main
methods used in developed countries for estimating hazards in health care systems ................. 418
Table 6, Chapter 36. Estimated costs of systems for detecting patient safety problems in one
hospital ........................................................................................................................................ 420
Table 1, Chapter 38. Literature on simulation training for central venous catheterization ........ 448
Table 2, Chapter 44. Strongly encouraged patient safety practices ............................................ 504
Table 3, Chapter 44. Encouraged patient safety practices .......................................................... 504
xv
Figures
Figure A. Framework for evidence assessment of patient safety practices .............................. ES-3
Figure 1, Chapter 2. Overview of the project ................................................................................. 7
Figure 2, Chapter 2. Framework for evidence assessment of patient safety practices ................. 11
Figure 3, Chapter 2. The evidence review process ....................................................................... 15
Figure 1, Chapter 5. Official do not use list .............................................................................. 41
Figure 1, Chapter 9. Lifecycle of the urinary catheter .................................................................. 74
Figure 2, Chapter 9. Summary of CAUTI and urinary catheter outcomes from 14 studies ......... 79
Figure 3, Chapter 9. Summary of CAUTI and urinary catheter outcomes from 12 additional
studies ........................................................................................................................................... 80
Figure 1, Chapter 13. Screenshot of adoption and diffusion of the WHO surgical safety
checklist ...................................................................................................................................... 133
Figure 1, Chapter 14. Example of observed to expected ratio reporting for American College of
Surgeons national surgical quality improvement program ......................................................... 141
Figure 2, Chapter 14. Geographic distribution of American College of Surgeons national surgical
quality improvement program participating sites ....................................................................... 148
Figure 3, Chapter 14. The 12 steps to implement the national surgical quality improvement
program ....................................................................................................................................... 151
Figure 1, Chapter 19. Multi-systemic fall prevention model ...................................................... 183
Figure 2, Chapter 19. Meta-analysis from Oliver et al. 2006 for multifaceted interventions
in hospital falls (random effects model) ..................................................................................... 185
Figure 1, Chapter 22. Short-term mortality in studies of intensive insulin therapy, by inpatient
setting and condition ................................................................................................................... 237
Figure 2, Chapter 22. Risk for hypoglycemia in studies of intensive insulin therapy in various
inpatient settings ......................................................................................................................... 239
Figure 1, Chapter 25. Overview of medication reconciliation.................................................... 271
Figure 2, Chapter 25. Median and interquartile range for the number of clinically significant
unintentional discrepancies per patient for the 13 included interventions .................................. 279
Figure 3, Chapter 25. Emergency department visits and hospitalizations within 30 days
of discharge in three randomized controlled trials ...................................................................... 280
Figure 1, Chapter 31. SEIPS model of work system and patient safety ..................................... 330
Figure 1, Chapter 34. Hospital organization, nursing organization, and patient outcomes ........ 373
Figure 2, Chapter 34. Despins model on patient risk detection ................................................. 373
Figure 3, Chapter 34. Tourangeaus model on determinants of 30-day mortality...................... 374
Figure 4, Chapter 34. Thornlows model on cascade iatrogenesis: postoperative respiratory
failure .......................................................................................................................................... 375
Figure 5, Chapter 34. Pooled odds ratio of quartiles of death by nurse staffing levels .............. 376
Figure 1, Chapter 35. Interventions by type ............................................................................... 389
Figure 2, Chapter 35. Intervention studies by year ..................................................................... 390
Box
Box 1. Taxonomy of interventions to improve transitional care at hospital discharge .............. 426
xvi
Appendixes
Appendix A. Original List of Patient Safety Practices
Appendix B. AMSTAR: A Measurement Tool To Assess Systemic Reviews
Appendix C. Literature Searches and Topic-Specific Methods
Appendix D. Supplementary Evidence Tables
(*NEW) refers to a new review that was not addressed in the 2001 report, Making Health Care
Safer: A Critical Analysis of Patient Safety Practices.
xvii
Executive Summary
Background
The 1999 Institute of Medicine report To Err is Human: Building a Safer Health System, is
credited by many with launching the modern patient safety movement.1 A year after this report
was published, as part of its initial portfolio of patient safety activities, the Agency for
Healthcare Research and Quality (AHRQ) commissioned a group from the University of
California, San Francisco-Stanford Evidence-based Practice Center (EPC) to analyze evidence
behind a diverse group of patient safety practices (PSPs) that existed at that time.
The resulting 2001 report, Making Health Care Safer: A Critical Analysis of Patient Safety
Practices,2 hereafter referred to as Making Health Care Safer, was both influential and
controversial. A significant number of copies of the report were distributed by AHRQ, and it
became a cornerstone of other efforts (such as the National Quality Forums 34 Safe Practices
for Better Healthcare list)3 to rank safety practices by strength of evidence. However, the low
rankings given to some popular safety practices, such as computerized order entry, raised
fundamental questions about the role of evidence-based medicine in quality and safety practices.
Since the Making Health Care Safer report was published, the safety field has matured.
Regulators and accreditors encourage health care organizations to adopt safe practices and to
avoid adverse events that are considered wholly or largely preventable. A significant amount of
money and person-hours have been invested in efforts to improve safety, and almost all healthcare delivery organizations regard safety as a primary strategic priority.
However, evidence indicates that progress has not matched the efforts and investment. Some
patient safety practices (PSPs) have resulted in unintended consequences, whereas others have
been shown to be highly context dependent, working effectively in a research setting but failing
during broader implementation. In the past 2 years, three studies have found high rates of
preventable harm in hospitals,4-6 one of which found no improvement in adverse event rates from
2003 to 2008.
Against this backdrop, AHRQ commissioned an updated research report on the state of PSPs.
Because many of the project team members and much of the methodology were drawn from the
initial Making Health Care Safer project, and because most of the relevant practices were
reviewed then, we see this report as a natural sequel to the 2001 report. However, because of the
burgeoning literature relevant to patient safety and the limits of budget and time, we chose to
examine a subset of PSPs (chosen through methods described below). Moreover, part of the
maturation of the safety field has included a deeper appreciation of the importance of context in
patient safety practices, a topic examined by our research team in the 2010 report, Assessing the
Evidence for Context-Sensitive Effectiveness and Safety of Patient Safety Practices: Developing
Criteria, hereafter referred to as Context Sensitivity.7 Accordingly, this report emphasizes
matters of context and generalizability, as well as unintended consequences, to a greater degree
than the 2001 Making Health Care Safer report.
Objectives
The goal of this project was to conduct a systematic literature review evaluating the evidence
for a large number of patient safety practices.
ES-1
Analytic Framework
For this report, we adopted the definition of a PSP used in the 2001 Making Health Care
Safer report:
A Patient Safety Practice is a type of process or structure whose
application reduces the probability of adverse events resulting from
exposure to the health care system across a range of diseases and
procedures.
The framework for considering the evidence regarding a PSP was worked out as part of the
report on Context Sensitivity.7 One of the principal challenges in the review of PSPs has been
addressing the question of what constitutes evidence for PSPs. Many practices intended to
improve quality and safety are complex sociotechnical interventions whose targets may be entire
health care organizations or groups of providers, and these interventions may be targeted at rare
events. To address the challenge regarding what constitutes evidence, we recognize that PSPs
must be evaluated along two dimensions: the evidence regarding the outcomes of the safe
practices, and the contextual factors influencing the practices use and effectiveness.
These dimensions are represented in Figure A, which depicts a sample PSP that consists of a
bundle of components (the individual boxes), and the context within which the PSP is embedded.
Important evaluation questions, as depicted on the right in the figure, include effectiveness and
harms, implementation, and adoption and spread. We then apply criteria to evaluate the four
factors that together constitute quality (depicted as puzzle pieces in the bottom half of the figure.
They include:
1. Constructs about the PSP, its components, context factors, outcomes, and ways to
accurately measure these constructs
2. Logic model or conceptual framework about the expected relationships among these
constructs
3. Internal validity to assess the PSP results in a particular setting
4. External validity to assess the likelihood of being able to garner the same results in
another setting
We then synthesize this information into an evaluation of the strength of the evidence for a
particular PSP.
ES-2
Figure A. Framework for evidence assessment of patient safety practices
The principal results of the Context Sensitivity report included the following key points.
Whereas controlled trials of PSP implementations offer investigators greater control of
sources of systematic error than do observational studies, trials often are not feasible in
terms of time or resources. Also, controlled trials are often not possible for PSPs
requiring large-scale organizational change or PSPs targeted at very rare events.
Furthermore, the standardization imposed by the clinical trial paradigm may stifle the
adaptive responses necessary for some quality improvement or patient safety projects.
Hence, researchers may need to use designs other than randomized controlled trials to
develop strong evidence about the effectiveness of some PSPs.
Regardless of the study design chosen for an evaluation, components that are critical for
evaluating a PSP in terms of how it worked in the study site, and whether it might work
in other sites, include the following:
o Explicit description of the theory for the chosen intervention components, and/or an
explicit logic model for why this PSP should work
o Description of the PSP in sufficient detail that it can be replicated, including the
expected change in staff roles
o Measurement of contexts
o Explanation, in detail, of the implementation process, the actual effects on staff roles,
and changes over time in the implementation or the intervention
o Assessment of the impact of the PSP on outcomes and possible unexpected effects
(including data on costs, when available)
o For studies with multiple intervention sites, assessment of the influence of context on
intervention and implementation effectiveness (processes and clinical outcomes)
High priority contexts for assessing any PSP implementation include measuring and
information for each of the following four domains:
o Structural organizational characteristics (such as size, location, financial status,
existing quality and safety infrastructure)
o External factors (such as regulatory requirements, the presence in the external
environment of payments or penalties such as pay-for-performance or public
ES-3
reporting, national patient safety campaigns or collaboratives, or local sentinel patient

safety events)
o Patient safety culture (not to be confused with the larger organizational culture),
teamwork, and leadership at the level of the unit
o Availability of implementation and management tools (such as staff education and
training, presence of dedicated time for training, use of internal audit-and-feedback,
presence of internal or external individuals responsible for the implementation, or
degree of local tailoring of any intervention)
These principles guided our search for evidence, and the way in which we presented our
findings in this report.
Methods
We divided the project into three phases: topic refinement, the evidence review, and the
critical review and interpretation of the evidence. The project team performed topic refinement
and conducted the critical review of the evidence jointly with the Technical Expert Panel (TEP),
which had also participated in the Context Sensitivity project. This TEP included many of the
key patient safety leaders in the United States, Canada, and the United Kingdom: experts in
specific PSPs and evaluation methods and persons charged with implementing PSPs in hospitals
and clinics.
Topic Refinement
Because the goals of the project were to assess the evidence of the effectiveness of new safe
practices and the evidence of implementation for current safe practices, most PSPs were eligible
for this review. Thus, our first task was to refine the scope of the topic to fit within the timeframe
and budget of the project, a task undertaken by the project team and the TEP. To accomplish this
task, we created an initial list of 158 PSPs that we considered potentially eligible for inclusion.
Through a process of internal team triage, group discussion with the TEP, and formal TEP votes,
we narrowed the list to 41 PSPs for which a review of evidence was judged likely to be most
helpful to providers, policymakers, and patients. However, this number of PSPs was still too
large for us to review the evidence comprehensively within the timeframe. For that reason, we
asked our TEP whether breadth or depth was likely to be more valuable for stakeholders; in
other words, we asked whether the review should focus on fewer topics in more detail or cover
all topics but with less detail. Our TEP recommended a hybrid approach, in which some topics
would be reviewed in depth, whereas other topics would receive only a brief review.
Topics could be considered as needing only a brief review for several reasons: the PSP is
already well established; stakeholders need to know only whats new since the last time a topic
was reviewed in depth; new evidence suggests the PSP may not be as effective as originally
believed, so it is no longer a priority PSP; or the PSP is emerging with little evidence
accumulated. We ultimately ended up with 18 in-depth reviews and 23 brief reviews.
ES-4
Evidence Reviews
In-Depth Reviews
Overall approach. For many of the 18 topics designated to receive an in-depth review, a
systematic review was likely to exist. Thus, a search to identify existing systematic reviews was
usually the project teams first step. To assess the potential utility of such reviews, we followed
the procedures proposed by Whitlock and colleagues,8 which essentially meant addressing the
following two questions: (1) Is the existing review sufficiently on topic to be of use? (2) Is the
existing review of sufficient quality for us to have confidence in the results?
If an existing systematic review was judged to be sufficiently on topic and of acceptable
quality, we took one of two steps. We either performed an update search; that is, we searched
databases for new evidence published since the end date of the search in the existing systematic
review. Or, we conducted a search for signals for updating. Such searches generally followed
the criteria proposed by Shojania and colleagues.9 The searches involved a search of high-yield
databases and journals for pivotal studies that could be a signal that a systematic review is outof-date. Any evidence identified via the update search or the signals search was added to the
evidence base from the existing systematic review.
Some PSPs had no existing systematic reviews, while other PSPs had prior reviews that were
either not sufficiently relevant or were not of sufficient quality to be used. In those situations, we
conducted new searches using guidance as outlined in AHRQs Methods Guide for
Effectiveness and Comparative Effectiveness Reviews.10
Evidence about context, implementation, and adoption are key aspects of this review. We
searched for evidence on these topics in two ways:
We looked for and extracted data about contexts and implementation from the articles
contributing to the evidence of effectiveness.
We identified implementation studies from our literature searches. Implementation
studies focus on the implementation process, particularly the elements demonstrated or
believed to be of special importance for the success, or lack of success, of the
intervention. To be eligible, implementation studies needed to either report or be linked
to reports of effectiveness outcomes.
Reporting format. We took the format for in-depth reviews from AHRQs Context
Sensitivity report. Table A outlines the format of the in-depth reviews.
ES-5
Table A. Format for in-depth reviews

This section briefly sketches the nature of the target for the Patient Safety Practice.
What is the Patient Safety Practice?
This section describes the practice or practices proposed and evaluated.
Why should this Patient Safety Practice work?
This section describes what has been written about the basis for a proposed Patient Safety Practice, such
as an underlying theory, a logic model for how it should work, or prior data.
What are the beneficial effects of the Patient Safety Practice?
This section provides the review of the evidence of effectiveness, and is the section most similar to
traditional Evidence-based Practice Center reports.
What are the harms of the Patient Safety Practice?
This section contains the evidence of harms. Unlike reviews of most clinical interventions, evaluating
potential harms is not a routine part of Patient Safety Practice evaluations. Thus, for most topics, this section
is underdeveloped.
How has the Patient Safety Practice been implemented, and in what contexts?
This section describes what has been reported about how to implement the Patient Safety Practice and the
range of institutions or contexts of where it has been implemented. When there is sufficient evidence,
implementation studies are evaluated qualitatively for themes regarding effective implementation.
This section describes the evidence of costs of implementing the Patient Safety Practice, or, in some cases,
cost-effectiveness analyses that have been performed.
Are there any data about the effect of context on effectiveness?
This section describes the evidence about whether or not the Patient Safety Practice has been shown to
have differential effectiveness in different contexts. The Context Sensitivity project defined important
contexts for Patient Safety Practices in four domains: external factors (e.g., financial or performance
incentives or Patient Safety Practice regulations); structural organizational characteristics (e.g., size,
organizational complexity, or financial status); safety culture, teamwork, and leadership involvement; and
availability of implementation and management tools (e.g., organizational training incentives).11
Brief Reviews
Brief reviews are not full systematic reviews. The goals of the brief reviews covered in this
report varied by PSP according to the needs of stakeholders. The assessment could focus on
either information about effectiveness of an emerging PSP or implementation of an established
PSP; alternatively, the review could explore whether new evidence calls into question the
effectiveness of an existing PSP. Thus, the methods for the brief reviews differed by topic.
However, in general, brief reviews were conducted by a content expert who worked with the
project team. The brief reviews involved focused literature searches for evidence relevant to the
specific need. The evidence was then narratively summarized in a format that also varied with
the particular goal.
Evidence Summary
We judged that users of this report would want a summary of the evidence for each topic.
Such summary messages may facilitate an uptake of the findings. The project team developed
the following summary domains with input from the TEP.
ES-6
Scope of the problem. In general, we addressed two issues: the frequency of the safety problem,
and the severity of each average event. For benchmarks, we regarded safety problems that occur
approximately once per 100 hospitalizations as common; examples include falls, venous
thromboembolism (VTE), potential adverse drug events, or pressure ulcers. In contrast, events an
order of magnitude or more lower in frequency were considered rare; such events include
inpatient suicide, wrong-site surgery, and surgical items being left inside a patient. The scope
must also consider the severity of each event; for instance, most falls do not result in injury, and
most potential adverse drug events do not result in clinical harm. However, each case of inpatient
suicide or wrong-site surgery is devastating.
Strength of evidence for effectiveness. This assessment follows a framework for strength of
evidence that the project team adapted from existing EPC Methods guidance12 to increase the
relevance to patient safety practices. This means we included in strength of evidence assessments
evidence about context, implementation, and the use of theory or logic models, in addition to
standard EPC criteria on inconsistency, in precision, and the possibility of reporting bias.
Evidence on potential for harmful unintended consequences. Most PSP evaluators have not
explicitly assessed the possibility of harm. Consequently, this domain includes evidence of both
actual harm and the potential for harm. The ratings on known or potential harms ranged from
high risk of harm to low (or negligible) risk of harm; in some cases, the evidence was too sparse
to provide a rating.
Estimate of costs. This domain is speculative, because most evaluations do not present cost data.
However, we believed that providing at least a rough estimate of cost would be beneficial
information to include in this report. Therefore, we used the following categories and
benchmarks to provide a rough estimate of cost, noting, where necessary, the factors that might
cause cost estimates to vary.
Low cost. PSPs that do not require hiring new staff or large capital outlays but instead
involve training existing staff and purchasing some supplies. Examples include most fall
prevention programs, VTE prophylaxis, and medical history abbreviations designated as
Do Not Use.
Medium cost. PSPs that might require hiring one or a few new staff members, have
modest capital outlays, or incur ongoing monitoring costs. Examples include some fall
prevention programs, many clinical pharmacist interventions, and participation in the
American College of Surgeons outcomes reporting system ($135,000/year).13
High cost. PSPs that require hiring substantial numbers of new staff, have considerable
capital outlays, or both. Examples include computerized order entry (because it requires
an electronic health record), having to hire many nurses to achieve a certain nurse-topatient ratio, or facility-wide infection control procedures (estimated at $600,000 year for
a single intensive care unit).14
Implementation issues. This section summarizes how much we know about how to implement
the PSP and how difficult it is to implement. To approach the question of how much we know,
we considered the available evidence about implementation, the existence of data about the
effect and influence of context, the degree to which a PSP has been implemented, and the
presence of implementation tools, such as written materials and training manuals.
ES-7
For the question of implementation difficulty, we used three categories: difficult, for PSPs
that require large scale organizational change; not difficult, for PSPs that require protocols for
drugs or devices, such as those needed to reduce radiation exposure or to help prevent stressrelated gastrointestinal bleeding; and moderate, for PSPs falling between the extremes.
Critical Review and Interpretation of Evidence

The TEP reviewed the results of the evidence review performed by the project team both in a
written draft document and at a face-to-face meeting in January 2012. One outcome of this
review was a set of recommendations about priorities for PSP adoption.
Results
We completed 18 in-depth reviews and 23 brief reviews. Table B summarizes the findings
according to the five main issues previously described (scope, strength of evidence, harms, costs,
and implementation). The table is organized into two main sections: PSPs aimed at a specific
(single) patient safety target, such as adverse drug events, or general clinical topics, such as
preventing pressure ulcers; and PSPs designed to improve the overall system or to address
multiple patient safety targets, such as nurse-staffing ratios or computerized provider order entry.
In some cases, the text in the PSP column differs slightly from the chapter heading for that PSP.
This is due to the desire by our TEP to include the target safety problem in the table (if targeted
at a specific safety problem), more specification, or an example of the PSP (e.g., adding such as
a centralized display of consolidated data to the PSP designated as operating room integration
and display systems).
Table B. Summary table*
Patient Safety Practice
Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Practices Designed for a Specific Patient Safety Target
Adverse Drug Events
High-alert drugs: patient safety
Common/Moderate
practices for intravenous
anticoagulants;
in-depth review
Use of clinical pharmacists to
Common/Low
prevent adverse drug events;
brief review
The Joint Commissions Do Not
Common/Low
Use list; brief review
Smart infusion pumps; brief
Common/Low
review
Infection Control
Barrier precautions, patient
Common/Moderate
isolation, and routine surveillance
for the prevention of healthcareassociated infections; brief
review
Interventions to improve hand
Common/Moderate
hygiene compliance; brief review
Strength of
Evidence for
Effectiveness
of the PSPs
Evidence or
Potential for
Harmful
Unintended
Consequences
Estimate of
Cost
Implementation
Issues:
How Much Do We
Know?/How Hard Is
it?
Low
Low-to-moderate
Low
Little/Moderate
Moderate-tohigh
Low
High
Little/Moderate
Low
Negligible
Low
Low
Low
Moderate
Little/Probably not
difficult
Moderate/Moderate
Moderate
Moderate
(isolation of
patients)
Moderate-tohigh
Moderate/Moderate
Low
Low
Low
Moderate/Moderate
ES-8
Table B. Summary table* (continued)

Reducing unnecessary urinary

catheter use and other strategies
to prevent catheter-associated
urinary tract infections; brief
review
Prevention of central lineassociated bloodstream
infections; brief review
Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Common/Moderate
Strength of
Evidence for
Effectiveness
of the PSPs
Moderate-tohigh
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low
Implementation
Issues:
How Much Do We
Know?/How Hard Is
it?
Moderate/Moderate
Common/Moderate
Moderate-tohigh
Low
Low-tomoderate
Ventilator-associated
Common/High
pneumonia; brief review
Interventions to allow the reuse
Common/Low
of single use devices; brief
review
Surgery, Anesthesia, and Perioperative Medicine
Preoperative checklists and
Common/Moderate
anesthesia checklists to prevent
a number of operative safety
events, such as surgical site
infections and wrong site
surgeries; in-depth review
The use of ACS-NSQIP report
Common/High
cards and outcome
measurements to decrease
perioperative morbidity and
mortality; in-depth review
New interventions to prevent
Rare/Low
surgical items from being left
inside a patient; brief review
Moderate-tohigh
Low
Low
Low
Low-tomoderate
Low
High
Negligible
Low
A lot/Moderate
Moderate-tohigh
Low
Moderate
Moderate/Moderate
Low
Negligible
Little
Operating room integration and

display systems, such as a
centralized display of
consolidated data; brief review
Use of beta blockers to prevent
perioperative cardiac events;
brief review
Common/Low-tohigh
Low
Negligible
Low if it
simply
involves
more
frequent
manual
counting;
high if RFID
is used
Moderate
Common/High
High (death,
stroke,
hypotension, and
bradycardia)
Low
NA
Use of real-time ultrasound

guidance during central line
insertion to increase the
proportion correctly placed on
the first attempt; brief review
Common/Low-tomoderate
High evidence
harms may
equal or
exceed
benefits
High
Negligible
Low-tomoderate
A lot/Moderate
ES-9
Moderate-to-difficult/
Not difficult
(implementation of a
bundle)-to-moderate
(understanding
organization culture
and context)
Moderate/Moderate
A lot/Not difficult
Moderate/Moderate

Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Safety Practices for Hospitalized Elders
Multicomponent interventions to
Common/Low
prevent in-facility falls; in-depth
review
prevent in-facility delirium; indepth review
General Clinical Topics
Multicomponent initiatives to
prevent pressure ulcers; in-depth
review
Inpatient, intensive, glucose
control strategies to reduce
death and infection; in-depth
review
Interventions to prevent contrastinduced acute kidney injury; indepth review
Rapid-response systems to
prevent failure-to-rescue; indepth review
Medication reconciliation
supported by clinical
pharmacists; in-depth review
Identifying patients at risk for
suicide; brief review
Strategies to prevent stressrelated gastrointestinal bleeding
(stress ulcer prophylaxis); brief
review
Strategies to increase
appropriate prophylaxis for
venous thromboembolism; brief
review
Preventing patient death or
serious injury associated with
radiation exposure from
fluoroscopy and computed
tomography through technical
interventions, appropriate
utilization, and use of algorithms
and protocols; brief review
Ensuring documentation of
patient preferences for lifesustaining treatment, such as
advanced directives; brief review
Increasing nurse-to-patient
staffing ratios to prevent death;
in-depth review
Strength of
Evidence for
Effectiveness
of the PSPs
High
Evidence or
Potential for
Harmful
Unintended
Consequences
Estimate of
Cost
Implementation
Issues:
How Much Do We
Know?/How Hard Is
it?
Moderate
Moderate/Moderate
Moderate
Moderate/Moderate
Common/Low
Moderate
Moderate
(increased use of
restraints and/or
sedation)
Low
Common/Moderate
Moderate
Negligible
Moderate
Moderate/Moderate
Common/Moderate
Moderate-tohigh evidence
it doesnt help
High
(hypoglycemia)
Low-tomoderate
NA
Common/Low
Low
Negligible
Low
Little/Not difficult
Common/High
Moderate
Low
Moderate
Moderate/Moderate
Common/Low
Moderate
Low
Moderate
Moderate/Moderate
Rare/High
Low
Low
Moderate
Little/Moderate
Rare/Moderate
Moderate
Moderate
(pneumonia)
Moderate
Common/Moderate
High
Moderate
(bleeding)
Low
Little/Moderate
Rare/High
Moderate
Negligible
Low
Moderate/Not difficult
Common/Moderate
Moderate
Low
Low
Moderate/Moderate
Common/High
Moderate
Low
High
A lot/Not difficult
ES-10

Scope of the
Strength of
Problem Targeted
Evidence for
by the PSP
Effectiveness
(Frequency/
of the PSPs
Severity)
Practices Designed To Improve Overall System/Multiple Targets
Increasing nurse-to-patient staff
Common/High
Low
ratios to prevent falls, pressure
ulcers, and other nursing
sensitive outcomes (other than
mortality); in-depth review
Incorporation of human factors
Potentially
Not assessed
and ergonomics in the design of
applicable to all
systematically,
health care practices by hiring an patient safety
but moderateexpert or training clinicians in
problems
to-high
human factors; in-depth review
evidence for
some specific
applications
Promoting engagement by
Common
Emerging
patients and families to reduce
practice (few
adverse events (such as patients
studies
encouraging providers to wash
available)
their hands); in-depth review
Interventions to promote a
Common/Low-toLow
culture of safety; in-depth review
high
Evidence or
Potential for
Harmful
Unintended
Consequences
Estimate of
Cost
Implementation
Issues:
How Much Do We
Know?/How Hard Is
it?
Low
High
A lot/Not difficult
Negligible
Moderate
A lot/Moderate
Uncertain
Low
Little/Moderate
Uncertain
Moderate/Not difficultto-moderate (varies

with intervention)
Varies
Emerging
practice (few
studies
available)
Low
Uncertain
Lowtomoderate
(varies)
Varies
Negligible
High
Moderate/Difficult
Low
Negligible
Moderate-tohigh
Little/Difficult
Uncertain
Moderate
Moderate
Common/Moderate
Moderate-tohigh for
specific topics
Moderate
Negligible
Low
Common/High
Moderate
Low
Moderate
Common/Moderate
Low-tomoderate
Low-to-moderate
High
Moderate/Moderateto-difficult
Moderate/Difficult
Common/Moderate
Low
Low
Low
Patient safety practices targeted

at diagnostic errors; in-depth
review
Common/High
Monitoring patient safety

problems; in-depth review
Interventions to improve care
transitions at hospital discharge;
in-depth review
Use of simulation-based training
and exercises; in-depth review
Common/Low-tohigh
Common/Moderate
Obtaining informed consent from

patients to improve patient
understanding of potential risks
of medical procedures; brief
review
Team-training in health care;
brief review
Computerized provider order
entry (CPOE) with clinical
decision support systems
(CDSS); brief review
Interventions to prevent tubing
misconnections; brief review
Common/Moderateto-high
ES-11

Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Common/Moderate
Strength of
Evidence for
Effectiveness
of the PSPs
Evidence or
Estimate of
Implementation
Potential for
Cost
Issues:
How Much Do We
Harmful
Know?/How Hard Is
Unintended
Consequences
it?
Limiting trainee work hours; brief
Low
Moderate (at
High
Moderate/Difficult
review
least); includes
lack of training
time
Abbreviations: ACS NSQIP=American College of Surgeons National Surgical Quality Improvement Program; NA = not
available; PSP: Patient Safety Practice; RFID = radio-frequency identification.
*In some cases, the text in the PSP column differs slightly from the chapter heading for that PSP. This difference is attributable
to our Technical Expert Panels desire to include the target safety problem (if the practice is in fact targeted at a specific safety
problem), more specification, or an example of the PSP (e.g., adding such as a centralized display of consolidated data to the
PSP designated as operating room integration and display systems).
Rating Scales:
Scope of the problem targeted by the PSP (frequency/severity): frequency = rare or common; severity = low, moderate, or high.
Strength of evidence for effectiveness of the PSPs: low, moderate, or high.
Evidence or potential for harmful unintended consequences: negligible, low, moderate, or high.
Estimate of cost: low, moderate, or high.
Implementation issues: How much do we know? = little, moderate, or a lot; How hard is it? = not difficult, moderate, or difficult.
Discussion
Since the 2001 report, Making Health Care Safer, a vast amount of new information on
PSPs has emerged. Compared with a decade ago, more agreement is now evident on what
constitutes evidence of effectiveness and the importance of implementation and context. In this
review, we determined that the strength of evidence was at least moderate for 20 PSPs, or about
half of those reviewed. For 26 of the PSPs, we judged that evidence of at least moderate strength
was available on how to implement them.
Thus, sufficient evidence exists about effectiveness and implementation to permit our TEP
members to conclude that some PSPs are ready to be strongly encouraged for adoption by
health care providers. Their assessments were based explicitly on the combination of the
available evidence with their expert judgment in interpreting the evidence. The 10 strongly
encouraged PSPs are listed in Table C.
Table C. Strongly encouraged patient safety practices
Preoperative checklists and anesthesia checklists to prevent operative and post-operative events
Bundles that include checklists to prevent central line-associated bloodstream infections
Interventions to reduce urinary catheter use, including catheter reminders, stop orders, or nurse-initiated
removal protocols
Bundles that include head-of-bed elevation, sedation vacations, oral care with chlorhexidine, and subglotticsuctioning endotracheal tubes to prevent ventilator-associated pneumonia
Hand hygiene
Do Not Use list for hazardous abbreviations
Multicomponent interventions to reduce pressure ulcers
Barrier precautions to prevent healthcare-associated infections
Use of real-time ultrasound for central line placement
Interventions to improve prophylaxis for venous thromboembolisms
The TEP members concluded that several other PSPs had sufficient evidence of effectiveness
and implementation, and that they should be encouraged for adoption. The 12 encouraged
PSPs are listed in Table D.
ES-12
Table D. Encouraged patient safety practices
Multicomponent interventions to reduce falls

Use of clinical pharmacists to reduce adverse drug events
Documentation of patient preferences for life-sustaining treatment
Obtaining informed consent to improve patients understanding of the potential risks of procedures
Team training
Practices to reduce radiation exposure from fluoroscopy and computed tomography scans
Use of surgical outcome measurements and report cards, like the American College of Surgeons National
Surgical Quality Improvement Program
Rapid response systems
Utilization of complementary methods for detecting adverse events/medical errors to monitor for patient
safety problems
Computerized provider order entry
Use of simulation exercises in patient safety efforts
The 22 PSPs in Tables C and D represent practices that health care providers can consider for
adoption now. This recommendation particularly applies to the 10 strongly encouraged
practices. For these practices, at least in the judgment of our TEP, there is sufficient knowledge
to implement them, and that doing so will likely result in safer care. Future evaluations will
likely further the knowledge of how best to implement the practices to make them most effective.
However, in the meantime, our TEP believes that providers should not delay their consideration
of adopting these practices, as enough is known now to permit health care systems to move
forward.
Limitations
Because of limited resources and time, the current report does not cover the entire patient
safety field, which has grown exponentially since the last report, both in the number of potential
PSPs and in the amount of data about individual PSPs. For that reason, we used an explicit and
transparent process to select which PSPs to evaluate, and our final list of 41 (from the more than
150 candidates) included the PSPs we felt were of highest priority to policymakers and
providers.
Secondly, we did not perform in-depth reviews for all 41 PSPs. To maximize use of the
available time and resources, we tailored our methods to the needs of our stakeholders. In
particular, we targeted the 18 PSPs that were of the greatest interest to our stakeholders, or for
which we likely had the most new information for in-depth reviews. The remaining 23 PSPs
received brief reviews. It is important to note that the decisions about which PSPs would receive
which level of scrutiny and analysis were made by a broadly representative stakeholder
committee.
Thirdly, the in-depth reviews, although thorough, did not conform to all of the criteria for
conducting an evidence review as presented in the Institute of Medicines report, Finding What
Works in Health Care: Standards for Systematic Reviews,15 or to all the criteria in AHRQs
Methods Guide for Effectiveness and Comparative Effectiveness Reviews9; for example, we
did not publicly post a protocol for each of the individual reviews. We used our collective
experience as EPC team members to adapt existing EPC methods that best preserved the essence
of a systematic review, while allowing for the completion of 18 in-depth reviews within 9
months and within the available budget.
Additionally, over time, we will likely improve our methods for assessing evidence regarding
how patient safety interventions affect health care processes and outcomes. The methods we used
ES-13
for this report incorporate new perspectives regarding the importance of implementation and
context, which was the focus of the Context Sensitivity report; likewise, in the future, we
expect to increase our understanding of the interactions between multiple intervention,
implementation, and organizational variables and how the variables influence safety outcomes. If
future research reveals that these variables interact in ways that our current understanding of
theory and logic models cannot explain, we will need to modify the methods for evaluating PSPs
again.
Lastly, we relied on the judgment of our TEP at every important step of the project.
Therefore, the results are as much a product of these judgments as are our systematic review
methods. Hence, our results might be sensitive to the selection of particular experts on our TEP.
However, we mitigated this potential bias by including more than double the number of experts
on our TEP as we typically would for an EPC review, which allowed us to include a diverse set
of stakeholders from the U.S., Canada, and the United Kingdom. Stakeholders included PSP
developers and evaluators, patient safety policymakers, and experts in design and evaluation
methods. Rather than regarding the tight linkage between the needs of the stakeholders and the
work of the EPCs as a limitation, we view it as a strength that increases the likelihood that the
results of the review will be meaningful to providers, payors, and patients, and that the reports
results will lead to meaningful change.
Future Research Needs

Despite over a decade of effort, there is little evidence that patient outcomes (broadly
measured) have significantly improved, yet there has been some success (generally in efforts to
reduce one type of harm, usually using one method of improvement). For example, efforts have
focused on reducing blood stream infections, improving teamwork, or enhancing patient
engagement.
If health care is to make significant improvements in patient safety, research should inform
and guide these efforts. We have learned much about how to improve safety, yet we need to
learn much more. Acquiring this knowledge will require investments in patient safety research,
including investing in basic methodological research. To date, investments in patient safety
research have fallen far short of the magnitude of the problem.
To achieve progress in improving patient safety, research is needed in a number of areas,
including the following:
Basic patient safety research to develop new tools and measures, and research to ensure
that the tool matches the problem
A larger number of valid measures of patient safety
Better methods to measure context and how an intervention was implemented
Methods to identify and provide the necessary skills, resources, and accountability (e.g., a
safety management infrastructure) at each level of the health care system
More effective and less burdensome methods of improvement so that clinicians, researchers, and
administrators can work on reducing all potential patient harms, rather than a select few.
ES-14
References
1.
2.
3.
4.
Kohn L, Corrigan J, Donaldson M, eds. To

Err is Human: Building a Safer Health
System. Committee on Quality of Health
Care in America, Institute of Medicine.
Washington, DC: The National Academies
Press; 2000.
Shojania KG, Duncan BW, McDonald KM,
et al., eds. Making Health Care Safer: A
Critical Analysis of Patient Safety Practices.
No. 43. (Prepared by the University of
California at San FranciscoStanford
Evidence-based Practice Center under
Contract No. 290-97-0013.) AHRQ
Publication No. 01-E058. Rockville, MD:
Agency for Healthcare Research and
Quality. July 2001.
www.effectivehealthcare.ahrq.gov.
National Quality Forum. Safe Practices for
Better Healthcare: 2010 Update.
www.qualityforum.org/Publications/2010/04
/Safe_Practices_for_Better_Healthcare_%E
2%80%93_2010_Update.aspx. Accessed
December 13, 2011.
Classen DC, Resar R, Griffin F, et al.
Global trigger tool shows that adverse
events in hospitals may be ten times greater
than previously measured. Health Aff
(Millwood). 2011;30(4):581-9. PMID:
21471476.
5.
Landrigan CP, Parry GJ, Bones CB, et al.

Temporal trends in rates of patient harm
resulting from medical care. N Engl J Med.
2010;363(22):2124-34. PMID: 21105794.
6.
Levinson DR. Adverse Events in Hospitals:

National Incidence Among Medicare
Beneficiaries. OEI-06-09-00090. Office of
Inspector General. Department of Health
and Human Services. November 2010.
7.
Shekelle P, Pronovost P, Wachter R, et al.

Assessing the Evidence for ContextSensitive Effectiveness and Safety of Patient
Safety Practices: Developing Criteria.
(Prepared by the Southern California-RAND
Contract No. 290-2009-10001C). AHRQ
Publication No. 11-0006-EF. Rockville,
MD: Agency for Healthcare Research and
Quality. December 2010.
ES-15
8.
Whitlock EP, Lin JS, Chou R, et al. Using

existing systematic reviews in complex
systematic reviews. Ann Intern Med.
2008;148(10):776-82. PMID: 18490690.
9.
Shojania KG, Sampson M, Ansari MT, et al.

How quickly do systematic reviews go out
of date? A survival analysis. Ann Intern
Med. 2007;147(4):224-33. PMID:
17638714.
10.
Methods Guide for Effectiveness and

Comparative Effectiveness Reviews. AHRQ
Publication No. 10(12)-EHC063-EF.
Rockville, MD: Agency for Healthcare
Research and Quality. April 2012.
11.
Taylor SL, Dy S, Foy R, et al. What context

features might be important determinants of
the effectiveness of patient safety practice
interventions? BMJ Qual Saf.
2011;20(7):611-7. PMID: 21617166.
12.
Owens DK, Lohr KN, Atkins D, et al.

AHRQ series paper 5: grading the strength
of a body of evidence when comparing
medical interventions--agency for healthcare
research and quality and the effective healthcare program. J Clin Epidemiol. 2010
May;63(5):513-23. PMID: 19595577.
13.
Maggard-Gibbons M. Chapter 14. Use of

Report Cards and Outcome Measurements
To Improve Safety of Surgical Care:
American College of Surgeons National
Quality Improvement Program in Making
Health Care Safer II: An Updated Critical
Analysis of the Evidence for Patient Safety
Practices. Comparative Effectiveness
Review No. 211. (Prepared by the Southern
California-RAND Evidence-based Practice
Center under Contract No. 290-2007-10062I.) AHRQ Publication No. 13-E001-EF.
Research and Quality. March 2013.
14.
Shekelle PG. Chapter 34. Effect of Nurse-toPatient Staffing Ratios on Patient Morbidity
and Mortality in Making Health Care Safer
II: An Updated Critical Analysis of the
Evidence for Patient Safety Practices.
Comparative Effectiveness Review No. 211.
(Prepared by the Southern California-RAND
Contract No. 290-2007-10062-I.) AHRQ
Publication No. 13-E001-EF. Rockville,

Quality. March 2013.
15.
ES-16
Committee on Standards for Systematic

Reviews of Comparative Effectiveness
Research, Institute of Medicine. Finding
What Works in Health Care: Standards for
Systematic Reviews. Washington, DC: The
National Academies Press; 2000.
Part 1. Overview
Chapter 1. Introduction
The modern patient safety movement is believed by many to trace its origins to the 1999
publication of the groundbreaking report, To Err is Human by the Institute of Medicine.1 This
report, which highlighted the 44,000 to 98,000 deaths per year from medical errors in the United
States (U.S.) (the equivalent of the fatalities that would result from the crash of a jumbo jet a
day), galvanized the public and resulted in the focus, of widespread media and legislative
attention, for the first time, on the issue of patient safety. Parallel reports from other countries
were similarly influential.2
As part of its initial portfolio of patient safety activities, the Agency for Healthcare Research
and Quality (AHRQ) commissioned a team from the University of California, San Francisco
(UCSF)-Stanford University Evidence-Based Practice Center to analyze the evidence behind a
diverse group of patient safety practices (PSPs) in useor conceptualizedat that time. The
reportMaking Health Care Safer: A Critical Analysis of Patient Safety Practices2 (MHCS)
was published in 2001.3 The report analyzed nearly 80 different safety practices on several
dimensions, including potential impact, supporting evidence, and costs and complexity of
implementation. Based on these evidence reviews, practices were ultimately rated on both impact
and evidence, as well as prioritization for future research.
MHCS was immediately both influential and controversial. Several hundred thousand copies
of the report were distributed by AHRQ, and it became a lynchpin for other efforts (such as the
National Quality Forums Safe Practices list) to describe PSPs through the lens of evidencebased medicine. The controversy was generated by the reports rankings of PSPsin particular,
the relatively low rankings for certain popular practices such as computerized order entry
which raised fundamental questions about the role of evidence in assessing the value of quality
and safety practices, questions that continue to be debated to this day.4-7
In 2001, hospitals and health care organizations were under relatively little pressure to
implement safety practices. A decade later, the stakes have grown far higher.8 Regulators and
accreditors are pushing health care organizations to adopt various safe practices or to avoid
particular adverse events that are considered wholly or largely preventable. Many payers,
including the Centers for Medicare & Medicaid Services, have embedded patient safety into payfor-performance and no pay for errors initiatives. Billions of dollars and millions of personhours have been invested in a variety of efforts to improve safety, and virtually every health care
delivery organization now identifies patient safety as one of its top strategic priorities.
Yet the evidence indicates that our progress in eradicating medical errors has not matched the
efforts and financial resources invested in implementing PSPs. Studies of some practices that
have tremendous intuitive appeal, such as reducing resident duty hours and implementing rapid
response teams, have yielded conflicting results.9,10 Many examples of unintended consequences
of safety practices have emerged,11 and the successful implementation of safety practices has
been shown to be highly context dependent,12 often working effectively in some hospitals but not
others. Although a national initiative to improve safety in the United Kingdom found some
evidence of improvement, control hospitals improved as much as those that participated in a
vigorous intervention.13 Three recent U.S. studies have demonstrated continuing high rates of
preventable harm in hospitals;14-16 one of these studies showed evidence of no improvement in
adverse event rates from 2003 to 2008.14
Against this backdrop, AHRQ, believing that the time has come to re-examine the state of the
evidence supporting a wide variety of PSPs, commissioned a team led by investigators at RAND
Health, UCSF, and Johns Hopkins to reexamine the evidence behind key PSPs. Many of the
individuals engaged in this task participated in producing the original MHCS report, the MCHS
methodology formed the cornerstone of the present effort, and many of the practices examined
for this report were those previously reviewed in 2001. Thus, we see the present report as a
natural sequel to MHCS.
Because of the burgeoning literature relevant to patient safety and the limits of budget and
time, we selected a subset of PSPs to examine for this present report (chosen through methods
described in Chapter 2) rather than attempt, as we did in 2001, to review all PSPs. Moreover, the
maturation of the safety field has led to a deeper appreciation of the importance of context in
PSPs, a topic examined by our research team in our 2010 report, Assessing the Evidence for
Context-Sensitive Effectiveness and Safety of Patient Safety Practices: Developing Criteria.17
Accordingly, this report emphasizes matters of context and generalizability, as well as
unintended consequences, to a greater degree than did MHCS.
Who Will Use This Report, and for What Purpose?

We envision that this report will be useful to a wide audience.
Policymakers may use its contents and recommendations to promote or fund the implementation
of particular practices. Similarly, leaders of local health care organization (including hospitals,
medical groups, or integrated delivery systems) may use the data and analyses to choose which
practices to consider implementing or further promote at their institutions. Because of
consumers keen interest in patient safety, the connection between the emergence of an evidencebased practice and the enactment of an associated accreditation standard, regulation, public
reporting requirement, or payment-based initiative is much tighter for PSPs than it is for clinical
practices. This makes it particularly crucial that policymakers have good data on which to base
their decisions.
Clinicians are increasingly being asked to participate in patient safety activities and want to
know the evidence supporting PSPs that they are being asked to help implement. For trainees
and teachers, patient safety is now seen as foundational to the education of doctors, nurses,
pharmacists, health care administrators, and other ancillary health care personnel. We hope that
trainees and practicing clinicians alike will find the material both interesting and relevant to their
day-to-day practices.
Researchers will find a wealth of potential research opportunities. Those who fund research,
including (but not limited to) AHRQ, will find that we explicitly identified future research needs.
As our understanding of the patient safety field has matured, researchers have become
increasingly aware of the complexity of PSPs. For example, the widespread enthusiasm for the
use of checklists18 (which was largely absent in 2001) has led to cautionary notes from several
patient safety leaders regarding the degree to which even seemingly simple PSPs are dependent
on culture change and local context.19,20
Patient safety professionals, meaning people directly involved in improving patient safety and
those working in organizations focused on quality and patient safety, overlap with each of the
three groups above, but deserve their own designation here, as they may be the most frequent and
intense users of this report as they seek to improve patient safety at their own institutions.
Finally, while this volume is not primarily written for patients and their families, both groups
have become increasingly involved in patient safety efforts in a variety of ways. We welcome
such engagement and believe that patients, families, and their advocates can help advance efforts
to prevent harm.
A decade ago, our early enthusiasm for patient safety was accompanied by a hopeand
some magical thinkingthat finding solutions to medical errors would be relatively
straightforward. Simply adopt some techniques drawn from aviation and other safe industries,
build strong information technology systems, and improve culture, and, the hope went, patients
would immediately become safer in hospitals and clinics everywhere.
We now appreciate the naivety of this point of view. Making patients safe will require
ongoing efforts to improve practices, training, information technology, and culture. It will need
top-down resources and leadership, accompanied by bottom-up wisdom and innovation. It will
depend on a strong policy environment that creates appropriate incentives, while avoiding an
environment in which providers enthusiasm and creativity are sapped by an overly rigid,
prescriptive bureaucracy and set of rules.
While we have become more sophisticated about the challenges of keeping patients safe over
the past decade, the fundamentals have not changed: we need good and well-trained people,
armed with good data, operating under good policies, working under good leaders to do the right
things for patients. We hope this report contributes to these efforts by helping to identify those
right things.
References
1.
2.
3.
4.
Kohn L, Corrigan J, Donaldson M, eds. To

System . Washington, DC: Committee on
Quality of Health Care in America, Institute
of Medicine: National Academy Press;
2000.
Donaldson L. An Organisation with a
Memory: Report of an Expert Group on
Learning from Adverse Events in the NHS
Chaired by the Chief Medical Officer.
London: The Stationery Office; 2000.
No. 43, AHRQ Publication No. 01-E058.
Research and Quality; 2001.
et al. Safe but sound: patient safety meets
evidence-based medicine. JAMA
2002;288:508513.
5.
Leape LL, Berwick DM, Bates DW. What

practices will most improve safety?
Evidence-based medicine meets patient
safety. JAMA 2002;288:501507.
6.
Auerbach AD, Landefeld CS, Shojania KG.

The tension between needing to improve
care and knowing how to do it. N Engl J
Med 2007;357:608613.
7.
Berwick DM. The science of improvement.

JAMA 2008;299:11821184.
8.
Wachter RM. Patient safety at ten:

unmistakable progress, troubling gaps.
Health Aff (Millwood). 2010;29:165173.
9.
Shetty KD, Bhattacharya J. Changes in

hospital mortality associated with residency
work-hour regulations. Ann Intern Med
2007;147:7380.
10.
Chan PS, Jain R, Nallmothu BK, et al. Rapid

response teams: a systematic review and
meta-analysis. Arch Intern Med
2010;170:1826.
11.
Sittig DF, Singh H. Defining health

information technologyrelated errors: new
developments since to err is human. Arch
Intern Med 2011;171:12811284.
12.
Shekelle PG, Pronovost PJ, Wachter RM, et

al. Advancing the science of patient safety.
Ann Intern Med 2011;154:693696.
13.
Benning A, Dixon-Woods M, Nwulu U, et

al. Multiple component patient safety
intervention in English hospitals: controlled
evaluation of second phase. BMJ
2011;342:d199.
14.

resulting from medical care. N Engl J Med
2010;363:21242134.
15.
Classen DC, Resar R, Griffi n F, et al.

Global Trigger Tool shows that adverse
(Millwood) 2011;30:581589.
16.
Levinson DR. Adverse Events in Hospitals:

National Incidence Among Medicare
Beneficiaries. Washington, DC: U.S.
Department of Health and Human Services,
Office of the Inspector General; November
2010. Report No. OEI-06-09-00090.
17.

al. Assessing the Evidence for ContextSensitive Effectiveness and Safety of Patient
Contract Final Report. AHRQ Publication
No. 11-0006-EF, December 2010. Prepared
under Contract No. HHSA-290-200910001C. Agency for Healthcare Research
and Quality, Rockville, MD.
www.ahrq.gov/qual/contextsensitive/
18.
Pronovost P, Needham D, Berenholtz S, et

al. An intervention to decrease catheter
related bloodstream infections in the ICU. N
Engl J Med 2006;355:27252732.
19.
Bosk CL, Dixon-Woods M, Goeschel CA, et

al. Reality check for checklists. Lancet
2009;374:444445.
20.
Dixon-Woods M, Bosk CL, Aveling EL, et

al. Explaining Michigan: developing an ex
post theory of a quality improvement
program. Milbank Q 2011;89:167205.
Chapter 2. Methods
Topic Development
This topic was nominated by leaders of the Agency for Healthcare Research and Qualitys
Patient Safety Portfolio, part of the Center for Quality Improvement and Patient Safety.
The original goals of the project were stated as follows:
The analysis shall build on and expand upon earlier evidence
reports and current listing of Safe Practices by the National Quality
Forums (NQF) Safe Practices for Better Healthcare 2010
Update. The analysis shall focus on the collection of evidence of
the effectiveness of new safe practices that have been developed
but not included in the 2010 update, evidence of implementation of
current and new safe practices and the adoption of safe practices
by health care providers. This analysis shall include the review of
scientific literature, other appropriate analyses, and extensive peer
review of the draft report. The final report of this project will be
used by AHRQ for strategic planning in its patient safety portfolio
for future project development, implementation of safe practices.
The report will also be used by external organizations such and the
NQF, Joint Commission and others in their patient safety efforts.1
The preliminary Key Questions, pending topic refinement, were organized into three
categories.
Design, Development and Testing of New Patient Safety Practices
What new patient safety practices (PSPs) have been developed since 2001 and/or are not
included in the NQF safe Practice list in 2010?
What is the nature of the safety practice i.e. clinical, organizational, or behavioral?
What is the intended risk that the practice is designed to prevent or mitigate?
Describe how the practice is a bundle of individual components or practices, if
applicable.
What is the intended setting for the practice, i.e., in patient, ambulatory, combination,
specialty, or clinical domain, and organizational setting?
What are the nature, quality, and weight of evidence of the practices effectiveness?
Implementation of Patient Safety Practice
Was the safety practice implemented outside the developing institution?

What were the contextual settings in which it was implemented?
What were the issues, barriers, problems, successes, and failures in the implementation of
the practice?
What modifications and/or customizations were made (if any) in the implementation
process?
What are the different implementation settings outside the developing institution that
have been reported for this practice?
Describe how the practice has been sustained in its use after initial implementation.
Was there any external support for the implementation process, e.g., AHRQ technical
support, use by a collaborative, or quality improvement organization (QIO)?
Adoption/Diffusion
What is the extent to which the practice has been adopted by multiple institutions or
organizations outside the developing institution?
Was there any organized activity or program to support the diffusion of this innovation or
practice?
What, if any, evidence exists on the sustained use of the practice?
Has the practice become a requirement for use by any accreditation or credentialing
agency or organization?
Project Overview
An overview of the project is depicted in Figure 1. A key aspect of this project is the active
participation of a Technical Expert Panel (TEP) comprising a large number of patient safety
stakeholders and evaluation methods experts. We retained the participation of the TEP that had
participated in a prior AHRQ-supported project, Assessing the Evidence for Context-Sensitive
Effectiveness and Safety of Patient Safety Practices: Developing Criteria (hereafter referred to
as Context Sensitivity). The TEP comprised many of the key patient safety leaders in the
United States., Canada, and the United Kingdom, including experts in specific PSPs, as well as
experts in evaluation methods and people charged with implementing PSPs in hospitals and
clinics.
We divided the project into three phases: topic refinement, the evidence review, and the
critical review and interpretation of the evidence. The project team conducted the topic
refinement and the critical review and interpretation of the evidence jointly with the TEP; the
project team performed the evidence review.
Topic Refinement
Because the goals of the project were to assess the evidence of the effectiveness of new safe
practices and the evidence of implementation of currentsafe practices, practically all PSPs
were potentially eligible for inclusion in this review. Thus our first task was to refine the scope
of the topic to something that was achievable within the timeframe and budget for the project;
this task was undertaken by the project team and the TEP. Figure 1 presents an overview of how
this task was accomplished. We first compiled a list of potential PSPs for the review, starting
with the 79 topics in the MHCS report (2001)2 and adding practices from the National Quality
Forums 2010 Update, the Joint Commission, and the Leapfrog Group; practices identified in an
initial scoping search; and those suggested by our TEP. This effort resulted in an initial list of
158 potential PSPs (see Appendix A).
Figure 1, Chapter 2. Overview of the project
We then conducted an internal project team process that included amalgamation of some topics
and renaming of others, resulting in 96 PSPs. Internal project team triage resulted in our
identifying 35 PSPs we believed must be included, 48 PSPs about which we were unsure, and 13
that we believed could be excluded or folded into other PSPs that were on our include list
(Table 1). As indicated, we incorporated some of those 13 topics into other topics, such as the
monitoring topics. We excluded others that we judged to represent more of a quality issue than a
patient safety issue (such as pneumococcal vaccination interventions and regionalizing surgery to
high volume centers), whereas we judged others to be too late (warfarin interventions, in light of
the emergence of new oral anticoagulants) or too early in development (radio-frequency
identification [RFID] devices attached to wandering patients) for consideration.
Table 1, Chapter 2. Initially excluded topics

Topic
Use of computer monitoring for potential ADEs (ADEs
related to targeted classes (analgesics, potassium
chloride, antibiotics, heparin) (focus on detection))
Anticoagulation services and clinics for coumadin
(Adverse events related to anticoagulation)
Team Comment
Seems this could be part of a broader focus on patient
safety reporting systems.
This is likely to become less important in the future with a
move to non-coumadin-based anticoagulants such as
dabigatrin, which do not require the same degree of
monitoring.
We could include if the policy recommendation (i.e., the
intervention) was to implement this type of policy. This
became a safety practice when Leapfrog included it,
but its just as easy to argue that its quality rather than
safety.
Would bundle in preventing SSI.
Localizing specific surgeries and procedures to high

volume centers (Mortality associated with surgical
procedures)
Maintenance of perioperative normothermia (Surgical site

infections)
Use of supplemental perioperative oxygen (Surgical site
infections)
Intraoperative monitoring of vital signs and oxygenation
(Critical events in anesthesia)
Methods to increase pneumococcal vaccination rate
(Pneumococcal pneumonia)
Pain management (overall topic)
Non- pharmacologic interventions to relieve postoperative pain (e.g., relaxation, distraction)
Endoscope reprocessors (Healthcare-associated
infections)
Laser resistant endotracheal tubes (Surgical fire)
Surgical and exam gloves (i.e., to prevent infection from
clinician to patient)
RFID-type tracking of patient location (e.g., for
wandering) (Wandering and elopement in
patients/residents with dementia, or infant abduction)
Could add to info on OR data integration and display

systems
Seems more like a quality issue than a safety issue.
Probably not a PSP.
Include under reprocessing topic
Not sure if covered in other topic.

Interesting topic, but no evidence yet that the team
knows of.
We then sought input from our TEP about these decisions, offering them the opportunity to
change any of the include/exclude decisions, and asked for formal votes on the 49 PSPs
classified an Unsure.
This effort resulted in 48 PSPs judged to be of highest priority in terms of the need for an
evidence review of effectiveness, implementation, or adoption, still too large a number of topics
to review comprehensively within the given timeframe. Therefore, we asked our TEP to assess
whether breadth or depth was likely to be more valuable for stakeholdersin other words,
we asked whether the review should focus on fewer topics in more detail or cover all topics but
in less detail. Our TEP recommended a hybrid approach in which some topics would be
reviewed in depth, whereas other topics would receive only a brief review. Topics could be
considered to need only a brief review for several reasons: the PSP is already well-established;
stakeholders need to know only whats new? since the last time this topic was reviewed in
depth; new evidence suggests the PSP may not be as effective as originally believed, so it is no
longer a priority safety practice to implement; or it is an emerging PSP with limited evidence yet
accumulated about it.
For each of the 48 topics, we then solicited formal input from our TEP about the need for an
in-depth review, a brief review, or no review at all. Table 2 presents the results in terms of the
proportion of TEP members who recommended a topic undergo an in-depth review, a brief
review, or no review at all. We designated all topics that received 50 percent or greater support
for an in-depth review to be reviewed in depth; all other topics were designated for brief reviews.
No topic on the list received 50 percent or greater support for no review at all. The list underwent
further modification, as some PSPs originally designated as separate topics were judged to be
sufficiently similar to be covered together in one review; examples included the topics related to
transitions in care and those related to monitoring.
A final set of modifications to this scope occurred during the course of the reviews.
Our PSP topic on pressure ulcers was modified to focus solely on implementation, as an
EPC review of the effectiveness of pressure ulcer prevention interventions is currently
underway.
We combined the topics, diagnostic errors and notification of test results to patients
into a single in-depth review.
The body of literature on simulation methods was sufficiently large that we treated it as
an in-depth review.
The review topics were then divided among the participating EPCs. Weekly teleconference
calls and email were used to promote common practices in the review process.
Table 2, Chapter 2. Proportion of technical expert panelists expressing a preference for the level
of evidence review for each PSP
PSP*
In-Depth
In-Depth
Handoff - (Transitions in care)
Rapid response teams
Fall prevention strategies and
interventions to reduce the use of
restraints
Diagnostic errors - meta-cognition,
computerized decision support
Protocols for notification of test results to
patients
Geriatric/delirium programs
Monitoring for patient safety problems
Preventing ventilator-associated
pneumonia
Pressure ulcer prevention
Promoting a culture of safety
Universal protocol/preoperative checklist
(surgical safety)
Report cards/outcomes measurement
like NSQIP (surgical safety)
Nurse staffing patterns and ratios
Other interventions targeting improved
transitions in care (Transitions in care)
Intensive insulin therapy for glycemic
control
Use of preoperative anesthesia
checklists (Complications due to
anesthesia equipment failures)
Protocols for high risk drugs, e.g.,
nomograms for heparin
Interventions to prevent contrast-induced
renal failure
The patients role in preventing errors
Brief Review
No Review
79%
71%
67%
21%
29%
20%
0%
0%
13%
64%
29%
7%
64%
21%
14%
64%
29%
7%
64%
57%
7%
36%
29%
7%
57%
21%
21%
57%
53%
36%
33%
7%
13%
50%
43%
7%
50%
43%
7%
50%
36%
14%
50%
36%
14%
50%
36%
14%
50%
29%
21%
50%
29%
21%
50%
36%
14%
50%
21%
29%
Table 2, Chapter 2. Proportion of technical expert panelists expressing a preference for the level
of evidence review for each PSP (continued)
PSP*
In-Depth
Brief Review
No Review
Human factors as a general topic,
50%
13%
38%
focus still to be more precisely defined
Brief Review
CPOE and clinical decision support
47%
47%
7%
systems (CDSS)
Bundles and checklists as a general
47%
40%
13%
strategy (not just for specific indications)
Simulator-based training
46%
31%
23%
Prevention of surgical items left inside
43%
50%
7%
patient (surgical safety)
Medication administration
43%
50%
7%
Display systems
43%
43%
14%
Hand washing + interventions to improve
36%
50%
14%
hand washing compliance
Perioperative beta- blockers
36%
57%
7%
VTE prophylaxis and methods for
36%
50%
14%
implementation
Team training/team practices
36%
43%
21%
Limiting individual providers hours of
36%
50%
14%
service
Smart pumps and other protocols for
36%
50%
14%
infusion pumps
Device-related strategies for preventing
36%
43%
21%
tubing misconnections
Clinical pharmacist consultation services
36%
43%
21%
Prevention of nosocomial UTIs
33%
53%
13%
Use of real-time ultrasound guidance
33%
60%
7%
during central line insertion
Patient understanding/informed consent
29%
36%
36%
(possibly includes health literacy)
Interventions for central venous catheter29%
50%
21%
related blood infections
Patient death or serious injury associated
with prolonged fluoroscopy with
29%
36%
36%
cumulative dose
Death among surgical patients with
serious treatable complications (failure to
29%
36%
36%
rescue)
Barrier precautions, patient isolation,
routine surveillance for patients at
21%
71%
7%
admission
Identifying patients at risk for suicide
21%
64%
14%
Sign your site protocols - potentially
14%
79%
7%
part of checklists
Processes related to reprocessing
14%
50%
36%
single-use medical devices
Do not use abbreviations, acronyms,
symbols, and dose designation
14%
79%
7%
campaign
Ensure documentation of patients
7%
50%
43%
preferences for life-sustaining treatment
Strategies to prevent stress-related
7%
50%
43%
gastrointestinal bleeding
*The topic titles listed in this table were the exact titles the TEP considered in their decision-making; some of these PSP topics or
titles underwent further revisions to their final title between that assessment and this final report.
10
Evidence Assessment Framework

The framework for our consideration of the evidence regarding a PSP was worked out as part
of the prior AHRQ Context Sensitivity project.3 A principal challenge in previous reviews of
PSPs has been addressing the question of what constitutes evidence for PSPs. Many practices
designed to improve quality and safety are complex sociotechnical interventions whose targets
may be entire health care organizations or groups of providers, and these interventions may be
targeted at extremely rare events. To address the challenge regarding what constitutes evidence,
we recognize that PSPs must be evaluated along two dimensions: (1) the evidence regarding the
outcomes of the safety practices, and (2) the contextual factors that influence the practices use
and effectiveness.
Figure 2 presents this framework, depicting a generic PSP that consists of a bundle of
components (the individual boxes) and the context within which the PSP is embedded. Important
evaluation questions, as depicted on the right, concern effectiveness and harms, implementation,
and adoption and spread. We then apply criteria to evaluate each of four factors that together
constitute equality (depicted as puzzle pieces in the bottom half of the figure):
1. Constructs about the PSP, its components, context factors, outcomes, and ways to
measure these constructs accurately;
2. Logic model or conceptual framework about the expected relationships among these
constructs;
3. Internal validity to assess the PSP results in a particular setting; and
4. External validity to assess the likelihood of being able to garner the same results in
another setting.
We then synthesize this information into an evaluation of the strength of the evidence about a
particular PSP.
Figure 2, Chapter 2. Framework for evidence assessment of patient safety practices
The principal results of the Context-Sensitivity project included the following key points.
Whereas controlled trials of PSP implementations offer investigators greater control of
sources of systematic error than do observational studies, trials often are not feasible, in
terms of time or resources. Also, controlled trials are often not possible for PSPs
11
requiring large-scale organizational change or PSPs targeted at very rare events.

Furthermore, the standardization imposed by the clinical trial paradigm may stifle the
adaptive responses necessary for some quality improvement or patient safety projects.
Hence, researchers need to use designs other than RCTs to develop strong evidence about
the effectiveness of PSPs.
Regardless of the study design chosen for an evaluation, components that are critical for
evaluating a PSP in terms of how it worked in the study site and whether it might work in
other sites include the following:
o Explicit description of the theory for the chosen intervention components, and/or an
explicit logic model for why this PSP should work;
o Description of the PSP in sufficient detail that it can be replicated, including the
expected change in staff roles;
o Measurement of contexts;
o Explanation, in detail, of the implementation process, the actual effects on staff roles,
and changes over time in the implementation or the intervention;
o Assessment of the impact of the PSP on outcomes and possible unexpected effects
(including data on costs, when available); and
o For studies with multiple intervention sites, assessment of the influence of context on
intervention and implementation effectiveness (processes and clinical outcomes).
High-priority contexts for assessing any PSP implementation include measuring and
information for each of the following four domains:
o Structural organizational characteristics (such as size, location, financial status,
existing quality and safety infrastructure);
o External factors (such as regulatory requirements, the presence in the external
environment of payments or penalties such as pay-for-performance or public
reporting, national patient safety campaigns or collaboratives, or local sentinel patient
safety events);
o Patient safety culture (not to be confused with the larger organizational culture),
teamwork, and leadership at the level of the unit; and
o Availability of implementation and management tools (such as staff education and
training, presence of dedicated time for training, use of internal audit-and-feedback,
presence of internal or external people responsible for the implementation, or degree
of local tailoring of any intervention).
These principles guided our search for evidence and the way we present our findings in this
report (see Table 3).
12
Table 3, Chapter 2. Format for in-depth reviews

This section describes what has been written about the basis for a proposed Patient Safety Practice, such
as an underlying theory, a logic model for how it should work, or prior data.
This section provides the review of the evidence of effectiveness, and is the section most similar to
traditional Evidence-based Practice Center reports.
This section contains the evidence of harms. Unlike reviews of most clinical interventions, evaluating
potential harms is not a routine part of Patient Safety Practice evaluations. Thus, for most topics, this section
is underdeveloped.
This section describes what has been reported about how to implement the Patient Safety Practice and the
range of institutions or contexts of where it has been implemented. When there is sufficient evidence,
implementation studies are evaluated qualitatively for themes regarding effective implementation.
This section describes the evidence of costs of implementing the Patient Safety Practice, or, in some cases,
cost-effectiveness analyses that have been performed.
This section describes the evidence about whether or not the Patient Safety Practice has been shown to
have differential effectiveness in different contexts. The Context Sensitivity project defined important
contexts for Patient Safety Practices in four domains: external factors (e.g., financial or performance
incentives or Patient Safety Practice regulations); structural organizational characteristics (e.g., size,
organizational complexity, or financial status); safety culture, teamwork, and leadership involvement; and
4
availability of implementation and management tools (e.g., organizational training incentives).
Evidence Review Process

As already noted, this report presents two types of evidence reviews: in-depth reviews and
brief reviews. In this section, we describe the general methods for each type of review. The
details of the review processes for individual topics (for example, the search strategies and flow
of articles) varied by topic and are described in Appendix C. The evidence reviews were
conducted by the project team. Figure 3 presents an outline of the general methods for each type
of review.
In-Depth Reviews
Many of the 18 topics designated for an in-depth review were likely to have been the subject
of a previous systematic review; thus, the review process usually began with a search to identify
existing systematic reviews. To assess their potential utility, we followed the procedures
proposed by Whitlock and colleagues5 which essentially meant addressing the following two
questions:
Is the existing review sufficiently on topic to be of use? and
Is it of sufficient quality for us to have confidence in the results?
13
Assessment of whether a review was sufficiently on topic was a subjective judgment based
on the patients-intervention-comparators-outcomes-timeframe (PICOT) focus of the existing
review. To assess the quality of the systematic review, we, in general, used the AMSTAR criteria
(see Appendix B).6 If an existing systematic review was judged to be sufficiently on topic and
of acceptable quality, then based on that review, the following searches were undertaken:
A full update search, in which databases were searched for new evidence published since
the end date of the search in the existing systematic review; and/or
A search for signals for updating, according to the criteria proposed by Shojania and
colleagues,7 which involved a search of high-yield databases and journals for pivotal
studies whose results might be a signal that a systematic review is out-of-date.
Based on the results of these searches, the existing review was supplemented with newer
evidence or considered to be up-to-date.
Any evidence identified via the update search or the signals search was added to the
evidence base from the existing systematic review.
For some topics, no systematic review could be identified, or those that were identified were
either not sufficiently relevant or not of sufficient quality to be used. In those situations, new
searches were done using guidance as outlined in the EPC Methods Guide.8
As indicated above, evidence about context, implementation, and adoption are key aspects of
this review. We searched for evidence on these topics in two ways:
We looked for and extracted data about contexts and implementation from the articles
contributing to the evidence of effectiveness;
We identified implementation studies from our literature searches. Implementation
studies focus on the implementation process, especially those elements of the implementation
demonstrated or believed to be of particular importance for the success, or lack of success, of the
intervention. To be eligible, implementation studies needed to either report, or be linked to
reports of, effectiveness outcomes.
Brief Reviews
Brief reviews are explicitly not full systematic reviews or updates. The goals of the brief
reviews varied by PSP, according to the needs of stakeholders. The assessment could focus
primarily on information about effectiveness of an emerging PSP or implementation of an
established PSP; alternatively, the review could explore whether new evidence calls into
question the effectiveness of an existing PSP. Thus, the methods used to conduct the brief
reviews varied according to the various goals of the reviews. . However, in general, brief reviews
were conducted by an expert in the topic in collaboration with the project team, and involved
focused literature searches for evidence relevant to the specific need. This evidence was then
narratively summarized in a format that also varied with the particular goal.
14
Figure 3, Chapter 2. Evidence review process
Assessing Quality of Individual Studies

In general, to assess the quality, or risk of bias, of individual studies contributing evidence of
effectiveness to in-depth reviews, we used the criteria published on the Cochrane Effective
Practice and Organisation of Care (EPOC) Web site.9 This Cochrane Group is devoted to
reviews of interventions designed to improve the delivery, practice, and organization of health
care. Thus, it uses quality/risk of bias assessment instruments that are applicable to numerous
study designs; criteria are available for controlled-before-and-after studies and for time series
studies, as well as for randomized trials.
For the many topics included in this review for which we identified an existing systematic
review as a starting point for our review, we accepted the original reviews assessment of the
quality/risk of bias of included studies. In other words, we did not re-score the original studies
included in an existing systematic review for risk of bias. A consequence of this decision is that
we did not apply the EPOC criteria to assess quality/risk of bias for some topics in this report,
15
but instead relied on the criteria originally chosen for that review, for example the criteria of the
U.S. Preventive Services Task Force.
Implementation studies were not assessed for their quality, as we lacked evidence or expert
opinion about the criteria for such an assessment.
Assessing Strength of Evidence for a Patient Safety Practice

Table 4 shows the scheme we employed for assessing the strength of the body of evidence
regarding a specific PSP. This scheme starts with elements taken from the EPC Methods Guide
on strength of evidence,10 which itself borrows largely from the GRADE scheme,11,12 and
incorporates elements about theory, implementation, and context taken from the prior AHRQ
Context Sensitivity report.3 It includes an assessment of the risk of bias, by whatever criteria
were used for a particular PSP, and then adjusts the strength up or down based on standard
GRADE criteria and on criteria about the use of theory and description of implementation. The
points for scoring are meant only as a guide. Implementation studies were not assessed for
strength of evidence.
Table 4, Chapter 2. Criteria for assigning strength of evidence for effectiveness/harms questions
What does the evidence show about the effectiveness of this PSP among those at risk?
Individual study risk-of-bias score: Low (+4); Moderate (+3); High (+2); for Cochrane/EPOC Risk of Bias
instrument, suggest zero No answers = Low risk, one to two No answers = Moderate risk, and three or
more No answers = High risk; suggest taking the median or average as the overall risk of bias for the
evidence base.
Across all study types, decrease score if:
Important inconsistency across studies (-1)

Serious imprecision (-1)
High probability of reporting bias (-1)
No explanation in any of the studies of why the PSP might work, either in terms of theory, logic
models, or prior success in other fields or in pilot studies (-1)
PSP not described in sufficient detail to permit replication (-1)
Across all study types, increase score if:
Very strong effect in majority of studies (+1)

All plausible residual confounding would reduce a demonstrated effect or would suggest a spurious
effect if no effect was observed (+1)
Use of theory/logic models, assessment of contexts, reporting of implementation process, and
fidelity of implementation (+1)
For observational studies, increase score if:
Use of observational study designs of stronger internal validity (controlled before-and after, time
series, statistical process control) (+1)
If evidence allows a conclusion, then strength of evidence should be scored as follows:
+4 = High
+3 = Moderate
+2 = Low
If evidence does not permit a conclusion then the strength of evidence = insufficient
16
Summarizing the Evidence

We expected that users of this report would want a summary of the evidence for each topic.
Such summary messages may facilitate uptake of the findings. We summarized the evidence
according to the following domains:
Scope of the problem. In general, we addressed two issues: (1) the frequency of the safety
problem, and (2) the severity of each average event. For benchmarks, we regarded safety
problems that occur approximately once per 100 hospitalized patients, as common; examples
include falls, venous thromboembolism (VTE), potential adverse drug events, or pressure ulcers.
In contrast, events an order of magnitude or more lower in frequency were considered rare;
such events include inpatient suicide and surgical items left inside the patient. The scope must
also consider the severity of each event: most falls do not result in injury, and most potential
adverse drug events do not result in a clinical harm. However, each case of inpatient suicide or
wrong site surgery is devastating.
Strength of evidence for effectiveness. In general, this assessment follows the framework for
strength of evidence presented above.
Evidence on potential for harmful unintended consequences. Most PSP evaluators have not
explicitly assessed the possibility of harm. Consequently, this domain includes evidence of both
actual harm and the potential for harm. The ratings on known or potential harms ranged from
high risk of harm to low (or negligible); in some cases, the evidence was too sparse to provide a
rating.
Estimate of costs. This domain is speculative, because most evaluations do not present cost data.
However, we judged that readers would want at least a rough estimate of cost. Therefore, we
used the following categories and benchmarks, noting in places the factors that might cause cost
estimates to vary.
Low cost: PSPs that did not require hiring new staff or large capital outlays, but instead
involved training existing staff and purchasing some supplies. Examples would include
most falls prevention programs, VTE prophylaxis, or medical history abbreviations
designated, Do Not Use.
Medium cost: PSPs that might require hiring one or a few new staff, and/or modest
capital outlays or ongoing monitoring costs. Examples would include some falls
prevention programs, many clinical pharmacist interventions, or participation in the
American College of Surgeons Outcomes Reporting System ($135,000/year).
High cost: PSPs that required hiring substantial numbers of new staff, considerable
capital outlays, or both. Examples would include computerized order entry (because it
requires an electronic health record), having to hire many nurses to achieve a certain
nurse-to-patient ratio, or facility-wide infection control procedures (estimated at
$600,000 year for a single intensive-care unit [ICU]).
Implementation issues. This section summarizes how much we know about how to implement
the PSP, and how difficult it is to implement. To approach the question of how much we know,
we considered the available evidence about implementation, the existence of data about the
effect of context and the influence of context, the degree to which a PSP has been implemented,
17
and the presence of implementation tools such as written implementation materials or training
manuals.
For the question of implementation difficulty, we use three categories: difficult for PSPs that
required large scale organizational change; not difficult for PSPs that required protocols for
drugs or devices such as those to reduce radiation exposure or to help prevent stress-related
gastrointestinal bleeding; and moderate for PSPs falling between the extremes.
Setting Priorities for Adoption of Patient Safety Practices

After obtaining critical input from our TEP about the dimensions and benchmarks used for
summarizing the evidence, we next solicited their views on whether the evidence was sufficient
at present to encourage wider adoption of some of the PSPs. Specifically, we asked our TEP the
following questions:
We are asking for your global judgment of the priority for adoption
of the PSPs that are included in our report. By global judgment,
we mean that you will be making a summary judgment, which
considers all the factors discussed in the chapters and listed in the
summary table (the magnitude of the current safety problem [in
terms of frequency and severity], the degree to which the PSP can
improve safety outcomes, any potential for unintended
consequences, what we know and how hard it is to implement the
PSP, and the cost) plus your own experience as a researcher,
provider, policymaker, or PSP developer. We have chosen a fourcategory scheme for this judgment:
THIS PSP SHOULD BE STRONGLY ENCOURAGEDWe
know enough now that if we were choosing a hospital (or nursing
home or ambulatory care center, etc.) to get care from, we would
choose a hospital (or nursing home or ambulatory care center, etc)
that was implementing this PSP over one which was not. Another
way of thinking about this might be: unless the hospital (or nursing
home or ambulatory care center, etc) knows its outcomes for this
safety problem are already excellent (or the safety problem is not
relevant for the setting, such as failure-to-rescue in an ambulatory
care center), then it ought to be implementing this PSP. We would
expect over the next 3 years that most organizations would
implement this PSP, even if it has substantial cost. Most does not
have a precise definition but it does not mean 51% nor does it
mean 95%. Lets say it means about 70-80%.
THIS PSP SHOULD BE ENCOURAGEDThis is a PSP that
wed like to be implemented at the hospital (or nursing home or
ambulatory care center, etc.) where we would receive our care, but
theres just enough uncertainty about the effect, or concern about
the cost, or some other factor, to keep us from putting it on the
strongly encouraged list. We would expect that over the next 3
years many organizations would implement this PSP, and high cost
might be a significant factor in an organizations decision.
18
THIS PSP IS STILL DEVELOPMENTALTheres still more

that needs to be known about this PSP before we should be
encouraging health care providers to adopt it. Organizations
implementing these PSPs should be encouraged to publish
evaluations of their implementation and effectiveness in order to
increase the evidence base for the PSP.
THIS PSP SHOULD BE DISCOURAGEDThis PSP is one
where were pretty sure the cost or difficulty implementing it is not
worth the potential benefit, or even that the harms or potential for
harms exceeds the evidence of benefit.
As in prior group judgment processes, we also provide a response
option I DO NOT WANT TO RATE THIS PSP so that people
are not forced to make decisions about PSPs they feel unprepared
to assess, AND we can distinguish between that decision and an
inadvertent skipped PSP.
We received input from 19 of the 21 members of the TEP; the remaining two declined to rate
the PSPs because they judged that making these kinds of clinical and policy decisions was not
within their area of expertise. Based on the judgments of the panelists, we classified the PSPs
according to the following rules:
Strongly Encouraged: To be classified as strongly encouraged, a PSP had to receive a
rating of strongly encourage or encourage from 75 percent or more of the technical
experts, no TEP member could rate the PSP as this PSP should be discouraged, and a
majority of the strongly encourage/encourage ratings had to be strongly encourage.
Encouraged: To be classified as encouraged, a PSP had to receive a rating of strongly
encourage or encourage from 75 percent or more of the technical experts, and a
majority of the strongly encouraged/encourage ratings had to be encourage.
In any such process, the thresholds are somewhat arbitrary and can magnify the apparent
impact of small differences in ratings. Therefore, we also assessed PSP at the threshold between
strongly encourage and encourage (two PSPs received equal numbers of votes for each
category) and the threshold between encourage and no rating (four additional PSPs). For these
additional ratings, we used a four-person subset of our TEP, the people actually responsible for
policymaking or implementing PSPs. For each of our threshold PSPs, we judged that three of
these four technical experts needed to either encourage or strongly encourage the PSP, to
retain its strongly encouraged or encouraged Classification. This determination resulted in
one PSP being down-rated from strongly encouraged to encouraged, and affirmed that all
four PSPs that made it by one vote should be classified as encouraged.

To assess future research needs with respect to PSPs, we first devoted 2 hours of discussion
time at the face-to-face meeting of the TEP to this topic. Two project team members recorded
both general and specific topics for future research that the TEP discussed. From these notes we
obtained themes or domains that we used to organize the future research needs. To these we
added future research needs for specific PSPs suggested by the individual team members who
reviewed the literature on those PSPs. We then sought input from the TEP regarding which
19
future research needs were highest priority, and classified as high priority those topics receiving
more than 50 percent support.
Peer and Public Review Process

The draft of this report was posted for public comment and sent to six peer reviewers and our
TEP for review.
References
1.
2.
3.
4.
5.
Evidence-based Practice Center Systematic

Review Protocol: Critical Analysis of the
Evidence for Patient Safety Practices.
Research and Quality. November 9, 2011.
http://effectivehealthcare.ahrq.gov/searchfor-guides-reviews-andreports/?pageaction=displayproduct&produc
tid=840.
Quality. July 2001.
(Prepared under Contract No. HHSA-2902009-10001C.). AHRQ Publication No. 110006-EF. Rockville, MD: Agency for
Healthcare Research and Quality. December
2010. www.effectivehealthcare.ahrq.gov.
interventions? BMJ Qual Saf.
2011;20(7):611-7.21617166.
Whitlock EP, Lin JS, Chou R, et al. Using
existing systematic reviews in complex
systematic reviews. Ann Intern Med.
2008;148(10):776-82.18490690.
20
6.
Shea BJ, Grimshaw JM, Wells GA, et al.

Development of AMSTAR: a measurement
tool to assess the methodological quality of
systematic reviews. BMC Med Res
Methodol. 2007;7:10.17302989.
7.

Med. 2007;147(4):224-33.17638714.
8.
Relevo R, Balshem H. Methods Guide for

Effectiveness and Comparative
Effectiveness Reviews. Chapter 3. Finding
Evidence for Comparing Medical
Interventions. AHRQ Publication No.
10(12)-EHC063-EF. Rockville, MD:
April 2012.
9.
Cochrane Effective Practice and

Organisation of Care Group (EPOC). 2011
cited July 20, 2012;
http://epoc.cochrane.org/sites/epoc.cochrane
.org/files/uploads/EPOC%20Study%20Desi
gns%20About.pdf.
10.
Owens DK, Lohr KN, Atkins D, et al.

AHRQ series paper 5: grading the strength
of a body of evidence when comparing
medical interventions--agency for healthcare
research and quality and the effective healthcare program. J Clin Epidemiol.
2010;63(5):513-23.19595577.
11.
Atkins D, Eccles M, Flottorp S, et al.

Systems for grading the quality of evidence
and the strength of recommendations I:
critical appraisal of existing approaches The
GRADE Working Group. BMC Health Serv
Res. 2004;4(1):38.15615589.
12.
Grading of Recommendations Assessment,

Development and Evaluation (GRADE)
Working Group.
www.gradeworkinggroup.org. Accessed
July 20, 2012.
Part 2. Evidence Reviews of Patient Safety Practices

The following pages contain the evidence reviews for 41 patient safety practices or
approaches to care. They are organized as follows:
Practices designed for a specific patient safety target
Practices designed to improve the overall system/multiple targets
Within the section Practices Designed for a Specific Patient Safety Target the topics are
organized according to the target:
Adverse drug events
Infection control
Surgery, anesthesia, and perioperative medicine
Safe practices for hospitalized elders
General clinical topics
Within each subsection, the topics are organized as follows:
In-depth reviews
Brief reviews
In-depth reviews are presented in the following format:
This section describes what has been written about the basis for a proposed Patient Safety Practice, such as an
underlying theory, a logic model for how it should work, or prior data.
This section provides the review of the evidence of effectiveness, and is the section most similar to traditional
Evidence-based Practice Center reports.
This section contains the evidence of harms. Unlike reviews of most clinical interventions, evaluating potential harms
is not a routine part of Patient Safety Practice evaluations. Thus, for most topics, this section is underdeveloped.
This section describes what has been reported about how to implement the Patient Safety Practice and the range of
institutions or contexts of where it has been implemented. When there is sufficient evidence, implementation studies
are evaluated qualitatively for themes regarding effective implementation.
This section describes the evidence of costs of implementing the Patient Safety Practice, or, in some cases, costeffectiveness analyses that have been performed.
This section describes the evidence about whether or not the Patient Safety Practice has been shown to have
differential effectiveness in different contexts. The Context Sensitivity project defined important contexts for Patient
Safety Practices in four domains: external factors (e.g., financial or performance incentives or Patient Safety Practice
regulations); structural organizational characteristics (e.g., size, organizational complexity, or financial status); safety
culture, teamwork, and leadership involvement; and availability of implementation and management tools (e.g.,
organizational training incentives).*
* Taylor SL, Dy S, Foy R, et al. What context features might be important determinants of the effectiveness of patient safety
practice interventions? BMJ Qual Saf. 2011;20(7):611-7.21617166
21
Brief reviews use a different format, that varies somewhat depending on the topic. The general
format for brief reviews is: What is (are) the patient safety practice(s)?; How has the patient
safety practices been implemented?; What have we learned about the practice(s)? Brief update
reviews are topics that were covered in Making Health Care Safer 2001 and use a format
designed for reader to identify whats new since then.
References. Each chapter is individually referenced for convenience.
22
Part 2a. Practices Designed for a Specific Patient

Safety Target
Section A. Adverse Drug Events
Chapter 3. High-Alert Drugs: Patient Safety Practices
for Intravenous Anticoagulants
Elizabeth Pfoh, M.P.H.; David Thompson, D.N.Sc., M.S., R.N.; Sydney Dy, M.D., M.Sc.
How Important Is the Problem?

High-alert medications are defined as medications that are the most likely to cause significant
patient harm, even when used correctly. These medications are more likely to be associated with
harm due to issues such as narrow therapeutic ranges (increasing the potential for a prescribing
error), and also cause more significant harm when an error does occur because of the significant
nature of the potential adverse effects such as bleeding or hypoglycemia.1,2 Many of these
medications are also more likely to be associated with dosing errors, due to issues such as the
need to frequently calculate dosing based on weight. A study evaluating adverse drug events
found that high-alert medications accounted for 48 percent of the events.3
The Institute of Safe Medical Practices identifies the top high-alert medications to be insulin,
opioids, injectable potassium chloride (or phosphate), intravenous anticoagulants (heparin), and
sodium chloride solutions above 0.9 percent, due to both common use and significance of
associated harm.1 Other high-alert medications include chemotherapeutic agents and sedatives.
From 1997 to 2007, 9.3 percent of all hospital sentinel events were medication-related, and
anticoagulants made up 7.2 percent of medication events. Unfractionated heparin was the
anticoagulant most frequently involved in these events.4 Administration errors (e.g., dosing and
timing), omission, and prescribing errors constituted approximately 70 percent of heparin errors.5
What Is the Patient Safety Practice?

In the 2001 Making Health Care Safer report, this PSP was conceptualized as Protocols
for High-Risk Drugs: Reducing Adverse Drug Events Related to Anticoagulants.6 The rationale
for focusing on anticoagulants was that, although a number of other classes of medications have
been identified as high-risk, and some recommendations to reduce risks apply to multiple
classes of medications, the effectiveness of interventions to reduce risks associated with other
medications have not been as extensively evaluated as interventions to reduce risks associated
with heparin. Because interventions to reduce adverse events may differ significantly by drug
type, and the focus on anticoagulants in the inpatient setting is mainly on heparin, this review
focused on heparin, the most commonly used intravenous anticoagulant, as an illustrative
example rather than addressing issues for high-alert medications overall.
The original report reviewed two types of interventions for heparin:
The implementation of dosing protocols or nomograms, which normally include standard
initial doses and instructions for monitoring and adjusting doses
Inpatient anticoagulation services, which provide pharmacist input on dosing and
monitoring
23
Weight-based nomograms use actual patient body weight to calculate an optimum dose that
is patient-specific. In contrast, physician dosing without nomograms often does not account
accurately for patient characteristics.
This current report systematically reviewed the literature to identify effectiveness studies of
any intervention with a goal to reduce adverse events related to intravenous heparin in the
inpatient setting that had a comparison group and was not a qualitative study. Since this PSP is
currently most often conceptualized as focusing on intravenous administration as the most highrisk route, we did not include subcutaneous or oral anticoagulant administration in this review.
Intravenous anticoagulants are particularly high risk because dosing is complex and the
therapeutic range is particularly narrow. This narrow range increases the opportunity for harm.6-9
Although bleeding can occur even at therapeutic doses of heparin, it is much more likely when
the dose is excessive or inadequately monitored. Unfractionated heparin, which is given
intravenously, is widely used as the drug of choice for a variety of clinical conditions where
rapid and closely monitored anticoagulation is needed, such as acute coronary syndromes.10
However, since the 2001 publication of Making Health Care Safer, low-molecular-weight
heparinswhich have a less complex dosing regimen, are given subcutaneously, and have been
shown to have equivalent efficacy for many indicationshave widely replaced unfractionated
heparin for some clinical conditions such as venous thromboembolism (VTE) prophylaxis.
A wide variety of safety practices are recommended to increase patient safety for intravenous
anticoagulants in general. These practices include limiting the number and dosage of high alert
drugs prescribed (to ensure that only patients who are most likely to benefit receive the
medications or that lower-risk options are used whenever possible), having independent system
checks and balances in place to identify and prevent dosing errors, and having a transparent error
reporting system to aid in the development and implementation of system changes.1,2 Other
practices include removing high-alert medications from nursing units and floor stock,
standardizing medication doses, using single doses or pre-mixed solutions, labeling different
strength solutions clearly to avoid mixups (e.g., Heplock packaging), provider education and
drug-administration protocols and decision support tools that involve double-checking of the
drug and dosing, pump-setting, and dosage.4 Health information technology tools may help
reduce errors associated with high-alert medications by preventing significant overdoses (e.g.
tenfold errors in dosing) and verifying that the correct medication is being administered.11,12
However, the level of effectiveness of health information technology may vary.13,14 Specific
heparin patient-safety practices reviewed here include dosing nomograms and weight-based
dosing interventions, with and without the use of health information technology tools as part of
the intervention.
Why Should This Patient Safety Practice Work?

Numerous patient factors, particularly patient weight, can influence the dosing needs for
heparin. Bleeding risk increases as the dose increases and with inappropriately high dosing.
Patients on intravenous heparin have multiple risk factors for bleeding that may also affect
dosing needs: they often have high acuity conditions such as recent stroke, or are undergoing
high-risk procedures such as coronary artery bypass or continuous hemodialysis. In addition,
dosing ranges for heparin vary by indication; physicians often tend to be conservative and
underdose heparin when not using standard nomograms.10 For these reasons, standardization of
dosing and monitoring of subsequent anticoagulation are vital.
24
Heparin-induced adverse effects not related to dosing issues (e.g., heparin-induced

thrombocytopenia) are also important considerations in heparin use, but are not generally
considered patient safety events and were not included in the scope of this review.
What Are the Beneficial Effects of the Patient Safety Practice?

The original Making Health Care Safer report6 found six studies, mostly of low quality, on
heparin nomograms. All showed a statistically significant improvement in time to achievement
of, or proportion of patients with, appropriate anticoagulation. Two low-quality studies of
inpatient anticoagulation services also showed statistically significant improvements in
anticoagulation. All studies either did not evaluate bleeding outcomes or did not have a sufficient
sample size to measure these outcomes. Four of the six studies of nomograms did show a
statistically significant increase in the proportion of patients with partial thromboplastin time
(PTT) values above the normal range (and therefore at increased risk for over-anticoagulation
and bleeding complications).
For this review, a total of 1,960 unique abstracts were captured by the search strategy. Of
these, 1,936 were excluded during the abstract screening phase. Seven articles met the inclusion
criteria for intervention studies evaluating the effectiveness of interventions to improve the safety
of intravenous heparin administration, published after the Making Health Care Safer report
(Table 1). We did not identify any additional recent systematic reviews of high-alert medications
or heparin. We identified five studies evaluating the use of weight-based nomograms, all
published between 2001 and 2005. The only randomized, controlled trial was by Toth and
colleagues, who developed a weight-based nomogram for heparin dosing in transient ischemic
attack (TIA) and/or stroke.15 Out of 206 patients, total complications were significantly reduced
using the nomogram (9 pre [8.5%] vs. 2 post [2%] p=0.04). Additionally, time to
supratherapeutic activated PTT (aPTT) (i.e., adequate anticoagulation) was reduced (1.1 with
nomogram vs. 1.6 without nomogram; p=0.01) and time to therapeutic-range aPTT (i.e.,
therapeutic anticoagulation) was reached with fewer adjustments (18 with the intervention vs. 13
for the control group; p<0.01). Zimmermann and colleagues also used a pre-post design to assess
the effect of a weight-based nomogram for 173 patients with acute coronary syndromes.16
Median time to first therapeutic aPTT was reduced from >24 to 8.75 hours (p<0.001) and the
mean number of aPTT tests decreased from 4.15 (SD.83) to 3.62(SD.85) (p=0.002). Oyen and
colleagues conducted a pre-post study of 419 patients evaluating the implementation of a
computerized nomogram for acute coronary syndromes targeted at nurses and found
improvements in anticoagulation outcomes (percentage of a PTT in goal range 44% with the
nomogram vs. 27% without); data on complications were not reported. Baird and colleagues17
used a pre-post study design in a small patient sample (n=68) to test an evidence-based
nomogram that was developed with a team of nurses, doctors, and a pharmacist; no statistics
were reported. Finally, Fraipont et al developed a nurse-directed weight-based nomogram in a
very small study (total n=38); the study found that there were no statistically significant
differences in anticoagulation outcomes or complications between the intervention and control
groups.18
The remaining two, more recent, papers assessed the impact of technology along with
processes and procedures for the use of the technology on heparin administration safety. A 2011
study by Prusch and colleagues aimed to improve medication safety through the use of intelligent
infusion devices (IIDs), a bar-code-assisted medication administration system, and an electronic
medication administration record system using a pre-post design. Monthly compliance with the
25
telemetry drug library increased from 56.5 percent (SD: 1.5%) pre-intervention to 72.1 percent
(SD: 2.1%) post-intervention (p<0.001), and the number of telemetry manual pump edits
decreased (56.9 [SD: 12.8] to 14.7 [SD: 3.9]; p<0.001).13 Finally, Fanikos and colleagues
assessed the impact of a smart infusion device with a hospital-determined drug library and
programmable software on anticoagulation errors using a pre-post design. After reviewing a total
of 14,012 administered doses of heparin in 3,674 patients, the software generated a total of 501
heparin alerts in 246 patients. No significant difference in anticoagulation errors was found as a
result of the intervention (49 pre- vs. 48 post-intervention).14
Table 1, Chapter 3. Summary tableheparin effectiveness studies
Author, Year
17
Study
Design
Description of PSP
Baird, 2001
Single protocol for heparin administration
Pre-post
Fanikos,
14
2007
Fraipont,
18
2003
Smart pump; drug library with point-of-care

decision support; programmable alert
Nurse-directed weight-based nomogram
Pre-post
Computerized nomogram for acute coronary

syndromes
Pre-post
Oyen, 2005
19
13
Prusch, 2011 Intelligent infusion devices (IIDs), bar-codeassisted medication administration system,
and electronic medication administration
record system
15
Toth, 2002
Weight-based nomogram for transient

ischemic attack and/or stroke
Zimmermann, Weight-based nomogram for acute coronary

16
2003
syndromes
Pre-post
Pre-post
Outcomes: Benefits
Dosing and time to
anticoagulation: No statistics
reported
Anticoagulation medication
errors: No significant differences
Time to therapeutic
anticoagulation, complications:
not significant
Therapeutic anticoagulation
significantly improved,
complications not reported
Telemetry drug library monthly
compliance and manual pump
edits: Statistically significant
improvement
RCT
Total complications,
overanticoagulation, time to
anticoagulation all statistically
significant improvement
Pre-post
Time to anticoagulation
significant; complications not
significant
In terms of evidence grading, the strength of evidence for this topic was low. Risk of bias
was high due to study design issues: Only one study was an RCT.14 Results were inconsistent,
with half of the studies reporting no statistically significant findings; several studies were too
small to measure outcomes meaningfully. Many studies did not report patient safety outcomes,
but instead reported the outcomes for process measures such as time to therapeutic
anticoagulation or compliance with a drug library; many studies that did report complications or
errors did not have sufficient sample size. Finally, regarding precision, a number of different
outcome measures were used, so no conclusions could be made (see Evidence Table on risk of
bias in Appendix D).
What Are the Harms of the Patient Safety Practice?

Neither the original report nor our updated review found studies that reported on harms of the
Patient Safety Practice.
26
How Has the Patient Safety Practice Been Implemented, and in

What Contexts?
The effectiveness studies included older studies on weight-based nomograms and
anticoagulation services and newer studies on intelligent infusion devices and other electronic
medication systems, in various populations and types and sizes of hospitals (e.g., community and
teaching). One United States study from 2000 that evaluated the use of a weight-based
nomogram found that utilization was extremely low at approximately 10 percent. Further,
utilization was not improved after an intervention that included education as well as configuring
the computerized order entry system to allow physicians to choose either the weight-based
nomogram, or traditional heparin ordering.20
Are There Any Data About Costs?

Implementation of heparin nomograms is feasible, although institutions often develop their
own systems rather than adapting existing nomograms. The original report found one study that
concluded that the costs of frequent monitoring were offset by the reduction in the number of
heparin boluses required.6 One of the nomogram studies identified in our update search16 found a
statistically significant decrease in the number of monitoring blood tests required, which would
reduce the costs to manage patient care.
Are There Any Data About the Effect of Context on Effectiveness?

Data regarding the impact of context on effectiveness is limited. The evidence found in the
studies mentioned above could be divided into three categories: leadership, organizational
characteristics, and administration tools.
Two studies commented on the impact of leadership on the effectiveness of the intervention.
Baird and colleagues reported that leadership was important for protocol development.17 Prusch
and colleagues reported that executive sponsorship and oversight as well as the support of the
pharmacy and therapeutics committee were key to effectiveness.13
Regarding organizational characteristics, one study cited the impact of a multidisciplinary
team and a relationship between the hospital and the intelligent diffusion device vendors on the
development of interoperability between systems.13 Another study19 found that a computerized
nomogram provided greater levels of standardization than a paper-based form, since the paperbased form was altered by providers more than 50 percent of the time. Additionally, the
computerized version was able to provide feedback on patient states, which improved patient
monitoring and the evaluation of the nomogram. Therefore, through the implementation of the
computerized nomogram, the heparin dosing protocols and monitoring practices were
standardized.
Three studies mentioned external implementation tools, but no details of how these
implementation tools affected effectiveness (overall effectiveness results are described above in
the section on beneficial effects). Prusch and colleagues used new medication administration
technology developed with frontline nurses and pharmacists. Historical data were analyzed to
ensure the drug library had optimal dosing limits and medications. Finally, the technology was
pilot tested prior to implementation.13 Fanikos and colleagues used the software in the smart
infusion device to establish limits for rates programmed into the.14 Fraipont and colleagues used
the previously developed Raschke nomogram21 in their study.18
27
Finally, we identified one additional study which did not meet our inclusion criteria for
reporting effectiveness data but took a broader human factors approach to improving heparin
safety by improving administration. Harder and colleagues evaluated the human factors
associated with improving the safety of heparin administration.22 After completing interviews
with the staff, the authors offered suggestions for improving the heparin administration process
in order to make the computerized heparin dosing interface more user-friendly (e.g.,
automatically converting English and metric measurements.) Iterative refinements were made to
the system after the initial modifications, and an educational program was rolled out to inform
providers about the new heparin administration process.22
Conclusions and Comment

In conclusion, we found low strength of evidence that patient safety practices, including
nomograms and new intelligent medication administration, dosing, and monitoring technology,
can improve outcomes for the use of intravenous heparin (Table 2). Through our systematic
review, we identified no studies of nomograms published after 2005 and no studies of inpatient
anticoagulation services published since 2000, although both the use of protocols (e.g.,
computerized order entry) and indications for heparin have changed dramatically since that time
including concerns regarding dosing in obese patients.23 Only two studies evaluated new
technology, and no studies evaluated other types of interventions to improve heparin safety.
Study quality was generally low, and many studies had small sample sizes, usually insufficient
for the detection of the impact of interventions on complications of heparin administration. We
did not identify any studies evaluating the harms of these patient safety practices, although there
could be some potential harm from errors caused by misunderstanding of protocols or
miscommunication with anticoagulation services, which could also lead to errors in dosing.
Although the standardization of dosing protocols, accomplished with the input of front-line
personnel, is an important component of increasing safety and has been shown to improve the
effectiveness of heparin administration, few studies have evaluated these protocols and had
sufficient sample size for patient safety outcomes. Significant barriers also exist to implementing
these protocols, and no studies have demonstrated the impact of interventions to increase their
use by health care providers. Only a few, small, low-quality studies evaluated other types of
interventions to improve the safety of inpatient anticoagulation, such as human factors,
anticoagulation services, or new technology, such as computerized order entry or intelligent
infusion devices. Because intravenous anticoagulants are one of the most common sources of
patient harm from safety issues with high-alert drugs, research on interventions to improve their
safety should be a priority. Further study is needed to evaluate the implementation, effectiveness,
and context factors for patient safety practices for intravenous heparin, especially in regards to
use of new technological tools.
Table 2, Chapter 3. Summary table
Scope of the
Strength of
Problem Targeted by Evidence for
the PSP
Effectiveness
(Frequency/Severity) of the PSPs
Common/Moderate
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Low-to-moderate
28
Estimate of
Cost
Low
Implementation Issues:
How Much do We
Know?/How Hard Is it?
Little/Moderate
References
1.
The Institute of Safe Medical Practices.

ISMPs List of High Alert Medications.
2012.
www.ismp.org/tools/highalertmedications.p
df. Accessed January 31, 2012.
2.
Institute for Healthcare Improvement. HighAlert Medication Safety.

www.ihi.org/explore/HighAlertMedicationS
afety/Pages/default.aspx. Accessed
December 13, 2011.
3.
Winterstein AG, Hatton RC, Gonzalez-Rothi

R, Johns TE, Segal R. Identifying clinically
significant preventable adverse drug events
through a hospitals database of adverse
drug reaction reports. Am J Health Syst
Pharm 2002; 59(18):1742-9.
4.
5.
6.
Joint Commission. Preventing Errors related

to commonly used anticoagulants. Issue 41.
September 24, 2008.
www.jointcommission.org/sentinel_event_al
ert_issue_41_preventing_errors_relating_to_
commonly_used_anticoagulants/. Accessed
December 5, 2011.
Santell, JP. Medication Errors Involving
Heparin: Findings from a National
Reporting Program. Ed. Schneider, PJ. In
Improving Heparin Safety. Proceedings
from The Ninth Conference CareFusion
Center for Safety and Clinical Excellence
March 13-14, 2008, San Diego, CA.
www.carefusion.com/safety-clinicalexcellence/conferences/ninth-invitedconference.aspx Accessed October 13, 2011.
Gandhi TK, Shojania KG, Bates DW.
Chapter 9. Protocols for High-Risk Drugs:
Reducing Adverse Drug Events Related to
Anticoagulants. In Making Health Care
Safer: A Critical Analysis of Patient Safety
Practices. Evidence Report/Technology
Assessment No. 43. (Prepared by the
University of California at San Francisco
Stanford Evidence-based Practice Center
under Contract No. 290-97-0013.) AHRQ
Quality. July 2001.
29
7.
Hirsh J, Anand SS, Halperin JL, Fuster V.

AHA Scientific Statement: Guide to
anticoagulant therapy: heparin: a statement
for healthcare professionals from the
American Heart Association. Arterioscler
Thromb Vasc Biol 2001; 21(7):E9-9.
8.
Wells PS, Holbrook AM, Crowther NR,

Hirsh J. Interactions of warfarin with drugs
and food. Ann Intern Med 1994;
121(9):676-83.
9.
Holbrook AM, Pereira JA, Labiris R et al.

Systematic overview of warfarin and its
drug and food interactions. Arch Intern Med
2005; 165(10):1095-106.
10.
Hirsh J, Raschke R. Heparin and lowmolecular-weight heparin: the Seventh

ACCP Conference on Antithrombotic and
Thrombolytic Therapy. Chest 2004; 126(3
Suppl):188S-203S.
11.
Poon EG, Keohane CA, Yoon CS et al.

Effect of bar-code technology on the safety
of medication administration. N Engl J Med
2010; 362(18):1698-707.
12.
Keohane CA, Hayes J, Saniuk C, et al.

Intravenous medication safety and smart
infusion systems: lessons learned and future
opportunities. J Infus Nurs 2005; 28(5):3218.
13.
Prusch AE, Suess TM, Paoletti RD, et al.

Integrating technology to improve
medication administration. Am J Health Syst
Pharm 2011; 68(9):835-42.
14.
Fanikos J, Fiumara K, Baroletti S, et al.

Impact of smart infusion technology on
administration of anticoagulants
(unfractionated Heparin, Argatroban,
Lepirudin, and Bivalirudin). Am J Cardiol
2007; 99(7):1002-5.
15.
Toth C, Voll C. Validation of a weightbased nomogram for the use of intravenous

heparin in transient ischemic attack or
stroke. Stroke 2002; 33(3):670-4.
16.
Zimmermann AT, Jeffries WS, McElroy H,

et al. Utility of a weight-based heparin
nomogram for patients with acute coronary
syndromes. Intern Med J 2003; 33(1-2):1825.
17.
Baird RW. Quality improvement efforts in

the intensive care unit: development of a
new heparin protocol. Proc (Bayl Univ Med
Cent) 2001; 14(3):294-6; discussion 296-8.
18.
Fraipont V, Lambermont B, Moonen M, et

al. Annales Francaises dAnesthesie et de
Reanimation: Comparison of a nursedirected weight-based heparin nomogram
with standard empirical doctor-based
heparin dosage. Ann Fr Anesth Reanim.
2003 Jul;22(7):591-4.
19.
Oyen LJ, Nishimura RA, Ou NN, et al.

Effectiveness of a computerized system for
intravenous heparin administration: using
information technology to improve patient
care and patient safety. Am Heart Hosp J
2005; 3(2):75-81.
20.
Balcezak TJ, Krumholz HM, Getnick GS, et

al. Utilization and effectiveness of a weightbased heparin nomogram at a large
academic medical center. Am J Manag Care
2000; 6(3):329-38.
21.
Raschke RA, Reilly BM, Guidry JR, et al.

The weight-based heparin dosing nomogram
compared with a standard carenomogram.
A randomized controlled trial. Ann Intern
Med 1993; 119(9):874-81.
22.
Harder KA, Bloomfield JR, Sendelbach SE

et al. Improving the Safety of Heparin
Administration by Implementing a Human
Factors Process Analysis. 2005.
23.
Barletta JF, DeYoung JL, McAllen K, et al.

Limitations of a standardized weight-based
nomogram for heparin dosing in patients
with morbid obesity. Surg Obes Relat Dis
2008; 4(6):748-53.
30
Chapter 4. Clinical Pharmacists Role in Preventing Adverse

Drug Events: Brief Update Review
Introduction
In our original report, Making Health Care Safer 2001, Kaushal and Bates noted that over
770,000 people were harmed or died in hospitals annually from adverse drug events (ADE),1-4
with incidence rates in hospital-based studies ranging from 2 to 7 per 100 admissions.1,5-7 In the
outpatient setting, as they also noted, one study on adults estimated the ADE incidence rate at 3
percent.8 The purpose of this review is to update the data on the incidence of ADEs in hospital
settings and to review measures aimed at preventing these events, including the role of the
clinical pharmacist. We searched the literature from 2001 to 2011 and included studies most
relevant to clinical pharmacist interventions on medication errors and adverse drug events in
various health care settings. Our focus was on studies that to some degree addressed the possible
association between clinical pharmacist activities and improved prescribing practices and/or
assessed whether such activities might lead to reduced medication errors and adverse drug
events.
What is the Role of the Clinical Pharmacist in Preventing Adverse

Drug Events?
There have been various patient safety initiatives implemented that involve pharmacists with
the goal of reducing ADEs. These initiatives are often based on the premise that clinical
pharmacists can play an important role in intercepting and acting on possible prescribing errors
and/or recognizing drug-related problems before injury, or further injury, can occur. This
concept has been tested in a variety of settings in a variety of ways.
In the original report, Kaushal and Bates4 noted that in a seminal study by Leape and
colleagues,9 a clinical pharmacist participating in an intensive care unit team led to a
statistically significant 66% decrease in preventable ADEs due to medication ordering. Another
study suggested that ward-based clinical pharmacists may benefit inpatient medication use safety
and quality.10 A single study in a geriatric population found a decrease in medication errors at the
time of inpatient discharge when clinical pharmacists were involved.11 Based on a meta-analysis,
clinical pharmacists were considered to have a modest effect on maintaining acceptable drug
ranges.12 In the ambulatory setting, the authors noted that clinical pharmacists may have positive
impacts on a variety of chronic diseases (hypertension, hypercholesterolemia, chronic heart
failure, and diabetes).13 However, these ambulatory studies had significant limitations and
potential biases, making generalizations problematic.4
At the time of the first review,4 the authors noted that, in two studies, physicians were
receptive to and often acted on clinical pharmacist interventions9,14 attesting to the often
collaborative relationship between the two groups. Overall, Kaushal and Bates concluded that,
Given the other well-documented benefits of clinical pharmacists and the promising results in
the inpatient setting, more focused research documenting the impact of clinical pharmacist
interventions on medication errors and ADEs is warranted.4
31
What Have We Learned About the Role of Clinical Pharmacists?

Recent Reviews and Systematic Evaluations Suggest Clinical
Pharmacists Improve Medication Management
Since the 2001 report, several new systematic reviews, have addressed the role of clinical
pharmacists in different clinical settings. The largest such review was Kaboli and colleagues15
(AMSTAR score 7 positive of 9 relevant domains). This review included studies from 1985 to
2005 that assessed clinical pharmacists interventions in inpatient care. Eligible studies were
those using concurrent controls or time series design, and measuring a number of different
outcomes.
Thirty six studies contributed evidence to the review, including 10 studies of pharmacists
participation on rounds, 11 studies of their participation in medication reconciliation, and 15
studies of drug-specific services (e.g. coumadin, antibiotics). The review was narrative, and
concluded that the evidence supports the use of clinical pharmacists in the inpatient setting to
improve the quality, safety and efficiency of care, although noting that the evidence base is still
limited by small sample size, many studies were conducted at only a single institution, and most
studies have differing measures of outcome.
Three other reviews dealt with clinical pharmacists benefit in the care of elderly adults, in
nursing homes, and pediatric patients.
Hanlon and colleagues16 found a number of benefits for elderly adults, in a variety of
settings, in optimizing prescribing (i.e., improving quality of pharmaceutical care) and reducing
drug-related problems. While there was scant evidence on reducing adverse drug events, they
commented on the difficulty in designing a study that would show ADR reduction, noting that to
detect a 25% decrease in adverse effects, due to a pharmacist intervention, would require
randomizing at least 800 to 1400 elderly patients. This review scored 4 of 9 relevant AMSTAR
domains. In a narrative review of interventions in nursing homes, Marcum and colleagues
included five randomized controlled studies assessing the impact of clinical pharmacists on
various outcomes, including drug-related adverse events; they also included two studies with a
pharmacist or pharmacologist as part of a multidisciplinary approach. While some studies
showed significant differences in the numbers and/or choices of (or changes in) drugs, clinical
outcomes--measured in various ways--were mixed, tending overall to show inconsistent and/or
nominal impacts.17 This review scored 6 of 9 relevant AMSTAR domains. Sanghera and
colleagues18 noted that pharmacists provide important improvements on drug therapy for
children. Many of the 18 studies in the review were older, and methodologies differed (e.g.,
measuring outcomes in various ways, by various designs and definitions), but an overall positive
impact was consistently seen in the studies reviewed. Most of the studies were in the inpatient
setting, and only three were in the outpatient area. Even so, the review highlighted that
pharmacists play a crucial role in detecting and correcting medication errors, such as dosing
mistakes, sometimes potentially lethal ones. The authors concluded, pharmacists reviewing
medication charts is very important in identifying medication-related problems; hence it is likely
to be the most effective factor in improving drug therapy in children. It should be kept in mind
that many of the studies pre-dated the electronic era. This review scored 7 of 9 relevant
AMSTAR domains.
Another review, by Cohen and colleagues,19 included 16 studies of pharmacist activities in
the Emergency Department (AMSTAR score 6 positive of 9 relevant domains). Again noted was
the wide diversity of tasks in which pharmacists were engaged, including (but not limited to)
32
providing drug information, patient counseling, precepting, toxicology case assistance and
various forms of therapeutic consultations, interventions and managements, including medication
error prevention (though included studies were limited in this latter regard).
By and large, these reviews support clinical pharmacist activities in improving medication
management. In general, three issues emerge from the literature. First, clinical pharmacists are
engaged in a multitude of patient level activities, including recognizing, intercepting, and
documenting drug-related problems, as well as assisting in optimizing pharmaceutical choices
for patients and, in some cases, engaging in specific interventions or in specific disease
management practices. Second, it is problematic to accurately capture all that pharmacists do at
either an individual patient level or at an organization level,20 which makes it that much more
difficult to assess their impact, especially since clinical pharmacists do not work in isolation but
rather with other clinicians and, frequently, within hospitals or health care systems or settings.
Third, studies that attempt to show the benefit of pharmacists engaged in various activities from
a larger vantage point (e.g., assessing whether adding a pharmacist to a ward team reduces
medication errors or adverse drug events) often have challenges in their interpretation, including
lack of concurrent control groups, indeterminate definitions of suboptimal prescribing, varying
definitions of medication errors and preventable adverse drug events, different methods of error
and event capture and reporting, and varying clinical outcome assessments. Even so, while
individual studies do not always demonstrate benefits from an organizational perspective, the
body of work suggests that pharmacists provide substantial value to patient care, clinical teams,
institutions, and health care organizations.
Original Studies Not Included in the Systematic Reviews Show that

Interventions With Clinical Pharmacists Tend to Reduce Adverse
Events
As with the systematic reviews we again focused on studies that attempted to address the
relationship between clinical pharmacist activities and improved prescribing and/or a reduction
in adverse events. We identified eight new studies not included in the systematic reviews already
discussed. Of note, many of the more recent studies have had limited success in overcoming
some of those methodological issues seen in some of the older studies. As above, we focused on
studies from the United States and other English speaking countries. The studies are summarized
in Table 1, Chapter 4.
Table 1, Chapter 4. Summary of studies
Study, Year
21
Kaushal, 2008
Population and
Controls
Pediatric ICU or
general ward with
paper charting;
matched units did
not receive
intervention
Intervention
Part or full-time
clinical pharmacist
rounding and
monitoring drug
dispensing, storage,
and administration
33
Outcomes
Measured and
Timing
Medication errors
and adverse events
pre/post, identified
by nurse and
reviewed by 2
blinded physician
reviewers; 6-8 weeks
baseline, 3-month
intervention period
Findings
Full-time clinical
pharmacist
decreased
medication errors
(29 to 6 per 1000
patient days);
increase in
medication errors in
controls; part-time
pharmacists did not
decrease error rate.
Table 1, Chapter 4. Summary of studies (continued)

Study, Year
Wang, 2007
Population and
Controls
Intervention
22
Pediatrics unit of a
community teaching
hospital
Addition of CPOE to
existing clinical
pharmacist system
23
General medical ICU
Inclusion of clinical
pharmacist in
rounding
Clinically important
drug-drug
interactions pre/post
over a 10-week
period
Cardiac ICU
Rounding and
participation in
patient-oriented
activities (e.g., taking
medication histories,
discharge
counseling), and
provider level
activities (e.g., giving
in-service talks to
house staff and
communicating with
physician and
nursing staff)
Pharmacist testing/
recommendations
regarding patients on
antipsychotics who
had movement
disorder complaints
or who were taking
drugs to counter
movement disorders
Medication error
interventions (e.g.,
dose or medication
changes, missing
medications, allergydrug
contraindications)
pre/post over 5 years
Pharmacist-run
education program
on medication orders
and IV fluid review
implemented at
month 4 of 12
months plus other
process changes
Medication errors
pre/post; case
finding by incident
reporting
Rivkin, 2011
24
LaPointe, 2003
25
Stoner, 2000
Simpson, 2004
Outpatient
psychiatric setting
(235 sets of
evaluations in 83
patients on antipsychotics)
26
Neonatal ICU
34
Outcomes
Measured and
Timing
Medication errors,
near misses, and
adverse events over
a 3-month period
Movement disorder
(extrapyramidal)
symptoms
Findings
Clinical pharmacist
intercepted 78% of
111 potentially
serious prescribing
errors but none of 32
harmful
administrative errors
and few of the
transcribing (6/25) or
monitoring errors
(3/7)
Drug interaction
rates decreased
significantly ( 65%)
when compared
retrospectively
(historically) to a 10week period earlier
in the year
Incidence of
medication errors
increased from
around 15 to nearly
24 per 100
admissions, and a
higher trend was
seen during times of
house staff transition
A majority of
recommendations
(82% of 130
evaluations) were
followed by
clinicians; of these,
93% led to a
resolution or
reduction in
extrapyramidal
symptoms
Significant decrease
in medication errors
(from 24 to 5 per
1,000 neonatal
activity days/month);
error rate increased
during summer
staffing change

Study, Year
Population and
Controls
Intervention
Outcomes
Measured and
Timing
Adverse drug
reactions (ADRs)
Bond, 2006
27
584 hospitals
encompassing
>35,000 Medicare
patient stays
Pharmacy staffing
and presence or
absence of various
pharmacy services
Bond, 2007
28
885 U.S. hospitals

with data on 2.8
million Medicare
patients
14 different clinical
pharmacy services
and several staffing
models
Severity-adjusted
mortality rates
Large rural hospital

Emergency
Department
Review of
medication orders
and identification of
errors via
retrospective review
by an independent
reviewer.
Pharmacists also
documented their
interventions.
Medication Errors, 1
month when
pharmacist was not
present to check
medication orders
versus 1 month
when pharmacist (s)
was (were) present;
time periods for
assessment were
one year apart
Brown, 2008
29
35
Findings
Pharmacist
involvement in 8
services (in-service
education, drug
information services,
adverse drug
reaction
management, drug
protocol
management,
cardiopulmonary
resuscitation teams,
medical rounds and
completing
admission drug
histories) as well as
higher staffing rates
decreased ADRs;
however, pharmacist
participation in total
parenteral nutrition
teams increased
ADRs
In-service education,
drug information,
adverse drug
reaction monitoring;
participation in drug
protocol
management,
cardiopulmonary
resuscitation teams
and medical rounds;
and completing
admission drug
histories were
associated with
reduced mortality as
were two staffing
variables
Pre-post analysis
showed significant
decrease (66.6%)
from error rates of
approximately 16 to
5 per one hundred
medications orders

Study, Year
Rothschild, 2010
Cesarz, 2012
31
Population and
Controls
30
Intervention
Four academic
Emergency
Departments
Observational study
in which pharmacy
residents conducted
226 sessions (787
hours) of observing
pharmacist activities;
the study included
over 17,000
medications ordered
or administered to
nearly 6,500 patients
An academic
medical centers 32bed Emergency
Department, serving
pediatric and adult
populations
Prospective
observational study
looking at activities
of four pharmacists
during relevant shifts
in reviewing
discharge
prescriptions. Data
collection was over a
3 week period and
used standardized
forms for reporting
interventions. All
recommendations
were provided to the
ordering physician
who made the
determination to
change a
prescription
Outcomes
Measured and
Timing
Identification of
medication errors at
various stages of
prescribing or
administration by
unblinded,
continuous
observation. Data
collection was via
templated forms.
Captured elements
included errors of
interest, ranging
from those
intercepted before
reaching the patient
to caught after
reaching the patient
but before harm
could occur to
ameliorated adverse
events (collectively
these together were
known as recovered
medication errors) .
Case reviewers
independently
assessed suspected
error interventions.
Self-report of
interventions on
discharge
prescriptions. An
independent
reviewer determined
whether the
intervention was
categorized as error
prevention or
therapeutic
optimization
Findings
Pharmacists
identified over 500
recovered
medication errors,
with an overall rate
of about 3 per 100
medications or about
8 per 100 patients.
Approximately 90%
were intercepted
before reaching the
patient.
Of 674 discharge
prescriptions
reviewed, ED
pharmacists
intervened on about
10%; roughly half of
the 68 interventions
(54%) concerned
error prevention.
A number of the studies contained design flaws that prevented ruling out the contribution of
other process modifications or even secular changes to the observed results. Nevertheless,
overall, these newer studies continue to support the important roles of clinical pharmacists in
reducing prescribing mishaps as well as in improving several patient-level outcomes in various
36
settings. With the exception of one study, studies in which pharmacists participated in a greater
number of clinical processes seemed to show stronger effects.
Clinical Pharmacist Interventions Show Little Potential for Harm

Virtually no study has shown an outright potential for harm, apart from an occasional
isolated finding such as an ADR rate increase with pharmacist participation on total parenteral
nutrition teams (a result that, given its oddity, must remain questionable).27 Theoretically
speaking, as noted in the original report,4 involvement of clinical pharmacists and
implementation of their review processes may result in some delays in dispensing medications.
But if these interventions reduce errors (and/or clarify prescribing), this outcome cannot truly be
considered a harm, though perhaps it is bothersome and time consuming for patients or
providers.
Benefits of Implementation May Outweigh Costs

In terms of resource utilization and costs, the decrease in ADRs that should result from
improved prescribing practices should lead to financial savings and/or mitigations in the costs of
care. However, information in that regard is limited and generally unclear. Of the two primary
studies noted in the 2001 report that estimated annual savings, one based on interventions in an
intensive care unit and another based on pharmacist activities in a large university hospital,
estimated savings ranged from $270,000 to almost $400,000 per year.4,9,32 Because of differences
in outcomes and how they are measured, true costs and/or savings are hard to gauge and, not
surprisingly, vary widely. For example, in a review of economic benefits from hospital-based
interventions by De Rijdt and colleagus,33 financial outcomes, generally stated in estimated
annualized savings, ranged anywhere from less than $10,000 to over $500,000, depending on the
study and the clinical or interim outcome measured as well as the method of financial evaluation
and whether pharmacist costs were included.33 From another perspective, Bond and Raehl28
estimated that the legal settlement costs avoided by the reduction in preventable deaths in the
patient population they studied (Medicare) would be nearly $2.4 billion for hospitals that
incurred adverse events. While cost or savings estimates depend on a set of assumptions as well
as the financial costs of pharmacists time and effort, these widely varying estimations bring
home the point that reduction in medication errors or preventable ADEs can have subsequent
down the line effects and that financial changes may accrue at a variety of levels depending on
the intervention and the seriousness of clinical outcomes (or outcomes avoided).33 A major driver
of the cost-effectiveness of a clinical pharmacist intervention is whether new pharmacists need to
be hired or if the program can be implemented by reallocation of existing resources and/or the
use of lower cost pharmacy technicians for some roles, and thus increase the availability of
clinical pharmacists to directly interact with patients and physicians.

Clinical pharmacists play important roles in a variety of health care settings, and their
activities appear to benefit individual patients as well as health care organizations in a multitude
of ways, many of which are difficult to isolate when studying whether these interventions
objectively lower medication errors or ADEs. Many of the studies are not methodologically
strong, and the literature lacks consistency and comparability. Nevertheless, systematic reviews
and recent evidence generally supports that pharmacist involvement in intensive care units,
particularly when engaging in bedside rounds improves medication management and/or reduces
37
medication errors and preventable ADEs. The existing data for other inpatient and for outpatient
care settings are also supportive of a role for pharmacists but less robust than in intensive care
units. Data from nursing homes are not as clear as for other settings, but, logically speaking,
since medication and prescribing errors occur in this setting, and patients are elderly and more
prone to polypharmacy, it is likely by analogy that drug safety in nursing homes will be
improved by clinical pharmacist interventions. Similarly, evidence from emergency departments
is limited but given the high intensity of care activities and of prescription utilization, it is logical
that benefits will accrue from pharmacist interventions. More and better designed studies should
help determine the magnitude of the benefit(s), to the extent that such benefits exist, in various
health care settings. A summary table is located in Table 2, Chapter 4.
Scope of the
Strength of
the PSP
Effectiveness
Common/Low
Moderate-tohigh
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
High
How Much do We
Little/Moderate
References
1.
Classen DC, Pestotnik SL, Evans RS, et al.

Adverse drug events in hospitalized patients.
Excess length of stay, extra costs, and
attributable mortality. JAMA. 1997 Jan 2229;277(4):301-6. PMID 9002492.
2.
Cullen DJ, Bates DW, Small SD, et al. The

incident reporting system does not detect
adverse drug events: a problem for quality
improvement. Jt Comm J Qual Improv. 1995
Oct;21(10):541-8. PMID 8556111.
3.
Cullen DJ, Sweitzer BJ, Bates DW, et al.

Preventable adverse drug events in
hospitalized patients: a comparative study of
intensive care and general care units. Crit
Care Med. 1997 Aug;25(8):1289-97. PMID
9267940.
4.
Kaushal R, Bates DE. Chapter 7. The

Clinical Pharmacists Role in Preventing
Adverse Events. In Making Health Care
Practices. Evidence Report/Technology
Assessment No. 43. (Prepared by the
University of California at San Francisco
Stanford Evidence-based Practice Center
under Contract No. 290-97-0013.) AHRQ
Quality. July 2001.
38
5.
Bates DW, Boyle DL, Vander Vliet MB, et

al. Relationship between medication errors
and adverse drug events. J Gen Intern Med.
1995 Apr;10(4):199-205. PMID 7790981.
6.
Bates DW, Cullen DJ, Laird N, et al.

Incidence of adverse drug events and
potential adverse drug events. Implications
for prevention. ADE Prevention Study
Group. JAMA. 1995 Jul 5;274(1):29-34.
PMID 7791255.
7.
Jha AK, Kuperman GJ, Teich JM, et al.

Identifying adverse drug events:
development of a computer-based monitor
and comparison with chart review and
stimulated voluntary report. J Am Med
Inform Assoc. 1998 May-Jun;5(3):305-14.
PMID 9609500.
8.
Gandhi TK, Burstin HR, Cook EF, et al.

Drug complications in outpatients. J Gen
Intern Med. 2000 Mar;15(3):149-54. PMID
10718894.
9.
Leape LL, Cullen DJ, Clapp MD, et al.

Pharmacist participation on physician
rounds and adverse drug events in the
intensive care unit. JAMA. 1999 Jul
21;282(3):267-70. PMID 10422996.
10.
Leach RH, Feetam C, Butler D. An

evaluation of a ward pharmacy service. J
Clin Hosp Pharm. 1981 Sep;6(3):173-82.
PMID 7298859.
11.
Lipton HL, Bero LA, Bird JA, et al. The

impact of clinical pharmacists consultations
on physicians geriatric drug prescribing. A
randomized controlled trial. Med Care. 1992
Jul;30(7):646-58. PMID 1614233.
12.
Ried LD, McKenna DA, Horn JR. Metaanalysis of research on the effect of clinical
pharmacokinetics services on therapeutic
drug monitoring. Am J Hosp Pharm. 1989
May;46(5):945-51. PMID 2729302.
13.
14.
15.
16.
17.
18.
19.
Beney J, Bero LA, Bond C. Expanding the

roles of outpatient pharmacists: effects on
health services utilisation, costs, and patient
outcomes. Cochrane Database Syst Rev.
2000(3):CD000336. PMID 10908471.
Hanlon JT, Weinberger M, Samsa GP, et al.
A randomized, controlled trial of a clinical
pharmacist intervention to improve
inappropriate prescribing in elderly
outpatients with polypharmacy. Am J Med.
1996 Apr;100(4):428-37. PMID 8610730.
Kaboli PJ, Hoth AB, McClimon BJ, et al.
Clinical pharmacists and inpatient medical
care: a systematic review. Arch Intern Med.
2006 May 8;166(9):955-64. PMID
16682568.
Hanlon JT, Lindblad CI, Gray SL. Can
clinical pharmacy services have a positive
impact on drug-related problems and health
outcomes in community-based older adults?
Am J Geriatr Pharmacother. 2004
Mar;2(1):3-13. PMID 15555474.
Marcum ZA, Handler SM, Wright R, et al.
Interventions to improve suboptimal
prescribing in nursing homes: A narrative
review. Am J Geriatr Pharmacother. 2010
Jun;8(3):183-200. PMID 20624609.
Sanghera N, Chan PY, Khaki ZF, et al.
Interventions of hospital pharmacists in
improving drug therapy in children: a
systematic literature review. Drug Saf.
2006;29(11):1031-47. PMID 17061909.
Cohen V, Jellinek SP, Hatch A, et al. Effect
of clinical pharmacists on care in the
emergency department: a systematic review.
Am J Health Syst Pharm. 2009 Aug
1;66(15):1353-61. PMID 19635771.
39
20.
Chedoe I, Molendijk HA, Dittrich ST, et al.

Incidence and nature of medication errors in
neonatal intensive care with strategies to
improve safety: a review of the current
literature. Drug Saf. 2007;30(6):503-13.
PMID 17536876.
21.
Kaushal R, Bates DW, Abramson EL, et al.

Unit-based clinical pharmacists prevention
of serious medication errors in pediatric
inpatients. Am J Health Syst Pharm. 2008
Jul 1;65(13):1254-60. PMID 18574016.
22.
Wang JK, Herzog NS, Kaushal R, et al.

Prevention of pediatric medication errors by
hospital pharmacists and the potential
benefit of computerized physician order
entry. Pediatrics. 2007 Jan;119(1):e77-85.
PMID 17200262.
23.
Rivkin A, Yin H. Evaluation of the role of

the critical care pharmacist in identifying
and avoiding or minimizing significant
drug-drug interactions in medical intensive
care patients. J Crit Care. 2011
Feb;26(1):104 e1-6. PMID 20646898.
24.
LaPointe NM, Jollis JG. Medication errors

in hospitalized cardiovascular patients. Arch
Intern Med. 2003 Jun 23;163(12):1461-6.
PMID 12824096.
25.
Stoner SC, Worrel JA, Jones MT, et al.

Pharmacist-designed and -implemented
pharmaceutical care plan for antipsychoticinduced movement disorders.
Pharmacotherapy. 2000 May;20(5):583-8.
PMID 10809346.
26.
Simpson JH, Lynch R, Grant J, et al.

Reducing medication errors in the neonatal
intensive care unit. Arch Dis Child Fetal
Neonatal Ed. 2004 Nov;89(6):F480-2.
PMID 15499135.
27.
Bond CA, Raehl CL. Clinical pharmacy

services, pharmacy staffing, and adverse
drug reactions in United States hospitals.
Pharmacotherapy. 2006 Jun;26(6):735-47.
PMID 16716127.
28.
Bond CA, Raehl CL. Clinical pharmacy

services, pharmacy staffing, and hospital
mortality rates. Pharmacotherapy. 2007
Apr;27(4):481-93. PMID 17381374.
29.
Brown JN, Barnes CL, Beasley B, et al.

Effect of pharmacists on medication errors
in an emergency department. Am J Health
Syst Pharm. 2008 Feb 15;65(4):330-3.
PMID 18238771.
32.
McMullin ST, Hennenfent JA, Ritchie DJ, et

al. A prospective, randomized trial to assess
the cost impact of pharmacist-initiated
interventions. Arch Intern Med. 1999 Oct
25;159(19):2306-9. PMID 10547170.
30.
Rothschild JM, Churchill W, Erickson A, et

al. Medication errors recovered by
emergency department pharmacists. Ann
Emerg Med. 2010 Jun;55(6):513-21. PMID
20005011.
33.
De Rijdt T, Willems L, Simoens S.

Economic effects of clinical pharmacy
interventions: a literature review. Am J
Health Syst Pharm. 2008 Jun
15;65(12):1161-72. PMID 18541687.
31.
Cesarz JL, Steffenhagen AL, Svenson J, et

al. Emergency Department Discharge
Prescription Interventions by Emergency
Medicine Pharmacists. Ann Emerg Med.
2012 May 24 PMID 22633338.
40
Chapter 5. The Joint Commissions Do Not Use List: Brief

Review (NEW)
Introduction
Medication errors stem from a variety of causes, including miscommunication between
prescribers and pharmacists in the form of misunderstood and/or illegible abbreviations. The
potential hazards of certain abbreviations started receiving heightened attention approximately
twenty years ago.1 Most notably, as one of its National Patient Safety Goals, the then named
Joint Commission on Accreditation of Healthcare Organizations (JCAHO, hereinafter referred to
as the Joint Commission for consistency) in 2003 announced that nine abbreviations and/or
shorthand notationsa Do Not Use list--should be banned in its accredited hospitals by April
2004.2,3 The list included the following inappropriate abbreviations: U or u instead of unit;
IU instead of International Unit; Q.D. or similar instead of once daily; Q.O.D or similar
instead of every other day, MS, MSO4 and MgSO4 instead of writing morphine sulfate or
magnesium sulfate; and use of zeros, either when trailing an ordinal number (1.0 instead of 1) or
lack of a zero before a decimal point (.9 instead of 0.9)2,4 (See Figure 1).
Figure 1, Chapter 5. Official do not use list
Figure taken from the JCAHO Web site.4

The Joint Commission: Joint Commission on Accreditation of Healthcare Organization. Facts about the Official Do Not
Use List. 2011. Reprinted with permission.
41
Avoiding potentially hazardous abbreviations was initially intended to pertain to handwritten

documents (e.g., written prescriptions), but the over-riding plan was to extend this stipulation to
all forms of patient-specific communications including printed, electronic or handwritten
materials, with targeted compliance rates of 90% for handwritten and electronic formats and
100% for printed material by 2005.2,8
As part of the initial Joint Commission safety program, health care organizations were to add
three abbreviations to their specific banned list, depending on the type of organization and their
own experiences with abbreviation errors; the Joint Commission provided an additional list of
abbreviations, symbols and acronyms for consideration.4 The Joint Commission is not the only
organization to provide lists or recommendations. The Institute for Safe Medication Practices
provides an even more extensive list for consideration5 and in 2006 began collaborating with the
Food and Drug Administration to reduce hazardous abbreviations.6,7
The magnitude of harm due to abbreviations and other shorthand notations such as acronyms
and symbols is not entirely clear. In a study completed after the Joint Commissions patient
safety goal was disseminated, Brunetti et al., using data from the United States Pharmacopeia
MEDMARX program which in turn uses the National Coordinating Council for Medication
Error Reporting and Prevention Index for Categorizing Medication Errors, found that between
2004 and 2006 a total of 29,974 medication errors out of 643,151 (4.7%) reported to the
MEDMARX program were associated with abbreviations.9 Of those with sufficient information
to ascertain a description of the error (n = 18,153), about 43% were due to using the term QD
(once daily). In addition, roughly 13% involved the abbreviation U (units), and approximately
13% cc (milliliter); nearly 10% used MSO4 or MS (morphine sulfate), and 3% HS (at
bedtime); almost 4% were attributed to decimal errors (e.g., no leading zero or a trailing zero).
Of the errors assessed, 0.3% led to patient harm, and most of those involved the abbreviation
U in some manner.
Most errors (81%) occurred during prescribing; not surprisingly, medical staff were
responsible for roughly 79% of abbreviation errors. Abbreviation use varied among staff groups,
with physicians often using sc, hs and cc. While the study was limited by the constraints
of voluntary reporting, the data suggest that relatively few abbreviations and notations are
responsible for perhaps 5% of related medication errorsand this number may well be larger
since not all errors are likely to be reported.
The purpose of this narrative literature review is to understand the degree to which health
care organizations have succeeded in implementing procedures to prevent inappropriate
abbreviations, and to identify which method(s) work well. We searched PubMed in October
2011 using major heading search terms abbreviation and safe or unsafe or adverse or harm for
English language articles published starting in the year 2000. Titles and abstracts were retrieved,
and relevant articles were retained for review. We expanded the search by using Google to
search for possibly pertinent articles and links; we identified additional articles by looking at
cited references from various publications. We focused on United States-based studies. Clinical
trials, observational studies, reviews, and anecdotal reports on implementation were our primary
resources and given priority in the order above.
What Are the Procedures for Reducing Prescribing Errors?

As Kuhn (2007) noted, there are three primary methods for addressing the safety issues
posed by abbreviations: education, enforcement and leadership.8 In addition, the advent of
electronic prescribing with clinical decision support may impact on abbreviation use.
42
Unfortunately, in all of these areas, the relevant United States literature is sparse, and
implementation efforts have had mixed results.
How Effective Have These Procedures Been?

Educating Providers to Reduce Potentially Unsafe Abbreviations. Abushaiqa et al. studied
the strategy to decrease six specified unsafe abbreviations (unit instead of U; microgram instead
of g; 3 times a week for TIW; avoiding the degree symbol for hour, and avoiding trailing zeros
and lack of leading zeros).3 The setting was a 340-bed hospital in Detroit. Educational materials
included pocket cards, chart dividers in patient charts, and traffic sign look-alike stickers.
Providers were sent memorandums and electronic mail. In-service programs were also
completed: prescribers using banned abbreviations or symbols were asked to clarify their orders
and received instruction on why to avoid banned abbreviations.
The evaluation period, including a baseline assessment, lasted from September 2003 to April
2004. Unsafe abbreviations dropped from about 20% in the pre-intervention phase to about 3%
by the end of the intervention period, with a total of over 20,000 orders reviewed. Sustainability
of the program was not addressed, but the authors noted that in April 2004 the facility started
utilizing the Joint Commissions Do Not Use list and in July 2004 the hospital no longer
accepted orders with unsafe abbreviations.3
On the other hand, Garbutt et al. focused on 20 safe prescribing behaviors using a multifaceted educational intervention at an urban teaching hospital in St Louis. The prescribing errors
included dangerous abbreviations such as potential dosing errors (e.g., trailing zeros, leading
zeros) and frequency measures (e.g., QD, QOD, TIW, HS). The intervention program included
an academic component (e.g., grand rounds or lecture format) as well as reminders and prompts
to emphasize desired prescribing practices. Overall, prescribing errors for surgical house staff
declined but paradoxically increased for medical house staff. Notably, neither group decreased
use of potentially hazardous abbreviations.10
Leonhardt and Botticelli studied an effort in Milwaukee, in 2003 to 2004, involving seven
independent health care organizations.11 The safety collaborative included local hospitals that
partnered with the local business community as well as retail pharmacies. The goal was to
completely eliminate nine abbreviations/shorthand notations from hospital medication orders and
five abbreviations/shorthand notations from outpatient prescriptions (including abbreviations
associated with units, once daily, every other day, trailing zeros and lack of leading zeros).
Interventions and strategies included banning the prohibited abbreviations, educational programs
(at various times during the intervention period) and providing informational materials (e.g.,
printed documents, wallet cards, posters); in addition, there was feedback to physicians who
continued to use banned abbreviations. In outpatient clinics the intervention was passive
education (i.e., newsletters).
The program improved prescribing for hospital-based medication orders but not for
outpatient-based prescriptions. More specifically, appropriate documentation (i.e., no banned
abbreviations or notations) rates, evaluated at thirteen hospitals, increased from approximately
62% at baseline to about 81% after the intervention (P < 0.0001). For clinic-based prescriptions,
evaluated at nine retail pharmacies, rates of appropriate prescriptions increased a non-significant
amount, from about 69% to 73% (P = 0.11).11
43
Leadership and Enforcement Effects on Abbreviation Use

We found no formal studies that isolated enforcement and/or leadership efforts, although the
Abushaiqa study clearly included some enforcement. There were some anecdotal success stories,
mostly after lack of success with educational programs. For example, at Childrens Hospitals and
Clinics in Minneapolis, prescribers were mandated to re-write orders with prohibited
abbreviations; no details were provided on the magnitude of the effect(s). Another hospital in
Tennessee contacted providers to ask for clarification of orders with designated abbreviations,
and a medical staff chairperson discussed abbreviations with individual prescribers identified as
using such; abbreviations in medication orders reportedly declined from around 30% to 6%. An
Ohio hospital retrospectively routed prescriptions that contained designated abbreviations
(apparently after filling the prescription) back to prescribers with feedback that the order had an
unacceptable abbreviation(s). This program reportedly had no noticeable decrease in
abbreviation use.12
Impact of Electronic Prescribing on Hazardous Abbreviations

Electronic prescribing provides a ready venue for focusing on abbreviation misuse. First,
electronic prescribing eliminates illegible handwriting. Second, clinical decision support may be
configured to prompt providers to avoid abbreviations and/or to auto-correct or translate
abbreviations to preferred terms (e.g., using Q.O.D. would yield every other day on the
prescription). However, there are limited data on how using electronic prescribing affects
abbreviation use.
In a small study of faculty providers practicing in an outpatient setting, Galt et al. conducted
a prospective, randomized controlled trial looking at how a personal digital assistant (PDA)
affected prescribing by 78 office-based primary care physicians.13 Practices were randomized to
either usual handwritten prescribing or to entering prescriptions using a PDA-based clinical drug
application. However, intervention offices could, when desired, use handwritten prescriptions.
Duplicate prescriptions were gathered by printing an extra electronic prescription or by using
carbon copies of written ones. The analysis compared the intervention group pre and post PDA
usethat is, during the period when handwritten prescriptions were used, and then during the
PDA use period, when physicians entered 43% of prescriptions via electronic means.
The study found that illegibility decreased from about 9% to 3% (though not to zero since not
all prescriptions were via PDA) and, among other errors, various abbreviations and shorthand
methods fell numerically (P-values not provided) including abbreviations for drug name (from
about 3% to 2% of errors), administration route (from about 63% to 37%), frequency (from
roughly 86% to 51%), and symbols on the prescription (from about 77% to 47%). In both time
periods, issues with zeros were relatively rare (< 1%); interestingly dosing abbreviations rose
from 61% to approximately 71%, as some of these were allowed in the application.13
Devine et al. studied the impact of a basic computerized provider order entry program in a
multispecialty clinic system in Washington State. Using a pre/post study design, evaluating
handwritten (pre-intervention) prescriptions from January to March to 2004 and electronic
prescriptions (post-intervention) from July 2005 to April 2006 at three retail pharmacies, they
found that illegible prescriptions decreased from just under 3% to less than 0.1% and
inappropriate abbreviations fell from around 5% to 0.4%.14
In a small prospective study of faculty providers practicing in an outpatient setting,
Abramson et al. found that reducing abbreviation error rates was the primary driver in reducing
overall prescribing errors when transitioning from an older to a newer electronic prescribing
44
system. The older, locally derived system had automatic conversion of inappropriate
abbreviations installed on some computers; it also allowed for free text entries on the ordering
template. It had minimal clinical decision support and did not send prescriptions directly to
pharmacies. The newer system had a commercially available clinical decision support package,
but did not auto-correct abbreviations. The system was able to send prescriptions to pharmacies.
The newer system included two alerts to providers when they entered and completed a
prescription containing an inappropriate abbreviation. In this yearlong study, data were available
on seventeen physicians in the academically affiliated clinic. Rates of inappropriate
abbreviations (per 100 prescriptions) fell from about 24 at baseline to just under 11 at 6 months
and then to approximately 6 at 1 year after implementation (p-values < 0.001). Interestingly,
non-abbreviation error rates rose at 12 weeks, but were similar at one year postimplementation.15
What Have We Learned About Procedures for Reducing

Prescribing Errors?
The U.S. literature on programs designed to reduce prescribing errors is sparse. Studies that
assessed the success of programs to educate providers report mixed results. We found no studies
that focused specifically on enforcement or leadership, but anecdotal reports are also mixed. No
studies address sustainability.
Electronic prescribing systems may hold promise. However the data on avoiding
abbreviations are limited, and it is not clear which technology or technologies will work best for
reducing shorthand methods of prescribing.

Abbreviations and other shorthand notations on prescriptions and orders increase the risk of
medication errors, and the majority of errors and subsequent harms are caused by relatively few
abbreviations or notations, and more specifically, QD (once daily), U (units), cc
(milliliter); MSO4 or MS (morphine sulfate), and HS (at bedtime); in addition, decimal errors
(e.g., no leading zero or a trailing zero) are also troublesome. Various organizations, most
notably the Joint Commission in the form of its Do Not Use list, have taken a strong stand
against using certain abbreviations. However, the available literature on various implementation
efforts is limited, and no clear route to success has been described. Moreover, we found no
studies that address sustainability of efforts and no studies on whether reducing abbreviations
leads to less patient harms, though logically this would seem to be the case.
All in all, abbreviations can lead to misunderstandings and miscommunications between the
prescribers and the pharmacists and in turn may lead to incorrect prescriptions being given to
patients. Most errors are caused by relatively few abbreviations. Harms from such errors are
uncommon but preventable. Although it is not clear how the Joint Commissions Do Not Use
List (or any other list of hazardous abbreviations) can best be implemented across the spectrum
of U.S. health care organizations it is important to note that there is no obvious patient harm to
implementing such a list and data, to the extent that it exists, suggests that avoiding certain
heightens prescribing safety. The cost and burden of implementation will depend on the
stringency and/or comprehensiveness of the method(s) used. For example, electronic prescribing
and decision support tools may offer the best chance of successfully reducing abbreviations on
the Do Not Use list. However, it will take some time before prescribers are universally using
45
these systems and the cost and effort is not insubstantial to newly utilizing electronic prescribing.
Another alternative would be enforcing a zero tolerance policy on handwritten prescriptions and
medication orders. However, this might create a substantial burden for prescribers and
pharmacists, particularly in the outpatient and retail pharmacy areas, not to mention mail out
facilities. In the meantime, a low-cost approach of implementation, such as through ongoing
education and/or feedback, focused on avoiding selected harmful abbreviations whenever and
wherever possible seems reasonable and feasible. A summary table is located at Table 1, Chapter
5.
Scope of the
Strength of
the PSP
Effectiveness
Common/Low
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Low
How Much do We
Little/Probably not difficult
References
1.
Cohen MR, Davis NM. Avoid dangerous Rx

abbreviations. Am Pharm.
1992;NS32(2):20-1.1546626.
2.
Thompson CA. JCAHO issues do-not-use

list of dangerous abbreviations. Am J Health
Syst Pharm. 2003;60(24):2540-2.14735769.
3.
Abushaiqa ME, Zaran FK, Bach DS, et al.

Educational interventions to reduce use of
unsafe abbreviations. Am J Health Syst
Pharm. 2007;64(11):1170-3.17519459.
4.
Joint Commission on Accreditation of

Healthcare Organizations. Facts about the
Official Do Not Use List. June 2011
[cited;
www.jointcommission.org/assets/1/18/Offici
al_Do_Not_Use_List_6_111.PDF.
5.
6.
7.
Institute for Safe Medication Practices.

ISMPs List of Error-Prone Abbreviations,
Symbols, and Dose Designations. 2011
[cited;
www.ismp.org/tools/errorproneabbreviation
s.pdf.
U.S. Food and Drug Administration. FDA
and ISMP Launch Campaign to Reduce
Medication Mistakes Caused by Unclear
Medical Abbreviations. June 14, 2006.
www.fda.gov/NewsEvents/Newsroom/Press
Announcements/2006/ucm108671.htm.
Baker D. Campaign to Eliminate Use of
Error-Prone Abbreviations. Hospital
Pharmacy. 2006;41(9):809-10.
46
8.
Kuhn IF. Abbreviations and acronyms in

healthcare: when shorter isnt sweeter.
Pediatr Nurs. 2007;33(5):392-8.18041327.
9.
Brunetti L. Abbreviations formally linked to

medication errors. Healthcare Benchmarks
Qual Improv. 2007;14(11):126-8.17966233.
10.
Garbutt J, Milligan PE, McNaughton C, et

al. Reducing medication prescribing errors
in a teaching hospital. Jt Comm J Qual
Patient Saf. 2008;34(9):528-36.18792657.
11.
Leonhardt KK, Botticelli J. Effectiveness of

a Community Collaborative for Eliminating
the Use of High-risk Abbreviations Written
by Physicians. Journal of Patient Safety.
2006;2:147-53.
12.
Traynor K. Enforcement outdoes education

at eliminating unsafe abbreviations. Am J
Health Syst Pharm. 2004;61(13):13145.15287220.
13.
Galt KA, Rule AM, Taylor W, et al. The

Impact of Personal Digital Assistant Devices
on Medication Safety in Primary Care
Implementation Issues. Advances in Patient
Safety 2005;3:247-63.21249997.
14.
Devine EB, Hansen RN, Wilson-Norton JL,

et al. The impact of computerized provider
order entry on medication errors in a
multispecialty group practice. J Am Med
Inform Assoc. 2010;17(1):78-84.20064806.
15.
Abramson EL, Malhotra S, Fischer K, et al.

Transitioning between electronic health
records: effects on ambulatory prescribing
safety. J Gen Intern Med. 2011;26(8):86874.21499828.
47
Chapter 6. Smart Pumps and Other Protocols for Infusion

Pumps: Brief Review (NEW)
Introduction
Medication errors represent a serious issue affecting the U.S. health care system, accounting
for the largest category of patient safety incidents within the larger category of medical errors.
One report estimated that at least 1.5 million preventable medication errors occur in the U.S.
each year.1 A list of high-alert medications (those with the highest potential for patient harm if
used in error) published by the Institute for Safe Medication Practices (ISMP) includes several
medications delivered by intravenous (IV) infusion (e.g., insulin, propofol, heparin).2
Because IV delivery is more rapid and leads to higher systemic concentrations of drugs
compared with other delivery methods, adverse drug effects tend to be more rapid and severe
when associated with IV infusion. Because traditional infusion pumps are typically programmed
in milliliters per hour (mL/hr) and volume-to-be-infused (VTBI) in mL, they are particularly
vulnerable to errors in drug administration and monitoring.1 Such errors include administration
of the wrong dose or the wrong drug as well as erroneous infusion to the wrong patient.
What Are the Practices for Reducing IV Medication Errors?

To address the shortcomings of infusion pumps, manufacturers have added technology to
recent models of general-purpose (large volume),3 syringe,4 and patient-controlled analgesia
(PCA) pumps5 specifically designed to prevent medication errors. Smart pumps include a
software program (also referred to as a dose error reduction system [DERS]) that provides a
customized drug library alerting users to predetermined minimum and maximum dose limits for
each drug.
The program provides soft alerts (also known as soft stops) that prompt users to reconsider a
given drug dosage but allow them to administer that dosage if they choose, as well as hard alerts
(or hard stops) that prevent users from going beyond the stated dose limits.1 These systems
permit the development of dosing limits for continuous and bolus deliveries, as well as clinical
advisories (point of care notifications) and area-wide default settings for alarm thresholds.
In addition, some smart pumps have incorporated barcode technology that allows verification
of patient identity, thereby preventing delivery of the wrong drug or delivery to the wrong
patient.6,7 One PCA pump offers an integrated bar code scanner for automatically locating the
correct drug entity (e.g., drug name and concentration), and a handful of hospitals have created
interfaces between their general purpose pump servers, barcode-enabled point of care (BPOC)
systems, and documentation systems to make sure that the pump is programmed according to the
medication order and that administration is automatically documented.6
Unlike traditional infusion pumps, smart pumps can alert health care workers when they have
selected inappropriate dosages for a given drug. Soft alerts have the shortcoming that they are
merely reminders that can be overridden by the user although overrides are captured in a DERS
log and can frequently be associated with a user. Hard alerts have the potential to be more
effective because they do not allow easy circumvention, although they can still be circumvented
by determined users (e.g., by bypassing the drug library and entering the infusion rate and
48
volume manually).8 A significant drawback is that inappropriately programmed hard alerts may
impede delivery of care, and circumvention of hard alerts can lead to serious errors.9
Smart pumps with DERS plus BPOC can additionally prevent drug delivery to the wrong
patients.10,11 As long as users comply with such alerts and prompts, smart pumps have the
potential to reduce the number of infusion errors. Compliance with safety features can be
improved by programming prompts that increase ease of use, and by emphasizing a culture of
safety within the organization. Smart pumps also contain a data log that can be used to identify
programming errors or show that the pump prevented adverse events.9
However, the basic limitation of smart pumps is that they can correct only errors of
administration; other types of medical errors can occur during ordering or prescribing,
dispensing, transcribing, and monitoring of patient response.10 For this reason, smart pumps
function best not as standalone devices but when integrated into a larger medication safety
system that connects them with computerized provider order entry (CPOE), BPOC, and
electronic medication administration records (eMARs).9 Such interconnected systems can target
not only errors of administration but also errors of ordering, dispensing, and transcription.10
How Have These Practices Been Implemented?

A recent systematic review by Hertzel and Sousa (2009) identified nine studies published
from 2003 to 2008 that assessed the use of smart pumps for prevention of medication errors. The
majority of studies evaluated smart pumps with soft alerts. The review summarized the study
findings and identified lack of user compliance with soft alerts as an important factor that
compromised the efficacy of smart pumps in the majority of studies. The authors concluded that
well-designed research is still lacking with respect to the effectiveness of smart pumps in
preventing medication errors.1 The most relevant studies mentioned in this review are
summarized in more detail below, along with more recent studies published subsequent to the
reviews publication date.
Smart Pumps With Soft Alerts

Nuckols et al. (2007) performed a retrospective review of 4,604 critically ill patients in ICUs
at two hospitals to determine how often preventable IV adverse drug events (ADEs) matched
smart pump safety features. These consisted of drug libraries with dose limits that triggered soft
alerts, which could be addressed or overridden. The study evaluated ADEs before and after smart
pump implementation. Of 100 preventable ADEs, only four (two before and two after smart
pump implementation) matched the safety features of smart pumps.12
Rothschild et al. (2005) performed a prospective time series study of smart pumps with
intervention (decision support on) and control (decision support off) periods to determine the
impact of integrated decision support on the incidence of medication errors and adverse drug
events in 735 cardiac surgery patients. Preventable adverse events (11 intervention, 14 control)
and non-intercepted potential adverse events (82 intervention, 73 control) did not differ
significantly between groups. Serious medication error rates were 2.41 and 2.03 per 100 patientpump days in the intervention and control periods, respectively (P = 0.124). Caregivers violated
infusion practice 25% of the time (571 infusions) by bypassing the drug library during the
intervention periods. Medications were administered without physician documentation 7.7% of
the time (intervention and control periods combined). The smart pumps were not programmed to
give hard alerts, which cannot be easily overridden; therefore, it was easy for caregivers to
override alerts or bypass the drug library. Poor caregiver compliance with the drug library and
49
dosage limits may have explained the lack of advantage of smart pump decision support in this
study.13 This study used an early version of smart pump technology that was opt-in rather than
opt-out, which made it easier for users to skip the library rather than look for it.
Larsen et al. (2005) performed a retrospective before-after study in pediatric patients that
compared medication infusion errors 12 months before and 12 months after adopting a new
protocol using a combination of smart pumps, standard drug concentrations, and humanengineered (user-friendly) medication labels. The smart pumps included a modifiable drug
library and provided soft alerts to users who attempted to use doses that exceeded the safety
limits. The infusion error rate dropped from 3.1 to 0.8 per 1000 doses from the pre-intervention
to the post-intervention period, a risk reduction of 2.3 (95% CI 1.1-3.4, P <0.001).14 However,
since this was a combination of three interventions, it is unclear what percentage of the error
reduction can be attributed to smart pumps alone. Data were obtained from the hospital-wideincident-reporting system, which tends to underreport errors, but the reported pre- and postintervention error rates should be representative of the relative number of errors.14
Adachi and Lodolce (2005) conducted a retrospective before-after study (one year preintervention, one year post-intervention) to determine whether a new intervention (revised
standard order sets and smart pumps with soft alerts) could reduce IV dosing and administration
errors. Although they found that only a small reduction occurred in overall dosing errors (59 to
46), a larger reduction occurred in pump-related errors (24 to 10, or from 41% to 22% of dosing
errors). Standard concentrations eliminated errors related to the wrong drug concentration. Nine
out of the 10 post-intervention pump programming errors occurred because users did not use the
pump software.15
Three uncontrolled studies illustrate compliance issues associated with smart pump soft
alerts. Eckel et al. (2006) reported a high frequency of programmings (44.4%) due to users
bypassing the drug library when selecting a drug. Furthermore, users overrode 88.5% of soft
alerts.16 Fields and Peterman (2005) reported 506 medication errors due to users overriding soft
alerts.17 However, a third study (Breland 2010) reported that a community hospital was able to
improve compliance with pump alerts from 33% (when smart pumps were first introduced) to
97% three years later.18
Smart Pumps With Soft and Hard Alerts

Schilling and Sandoval (2011) performed a retrospective before-after study (4 months preand 4 months post-intervention) of smart pumps with soft and hard alerts in a community
hospital setting. Use of rescue medications and heparin infusions decreased substantially from
pre- to post-intervention, and length of stay in patients receiving antimicrobial agents also
decreased substantially. Regarding dosage alerts, 86.2% were soft alerts and 13.8% hard alerts.
About 61% of soft alerts were overridden by users and 39.% were modified to comply with
accepted rates; users complied with every hard alert.19
Fanikos et al. (2007) conducted a retrospective before-after study evaluating the impact of a
smart pump with soft and hard alerts in an academic medical center. After reviewing
anticoagulation errors in 3,674 patients, the authors found no significant decrease in errors postintervention (49 pre vs. 48 post). This lack of difference may reflect the fact that only a relative
minority of events were infusion-related errors (19/97 total events). Infusion errors were
substantially higher in the period prior to smart pump implementation (15 errors) compared with
the post-intervention period (4 errors).20
50
Smart Pumps With Soft and Hard Alerts Plus Barcode Technology
Trbovich et al. (2010) conducted a simulation study comparing nurses ability to avoid
medication errors using a traditional pump, a smart pump, and a pump with an integrated bar
code scanner (the latter two had soft and hard alerts). The study was conducted in a laboratory
setting using patient mannequins with bar-coded wristbands and medication bags with bar-coded
labels containing patient ID; errors were assessed by type. Wrong drug errors did not differ
significantly by pump type. Patient ID errors were remedied by significantly more nurses using
pumps with barcode scanners (88%) than with the smart pumps without barcode scanners (58%)
or traditional pumps (46%). Significantly more nurses remedied critical overdose errors when
using pumps with barcode scanners (79%) and smart pumps without barcode scanners (75%) due
to hard alerts than with traditional pumps (38%). Wrong dose soft alerts did not result in
significant differences in fixing overdose errors among different pumps (errors remedied by 75%
of nurses using pumps with barcode scanners, 63% with smart pumps without barcode scanners,
and 50% with traditional pumps). This was because many nurses overrode soft alerts.7 While this
study provides perspectives on error rates, it does not faithfully simulate a clinical environment:
auto-programming in a clinical setting is limited at this time but is typically accomplished
through interfaces with BPOC systems instead of through printing medication labels with patient
ID.
Smart Pumps With Soft and Hard Alerts Integrated With Barcode
Technology and eMARs
Prusch et al. (2011) conducted a prospective before-after study evaluating a program
integrating intelligent infusion devices (IIDs) with a BPOC system and an eMAR system.21
Monthly compliance with the telemetry drug library increased from 56.5% pre to 72.1% post
intervention (p<0.001) and the number of telemetry manual pump edits decreased (56.9 to 14.7;
p<0.001). Pump programming errors related to i.v. unfractionated heparin occurred at a rate of
16.9 events/10,000 opportunities pre-implementation and 11.3 events /10,000 opportunities postimplementation, but the rate decrease was not statistically significant (P = 0.17). However, smart
pumps were used before and after the implementation period, the only difference being that the
smart pumps became fully integrated with BPOC and eMAR in the post-implementation period.
Therefore, the true impact of smart pumps on infusion error rates is unclear from this study.
None of the studies described above identified harms to patients that could be attributed
specifically to the use of smart pumps in place of traditional infusion pumps.
What Have We Learned About These Practices?

Implementation of smart pump technology by health systems and hospitals generally requires
considerable planning, including identification of stakeholders, evaluation of software
capabilities, evaluation of hospital-specific practices, decisions regarding standard operating
systems and procedures, building of drug libraries, and education of staff before the pumps can
be deployed.22 Successful implementation usually involves multidisciplinary teams that include
pharmacists, nurses, and physicians. With minor variations, this overall process has been
described in several published studies.17,18,23,24
In their guidelines for safe implementation and use of smart infusion pumps, ISMP identifies
several key steps necessary for implementation. These include:
51
Ownership of the process at the executive level (assessment of culture and budget
resources, forming a multidisciplinary team, performing a Failure Mode and Effects
Analysis [FMEA] to identify barriers to compliance)
Technological readiness (ensure that information technology [IT] systems can interface
with pumps and that IT staff levels are sufficient, update drug libraries and download
medication safety information efficiently [preferably via a wireless network], consider
wireless network communication upgrade if it is unavailable prior to smart pump
implementation)
Physical environment and equipment (ensure sufficient number of pumps, policies for
cleaning, storage, and distribution, short-term pump rental from outside vendors [if
necessary], ensure rental pumps are programmed with the renting facilitys drug library
and dose limits, ensure sufficient number of electrical outlets for pump operation in
patient areas and for recharging internal batteries when not in use)
Staff education (plan for several weeks of staff education, train super-users, ensure
ongoing education, explain purpose of and procedures for soft and hard stops, inform
staff about drug library updates, develop champions in each clinical area devoted to
safety culture, do smart-pump simulation exercises, emphasize benefits of smart pump
technology)
Specialized patient care areas (make plans to address needs of specific therapies or
patient care areas such as pediatrics/nursery, pain management, operating room,
oncology, emergency department, and patient transport)
Vendor support (to help define implementation timetable, provide sample drug libraries,
online tutorials, live telephone assistance, post-implementation follow-up visits,
assistance in data evaluation, and external support groups)
Rollout (prioritize sequence of patient care areas receiving pumps, select areas with
adequate staff and resources, select educators and champions from pilot units, vendor
support should be available, evaluate rollout process)8
Creation of safe and effective customized drug libraries is essential for proper utilization of
smart pumps. Institutions must evaluate their clinical practice when determining what drugs and
dosage limits to select for their library. Drug libraries should at least include all high-alert drugs
with standard concentrations as well as soft and hard stops for various dosage limits. Once drug
libraries have been developed, considerable time must also be devoted to maintaining and
updating the libraries. Wireless communication technology in an organizations infrastructure
allows easier adjustment or updating of drug libraries, which otherwise would require manually
updating each pump separately.8
Breland (2010) reported that a community hospital was able to increase compliance rates
with pump alerts from 33% at baseline (when smart pumps were first introduced) to 97% three
years later. This was done by having nursing directors and managers stress the importance of the
safety software and how it could improve patient safety. Compliance data were shared with staff
nurses and unannounced twice-weekly inspections were performed by pharmacy to determine
why safety software was not being used in individual cases. Continual reeducation and
customization of drug libraries for the needs of specific critical care areas (CCAs) also helped to
improve compliance. Compliance rates for individual CCAs were distributed to nursing
directors, who also emphasized to the staff the legal liability entailed in noncompliance. In
addition, a review of edits and overrides led to a drug library revision to eliminate unnecessary
52
alerts by changing some dosage limits to reflect actual dosing practices (which were determined
to be safe).18

The evidence supporting efficacy of smart pumps for prevention of medical errors is limited
by the relatively small number of studies and the use of observational study designs with
inherent susceptibility to bias (Table 1). In addition, most published studies have evaluated only
smart pumps with soft alerts; study findings are somewhat variable, ranging from suggesting no
effect to a limited effect of soft alerts in reducing the rate of medical errors. This appears to be
partly due to user compliance, which although somewhat variable among different institutions, is
usually low because users can easily override soft alerts. Hard alerts and barcode technology
should theoretically have more impact on error rates, but too few studies have evaluated these
features to judge their relative effectiveness. Smart pumps have the most potential to reduce
medication errors when integrated into a larger medication safety system that connects them with
CPOE, BPOC, and eMARs.
Implementation of smart pump technology by health systems and hospitals generally requires
considerable planning, including identification of stakeholders, evaluation of software
capabilities, evaluation of hospital-specific practices, decisions regarding standard operating
systems and procedures, building of drug libraries, and education of staff before the pumps can
be deployed. Successful implementation usually involves multidisciplinary teams that include
pharmacists, nurses, and physicians. Once drug libraries have been developed, considerable time
must also be devoted to maintaining and updating the libraries. Wireless communication
technology in an organizations infrastructure allows easier adjustment or updating of drug
libraries, which otherwise would require manually updating each pump separately.
Scope of the
Strength of
the PSP
Effectiveness
Common/Low
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate
References
1.
Hertzel C, Sousa VD. The use of smart

pumps for preventing medication errors. J
Infus Nurs 2009 Sep-Oct;32(5):257-67.
PMID: 20038875.
4.
ECRI Institute. Syringe infusion pumps with

dose error reduction systems. Health
Devices 2008 Feb;37(2):33-51. PMID:
18773865.
2.

ISMPs list of high-alert medications.
Horsham (PA): Institute for Safe Medication
Practices; 2011. 1 p.
5.
ECRI Institute. Infusion pumps, patientcontrolled analgesic. Plymouth Meeting

(PA): ECRI Institute; 2011 Sep. 41 p.
(Healthcare Product Comparison System).
3.
ECRI Institute. Large-volume infusion

pumps. The evolution continues. Health
Devices 2009 Dec;38(12):402-10. PMID:
21243910.
6.
ECRI Institute. Healthcare risk control Risk

Analysis. Supplement A, Pharmacy and
Medications 7. Plymouth Meeting (PA):
ECRI Institute; 2004 Jul. Bar-coded
medication administration systems.
53
7.
Trbovich PL, Pinkney S, Cafazzo JA, et al.

The impact of traditional and smart pump
infusion technology on nurse medication
administration performance in a simulated
inpatient unit. Qual Saf Health Care 2010
Oct;19(5):430-4. PMID: 20427310.
8.

Proceedings from the ISMP summit on the
use of smart infusion pumps: guidelines for
safe implementation and use. Horsham
(PA): Institute for Safe Medication
Practices; 2009. 19 p.
9.
10.
11.
12.
13.
14.
ECRI Institute. Healthcare Risk Control.

Volume 3, Medical technology 15.
Plymouth Meeting (PA): ECRI Institute;
2011 Jul. Executive Summary: Infusion
pumps. p. 1-9.
Rothschild JM, Keohane C. The role of bar
coding and smart pumps in safety. In:
AHRQ WebM&M [database online].
Rockville (MD): Agency for Healthcare
Research and Quality (AHRQ); 2008 Sep
[accessed 2011 Nov 22]. [4 p].
http://webmm.ahrq.gov/perspective.aspx?pe
rspectiveID=64.
Vanderveen T. IVs first: a new barcode
implementation strategy. In: Patient Safety
& Quality Healthcare - PSQH e-newsletter
[database online]. Marietta (GA): Lionheart
Publishing, Inc.; 2006 May-Jun [accessed
2011 Nov 22]. [9 p].
www.psqh.com/mayjun06/ivs.html.
Nuckols TK, Bower AG, Paddock SM, et al.
Programmable infusion pumps in ICUs: an
analysis of corresponding adverse drug
events. J Gen Intern Med 2008 Jan;23 Suppl
1:41-5. PMID: 18095043.
Rothschild JM, Keohane CA, Cook EF, et
al. A controlled trial of smart infusion
pumps to improve medication safety in
critically ill patients. Crit Care Med 2005
Mar;33(3):533-40. PMID: 15753744.
Larsen GY, Parker HB, Cash J, et al.
Standard drug concentrations and smartpump technology reduce continuousmedication-infusion errors in pediatric
patients. Pediatrics 2005 Jul;116(1):e21-5.
PMID: 15995017.
54
15.
Adachi W, Lodolce AE. Use of failure mode

and effects analysis in improving the safety
of i.v. drug administration. Am J Health Syst
Pharm 2005 May 1;62(9):917-20. PMID:
15851497.
16.
Eckel SF, Anderson P, Zimmerman C, et al.

User satisfaction with an intravenous
medication safety system. Am J Health Syst
Pharm 2006 Aug 1;63(15):1419-23. PMID:
16849706.
17.
Fields M, Peterman J. Intravenous

medication safety system averts high-risk
medication errors and provides actionable
data. Nurs Adm Q 2005 Jan-Mar;29(1):7887. PMID: 15779709.
18.
Breland BD. Continuous quality

improvement using intelligent infusion
pump data analysis. Am J Health Syst
Pharm 2010 Sep 1;67(17):1446-55. PMID:
20720244.
19.
Schilling MB, Sandoval S. Impact of

intelligent intravenous infusion pumps on
directing care toward evidence-based
standards: a retrospective data analysis.
Hosp Pract (Minneap) 2011 Aug;39(3):11321. PMID: 21881398.
20.
Fanikos J, Fiumara K, Baroletti S, Luppi C,

Saniuk C, Mehta A, Silverman J, Goldhaber
SZ. Impact of smart infusion technology on
2007 Apr 1;99(7):1002-5. PMID: 17398201.
21.
Prusch AE, Suess TM, Paoletti RD, Olin ST,

Watts SD. Integrating technology to
improve medication administration. Am J
Health Syst Pharm 2011 May 1;68(9):83542. PMID: 21515868.
22.
Cassano AT. IV medication safety software

implementation in a multi-hospital health
system. Hosp Pharm 2006;41:151-6.
23.
Siv-Lee L, Morgan L. Implementation of

wireless intelligent pump IV infusion
technology in a not-for-profit academic
hospital setting. Hosp Pharm 2007;42:83240.
24.
Keohane CA, Hayes J, Saniuk C, et al.

Intravenous medication safety and smart
infusion systems: lessons learned and future
opportunities. J Infus Nurs 2005 SepOct;28(5):321-8. PMID: 16205498.
Section B. Infection Control

Chapter 7. Barrier Precautions, Patient Isolation, and Routine
Surveillance for Prevention of Health Care-Associated
Infections: Brief Update Review
Marin Schweizer, Ph.D.
Introduction
Healthcare-associated infections are linked to high morbidity, mortality, and costs
worldwide. In 2002, an estimated 1.7 million healthcare-associated infections were seen in U.S.
hospitals, resulting in approximately 99,000 deaths.1 In 2005, 18,650 patients with methicillinresistant Staphylococcus aureus (MRSA) died, more than the number of Americans who died
from HIV/AIDS in that same year.2 In 2007, Clostridium difficile was ranked among the 20
leading causes of mortality among Americans over 65 years of age.3 Despite decades of infection
control interventions, health care-associated infections continue to be a major burden on U.S.
hospitals.4
Currently, there is a rising wave of new emergent healthcare-associated infections, including
multi-drug resistant strains of Acinetobacter baumannii and Klebsiella pneumoniae.
Additionally, reports of vancomycin-resistant S. aureus have appeared sporadically across the
Nation.5-7 No effective antibiotics are available for some strains of these pathogens, and few new
antibiotics are in the developmental pipeline. For example, since 2007, only two new antibiotics
have been developed. Thus, prevention, not treatment, is the most sustainable strategy to control
health care-associated infections.
Findings of Original Report

When Making Health Care Safer was first published in 2001, the main healthcareassociated pathogens of interest were vancomycin-resistant enterococci (VRE) and C. difficile.
Three types of barrier precaution interventions were actively being studied, including (1) gowns
and gloves for all contact with patients with VRE or C. difficile followed by immediate hand
hygiene, (2) use of dedicated or disposable examining equipment for patients with VRE or C.
difficile, and (3) patient and/or staff cohorting for patients with VRE or C. difficile.
Nearly all of the studies that assessed the effectiveness of barrier precautions were simple
before-after studies with small cohorts of patients. Additionally, these studies usually assessed a
large bundle of practices to prevent infections, thus it was difficult to elucidate which
components of the bundle were effective.
Although results varied, the majority of the studies demonstrated significant reduction in the
incidence of VRE or C. difficile following barrier precaution interventions. A review of the
literature published just before the publication of Making Health Care Safer noted that there
had been little progress in assessing the psychological effects of contact isolation. However, it
was noted that attending physicians may examine patient on barrier precautions less often. The
barrier precautions chapter of Making Health Care Safer concluded that barrier precaution
interventions are effective and called for future studies of the long-term efficacy of barrier
precaution interventions as well as the cost-effectiveness of barrier precaution interventions.
55
This update review focuses on what we have learned about infection prevention measures
and their effectiveness since the publication of the original report. We conducted a search of the
health care and health services literature for the time interval 2001 to 2011 and reviewed all
studies relevant to this topic.
What Are Infection Prevention Measures?

The reservoir for many healthcare-associated infections is primarily colonized or infected
patients. Transiently colonized health care workers and contaminated items in the environment
are often intermediates in the patient-to-patient transmission of these pathogens. Thus, breaking
transmission from these reservoirs is the most important strategy to prevent healthcare-associated
infections. Multiple interventions can prevent transmission. Vertical interventions, in which
specific organisms are targeted, include active surveillance plus contact isolation or nurse
cohorting. Horizontal interventions, in which all healthcare-associated infections are targeted,
include universal contact precautions in high-risk settings.8
Active Surveillance and Isolation

Active surveillance is the process of testing patients for asymptomatic colonization. Active
surveillance is usually only performed for MRSA or VRE, since these organisms have
established reservoirs and valid screening tests.9 Universal active surveillance entails testing all
admitted patients for colonization, while targeted active surveillance only tests patients at high
risk for colonization (e.g., patients who recently received antimicrobials).
Patients found to be colonized through active surveillance are then isolated from other
patients in order to prevent transmission. Isolation can be performed through nurse cohorting or
contact isolation. Nurse cohorting is defined as physical segregation of colonized or infected
patients from patients not known to harbor the specific pathogen in a distinct area of the same
ward, and nursed by designated staff.10 When a patient is placed on contact precautions, health
care workers are required to wear a gown and gloves when they come in contact with the patient
then remove the gown and gloves and wash their hands after the contact, to prevent transmission
to other patients via their hands or clothing.
Contact isolation includes contact precautions but the patient is also placed in a single room. If a
single room is not available, contact isolation can be performed by cohorting patients colonized
or infected with the same pathogen in the same room. Currently, most of the studies that assess
active surveillance or universal contact precautions have only assessed these interventions in
intensive care units (ICUs), since ICU patients are at high risk of healthcare-associated
infections.11-13
What Have We Learned About Infection Control Practices Since the

Original Report?
Increasing Resistance and Changing Epidemiology Among
Staphylococcus aureus
Since the publication of the Making Health Care Safer report in 2001, Staphylococcus
aureus has gained considerable attention due to a number of factors. First, healthcare-associated
56
methicillin-resistant S. aureus infections increased rapidly with a high mortality rate.2,7 However,
since 2007 rates of healthcare-associated MRSA have begun to decline.14 Second, communityassociated MRSA infections caused by the USA300 clone emerged between 1999 and 2001.15
USA300 MRSA has caused severe infections in previously healthy people with no prior contact
with the health care system, thus alarming both health care professionals and the general
public.15 Additionally, USA300 MRSA infections have not replaced healthcare-associated
MRSA infections (e.g. USA100), rather they have occurred as a separate epidemic leading to an
increasing number of MRSA infections.16 Third, isolated cases of vancomycin-resistant S. aureus
(VRSA), first recognized in 2002, have led to fears that failure to control VRE and MRSA
transmission may lead to a new epidemic of VRSA, which will be very difficult to treat.5-7
Hypervirulent Strains of Clostridium difficile Have Emerged

The epidemiology of C. difficile has also changed since the publication of the Making
Health Care Safer report. A hypervirulent strain known as PCR ribotype 027, restriction
endonuclease analysis group BI, and North American PFGE pulsotype 1 (027/BI/NAP1) has
emerged worldwide and is associated with increased morbidity and mortality.17,18 In fact, U.S.
mortality due to C. difficile increased from 793 deaths in 1999 to 6,372 deaths in 2007.3 Many
countries, including the United States, have also reported an increased incidence of communityassociated C. difficile infections among previously healthy people.17,19
What Methods of Infection Control Are Currently Being Studied?

There is great debate in the field of infection control over whether vertical or horizontal
approaches should be used to prevent healthcare-associated infections.8 Active surveillance, a
vertical approach because it focuses only on one organism, has been credited with the low rates
of morbidity and mortality from MRSA in northern Europe and in Western Australia.20,21
Proponents of active surveillance argue that active surveillance and isolation, which has
prevented spread of other nosocomial pathogens such as smallpox and severe acute respiratory
syndrome, can also be used to contain MRSA or VRE.20,22,23 Proponents of active surveillance
acknowledge that a single-pathogen approach is not ideal; however, current horizontal
approaches have not decreased healthcare-associated infection rates significantly.20 Furthermore,
active surveillance and isolation for asymptomatic carriers could prevent transmission of MRSA
or VRE through multiple routes such as directly from one patient to another, via health care
workers contaminated hands or clothing, and via the environment.24
In contrast, proponents of a horizontal approach argue that hospitals should implement
interventions that will decrease the spread of all healthcare-associated infections, which would
decrease the overall rate of healthcare-associated infections.8,25,26 Advocates of a horizontal
approach also argue that strategies focusing on active surveillance and contact isolation for
MRSA or VRE will not prevent spread of susceptible S. aureus or enterococcus, spread of other
resistant organisms, or endogenous infections in patients already colonized with MRSA or VRE.
Also, active surveillance programs that only assess one body site will miss colonization of other
body sites.25 The increasing burden of antibiotic-resistant infections, including highly
transmissible pathogens such as Acinetobacter baumannii, cannot currently be prevented through
active surveillance.27,28 Furthermore, the costs for active surveillance may decrease the funds
available to implement other important infection prevention interventions.26
Even current guidelines disagree over the use of active surveillance for MRSA or VRE. The
Society for Healthcare Epidemiology of America (SHEA) Guideline for Preventing Nosocomial
57
Transmission of Multidrug-Resistant Strains of Staphylococcus aureus and Enterococcus, as

well as Dutch and British guidelines, recommend routine screening of high-risk patients for
MRSA or VRE. However the Centers for Disease Control and Prevention Healthcare Infection
Control Practices Advisory Committee (CDC HICPAC) Guideline on Management of
Multidrug-Resistant Organisms in Healthcare Settings, as well as an Australian guideline,
recommend active surveillance as a targeted measure to be implemented only when the incidence
or prevalence of MRSA or VRE is not decreasing despite other infection control strategies.7,29-32
Evidence for Effectiveness of Infection Control Practices

Multiple systematic literature reviews concluded that the evidence for interventions for the
prevention and control of multidrug-resistant organisms were of poor quality and that definitive
recommendations could not be made.10,33-35 However, a large number of new articles have been
published on these topics including multiple studies with large patient populations and have not
been included in these systematic reviews.13,24,36,37
Four large studies have assessed the effectiveness of active surveillance plus contact isolation
for preventing spread of MRSA or VRE. Robicsek et al. performed a three-phase quasiexperimental study in three hospitals. Phase one was a baseline assessment in which no
intervention was performed. Phase two included surveillance for MRSA in ICUs and contact
isolation for MRSA carriers. Phase three expanded to whole-hospital universal surveillance for
MRSA, contact isolation for MRSA carriers, and decolonization of MRSA carriers with topical
mupirocin. These investigators demonstrated that the aggregate hospital-associated MRSA
disease prevalence density decreased by 36.2% (P=0.17) from baseline to ICU surveillance and
by 69.6% (P =0.03) from baseline to universal surveillance.36
Similarly, investigators in the Veterans Health Administration performed a quasiexperimental study to assess their nationwide MRSA Prevention Initiative. This initiative was
composed of an MRSA prevention bundle which included (1) hand hygiene promotion, (2) an
infection prevention culture change, and (3) whole-hospital universal surveillance for MRSA and
contact isolation for MRSA carriers. In their analysis of all 153 Veterans affairs hospitals, they
found that the rates of healthcare-associated MRSA infections declined by 45% in non-ICUs and
by 62% in ICUs after the Initiative was implemented.24
In contrast, Harbarth et al. implemented active surveillance for MRSA carriers in six surgical
wards while six other surgical wards served as a control. After a washout period, the intervention
and control wards were switched. MRSA carriers identified by active surveillance received a
bundled intervention which included contact isolation, adjustment of perioperative antibiotic
prophylaxis, and topical decolonization (nasal mupirocin ointment and chlorhexidine body
washing). This study did not find a significant change in MRSA infections (adjusted incidence
rate ratio, 1.20; 95% confidence interval, 0.85-1.69; P=0.29).37
Finally, the STAR*ICU Trial was a cluster-randomized trial of 18 ICUs. This study
randomized eight ICUs to standard of care and ten ICUs to a bundle that included universal
surveillance for MRSA and VRE, contact isolation for MRSA or VRE positive patients, and
universal gloving until surveillance culture results were negative for all other ICU patients. That
study found no difference between the intervention and control groups in terms of mean ICUlevel incidence of colonization or infection with MRSA or VRE per 1,000 patient-days (40.43.3
and 35.63.7 in the two groups, respectively; P = 0.35).13
These four studies differed in multiple ways. First, the two studies with positive results
assessed their interventions both in the ICUs and universally throughout the health care
58
institutions, while the two studies with negative results only assessed their interventions in ICUs
or surgical wards. Each study implemented a unique bundle in which the only common factor in
all four bundles was active surveillance plus contact precautions. For example, both the Harbarth
and Robicsek included nasal decolonization while the other two studies did not. The studies also
varied in how their laboratory testing was performed. For example, in the Veterans Health
Administration study, surveillance samples were tested at the local clinical microbiology
laboratory. In contrast, in the Star*ICU study, all surveillance samples were mailed to the
Clinical Microbiology Laboratory of the National Institutes of Health Clinical Center.
Interestingly, when comparing all four of these studies, the studies with negative results had
stronger study designs.
The studies above assessed active surveillance among ICU patients. Admission to the ICU is
a large risk factor for healthcare-associated infections, therefore, it may be cost-effective to
target only ICU patients for active surveillance rather than the entire hospital.38 The high cost of
active surveillance has led to multiple cohort studies with the goal of establishing a rule to
predict which patients are at high risk for MRSA or VRE colonization.39 A prediction rule would
help infection prevention staff determine which patients are likely to carry MRSA or VRE and,
thus, could transmit MRSA or VRE to other patients or could acquire an MRSA or VRE
infection. Ideally, screening the patients identified as high risk of colonization would be more
cost-effective and take less time than testing all patients for MRSA or VRE using traditional
active surveillance. Many prediction rules include recent admission to the hospital, which is a
strong predictor of MRSA and VRE colonization, with sensitivities ranging from 44% to 77%
and specificities ranging from 46% to 98%.38,40-45 Prediction rules have also included risk factors
for colonization such as prior operation, hemodialysis, prior history of MRSA or VRE, transfer
from long-term care facility, age, antimicrobial use during the past year, and a current wound. If
these prediction rules were applied, the proportion of MRSA or VRE colonized patients who
would be missed ranged from 15% to 43%.38,40-45 Thus, current prediction rules have had varying
success.
Similarly, three studies have created prediction rules to predict patients at high risk for C.
difficile infection.46,47 The first prediction rule included age, C. difficile infection pressure, recent
admission to the hospital, severity of illness score, days of high-risk antibiotic use, low albumin
level, ICU admission, and receipt of laxatives, gastric acid suppressors or antimotility drugs.46
The second rule only included the Waterlow score, a nursing tool routinely used to assess a
patients risk of developing a pressure ulcer.47 The third rule included age, hemodialysis and
length of ICU stay.48 The sensitivity of the C. difficile infection prediction rules ranged from
60% to 70% and the specificity ranged from 89% to 95%.46,47
Horizontal approaches to infection control could utilize contact precautions without the use
of expensive laboratory surveillance tests. A single ICU, quasi-experimental study of a bundle
which included universal contact precautions found that not only did this bundle stop an outbreak
of multidrug-resistant Acinetobacter baumannii, it also led to a decrease in MRSA acquisition
from 14% to 10%, and VRE acquisition from 21% to 9%.11 Two quasi-experimental studies
compared universal gloving (wearing a new pair of gloves for each patient) to active surveillance
and contact precautions in a single ICU.49,50 Active surveillance and contact precautions included
VRE and MRSA surveillance cultures on admission and every 4 days with contact precautions
for patients colonized or infected with VRE or MRSA. Both studies found no difference in
MRSA or VRE colonization no matter which intervention was implemented. However, one study
59
found an increase in nosocomial infection rates during the universal glove period, potentially due
to decreased compliance with hand hygiene after removal of gloves.49
Another horizontal approach would be to place patients at high risk for acquiring a
healthcare-associated infection under pre-emptive contact precautions.51,52 One ICU found that
their intubated patients were eight times more likely to acquire healthcare-associated MRSA
compared with non-ventilated patients, thus they performed a quasi-experimental study to assess
an intervention where all intubated patients were placed under pre-emptive contact precautions.
In the first phase of the study, active surveillance for MRSA was performed at ICU admission
and weekly with contact precautions for MRSA positive patients. In the second phase of the
study, active surveillance and contact precautions for MRSA remained, however all intubated
patients were also placed on contact precautions. This study found a decrease in healthcareassociated MRSA infections for both intubated patients (p=0.02) and in all ICU patients
(p<0.05).52
Less is known about optimal methods to prevent C. difficile transmission compared with
VRE and MRSA.53 Most studies of C. difficile prevention are simple quasi-experimental studies
that test a bundled intervention. Multiple recommendations and guidelines suggest contact
isolation for symptomatic C. difficile infected patients only.17,53,54 Contact isolation for C.
difficile infected patients should include single rooms with private toilets if possible.17 According
to the SHEA/IDSA Expert Panel, the only two approaches to preventing C. difficile with good
evidence to support them are wearing gloves when caring for an infected patient and
antimicrobial stewardship.17,54 No data currently support isolating asymptomatic C. difficile
carriers.53,54 An unresolved issue is whether to place symptomatic patients with a history of C.
difficile infection under contact precautions.17
Some Potential for Harm Is Associated With Contact Precautions

At the time that Making Health Care Safer was published, very few studies assessed the
potential harm associated with contact isolation. Recent studies, including a systematic literature
review, found that contact precautions have been associated with less patient-to-health care
worker contact, changes in systems of care that produce delays and more noninfectious adverse
events (e.g., falls, pressure ulcers), increased symptoms of depression and anxiety, and decreased
patient satisfaction with care.55-59
Costs and Implementation of Infection Prevention Interventions Have

Been Examined
Both vertical and horizontal interventions to prevent healthcare-associated infections require
upfront investments to pay for components of the intervention such as supplies (e.g., gowns and
gloves) and laboratory resources (e.g., tests, personnel).9 However, a business case can be made
for these interventions since the estimated median cost of a healthcare-associated infection
ranges from $26,424 to $34,657 for MRSA and from $17,1438 to $36,380 for VRE.60-64 Two
studies found that clinical active surveillance of ICU patients for VRE or MRSA colonization
was cost effective compared with the cost savings of preventing these infections.63,65 Similarly,
another study found that active surveillance and isolation for VRE colonization among high-risk
patients cost effective.66 A mathematical model compared whole hospital universal active
surveillance for MRSA to targeted active surveillance for MRSA and found that targeted
surveillance was more cost effective.67
60
The cost-effectiveness studies estimated that the cost of active surveillance and contact
isolation strategies for MRSA or VRE to range from $1,913 to $10,545 per month.63,65,66 The
mathematical model found that the average cost of targeted active surveillance of high risk
patients ranged from $4,100 to $12,508 per infection adverted depending on MRSA prevalence
and screening test used, while the average cost of universal active surveillance ranged from
$5,799 to $21,195 per infection adverted.67 When these costs were itemized, 13% to 99% of the
total cost was spent on specimen collection and laboratory testing while the remaining proportion
was spent on isolation (e.g., gowns, gloves, nurse time to don gowns and gloves).63,65-67 The vast
differences in these proportions were due to how labor costs were accounted for. Studies varied
as to how they assessed the cost of laboratory technologists, cost of nursing time to collect
swabs, and cost of nursing time to don and remove gowns and gloves.
Although cost-effectiveness analyses have not been performed for universal contact
precautions to prevent healthcare-associated infections, an analysis by Wenzel et al., compared
the cost-effectiveness of active surveillance and contact precautions for MRSA to a populationbased infection control approach. This analysis assumed that active surveillance for MRSA
would cost approximately $600,000 while the population-based approach would cost
approximately $300,000. If the active surveillance program reduced MRSA infections by 50%
and the population-based approach reduced healthcare-associated infections by 50%, then the
active surveillance program would save $245 million to $980 million nationally while the
population-based intervention would save $1.75 billion to $7 billion nationally.26
As with all health care interventions, health care worker support and implementation of the
intervention is necessary for the intervention to be successful. The STAR*ICU trial noted
suboptimal implementation of their interventions. That study demonstrated that when contact
precautions were specified, gloves were used for 82% of contacts, gowns for 77% of contacts,
and hand hygiene was performed after gloves were removed for 69% of contacts. Additionally,
when universal gloving was specified, gloves were used for 72% of contacts and hand hygiene
was performed after gloves were removed for 62% of contacts.13 The Veterans Health
Administrations MRSA initiative includes a dedicated MRSA coordinator at each acute care
hospital responsible for implementation of the initiative. From the beginning of the initiative in
2007 to the end of the study period in 2010, compliance with surveillance nasal screening for
MRSA increased, with the percentage of patients who were screened at admission rising from
82% to 96%, and the percentage who were screened at transfer or discharge rising from 72% to
93%. However adherence to contact precautions was not reported.24 Two studies by Bearman
and colleagues found that observed compliance was higher during a universal glove intervention
compared with observed compliance with contact precautions (gowns and gloves) during an
active surveillance plus contact precaution intervention. However, the studies found conflicting
results as to when hand hygiene compliance was greater. The first study found that the active
surveillance and contact precautions intervention was associated with greater compliance with
hand hygiene compared with hand hygiene compliance during the universal gloving
intervention.49 The second study, which included hand hygiene in-service trainings, found that
compliance with hand hygiene was higher during the universal gloving phase compared with the
active surveillance and contact precautions phase.50
Upcoming Studies
Of late, two multicenter cluster-randomized trials of contact precautions have been
implemented. The Cluster Randomized Trial of Hospitals to Assess Impact of Targeted versus
61
Universal Strategies to Reduce MRSA in ICUs (REDUCE MRSA trial) recently finished
collecting data on the effectiveness of the following strategies: (1) MRSA active surveillance of
ICU admissions, followed by contact isolation if positive, (2) MRSA active surveillance of ICU
admissions followed by nasal decolonization if positive, and (3) universal nasal decolonization
of ICU admissions without screening.68 The Benefits of Universal Glove and Gowning Study
(BUGG Study) is currently comparing the effectiveness of universal contact precautions to
standard of care in multiple ICUs in order to determine whether universal gowns and gloves
decrease the overall burden of healthcare-associated pathogens in the ICU setting. The results of
these studies should be available soon and will add to the growing body of evidence on
interventions to control healthcare-associated infections.12

Although many studies have been performed since the Making Health Care Safer report, there
is still much debate as to which interventions should be implemented to prevent healthcareassociated infections. Vertical interventions, such as active surveillance for MRSA or VRE, have
been studied the most; however, these studies have had conflicting results. Horizontal
approaches, such as universal gloving, have the potential to reduce the burden of all health careassociate pathogens; however these approaches have been understudied. Current evidence should
be considered by individual institutions to determine which interventions are right for their
institution based on their patient population, problem pathogens, and ability to implement
interventions.69 For example, universal active surveillance for MRSA may be optimal for
hospitals with endemic MRSA throughout their hospital, whereas ICU-level universal contact
precautions may be recommended for hospitals with multidrug-resistant Acinetobacter
baumannii transmission in their ICU. Interventions such active surveillance, contact precautions,
and contact isolation should not be performed alone. Rather, these interventions must be
performed in conjunction with other infection control interventions such as hand hygiene and
antimicrobial stewardship. In conclusion, high quality studies are still needed to determine the
optimal interventions to reduce healthcare-associated infections. A summary table is located at
Table 1, Chapter 7.
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Moderate
(isolation of
patients)
Estimate of
Cost
Moderate-tohigh
How Much do We
Moderate/Moderate
References
1.
2.
Klevens RM, Edwards JR, Richards CL, Jr.,

et al. Estimating health care-associated
infections and deaths in U.S. hospitals,
2002. Public Health Rep. 2007 MarApr;122(2):160-6. PMID 17357358.
62
Klevens RM, Morrison MA, Nadle J, et al.

Invasive methicillin-resistant
Staphylococcus aureus infections in the
United States. JAMA. 2007 Oct
17;298(15):1763-71. PMID 17940231.
3.
Xu JQ, Kochanek KD, Murphy SL, et al.

Deaths: Final data for 2007. National vital
statistics reports. National Center for Health
Statistics. Hyattsville, MD: 2010.
4.
Edwards JR, Peterson KD, Mu Y, et al.

National Healthcare Safety Network
(NHSN) report: data summary for 2006
through 2008, issued December 2009. Am J
Infect Control. 2009 Dec;37(10):783-805.
PMID 20004811.
5.
Centers for Disease Control and Prevention

(CDC). Staphylococcus aureus resistant to
vancomycin--United States. MMWR Morb
Mortal Wkly Rep. 2002 Jul 5;51(26):565-7.
6.

(CDC). Vancomycin-resistant
Staphylococcus aureus--Pennsylvania.
MMWR Morb Mortal Wkly Rep. 2002 Oct
11;51(40):902.
7.
Muto CA, Jernigan JA, Ostrowsky BE, et al.

SHEA guideline for preventing nosocomial
transmission of multidrug-resistant strains of
Staphylococcus aureus and enterococcus.
Infect Control Hosp Epidemiol. 2003
May;24(5):362-86. PMID 12785411.
8.
Wenzel RP, Edmond MB. Infection control:

the case for horizontal rather than vertical
interventional programs. Int J Infect Dis.
2010 Oct;14 Suppl 4:S3-5. PMID 20851010.
9.
Diekema DJ, Edmond MB. Look before you

leap: active surveillance for multidrugresistant organisms. Clin Infect Dis. 2007
Apr 15;44(8):1101-7. PMID 17366459.
10.
Cooper BS, Stone SP, Kibbler CC, et al.

Systematic review of isolation policies in the
hospital management of methicillin-resistant
Staphylococcus aureus: a review of the
literature with epidemiological and
economic modelling. Health Technol
Assess. 2003;7(39):1-194. PMID 14636487.
11.
12.
Wright MO, Hebden JN, Harris AD, et al.

Aggressive control measures for resistant
Acinetobacter baumannii and the impact on
acquisition of methicillin-resistant
Staphylococcus aureus and vancomycinresistant Enterococcus in a medical intensive
care unit. Infect Control Hosp Epidemiol.
2004 Feb;25(2):167-8. PMID 14994945.
ClinicalTrials.gov. [Internet]
www.clinicaltrials.gov. 2010.
63
13.
Huskins WC, Huckabee CM, OGrady NP,

et al. Intervention to reduce transmission of
resistant bacteria in intensive care. N Engl J
Med. 2011 Apr 14;364(15):1407-18. PMID
21488763.
14.
Kallen AJ, Mu Y, Bulens S, et al. Health

care-associated invasive MRSA infections,
2005-2008. JAMA. 2010 Aug
11;304(6):641-8. PMID 20699455.
15.
Chambers HF, Deleo FR. Waves of

resistance: Staphylococcus aureus in the
antibiotic era. Nat Rev Microbiol. 2009
Sep;7(9):629-41. PMID 19680247.
16.
Johnson JK, Khoie T, Shurland S, et al. Skin

and soft tissue infections caused by
methicillin-resistant Staphylococcus aureus
USA300 clone. Emerg Infect Dis. 2007
Aug;13(8):1195-200. PMID 17953091.
17.
Barbut F, Jones G, Eckert C. Epidemiology

and control of Clostridium difficile
infections in healthcare settings: an update.
Curr Opin Infect Dis. 2011 Aug;24(4):3706. PMID 21505332.
18.

(CDC). Surveillance for communityassociated Clostridium difficile-Connecticut, 2006. MMWR Morb Mortal
Wkly Rep. 2008 Apr 4;57(13):340-3.
19.

(CDC). Severe Clostridium difficileassociated disease in populations previously
at low risk--four states, 2005. MMWR Morb
Mortal Wkly Rep. 2005 Dec 2;54(47):12015.
20.
Farr BM, Jarvis WR. What works and what

doesnt for the control of methicillinresistant Staphylococcus aureus infection:
dogma and data. Clin Infect Dis. 2009 Sep
15;49(6):987-8; author reply 9-90. PMID
19694537.
21.
Vriens MR, Fluit AC, Troelstra A, et al. Is

more contagious than methicillin-susceptible
S. aureus in a surgical intensive care unit?
Sep;23(9):491-4. PMID 12269444.
22.
Farr BM. Doing the right thing (and figuring

out what that is). Infect Control Hosp
Epidemiol. 2006 Oct;27(10):999-1003.
PMID 17006804.
23.
Calfee DP, Giannetta ET, Durbin LJ, et al.

Control of endemic vancomycin-resistant
Enterococcus among inpatients at a
university hospital. Clin Infect Dis. 2003
Aug 1;37(3):326-32. PMID 12884155.
24.
Jain R, Kralovic SM, Evans ME, et al.

Veterans Affairs initiative to prevent
infections. N Engl J Med. 2011 Apr
14;364(15):1419-30. PMID 21488764.
25.
Kirkland KB. Taking off the gloves: toward

a less dogmatic approach to the use of
contact isolation. Clin Infect Dis. 2009 Mar
15;48(6):766-71. PMID 19196174.
26.
Wenzel RP, Bearman G, Edmond MB.

Screening for MRSA: a flawed hospital
infection control intervention. Infect Control
Hosp Epidemiol. 2008 Nov;29(11):1012-8.
PMID 18937571.
27.
Morgan DJ, Liang SY, Smith CL, et al.

Frequent multidrug-resistant Acinetobacter
baumannii contamination of gloves, gowns,
and hands of healthcare workers. Infect
Control Hosp Epidemiol. 2010
Jul;31(7):716-21. PMID 20486855.
28.
Marchaim D, Navon-Venezia S, Schwartz

D, et al. Surveillance cultures and duration
of carriage of multidrug-resistant
Acinetobacter baumannii. J Clin Microbiol.
2007 May;45(5):1551-5. PMID 17314222.
29.
Siegel JD, Rhinehart E, Jackson M, et al.

Management of multidrug-resistant
organisms in health care settings. Am J
Infect Control. 2007 Dec;35(10):S165-93.
30.
Australian guidelines for the prevention and

control of infection in healthcare.
Commonwealth of Australia [Internet].
www.nhmrc.gov.au. 2010.
31.
Wertheim HF, Ammerlaan HS, Bonten MJ,

et al. [Optimisation of the antibiotic policy
in the Netherlands. XII. The SWAB
guideline for antimicrobial eradication of
MRSA in carriers]. Ned Tijdschr Geneeskd.
2008 Dec 6;152(49):2667-71. PMID
19137966.
32.
Coia JE, Duckworth GJ, Edwards DI, et al.

Guidelines for the control and prevention of
meticillin-resistant Staphylococcus aureus
(MRSA) in healthcare facilities. J Hosp
Infect. 2006 May;63 Suppl 1:S1-44. PMID
16581155.
64
33.
McGinigle KL, Gourlay ML, Buchanan IB.

The use of active surveillance cultures in
adult intensive care units to reduce
methicillin-resistant Staphylococcus aureusrelated morbidity, mortality, and costs: a
systematic review. Clin Infect Dis. 2008 Jun
1;46(11):1717-25. PMID 18494098.
34.
Loveday HP, Pellowe CM, Jones SR, et al.

A systematic review of the evidence for
interventions for the prevention and control
of meticillin-resistant Staphylococcus aureus
(1996-2004): report to the Joint MRSA
Working Party (Subgroup A). J Hosp Infect.
2006 May;63 Suppl 1:S45-70. PMID
16616800.
35.
Backman C, Taylor G, Sales A, et al. An

integrative review of infection prevention
and control programs for multidrug-resistant
organisms in acute care hospitals: a socioecological perspective. Am J Infect Control.
2011 Jun;39(5):368-78. PMID 21429622.
36.
Robicsek A, Beaumont JL, Paule SM, et al.

Universal surveillance for methicillinresistant Staphylococcus aureus in 3
affiliated hospitals. Ann Intern Med. 2008
Mar 18;148(6):409-18. PMID 18347349.
37.
Harbarth S, Fankhauser C, Schrenzel J, et al.

Universal screening for methicillin-resistant
Staphylococcus aureus at hospital admission
and nosocomial infection in surgical
patients. JAMA. 2008 Mar
12;299(10):1149-57. PMID 18334690.
38.
Tacconelli E, Karchmer AW, Yokoe D, et

al. Preventing the influx of vancomycinresistant enterococci into health care
institutions, by use of a simple validated
prediction rule. Clin Infect Dis. 2004 Oct
1;39(7):964-70. PMID 15472847.
39.
Edmond M. Cost-effectiveness of perirectal

surveillance cultures for controlling
vancomycin-resistant Enterococcus. Infect
May;24(5):309-10; author reply 10-2. PMID
12785400.
40.
Riedel S, Von Stein D, Richardson K, et al.

Development of a prediction rule for
and vancomycin-resistant enterococcus
carriage in a Veterans Affairs Medical
Center population. Infect Control Hosp
Epidemiol. 2008 Oct;29(10):969-71. PMID
18702599.
41.
Robicsek A, Beaumont JL, Wright MO, et

al. Electronic prediction rules for
colonization. Infect Control Hosp
Epidemiol. 2011 Jan;32(1):9-19. PMID
21121818.
42.
Wibbenmeyer L, Williams I, Ward M, et al.

Risk factors for acquiring vancomycinresistant Enterococcus and methicillinresistant Staphylococcus aureus on a burn
surgery step-down unit. J Burn Care Res.
2010 Mar-Apr;31(2):269-79. PMID
20182380.
43.
44.
Morgan DJ, Day HR, Furuno JP, et al.

Improving efficiency in active surveillance
for methicillin-resistant Staphylococcus
aureus or vancomycin-resistant
Enterococcus at hospital admission. Infect
Dec;31(12):1230-5. PMID 21028966.
Furuno JP, McGregor JC, Harris AD, et al.
Identifying groups at high risk for carriage
of antibiotic-resistant bacteria. Arch Intern
Med. 2006 Mar 13;166(5):580-5. PMID
16534047.
45.
Harbarth S, Sax H, Fankhauser-Rodriguez

C, et al. Evaluating the probability of
previously unknown carriage of MRSA at
hospital admission. Am J Med. 2006
Mar;119(3):275 e15-23. PMID 16490475.
46.
Dubberke ER, Yan Y, Reske KA, et al.

Development and validation of a
Clostridium difficile infection risk
prediction model. Infect Control Hosp
Epidemiol. 2011 Apr;32(4):360-6. PMID
21460487.
50.
Bearman G, Rosato AE, Duane TM, et al.

Trial of universal gloving with emollientimpregnated gloves to promote skin health
and prevent the transmission of multidrugresistant organisms in a surgical intensive
2010 May;31(5):491-7. PMID 20350197.
51.
Tacconelli E. Screening and isolation for

infection control. J Hosp Infect. 2009
Dec;73(4):371-7. PMID 19699554.
52.
Matsushima A, Tasaki O, Tomono K, et al.

Pre-emptive contact precautions for
intubated patients reduced healthcareassociated meticillin-resistant
Staphylococcus aureus transmission and
infection in an intensive care unit. J Hosp
Infect. 2011 Jun;78(2):97-101. PMID
21474201.
53.
Dubberke ER, Gerding DN, Classen D, et al.

Strategies to prevent clostridium difficile
infections in acute care hospitals. Infect
Control Hosp Epidemiol. 2008 Oct;29 Suppl
1:S81-92. PMID 18840091.
54.
Cohen SH, Gerding DN, Johnson S, et al.

Clinical practice guidelines for Clostridium
difficile infection in adults: 2010 update by
the society for healthcare epidemiology of
America (SHEA) and the infectious diseases
society of America (IDSA). Infect Control
Hosp Epidemiol. 2010 May;31(5):431-55.
PMID 20307191.
55.
Morgan DJ, Diekema DJ, Sepkowitz K, et

al. Adverse outcomes associated with
Contact Precautions: a review of the
literature. Am J Infect Control. 2009
Mar;37(2):85-93. PMID 19249637.
47.
Tanner J, Khan D, Anthony D, et al.

Waterlow score to predict patients at risk of
developing Clostridium difficile-associated
disease. J Hosp Infect. 2009 Mar;71(3):23944. PMID 19162374.
56.
Day HR, Morgan DJ, Himelhoch S, et al.

Association between depression and contact
precautions in veterans at hospital
admission. Am J Infect Control. 2011
Mar;39(2):163-5. PMID 21356434.
48.
Garey KW, Dao-Tran TK, Jiang ZD, et al. A

clinical risk index for Clostridium difficile
infection in hospitalised patients receiving
broad-spectrum antibiotics. J Hosp Infect.
2008 Oct;70(2):142-7. PMID 18723249.
57.
Day HR, Perencevich EN, Harris AD, et al.

Do contact precautions cause depression? A
two-year study at a tertiary care medical
centre. J Hosp Infect. 2011 Oct;79(2):103-7.
PMID 21664000.
49.
Bearman GM, Marra AR, Sessler CN, et al.

A controlled trial of universal gloving versus
contact precautions for preventing the
transmission of multidrug-resistant
organisms. Am J Infect Control. 2007
Dec;35(10):650-5. PMID 18063129.
58.
Tarzi S, Kennedy P, Stone S, et al.

Methicillin-resistant Staphylococcus aureus:
psychological impact of hospitalization and
isolation in an older adult population. J Hosp
Infect. 2001 Dec;49(4):250-4. PMID
11740872.
65
59.
Catalano G, Houston SH, Catalano MC, et

al. Anxiety and depression in hospitalized
patients in resistant organism isolation.
South Med J. 2003 Feb;96(2):141-5. PMID
12630637.
60.
Perencevich EN, Stone PW, Wright SB, et

al. Raising standards while watching the
bottom line: making a business case for
infection control. Infect Control Hosp
17933084.
61.
62.
63.
64.
Filice GA, Nyman JA, Lexau C, et al.

Excess costs and utilization associated with
methicillin resistance for patients with
Staphylococcus aureus infection. Infect
Apr;31(4):365-73. PMID 20184420.
Cosgrove SE, Qi Y, Kaye KS, et al. The
impact of methicillin resistance in
Staphylococcus aureus bacteremia on patient
outcomes: mortality, length of stay, and
hospital charges. Infect Control Hosp
Epidemiol. 2005 Feb;26(2):166-74. PMID
15756888.
Shadel BN, Puzniak LA, Gillespie KN, et al.
Surveillance for vancomycin-resistant
enterococci: type, rates, costs, and
implications. Infect Control Hosp
17006814.
Stosor V, Peterson LR, Postelnick M, et al.
Enterococcus faecium bacteremia: does
vancomycin resistance make a difference?
Arch Intern Med. 1998 Mar 9;158(5):522-7.
PMID 9508230.
66
65.
Clancy M, Graepler A, Wilson M, et al.

Active screening in high-risk units is an
effective and cost-avoidant method to
reduce the rate of methicillin-resistant
Staphylococcus aureus infection in the
hospital. Infect Control Hosp Epidemiol.
2006 Oct;27(10):1009-17. PMID 17006806.
66.
Muto CA, Giannetta ET, Durbin LJ, et al.

Cost-effectiveness of perirectal surveillance
cultures for controlling vancomycinresistant Enterococcus. Infect Control Hosp
Epidemiol. 2002 Aug;23(8):429-35. PMID
12186207.
67.
Hubben G, Bootsma M, Luteijn M, et al.

Modelling the costs and effects of selective
and universal hospital admission screening
for methicillin-resistant Staphylococcus
aureus. PLoS One. 2011;6(3):e14783. PMID
21483492.
68.
Platt R, Takvorian SU, Septimus E, et al.

Cluster randomized trials in comparative
effectiveness research: randomizing
hospitals to test methods for prevention of
healthcare-associated infections. Med Care.
2010 Jun;48(6 Suppl):S52-7. PMID
20473200.
69.
Gasink LB, Brennan PJ. Isolation

precautions for antibiotic-resistant bacteria
in healthcare settings. Curr Opin Infect Dis.
2009 Aug;22(4):339-44. PMID 19461513.
Chapter 8. Interventions To Improve Hand Hygiene

Compliance: Brief Update Review
Elizabeth Pfoh, M.P.H.; Sydney Dy, M.D., M.Sc.; Cyrus Engineer, Dr.P.H.
Introduction
Healthcare-associated infections account for approximately 80,000 deaths per year in the
United States.1-3 A worldwide systematic review found that the incidence of healthcareassociated infections ranged from 1.7 to 23.6 per 100 patients. Hospital costs directly related to
healthcare-associated infections ranged from $28.4 to $33.8 billion in 2007 U.S. dollars.4 Yet
these infections are frequently preventable through hand hygiene.
Substantial epidemiologic evidence supports that hand hygiene reduces the transmission of
healthcare-associated pathogens and the incidence of health-care associated infections.5 The link
between hand hygiene and improvements in healthcare-associated infections is hard to prove
definitively in modern-day health care. However, the importance of hand hygiene is universally
acknowledged by organizations such as the Joint Commission, World Health Organization
(WHO) and Centers for Disease Control (CDC), which recommend or require hand hygiene
practices and interventions to improve hand hygiene compliance in order to reduce health careacquired infections.5-7 This review will therefore focus on interventions to improve compliance
with hand hygiene, rather than on the efficacy of hand hygiene for reducing healthcareassociated infections.
Compliance with hand hygiene practices among health care workers has historically been
very low, averaging 39 percent.5 The review on hand hygiene compliance and interventions
aimed at improving it that was conducted for the original 2001 Making Health Care Safer
report found that poor compliance has been documented in studies across hospital unit types and
in various other settings. Workers tend to underestimate the importance of compliance and often
overestimate their compliance with hand hygiene procedures.1 The report concluded that future
research studies needed to identify reasons for poor compliance and design sustainable
interventions that target these factors. The aim of this review is to assess the evidence for the
impact of interventions on hand hygiene compliance since that report.
What Is Hand Hygiene Compliance?

Hand hygiene is a general term for removing microorganisms with a disinfecting agent such
as alcohol or soap and water.1 Hand hygiene should be conducted by health care workers before
seeing patients, after contact with bodily fluids, before invasive procedures, and after removing
gloves.6 The WHO offers a slight variation by recommending five key moments when health
care workers should practice hand hygiene: before patient contact, before an aseptic task, after
bodily fluid exposure risk, after patient contact, and after contact with patient surroundings.5 The
National Quality Forums Safe Practices for Better Healthcare 2010 Update and the Joint
Commission recommend that organizations should implement CDC or WHO guidelines,
encourage staff compliance with guidelines with category II evidence, and ensure staff comply
with organizational rules regarding hand hygiene (see section below on implementation for
details).6,7
67
Monitoring health care workers compliance with hand hygiene practices is vital for
evaluating whether interventions are successful. WHO recommends using a validated
methodology for training observers to directly monitor hand-hygiene using My five moments
for hand hygiene.5 Other methods for monitoring include patient-observations, measuring of
hand hygiene product consumption (either by volume of product used or through electronic
counting devices), and electronic hand hygiene compliance monitoring systems (e.g. real-time
location systems, dedicated monitoring systems or video monitoring).8
Hand hygiene interventions include both single and multi-level interventions. These
interventions include staff and/or patient education and involvement, feedback initiatives,
cultural change, organizational change, social marketing, additional sinks or alcohol dispensers,
or a combination of the above.1,9
Advocates of hand-hygiene improvement interventions recommend that multimodal
interventions are needed to induce sustained hand-hygiene practice improvements, and should be
based on theories of behavior change. On the individual level, the intervention should target
provider education and motivation regarding hand-hygiene practices; on the interpersonal level,
patient empowerment and cues to action should reinforce proper hand-hygiene practices; and on
the organization level, organizational structure and philosophy needs to be supportive of proper
practices.5
How Have Interventions To Improve Hand Hygiene Compliance

Been Implemented?
Several major hand hygiene compliance programs have been developed and made publicly
available from the CDC, Institute for Healthcare Improvement, Joint Commission, and WHO,
and are widely implemented in health care institutions.
The CDC has published a guideline, interactive training and educational materials, and
posters for hand-hygiene compliance.10 The guideline provides suggestions for health care
worker educational and motivational programs; these suggestions include stating a rationale for,
and providing information regarding, when hand-hygiene is required; and providing proper hand
hygiene techniques, methods to maintain skin health, expectations of managers, and indicators
for glove use.11 The interactive tools include a set of PowerPoint slides and speaker notes that
provide background information on the importance of hand-hygiene, indications on when to use
hand-hygiene practices and how to properly clean ones hands, and educational/motivational
programs.12 Promotional posters aiming to demonstrate proper hand-hygiene and remind health
care workers of the importance of hand-hygiene are also available.6
The Institute for Healthcare Improvement, in collaboration with the CDC, the Association for
Professionals in Infection Control and Epidemiology, and the Society of Healthcare
Epidemiology of America, created a how-to guide on improving hand-hygiene among health
care workers for organizations. This guide includes evidence-based interventions, goal-setting
suggestions, evaluation suggestions, and measurement tools. The intervention is a multi-faceted
approach with four key aims: (1) to improve knowledge of proper hand hygiene practice; (2) to
have workers demonstrate hand hygiene knowledge; (3) to ensure the availability of alcoholbased rub and gloves at the point of care; and (4) to ensure that competency in hand hygiene is
regularly verified, compliance is monitored, and appropriate feedback loops are in place.13
The Joint Commission created a monograph to help health care organizations properly
measure hand hygiene performance. Content for the monograph came from examples of methods
and tools submitted through the Consensus Measurement in Hand Hygiene Project and published
68
literature.14 The monograph includes a comprehensive review of three measurement methods,

including surveys, measuring product use, and directly observing hand hygiene. Additional
information includes a review of ways to display data, intervention strategies, and additional
supplementary resources.
In 2009, the WHO published an extensive report, including a background on transmission of
infections, guidelines for proper hand-hygiene protocol; social, cultural, and behavioral aspects
of hand-hygiene; consensus recommendations; process and outcome measurement; and patient
involvement in hand-hygiene.5 A multimodal strategy was found to be necessary to improve
compliance; therefore recommendations for proper hand hygiene span different levels. For
providers, washing hands when visibly dirty, and using alcohol-based hand rub before and after
contact with a patient, contaminated surface, or medicine is critical. Additionally, they should
not wear artificial nails. Organizations should provide information to workers regarding handhygiene practices that reduce skin irritation and provide lotions or creams to minimize the
occurrence of skin irritation. When designing an intervention to increase proper hand hygiene, a
multi-faceted, multi-modal intervention should be used, practices should be monitored, and
feedback loops should be implemented. Health care administrators should ensure structural and
cultural factors are conducive to hand-hygiene practices, including ensuring access to alcoholbased hand-rub and/or a continuous water supply at the point of care, and making compliance
with a multi-faceted intervention an institutional priority.5 Individual factors, such as normative
beliefs (peer behavior), perceived control, and attitude (awareness of being observed) should also
be addressed since they were found to be important predictors of hand hygiene adherence. The
WHO provides training and education tools such as a template for creating an action plan, an
observation form for monitoring hand-hygiene compliance, training films, and educational
brochures. All tools were tested in eight official pilot sites in seven countries before being
finalized.5
What Have We Learned About Hand Hygiene Interventions?

A recent review determined that a successful hand hygiene educational program has several
key features. These features include reinforcement of hand hygiene messages; knowledge of
health care workers perceived importance of hand hygiene and its role in prevention of
healthcare-associated infections; monitoring and feedback of hand hygiene practices; practical
education tools; role-modeling by senior staff; and supportive infrastructure and management.
Interventions should be multimodal, and teaching methodology should be progressive and
include different types of methods. The educational program itself should be designed to include
local structure, priorities, and resources.15 Additionally, as stated above, across several studies,
the 2009 WHO report found hand hygiene practices should be multimodal, and structurally and
culturally tailored to improve compliance with hand hygiene.5
What Methods Have Been Used To Improve Hand Hygiene

Compliance?
The 2001 Making Health Care Safer report discussed studies that aimed to improve hand
hygiene through education, feedback, installation of sinks and alcohol-based solution, and
organizational changes.1 Making Health Care Safer included 14 non-randomized controlled or
before-after studies, 13 of which measured hand hygiene compliance through direct observation,
most in the intensive care unit setting. Interventions included increasing sink or alcohol-based
69
solution availability, education, and multifaceted interventions, including feedback. Ten studies
found a statistically significant increase in compliance, and four did not. Three studies evaluated
longer-term results and found that compliance rates decreased after the intervention ended.1
Impact of Interventions on Hand Hygiene Compliance

Since 2001, two major systematic reviews have been published on the impact of
interventions on hand-hygiene compliance.
A 2010 Cochrane systematic review (an update of a 2007 review) found insufficient evidence
that hand-hygiene interventions improve hand hygiene in the hospital setting.9 The review
included randomized controlled trials, controlled clinical trials, controlled before and after
studies, and interrupted time series analyses that met the criteria of the Cochrane Effective
Practice and Organization of Care Group from 1980-2009. Eligible outcomes included indicators
of compliance with hand hygiene or proxy indicators such as use of product; operating room
studies were excluded. Four studies were included, with one study finding a statistically
significant improvement in hand hygiene 4 months post-intervention, two studies finding a
statistically significant increase in product use which was sustained at one site for 2 years, and
one study finding no effect in the intervention compared with the control group 3 months postintervention. Studies focused on educational campaigns and promotion of guidelines, as well as a
multifaceted intervention to improve compliance. Simple substitution of a product with alcoholbased hand rub did not significantly increase product use.9
A 2008 systematic review addressed studies evaluating hand-hygiene interventions and
healthcare-associated infections in acute and long-term care settings (not the impact of the
interventions on compliance with hand hygiene).16 Studies included multifaceted initiatives,
introduction of new hand-hygiene products, and implementation of infection control practices
and policies, surveys, and electronic monitoring. The review included before and after studies
with and without control groups and cohort studies with no controls. Eighteen of 31 included
studies (58%) reported a statistically significant reduction in healthcare-associated infections
with the intervention compared with the control group; some studies also included other factors
that may have influenced the reductions in healthcare-associated infections.
Patient Engagement
A 2011 review by McGuckin and colleagues found evidence of the importance of patient
engagement or empowerment and multi-model strategies in hand-hygiene interventions. The
authors found that patient empowerment comprised patient participation, knowledge, skills, and
a facilitating environment for their participation in hand hygiene. The majority of patients agreed
that they would ask their health care workers to wash their hands (80% to 90%), especially if
encouraged to ask. However, the authors found scarce literature on the efficacy of patient
empowerment interventions to improve health care worker hand hygiene and were unable to
conduct a traditional evidence-based review.17

In conclusion, although it is well-accepted that hand hygiene is a critical patient safety
practice for reducing healthcare-associated infections, compliance with this practice is often low.
Well-developed tools are available for implementing hand hygiene interventions, although highquality evidence demonstrating which interventions are most effective is lacking. Reviews have
found that the results of hand hygiene compliance interventions were mixed, with effectiveness
70
waning over the long term. A recent systematic review focusing on higher quality evidence
found only four studies, three of which showed a significant impact. Another recent review
found mixed results for the impact of hand hygiene interventions on rates of healthcareassociated infections. A variety of interventions to improve hand hygiene are being implemented
and promoted by various U.S. and international organizations, particularly educational programs,
monitoring, and feedback. Interventions should be multimodal, addressing providers
knowledge, attitudes, and beliefs regarding hand hygiene, as well as strategies for behavioral
change, and should ideally be tailored to institutional needs as well as different provider groups
and health care situations. Health care administrators embarking on a hand hygiene intervention
should take advantage of the tools developed by the CDC and the WHO. New strategies, such as
patient engagement in hand-hygiene interventions, are an emerging area with only a few studies
assessing their effectiveness, and need further research on how best to implement them
effectively. Finally, research may be directed toward understanding the effectiveness of specific
elements of hand hygiene interventions, and the context in which they are implemented, in order
to understand which combinations lead most reliably to success. A summary table is located in
Table 1, Chapter 8.
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low
How Much do We
Moderate/Moderate
References
1.
Lautenbach E. Chapter 12. Practices to

Improve Handwashing Compliance. In
Making Healthcare Safer: A critical analysis
of patient safety practices. Shojania KG,
Duncan BW, McDonald KM, Wachter RM,
Markowitz AJ. Eds. Agency for Healthcare
Research and Quality. 2001.
http://archive.ahrq.gov/clinic/ptsafety.
Accessed November 17, 2011.
2.
Jarvis WR. Selected aspects of the

socioeconomic impact of nosocomial
infections: morbidity, mortality, cost, and
prevention. Infect Control Hosp Epidemiol
1996; 17(8):552-7.
3.
Klevens RM, Edwards JR, Richards CL Jr et

al. Estimating health care-associated
2002. Public Health Rep 2007;122(2):1606.
71
4.
Scott RD (2009) The direct medical costs of

healthcare-associated Infections in U.S.
hospitals and the benefits of prevention.
Atlanta, GA: Division of Healthcare Quality
Promotion National Center for
Preparedness, Detection, and Control of
Infectious Diseases Coordinating Center for
Infectious Diseases Centers for Disease
Control and Prevention.
5.
World Health Organization. WHO

Guidelines on Hand Hygiene in Health Care:
First Global Patient Safety Challenge. 2009.
www.who.int/gpsc/country_work/en/.
6.
Centers for Disease Control And Prevention.

Hand Hygiene in Health Care Settings:
Hand hygiene basics. May 19, 2011.
www.cdc.gov/handhygiene/Basics.html.
7.
National Quality Forum. Safe Practice 19:

Hand Hygiene. In Safe Practices for Better
Healthcare: 2010 Update.
/Safe_Practices_for_Better_Healthcare_%E
2%80%93_2010_Update.aspx. Accessed
November 18, 2011.
8.
Boyce JM. Measuring healthcare worker

hand hygiene activity: current practices and
emerging technologies. Infect Control Hosp
Epidemiol 2011; 32(10):1016-28.
9.
Gould DJ, Moralejo D, Drey N, Chudleigh

JH. Interventions to improve hand hygiene
compliance in patient care. Cochrane
Database Syst Rev 2010; (9):CD005186.
10.
Centers for Disease Control And Prevention.

Hand Hygiene in Health Care Settings:
Training. May 19, 2011
www.cdc.gov/handhygiene/training/interacti
veEducation/. Accessed November 16,
2011.
11.
Centers for Disease Control and Prevention.

Guideline for Hand Hygiene in Health-Care
Settings: Recommendations of the
Healthcare Infection Control Practices
Advisory Committee and the
HICPAC/SHEA/APIC/IDSA HandHygiene
Task Force. MMWR 2002;51(No. RR-16).
12.
Centers for Disease Control and Prevention.

Hand Hygiene in Healthcare SettingsCore2003.
www.cdc.gov/handhygiene/training.html.
13.
Institute for Healthcare Improvement. Howto Guide: Improving Hand Hygiene. 2006.
www.ihi.org/knowledge/Pages/ToolsHowto
GuideImprovingHandHygiene.aspx
14.
Joint Commission. Measuring Hand

Hygiene Adherance: Overcoming the
challenges. 2009.
www.jointcommission.org/Measuring_Hand
_Hygiene_Adherence_Overcoming_the_Ch
allenges_/
15.
Mathai E, Allegranzi B, Seto WH et al.

Educating healthcare workers to optimal
hand hygiene practices: addressing the need.
Infection 2010; 38(5):349-56.
72
16.
Backman C, Zoutman DE, Marck PB. An

integrative review of the current evidence on
the relationship between hand hygiene
interventions and the incidence of health
care-associated infections. Am J Infect
Control 2008; 36(5):333-48.
17.
McGuckin M, Storr J, Longtin Y, Allegranzi

B, Pittet D. Patient empowerment and
multimodal hand hygiene promotion: a winwin strategy. Am J Med Qual 2011;
26(1):10-7.
Chapter 9. Reducing Unnecessary Urinary Catheter Use and

Other Strategies To Prevent Catheter-Associated Urinary
Tract Infections: Brief Update Review
Jennifer Meddings M.D., M.Sc., Sarah L. Krein Ph.D., R.N., Mohamad G. Fakih M.D., M.P.H.,
Russell N. Olmsted M.P.H., C.I.C., Sanjay Saint M.D., M.P.H.
Introduction
Urinary tract infection has long been considered the most common healthcare-associated
infection (HAI), with the vast majority of these infections occurring after placement of the
convenient, often unnecessary,1-3 and easily forgotten urinary catheter.4 With an estimated one
million catheter-associated urinary tract infections5 (CAUTIs) per year, associated with an
additional cost of $676 per admission (or $2836 when complicated by bacteremia),6 it is not
surprising that CAUTIs were among the first hospital-acquired conditions selected for nonpayment by Medicare as of October 2008,7 and have been further targeted for complete
elimination8 as a never event, with a national goal to reduce CAUTI by 25% and reduce
urinary catheter use by 50% by 2014.9,10 These national initiatives renewed public and research
interest in the prevention of CAUTI, prompting updates of several comprehensive guidelines11-14
and reviews of strategies to prevent CAUTI released since the 2001 Making Health Care Safer
report.15
What Strategies May Prevent Catheter-Associated Urinary Tract

Infections?
Similar to other hospital-acquired infections such as central line-associated blood stream
infection (CLABSI) many CAUTI prevention strategies have been bundled into multimodal sets of interventions known as bladder bundles,16 consisting of educational interventions
to improve appropriate use and clinical skill in catheter placement, behavioral interventions such
as catheter restriction and removal protocols, and use of specific technologies such as the bladder
ultrasound. Despite some early success in implementing a bladder bundle16 to reduce urinary
catheterization rates,17 CAUTI prevention has proven challenging for several important reasons.
For example, monitoring urinary catheter use and CAUTI rates to inform and sustain urinary
catheter-related interventions is very resource intensive. Perhaps more importantly, improving
practice regarding urinary catheter placement and removal also requires interventions to change
the expectations and habits of nurses, physicians, and patients about the need for urinary
catheters.
To help organize and prioritize the many potential interventions to prevent CAUTI, we use
the conceptual model of the lifecycle of the urinary catheter18 to highlight that the highest yield
interventions to prevent CAUTI will target at least one of the four stages of the catheters
life. As illustrated in Figure 1, the lifecycle of the catheter (1) begins with its initial
placement, (2) continues when it remains in place, day after day, (3) ceases when it is removed,
and (4) may start over if another catheter is inserted after removal of the first one.
73
18
Figure 1, Chapter 9. Lifecycle of the urinary catheter

This conceptual model illustrates four stages of the urinary catheter lifecycle as targets for
interventions to decrease catheter use and catheter-associated urinary tract infections.
Meddings J, Saint S. Disrupting the Life Cycle of the Urinary Catheter. Clin Infect Dis. 2011; 52(11): 1291-3 by permission of
Oxford University Press.
Because avoiding unnecessary urinary catheter use is the most important goal in prevention
of CAUTI, this chapter reviews the evidence on two types of interventions that target
unnecessary urinary catheter use: (1) protocols and interventions to decrease unnecessary
placement of urinary catheters (catheter lifecycle stage 1), and (2) interventions that prompt
removal of unnecessary urinary catheters (catheter lifecycle stage 3).
The evidence summarized in this chapter was generated using a literature search conducted for a
prior systematic review and meta-analysis19 along with a focused update of the published peerreviewed literature (from August 2008 to February 2012) through a MEDLINE search for
intervention studies to reduce use of unnecessary urinary catheters in the acute care of adults. A
CINAHL database search was also performed for interventions developed and implemented by
nurses related to urinary catheter use. Studies were included if at least one outcome involving
catheter use or CAUTI events (Table 1) was reported as a result of the intervention, and with a
comparison group (either pre- vs. post-intervention or a separate control group).
74
19
Table 1, Chapter 9. Description of outcomes evaluated (adapted from the prior meta-analysis )
Number of CAUTI episodes per 1,000 catheter-days was recorded and a rate ratio was
calculated to compare pre- vs. post-intervention. When rates of both asymptomatic and
20
symptomatic CAUTI were reported separately, the rates of symptomatic CAUTI were used
19
for the meta-analysis.
Cumulative risk of CAUTI during hospitalization (i.e., the percentage of patients who
developed CAUTI) was also extracted for each study, and a risk ratio was calculated to
19
compare risks before and after the intervention for the meta-analysis.
Mean number of days of urinary catheter use per patient was recorded before and after
the intervention, and a standardized mean difference (SMD) was calculated to compare the
19
two groups for the meta-analysis.
Measures of Urinary Percentage of patient days in which the catheter was in place was calculated before
Catheter Use
and after the intervention, and a standardized mean difference (SMD) was determined for
19
each study for the meta-analysis.
Daily catheter prevalence, defined as the number of patients with catheters in place
during a specific time period, is reported for some of the more recent studies.
Need for Catheter
Re-catheterization need was extracted as the number and percent of patients who
Replacement
required replacement of a catheter after prior removal of an indwelling catheter.
The table in Appendix D summarizes the intervention studies described in this review, including study designs, patient
populations, and the interventions employed to avoid unnecessary catheter placement or to prompt catheter removal.
Meddings J, Rogers MA, Macy M, et al. Systematic review and meta-analysis: reminder systems to reduce catheter-associated
urinary tract infections and urinary catheter use in hospitalized patients. Clin Infect Dis. 2010;51(5):550-60 by permission of
Oxford University Press.
Measures of
Catheter-Associated
Urinary Tract
Infection (Cauti)
Development
What Strategies May Reduce Unnecessary Catheter Use?

Strategies To Avoid Unnecessary Placement of Indwelling Urinary
Catheters
Simply put, patients without urinary catheters do not develop CAUTI. Yet, multiple studies
show that between 21 and 63 percent1,3,21-24 of urinary catheters are placed in patients who do not
have an appropriate indication and therefore may not even need a catheter. Over the past decade,
several studies have employed interventions to decrease unnecessary catheter placement
(described in Appendix D Table). Although educational interventions are a common and
important first step to decrease inappropriate catheter use, more effective and potentially more
sustainable interventions go a step further by instituting restrictions on catheter placement.
Protocols that restrict catheter placement can serve as a constant reminder for providers about the
appropriate use of catheters, can suggest alternatives to indwelling catheter use (such as condom
catheters or intermittent straight catheterization), but perhaps most importantly, can generate
accountability for placement of each individual urinary catheter. A fairly typical approach for
developing a catheter restriction protocol is to begin with a basic list of appropriate catheter uses
(such as provided in the Centers for Disease Control and Preventions Healthcare Infection
Control Practices Advisory Committee (HICPAC) guideline11); this list (Table 2) can then be
tailored to include other indications based on local opinion and specialized patient populations.
75
Table 2, Chapter 9. Indications for indwelling urethral catheter use (from the 2009 CDCs
11
guideline )
A. Examples of Appropriate Indications for Indwelling Urethral Catheter Use
Patient has acute urinary retention or bladder outlet obstruction
Need for accurate measurements of urinary output in critically ill patients
Perioperative use for selected surgical procedures:
Patients undergoing urologic or other surgery on contiguous structures of genitourinary tract
Anticipated prolonged surgery duration; catheters inserted for this reason should be removed in post-anesthesia
care unit
Patients anticipated to receive large-volume infusions or diuretics during surgery
Need for intraoperative monitoring of urinary output
To assist in healing of open sacral or perineal wounds in incontinent patients
Patient requires prolonged immobilization (e.g., potentially unstable thoracic or lumbar spine, multiple traumatic
injuries such as pelvic fractures)
To improve comfort for end of life care if needed
B. Examples of Inappropriate Uses of Indwelling Catheters
As a substitute for nursing care of the patient or resident with incontinence
As a means to obtain urine for culture or other diagnostic tests when patient can voluntarily void
For prolonged postoperative duration without appropriate indications (e.g., structural repair of urethra or contiguous
structures, prolonged effect of epidural anesthesia, etc.)
The technology required to implement catheter placement restrictions ranges from low
technology strategies such as a hospital or unit policy on appropriate catheter placement or preprinted catheter orders with limited indications to higher technology strategies such as
computerized orders22,23,25 for catheter placement. Catheter restriction protocols have been a
common component of successful multi-modal interventions to decrease catheter use and/or
CAUTI rates, including hospital-wide23 interventions and interventions tailored for specific
environments such as the emergency department,21,26 inpatient units17 (including general
medical25,27,28-surgical29 wards and ICU29-33), and in the peri-procedural32 setting. Urinary
retention protocols22,28,29,32-34 are a type of catheter restriction protocols that often incorporate the
use of a portable bladder ultrasound22,28,32,34,35 to verify retention prior to catheterization, and
recommend use of intermittent catheterization rather than indwelling catheters to manage a
common and often temporary issue.
Strategies To Prompt Removal of Unnecessary Urinary Catheters

Urinary catheters are commonly left in place when no longer needed.3,24 In most hospitals,
four steps are required to remove a urinary catheter:18 (1) a physician recognizes the catheter is in
place, (2) the physician recognizes the catheter is no longer needed, (3) the physician writes the
order to remove catheter, and (4) a nurse removes the catheter. Thus, by default, hours and
sometimes days may pass before an unnecessary catheter is recognized and removed. Because
every additional day of urinary catheter use increases the patients risk of infectious and noninfectious catheter-related complications, interventions that facilitate prompt removal of
unnecessary catheters can have a strong impact. We describe below the evidence regarding
strategies that may accelerate or bypass some of these four steps to prompt catheter removal.
Perhaps the most important CAUTI prevention strategy after placement of the catheter is to
maintain awareness of the catheters existence (in lifecycle stage 2 of Figure 1), as health care
providers commonly forget the catheter is in place.4 Thus, a key step in prompting removal of
unnecessary catheters is frequently (by day or by shift) reminding nurses and physicians that the
catheter remains in place. Catheter reminder interventions include a daily checklist23,32,33,36,37 or
verbal/written reminder31,38-42 to assess continued catheter need, a sticker reminder on the
patients chart35,43,44 or catheter bag,45 or an electronic23 reminder that a catheter is still in place.
76
Reminder interventions can be generated by nurses, physicians or electronic order sets, and can
be targeted to remind either nurses or physicians about the catheter. Some reminder interventions
have employed nurses dedicated to detecting unnecessary catheters.23,35 Reminder interventions
can also serve to remind clinicians of appropriate catheter indications.
Unfortunately, reminder interventions can also be easy to ignore43 and catheters may remain
in place without action. The next type of intervention to prompt removal of unnecessary
catheters, which goes a step further, is a stop order that requires action. Stop orders prompt the
clinician (either nurse or physician) to remove the catheter by default after a certain time period
has elapsed or condition has occurred, unless the catheter remains clinically appropriate. For
example, catheter stop orders can be configured to expire in the same fashion as restraint or
antibiotic orders, unless action is taken by a clinician. Stop orders directed at
physicians23,25,28,30,42 require an order to be renewed or discontinued on the basis of review at
specific intervals, such as every 24 to 48 hours after admission or post-procedure. Stop orders
directed at nurses either require the nurse to obtain a catheter removal order from
physicians,27,32,46 or can empower nurses to remove the catheter without requesting a physician
order20,28,30,34,47-49 on the basis of an appropriate indication list. Admittedly, implementing a
nurse-empowered catheter removal protocol may be less effective than anticipated, as early
qualitative research of nurse-empowered interventions indicate some nurses are uncomfortable
with this autonomy49 and might not remove catheters as expected.
What Is the Impact of Strategies To Avoid Unnecessary Urinary

Catheter Use?
Impact of Interventions To Avoid Unnecessary Catheter Placement
Multiple before-and-after studies of interventions to decrease inappropriate catheter
placement (such as catheter placement restrictions and urinary retention protocols) have resulted
in a decrease in the use of urinary catheters,21-23,28,29,31,33 a lower proportion of catheters in place
without a physician order21,23,25,26 and a reduction in the proportion of catheters in place without
an appropriate indication.21,23,26,28
Impact of Reminder and Stop Order Interventions on Catheter Use

and CAUTIs
A systematic review and meta-analysis of 14 studies19 published prior to August 2008
(including nine reminder interventions and five stop order interventions) demonstrated that the
rate of CAUTI (episodes per 1,000 catheter-days) was reduced by 52 percent (p<0.001) with the
use of either a reminder or stop order. Based on this meta-analysis, reminders and stop orders
could result in large numbers of avoided CAUTI episodes per 1,000 catheter-days, particularly
when baseline rates of CAUTI are high (Table 3, adapted from a previous meta-analysis19).
Table 3, Chapter 9. Number of avoided CAUTI episodes per 1,000 catheter-days
Baseline rate of CAUTI
episodes per 1,000
catheter-days
5
10
20
Number of avoided CAUTI episodes per 1,000 catheter-days

anticipated by the type of intervention to prompt catheter removal
Reminder
Stop order
2.8
5.6
11.2
2.0
4.1
8.2
77
Overall
2.6 (95%CI, 1.63.6)
5.2 (95%CI, 3.2-7.2)
10.4 (95%CI, 6.4-14.4)
This meta-analysis19 also suggested that the mean duration of urinary catheterization
decreased by 37 percent, with 2.61 fewer days of catheterization per patient in the intervention
vs. control groups. The pooled standardized mean difference (SMD) in the duration of
catheterization was -1.11 overall (p=0.070); a statistically significant decrease in duration was
observed in studies that used a stop order (SMD -0.30; p=0.001) but not in those that used only a
reminder intervention (SMD -1.54; p=0.071).19 An update of the literature review since this
meta-analysis yielded 12 additional studies with reminder and/or stop order interventions. Figure
2 illustrates the major findings of the 14 studies for catheter use and CAUTI events as reported in
the prior meta-analysis;19 Figure 3 illustrates the major findings for the 12 subsequent studies,
including eight that reported measures of catheter use, and eight that reported CAUTI events.
78
Figure 2, Chapter 9. Summary of CAUTI and urinary catheter outcomes from 14 studies
Note: Summary comes from the 14 studies20,25,28-30,33,36-41,43,45 included in the 2010 meta-analysis.19
*Difference of p<0.05 reported between intervention and comparison group.
79
Figure 3, Chapter 9. Summary of CAUTI and urinary catheter outcomes from 12 additional studies
Note: Summary comes from 12 additional studies23,27,31,32,34,35,42,44,46-49 since the prior meta-analysis.19
*Difference of p<0.05 reported between intervention and comparison group.
80
Potential for Unintended Harm by Catheter Removal Interventions

Interventions that facilitate removal of urinary catheters do pose the risk of premature urinary
catheter removal, with patients then requiring unnecessary recatheterization; any catheterization
event is associated with procedure-related discomfort and other potential complications. Thus,
monitoring the need for re-catheterization is important to avoid unintended patient harm. In the
meta-analysis of reminder and stop order studies, only four of the 14 studies reported rates of recatheterization20,25,39,43 with low re-catheterization rates noted in both intervention and control
groups. None of the 12 more recent studies involving reminders or stop orders to prompt catheter
removal reported data on potential patient harm, such as premature removal.
Summary of Other Strategies To Prevent CAUTI

Several recent evidence-based guidelines11-14 have focused on preventing CAUTI and have
assessed the evidence and provided recommendations for implementing prevention strategies.
Key recommendations in the CDC guideline,11 in addition to appropriate catheter use (Table 2),
include (1) aseptic insertion of urinary catheters by properly trained personnel, using aseptic
technique and sterile equipment (with an exception being that clean technique is appropriate for
chronic intermittent catheterization), and (2) proper urinary catheter maintenance with a sterile,
closed drainage system permitting unobstructed urine flow. Aseptic insertion is primarily
recommended as a standard of care for which limited evidence exists. Stronger evidence
(epidemiological and clinical) supports the importance of a sterile, closed, unobstructed urinary
drainage system.
A more controversial topic is the use of antimicrobial catheters. Based on the current
evidence, the CDC guideline recommends11 that antimicrobial catheters should not be used
routinely to prevent CAUTI. It also suggests that further research is needed both on the effect of
silver-alloy coated catheters in reducing the risk of clinically significant CAUTI outcomes and
on the benefit of silver-alloy coated catheters in selected patients at high risk of infection.
Bundles of interventions are also an important strategy, as part of a multi-modal approach
that focuses efforts on high-yield interventions. For example, one strategy that includes several
of the components from the Bladder Bundle implemented by the Michigan Health and
Hospital Association (MHA) Keystone Center for Patient Safety & Quality is the ABCDE
approach:16
Adherence to general infection control principles is important (e.g., hand hygiene,
surveillance and feedback, aseptic insertion, proper maintenance, education).
Bladder ultrasound may avoid indwelling catheterization.
Condom catheters or other alternatives to an indwelling catheter such as intermittent
catheterization should be considered in appropriate patients.
Do not use the indwelling catheter unless you must!
Early removal of the catheter using a reminder or nurse-initiated removal protocol
appears warranted.
What Is the Cost of Implementing a CAUTI Prevention Program?

The cost of implementing a CAUTI prevention program will vary based on the level of
technology used (e.g., computerized vs. pre-printed catheter orders, and whether portable bladder
ultrasounds are purchased) and the time invested in implementing and evaluating the
interventions. Saint and colleagues, in their study of a written urinary catheter reminder
81
generated by a research nurse to remind physicians which of their inpatients had urinary
catheters,43 found that the intervention was either cost-neutral or modestly cost-saving depending
on the assumptions made. More recently, a study35 of five hospitals in the Netherlands employed
a multi-modal intervention including reminders in four hospitals, and a stop order in the fifth
hospital. The program was found to be cost-saving, with the mean amount saved being 537 (or
~$700) per 100 hospitalized patients.
What Methods Have Been Used To Improve the Implementation of

Interventions To Prevent Catheter-Associated Urinary Tract
Infections?
Because reducing unnecessary catheter use often requires changing well-established habits
and beliefs of nurses and physicians, the challenge of implementation should not be underestimated. To facilitate implementation of practices to prevent CAUTI, the Michigan Keystone
Bladder Bundle Initiative16 used the Johns Hopkins University collaborative model for
transformational change. This model is based in part on the four Es: Engage, Educate,
Execute, and Evaluate.50 During the Engage and Educate steps, hospitals were provided
information in multiple formats and a toolkit describing the intervention steps and outcomes
measures. In the Execute step, the hospital was strongly encouraged to choose one nurse
champion51 (for example, a case manager, nurse coordinator, or clinical nurse specialist) to lead
the initiative and organize a bladder bundle team, including at least one physician, and to
participate in workshops and conference calls with other participating hospitals to provide
additional expert content and practical coaching. Also during the Execute step, daily patient
rounds (which in some hospitals were called a catheter patrol) were recommended to assess
catheter presence and necessity, and provide feedback to specific units and re-evaluate strategies
in progress. Hospitals were also encouraged to implement more active strategies for prevention,
such as a catheter reminder system or promoting the use of catheter alternatives by developing
protocols or making sure the necessary supplies were readily available. In the Evaluate phase,
hospitals were asked to conduct a baseline assessment of catheter use (point prevalence) and
appropriate use according to specified indications and to conduct periodic reassessments to
assess progress and sustainability.
Implementation challenges within CAUTI prevention should be expected52 and managed
accordingly. Qualitative assessment focusing on HAI prevention has identified two important
potential barriers to healthcare-associated infection preventive efforts: active resisters and
organizational constipators.53 Active resisters are hospital personnel who vigorously and
openly oppose changes in practice, as a matter of habit or culture (e.g., just not how they were
trained). Management of active resisters often requires those in authority to mandate
compliance, collect data, and provide feedback.53 A champion who is influential or a peer of
the resisting staff may also help to overcome active resistance.51,52 Organizational constipators
are usually mid- or high-level executives who act as barriers to change by preventing or delaying
certain actions needed to implement new practices.53 Strategies to address an organizational
constipator are to include this person in early discussions to improve buy-in and motivation,
working around the person, or replacing the constipator.
A unique challenge to expect when implementing urinary catheter removal strategies is
reluctance by some nurses to remove the catheter,52 even when the nurse is empowered to do
so. In some cases, nurses may be active resisters due to disagreement with the catheter policy
82
and/or a desire to avoid the inconveniences and increased frequency of patient contact required
for the care of incontinence and catheter alternatives. Other nurses report they simply do not feel
comfortable49 removing the catheter without explicit orders from the physician, which is ironic
considering that many nurses place catheters without orders. Nursing comfort with catheter
removal can be increased49 with peer support and education, and may be facilitated by directly
addressing the workload concerns associated with the removal of indwelling catheters. Indeed, a
survey of nurses27 during implementation of a nurse-empowered catheter removal protocol
indicated increased nursing and patient satisfaction, despite the expected increase in workload.
Even though CAUTI is a very common healthcare-associated infection, Krein and colleagues
reported that CAUTI preventive practice use is lagging behind efforts to prevent central lineassociated bloodstream infection and ventilator-associated pneumonia,54 with room for
improvement in adopting catheter removal and CAUTI preventive strategies demonstrated again
in two recent large surveys of hospitals55 and ICUs.56 Fortunately, many resources exist
(www.Catheterout.org) to help hospitals develop and implement programs to decrease catheter
use and prevent CAUTI, including a range of tools and educational materials to address
implementation challenges. Hospital and unit-level leadership also play a key role in preventing
infection.57
Monitoring and Providing Feedback on Catheter Use and CAUTI

Rates
Inappropriate urinary catheter use is an easy habit to start and a difficult one to break.18
Consequently, many studies17,30 have emphasized the importance of on-going surveillance and
feedback as an intervention to reduce healthcare-associated infections such as CAUTI and
sustain prevention efforts. New national efforts to reduce CAUTI
(www.onthecuspstophai.org/stop-cauti/) incorporate periodic feedback to participating units on
urinary catheter use and CAUTI rates. The CAUTI rates evaluated include the National
Healthcare Safety Network (NHSN) and the newly described population-based rates.58 The
population-based CAUTI rate incorporates both the NHSN rate and the device utilization ratio,
to account for interventions focused on reduction in catheter use and improvements in placement
and maintenance.
Important next steps to address CAUTI involve developing strategies to decrease the effort
and resources required to monitor catheter use and CAUTI rates. Advanced informatics tools
have recently been shown to increase the impact of this feedback loop to the extent that rates of
CAUTI were lower in facilities that deployed these tools compared with those that did not.59
Careful selection or development of datasets used for implementing hospital payment changes
and public reporting for CAUTI events is also recommended. Unfortunately, the current
administrative data used to implement non-payment7 for hospital-acquired CAUTIs and to
publicly report hospital performance likely captures few CAUTI events, given documentation
and coding challenges60 to translate a urinary tract infection event from a medical record into
hospital-acquired CAUTI in the administrative datasets.

In summary, hospitals should strongly consider employing interventions to avoid
unnecessary catheter placement and to prompt removal of unnecessary catheters. These
interventions appear to be low cost, low risk and effective strategies to address a common
83
hospital-acquired infection in the United States, with some unique but not impossible challenges
for implementation. Furthermore, reducing indwelling catheter use addresses the noninfectious
complications of urinary catheter use such as catheter-related patient discomfort and immobility
(Table 4).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Moderate-tohigh
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low
How Much do We
Moderate/Moderate
References
1.
Gokula RR, Hickner JA, Smith MA.

Inappropriate use of urinary catheters in
elderly patients at a midwestern community
teaching hospital. Am J Infect Control.
2004;32(4):196-9. PMID 15175612.
2.
Fakih MG, Shemes SP, Pena ME, et al.

Urinary catheters in the emergency
department: very elderly women are at high
risk for unnecessary utilization. Am J Infect
Control. 2010;38(9):683-8. PMID
21034978.
3.
Jain P, Parada JP, David A, et al. Overuse of

the indwelling urinary tract catheter in
hospitalized medical patients. Arch Intern
Med. 1995;155(13):1425-9. PMID 7794092.
4.
Saint S, Wiese J, Amory JK, et al. Are

physicians aware of which of their patients
have indwelling urinary catheters? Am J
Med. 2000;109(6):476-80. PMID 11042237.
5.
Foxman B. Epidemiology of urinary tract

infections: incidence, morbidity, and
economic costs. Am J Med. 2002;113 Suppl
1A:5S-13S. PMID 12113866.
6.
Saint S. Clinical and economic

consequences of nosocomial catheter-related
bacteriuria. Am J Infect Control.
2000;28(1):68-75. PMID 10679141.
7.
Medicare program; changes to the hospital

inpatient prospective payment systems and
fiscal year 2009 rates. Fed Regist.
2008;73(161):48473-91. PMID 18956499.
84
8.
Cardo D, Dennehy PH, Halverson P, et al.

Moving toward elimination of healthcareassociated infections: a call to action. Am J
Infect Control. 2010;38(9):671-5. PMID
21058460.
9.
Department of Health and Human Services

Action Plan to Prevent HealthcareAssociated Infections, Appendix G. 2009;
www.hhs.gov/ash/initiatives/hai/actionplan/
hhs_hai_action_plan_final_06222009.pdf.
Accessed February 28, 2012.
10.

National Targets and Metrics: Monitoring
Progress Toward Action Plan Goals: A MidTerm Assessment. 2011;
www.hhs.gov/ash/initiatives/hai/nationaltarg
ets/index.html.
11.
Gould CV, Umscheid CA, Agarwal RK, et

al. Guideline for prevention of catheterassociated urinary tract infections 2009.
Infect Control Hosp Epidemiol.
2010;31(4):319-26. PMID 20156062.
12.
Hooton TM, Bradley SF, Cardenas DD, et

al. Diagnosis, prevention, and treatment of
catheter-associated urinary tract infection in
adults: 2009 International Clinical Practice
Guidelines from the Infectious Diseases
Society of America. Clin Infect Dis.
2010;50(5):625-63. PMID 20175247.
13.
Lo E, Nicolle L, Classen D, et al. Strategies

to prevent catheter-associated urinary tract
Control Hosp Epidemiol. 2008;29 Suppl
1:S41-50. PMID 18840088.
14.
Greene L, Marx J, Oriola S. An APIC guide

to the elimination of catheter-associated
urinary tract infections (CAUTIs):
developing and applying facility-based
prevention interventions in acute and longterm care settings. 2008.
23.
Knoll BM, Wright D, Ellingson L, et al.

Reduction of inappropriate urinary catheter
use at a Veterans Affairs hospital through a
multifaceted quality improvement project.
Clin Infect Dis. 2011;52(11):1283-90. PMID
21596671.
15.
Saint S. Chapter 15. Prevention of

Nosocomial Urinary Tract Infections.
Making Health Care Safer: A Critical
Analysis of Patient Safety Practices.
Evidence Report/Technology Assessment,
No. 43. AHRQ Publication No. 01-E058.
Quality, Rockville, MD. 2001.
www.ahrq.gov/clinic/ptsafety/pdf/chap15.pd
f. Accessed February 28, 2012.
24.
Gardam MA, Amihod B, Orenstein P, et al.

Overutilization of indwelling urinary
catheters and the development of
nosocomial urinary tract infections. Clin
Perform Qual Health Care. 1998;6(3):99102. PMID 10182561.
25.
Cornia PB, Amory JK, Fraser S, et al.

Computer-based order entry decreases
duration of indwelling urinary
catheterization in hospitalized patients. Am J
Med. 2003;114(5):404-7. PMID 12714131.
26.
Gokula RM, Smith MA, Hickner J.

Emergency room staff education and use of
a urinary catheter indication sheet improves
appropriate use of foley catheters. Am J
17980237.
27.
Rothfeld AF, Stickley A. A program to limit

urinary catheter use at an acute care hospital.
Am J Infect Control. 2010;38(7):568-71.
PMID 20381918.
28.
Topal J, Conklin S, Camp K, et al.

Prevention of nosocomial catheterassociated urinary tract infections through
computerized feedback to physicians and a
nurse-directed protocol. Am J Med Qual.
2005;20(3):121-6. PMID 15951517.
29.
Stephan F, Sax H, Wachsmuth M, et al.

Reduction of urinary tract infection and
antibiotic use after surgery: a controlled,
prospective, before-after intervention study.
16652311.
30.
Dumigan DG, Kohan CA, Reed CR, et al.

Utilizing national nosocomial infection
surveillance system data to improve urinary
tract infection rates in three intensive-care
units. Clin Perform Qual Health Care.
1998;6(4):172-8. PMID 10351284.
31.
Elpern EH, Killeen K, Ketchem A, et al.

Reducing use of indwelling urinary catheters
and associated urinary tract infections. Am J
Crit Care. 2009;18(6):535-41; quiz 42.
PMID 19880955.
16.
17.
18.
19.
20.
Saint S, Olmsted RN, Fakih MG, et al.

Translating health care-associated urinary
tract infection prevention research into
practice via the bladder bundle. Jt Comm J
Qual Patient Saf. 2009;35(9):449-55. PMID
19769204.
Fakih MG, Watson SR, Greene MT, et al.
Reducing inappropriate urinary catheter use:
a statewide effort. Arch Intern Med.
2012;172(3):255-60. PMID 22231611.
Meddings J, Saint S. Disrupting the life
cycle of the urinary catheter. Clin Infect Dis.
2011;52(11):1291-3. PMID 21596672.
Meddings J, Rogers MA, Macy M, et al.
Systematic review and meta-analysis:
reminder systems to reduce catheterassociated urinary tract infections and
urinary catheter use in hospitalized patients.
20673003.
Loeb M, Hunt D, OHalloran K, et al. Stop
orders to reduce inappropriate urinary
catheterization in hospitalized patients: a
randomized controlled trial. J Gen Intern
Med. 2008;23(6):816-20. PMID 18421507.
21.
Fakih MG, Pena ME, Shemes S, et al. Effect

of establishing guidelines on appropriate
urinary catheter placement. Acad Emerg
Med. 2010;17(3):337-40. PMID 20370769.
22.
Patrizzi K, Fasnacht A, Manno M. A

collaborative, nurse-driven initiative to
reduce hospital-acquired urinary tract
infections. J Emerg Nurs. 2009;35(6):536-9.
PMID 19914479.
85
32.
Fuchs MA, Sexton DJ, Thornlow DK, et al.

Evaluation of an evidence-based, nursedriven checklist to prevent hospital-acquired
catheter-associated urinary tract infections in
intensive care units. J Nurs Care Qual.
2011;26(2):101-9. PMID 21037484.
41.
Fakih MG, Dueweke C, Meisner S, et al.

Effect of nurse-led multidisciplinary rounds
on reducing the unnecessary use of urinary
catheterization in hospitalized patients.
2008;29(9):815-9. PMID 18700831.
33.
Reilly L, Sullivan P, Ninni S, et al.

Reducing foley catheter device days in an
intensive care unit: using the evidence to
change practice. AACN Adv Crit Care.
2006;17(3):272-83. PMID 16931923.
42.
34.
Schultz P, Aljawawdeh A, Hopp T. EB105:

with a nurse-driven protocol. Crit Care
Nurse. 2011;31(2):e42.
Seguin P, Laviolle B, Isslame S, et al.

Effectiveness of simple daily sensitization of
physicians to the duration of central venous
and urinary tract catheterization. Intensive
Care Med. 2010;36(7):1202-6. PMID
20237761.
43.
Saint S, Kaufman SR, Thompson M, et al. A

reminder reduces urinary catheterization in
hospitalized patients. Jt Comm J Qual
Patient Saf. 2005;31(8):455-62. PMID
16156193.
44.
Bruminhent J, Keegan M, Lakhani A, et al.

Effectiveness of a simple intervention for
prevention of catheter-associated urinary
tract infections in a community teaching
hospital. Am J Infect Control.
2010;38(9):689-93. PMID 21034979.
45.
Murphy D, Francis K, Litzenberger M, et al.

Reducing urinary tract infection: a nurseinitiated program. Pa Nurse. 2007;62(4):20.
PMID 18286841.
46.
Robinson S, Allen L, Barnes MR, et al.

Development of an evidence-based protocol
for reduction of indwelling urinary catheter
usage. Medsurg Nurs. 2007;16(3):157-61.
PMID 17849921.
47.
Gotelli JM, Merryman P, Carr C, et al. A

quality improvement project to reduce the
complications associated with indwelling
urinary catheters. Urol Nurs.
2008;28(6):465-7, 73. PMID 19241786.
48.
Voss AB. Incidence and duration of urinary

catheters in hospitalized older adults: before
and after implementing a geriatric protocol.
J Gerontol Nurs. 2009;35(6):35-41. PMID
19537293.
49.
Wenger JE. Cultivating quality: reducing

rates of catheter-associated urinary tract
infection. Am J Nurs. 2010;110(8):40-5.
PMID 20671495.
50.
Pronovost PJ, Berenholtz SM, Needham

DM. Translating evidence into practice: a
model for large scale knowledge translation.
BMJ. 2008;337:a1714. PMID 18838424.
35.
36.
van den Broek PJ, Wille JC, van Benthem

BHB, et al. Urethral catheters: can we
reduce use? BMC Urol. 2011;11:10. PMID
21605403.
Jain M, Miller L, Belt D, et al. Decline in
ICU adverse events, nosocomial infections
and cost through a quality improvement
initiative focusing on teamwork and culture
change. Qual Saf Health Care.
2006;15(4):235-9. PMID 16885246.
37.
Weitzel T. To cath or not to cath? Nurs.

2008;38(2):20-1. PMID 18223406.
38.
Apisarnthanarak A, Thongphubeth K,
Sirinvaravong S, et al. Effectiveness of
multifaceted hospitalwide quality
improvement programs featuring an
intervention to remove unnecessary urinary
catheters at a tertiary care center in
Thailand. Infect Control Hosp Epidemiol.
2007;28(7):791-8. PMID 17564980.
39.
40.
Crouzet J, Bertrand X, Venier AG, et al.

Control of the duration of urinary
catheterization: impact on catheterassociated urinary tract infection. J Hosp
Infect. 2007;67(3):253-7. PMID 17949851.
Huang WC, Wann SR, Lin SL, et al.
Catheter-associated urinary tract infections
in intensive care units can be reduced by
prompting physicians to remove
unnecessary catheters. Infect Control Hosp
Epidemiol. 2004;25(11):974-8. PMID
15566033.
86
51.
Damschroder LJ, Banaszak-Holl J,

Kowalski CP, et al. The role of the
champion in infection prevention: results
from a multisite qualitative study. Qual Saf
Health Care. 2009;18(6):434-40. PMID
19955453.
52.
Saint S, Kowalski CP, Forman J, et al. A

multicenter qualitative study on preventing
hospital-acquired urinary tract infection in
US hospitals. Infect Control Hosp
Epidemiol. 2008;29(4):333-41. PMID
18462146.
53.
Saint S, Kowalski CP, Banaszak-Holl J, et

al. How active resisters and organizational
constipators affect health care-acquired
infection prevention efforts. Jt Comm J Qual
Patient Saf. 2009;35(5):239-46. PMID
19480375.
54.
Krein SL, Kowalski CP, Hofer TP, et al.

Preventing hospital-acquired infections: a
national survey of practices reported by U.S.
hospitals in 2005 and 2009. J Gen Intern
Med. 2011 Dec 6 (online). PMID 22143455.
55.
Fink R, Gilmartin H, Richard A, et al.

Indwelling urinary catheter management and
catheter-associated urinary tract infection
prevention practices in Nurses Improving
Care for Healthsystem Elders hospitals. Am
J Infect Control. 2012 Jan 30 (online).
PMID 22297241.
56.
Conway LJ, Pogorzelska M, Larson E, et al.

Adoption of policies to prevent catheterassociated urinary tract infections in United
States intensive care units. Am J Infect
Control. 2012 Feb 6 (online). PMID
22317857.
57.
Saint S, Kowalski CP, Banaszak-Holl J, et

al. The importance of leadership in
preventing healthcare-associated infection:
results of a multisite qualitative study. Infect
Control Hosp Epidemiol. 2010;31(9):901-7.
PMID 20658939.
58.
Fakih MG, Todd Greene M, Kennedy EH, et

al. Introducing a population-based outcome
measure to evaluate the effect of
interventions to reduce catheter-associated
urinary tract infection. Am J Infect Control.
2011 Aug 24 (online). PMID 21868133.
87
59.
Palumbo AJ, Loveless PA, Moll ME, et al.

Evaluation of healthcare-associated
infection surveillance in Pennsylvania
hospitals. Infect Control Hosp Epidemiol.
2012;33(2):105-11. PMID 22227977.
60.
Meddings J, Saint S, McMahon LF, Jr.

Hospital-acquired catheter-associated
urinary tract infection: documentation and
coding issues may reduce financial impact
of Medicares new payment policy. Infect
Control Hosp Epidemiol. 2010;31(6):62733. PMID 20426577.
Chapter 10. Prevention of Central Line-Associated

Bloodstream Infections: Brief Update Review
Vineet Chopra, M.D., M.Sc.; Sarah L. Krein, Ph.D., R.N.; Russell N. Olmsted, M.P.H., C.I.C.;
Nasia Safdar, M.D., Ph.D.; Sanjay Saint, M.D., M.P.H.
Introduction
Central venous catheters (CVCs) are intravascular access devices that terminate within the
great vessels of the neck (superior or inferior vena cava, brachiocephalic veins, subclavian vein
or internal jugular vein), or a site proximal to the heart. CVCs are vital for the care of
hospitalized and critically ill patients, as they provide reliable venous access for clinical activities
such as blood sampling, infusion of medications, and hemodynamic measurement. However,
CVCs are also the leading cause of healthcare-associated bloodstream infections (BSIs) and are
frequently implicated in life-threatening illnesses.1,2 Infections associated with CVCs are
categorized in the literature as either central-line associated bloodstream infection (CLABSI),
or catheter-related bloodstream infection (CRBSI), based on whether surveillance or
ascertainment of infection is the desired goal. For instance, the Centers for Disease Control and
Preventions (CDC) National Healthcare Safety Network (NHSN) uses the CLABSI definition
for surveillance purposes, defining the term as a laboratory confirmed BSI in any patient with a
CVC present either at the time of, or within a 48-hour period before the detection of infection.3,4
Thus, the CDC-NSHN definition overestimates the true incidence of CRBSI, as some BSIs may
be due to infection at other sites (e.g., pneumonia or urinary tract infection) or at sites that are
difficult to detect (e.g., translocation from the gastrointestinal tract or mucositis following
chemotherapy). In contrast, CRBSI is a more precise and rigorous definition that requires either
(a) isolation of the same organism from the catheter and the peripheral blood, (b) simultaneous
quantitative blood cultures with a ratio of 5:1 or higher of those from the indwelling CVC
compared with peripheral blood, or (c) a differential time to positivity of CVC-derived versus
(vs.) peripheral blood culture positivity of more than 2 hours.5 The CRBSI definition is thus
largely used within the context of clinical care and research, whereas the term, CLABSI, is
implemented for epidemiologic surveillance.6 For the purposes of this review, we use the term
CLABSI to encompass both of these operational definitions.
Of the approximately 249,000 BSIs that occur in U.S. hospitals each year, 80,000 (32.2%)
occur in intensive care unit (ICU) settings.2 Because CVCs are more frequently used in ICUs
than in other areas of the hospital, and the strongest predictor of developing a BSI is the presence
of a CVC, the epidemiology of CLABSI has traditionally focused on the critically ill. With over
15 million catheter days in ICUs annually, the potential impact of CLABSI is substantial in this
population alone.6,7 However, in a survey of major medical centers, CVC use was identified in
24.4 percent of patients outside the ICU.8 Thus, millions of patients both in and out of ICU
settings are potentially at risk of developing CLABSI. Although the frequency of CLABSI
outside the ICU is largely unknown, Weber and colleagues found that the incidence of CLABSI
decreased when patients transitioned from ICUs to step-down units or non-ICU floors.9 Data
from the CDC-NHSN also suggest lower CLABSI rates in patients on hospital wards compared
with those in an ICU setting.10 Furthermore, recent evidence suggests that the incidence of
CLABSI in ICUs is significantly lesser than reported in 2001, likely due to a number of efforts
aimed at preventing this infection.11 These efforts notwithstanding, the increasing use of CVCs
88
such as peripherally inserted central catheters (PICCs) outside of ICUs may reflect an important
shift in the epidemiology of CLABSI to non-ICU settings.12 This change is highly relevant, as
lack of a uniform patient-care team and absence of comprehensive surveillance efforts in nonICU settings represent substantial obstacles to addressing CLABSI in these areas.
The economic burden of CLABSI is substantial. A recent analysis estimated that each
CLABSI episode independently increases length of hospitalization from 7 to 21 days, and adds
an attributable cost of about $37,000 (2002 dollars) per patient.13 The annual national cost of
caring for patients who develop CLABSI is estimated to range from $0.67 to $2.68 billion.13-15
Similar trends exist in European nations, where the incremental expenditure related to CLABSI
is estimated at 9,154 (18,241 [$24,558 in 2012 dollars vs. 9,087 [$12233]) per patient.16
Given this clinical and economic cost, investigators, policy-makers, and regulatory agencies in
the U.S. and abroad have devoted great efforts to curtail CLABSI over the past decade.17-19
CLABSIs are potentially preventable through the use of evidence-based practices.20 The
original Making Health Care Safer report examined the prevalence, strategies, and costs
associated with CLABSI prevention, and found that certain practices (e.g., the use of maximal
sterile precautions) were associated with both a decrease in CLABSI risk and reduced cost,
whereas others (e.g., intravenous antimicrobial prophylaxis) added expense without clear
benefit.21,22 In this review, we provide an update to the original report by highlighting the most
clinically and cost effective strategies associated with CLABSI prevention. To compile this
report, we performed a systematic review of the literature and searched multiple databases to
identify relevant studies published between 2000 and 2012 using terms such as Bacteremia,
Catheterization, Central Venous, and central line-associated bloodstream infection. Our
search strategy yielded a total of 1,087 unique manuscripts of which 337 articles were relevant
for this report.
What Practices Are Associated With CLABSI Prevention?

One of the most important advances in the science of CLABSI prevention has been the
identification of individual risk factors associated with this condition. These include (a) lengthy
hospitalization before venous catheterization; (b) prolonged duration of catheterization; (c) heavy
microbial colonization at the insertion site; (d) heavy microbial colonization of the catheter hub;
(e) femoral or internal jugular vein insertion (rather than subclavian vein); (f) operatorinexperience or lack of implementation of best practices during CVC insertion; (g) presence of
neutropenia; (h) total parenteral nutrition provided through the catheter; (i) inadequate
care/maintenance of the CVC after insertion; and (j) type of CVC.23-30 Strategies to prevent
CLABSI have evolved from targeting these variables.
The CDs Healthcare Infection Control Practices Advisory Committee (HICPAC) recently
updated their guidelines to summarize the evidence behind a number of practices associated with
CLABSI reduction.20 As with prior iterations, the update provides levels of recommendation for
each clinical practice based on the theoretical rationale, scientific data, applicability and impact
of the intervention. Based on the level of evidence in their support, recommendations are divided
into five categories, ranging from practices that are strongly recommended and supported by
well-designed experimental, clinical, or epidemiologic studies to those that are of unclear value
owing to insufficient evidence or lack of consensus regarding efficacy (Table 1). From a
conceptual standpoint, these practices can be classified as (a) interventions that may be
implemented at the time of CVC insertion; (b) practices best utilized after placement of a CVC;
and (c) institutional initiatives to reduce CLABSI.
89
Table 1, Chapter 10. Categories and recommendations for CLABSI reduction practices from the Healthcare Infection Control Practices
*
Advisory Committee of the Centers for Disease Control and Prevention
Recommendation
Hand hygiene prior to catheter
insertion
All inclusive catheter carts or kits
Maximal sterile barrier precautions

Chlorhexidine for skin anti-sepsis
Antimicrobial catheters
Subclavian vein insertion

Disinfect hubs and needle-less
connectors
Remove non-essential CVCs
Chlorhexidine cleansing
CVC dressing
Chlorhexidine sponge dressing
Topical antibiotic use

Antibiotic or anti-infective locks
Systemic antibiotic prophylaxis
Description
Decontaminate hands with either antiseptic-containing soaps or alcohol-based
gels/foams before inserting, repairing, replacing, or dressing a CVC
A catheter kit or cart contains all the equipment necessary for CVC insertion
(needle, guidewire, introducers, etc.), and ensures sterility by minimzing
interruptions during line placement
Use a cap, mask, sterile gown, sterile gloves, and a sterile full body drape when
inserting CVCs and PICCs or performing guidewire exchange(s)
Prepare clean skin with chlorhexidine preparation with alcohol before CVC
insertion and during dressing changes
Chlorhexidine/silver sulfadiazine or minocycline/rifampin-impregnated CVCs are
recommended only if the catheter is expected to remain in place 5 days or more
AND the CVC will be inserted in an environment where the CLABSI rate remains
high despite a comprehensive reduction strategy
Whenever possible, use the subclavian site, rather than the jugular or femoral
sites in adults
Minimize contamination risk by scrubbing the access site with an appropriate
antiseptic (chlorhexidine, povidone iodine, or 70% alcohol) prior to accessing the
CVC
Daily evaluation and prompt removal of CVCs that are no longer clinically
warranted is an important aspect of CLABSI prevention; routine replacement of
CVCs, PICCs, or hemodialysis catheters is not recommended
Daily cleansing using a 2% chlorhexidine solution or impregnated washcloth
rather than soap and water in ICU- and hemodialysis patients is recommended
Use either sterile gauze or sterile, transparent, semipermeable dressing to cover
the CVC site
The use of chlorhexidine-impregnated sponge dressings is recommended for
patients >2 months of age if the CLABSI rate is not decreasing despite
adherence to basic prevention measures, including education and training,
appropriate use of chlorhexidine for skin antisepsis and use of maximal sterile
barrier precautions
Topical antibiotic use may promote fungemia or bacteremia in non-dialysis
populations and is recommended only for hemodialysis catheter dressing
Instillation of supra-physiologic doses of an intravenous antibiotic or anti-infective
solution into a catheter lumen between periods of CVC access is recommended
only in those at high baseline risk for CLABSI
The use of oral or intravenous antibiotic therapy either during insertion or
following placement of a CVC is not recommended
90
Category of Recommendation
Category IB
Category IB
Category IB
Category IA
Category IA
Category IB
Category IA
Category IA
Category II
Category IA
Category 1B
Category IB
Category II
Category IB
Table 1, Chapter 10. Categories and recommendations for CLABSI reduction practices from the Healthcare Infection Control Practices
*
Advisory Committee of the Centers for Disease Control and Prevention (continued)
Recommendation
Educational interventions
Catheter bundles or checklists
Description
Education regarding appropriate indications, method of placement, and
surveillance for CLABSI are a critical component of a comprehensive CLABSI
prevention program
The use of five practices in unison at the time of CVC insertion, the bundle, is
recommended. These interventions include hand hygiene prior to insertion; use
of maximal sterile barrier precautions; chlorhexidine for skin antisepsis;
avoidance of the femoral site of insertion; and prompt removal of catheters when
no longer indicated
The use of trained personnel dedicated to the placement of CVCs in ICU and
hospitalized patients is recommended
Category of Recommendation
Category IA
Category IB
Use of specialized CVC insertion

Category IA
teams
*
Categories of recommendations:
Category IA: Strongly recommended for implementation and strongly supported by well-designed experimental, clinical, or epidemiologic studies;
Category IB: Strongly recommended for implementation and supported by some experimental, clinical, or epidemiologic studies and a strong theoretical rationale; or an accepted
practice (e.g., aseptic technique) supported by limited evidence.
Category IC: Required by State or Federal regulations, rules, or standards.
Category II: Suggested for implementation and supported by suggestive clinical or epidemiologic studies or a theoretical rationale.
Unresolved Issue: No specific recommendations exist due to conflicting or insufficient evidence.
Abbreviations: CLABSI = central line-associated bloodstream infection; CVC = central venous catheter; PICC = peripherally inserted central catheter.
Note: Adapted from OGrady, et al.20
91
Measures To Prevent CLABSI at Time of Central Venous Catheter

Insertion
Hand hygiene before catheter insertion. Hand hygiene is an important practice in the
prevention of CLABSI.31 Hand decontamination with either antiseptic-containing soaps or
alcohol-based gels/foams has consistently been shown to reduce CLABSI rates.32-34 A key
strategy in promoting hand hygiene involves educating staff who insert CVCs on the importance
of this practice. In a before-and-after study assessing the impact of an educational initiative on
hand hygiene, the incidence of CLABSI decreased from 3.9 per 1,000 catheter days to 1.0 per
1,000 catheter days (P< 0.001) following education on this topic in an ICU setting.35 As the most
common cause of CLABSI is entry of skin pathogens during CVC insertion and maintenance,
ensuring best practice during catheter placement and handling is crucial for CLABSI prevention
(Category IB).20,36
All-inclusive catheter-carts or kits. A study by Young and colleagues found that a systemsbased intervention featuring a catheter kit (that contained a large sterile drape and 2%
chlorhexidine gluconate) led to a significant reduction in CLABSI rates (11.3 per 1,000 CVCdays vs. 3.7 per 1000 CVCs, P<0.01) in a medical-surgical ICU.37 This approach has been
expanded upon by a number of other investigators to include not only a kit of essential items, but
a mobile cart that contains all of the equipment needed to insert CVCs.38,39 The use of an allinclusive cart or catheter kit minimizes interruptions related to non-availability of necessary
equipment and thus lends itself to CLABSI reduction by ensuring maintenance of a sterile field
during catheter insertion. Furthermore, the use of carts encourages a consistent approach to CVC
insertion by standardizing catheter types, guide-wires, needles, and other essential supplies.
Although the use of catheter-carts and kits is not specifically endorsed by the recent HICPAC
guidelines,20 they are pragmatic, relatively low-cost innovations that have been associated with
lower CLABSI rates.
Maximal sterile barrier precautions. Several studies have demonstrated that the use of
maximal barrier precautions including a cap, mask, sterile gown, gloves, and a sterile full-body
drape when inserting CVCs reduces CLABSI.6,37,40,41 Current HICPAC guidelines thus
recommend that maximal sterile barriers are used during insertion of all CVCs (Category IB).20
The cost-effectiveness of this practice in preventing CLABSI has been established, as the
expense of sterile barriers is dwarfed by the additional expense of CLABSI and its subsequent
care, even in resource-poor environments.42 Despite this evidence, a study of 10 ICUs in major
academic medical centers published in 2006 reported that fewer than 30 percent had
systematically adopted maximal sterile barrier precautions.43 However, a more recent national
survey of infection preventionists, which examined the use of evidence-based practices
(including maximal sterile barriers) in Federal and non-Federal hospitals between 2005 and
2009, found that the reported use of this practice is on the rise.44 This more recent finding
highlights the fundamental role of translating evidence into practice regarding CLABSI
prevention.
Chlorhexidine for skin antisepsis. In a systematic review and meta-analysis of 8 trials
involving 4,143 unique catheter insertions, skin antisepsis with chlorhexidine was found to be
92
associated with a 50 percent reduction in the subsequent risk of CLABSI compared with
povidone iodine.45 A formal economic evaluation by the same authors projected that although
costlier initially, the use of chlorhexidine over povidone iodine for insertion site disinfection and
CVC care would lead to a 1.6 percent decrease in the incidence of CLABSI, a 0.23 percent
decrease in the incidence of death, and a savings of $113 per catheter.46 Existing HICPAC
guidelines endorse the use of chlorhexidine gluconate for skin antisepsis prior to CVC insertion
(Category IA).20
Antimicrobial catheters. The utility of catheters impregnated with a variety of substances
including chlorhexidine-silver sulfadiazine, minocycline-rifampin, benzalkonium chloride, and
silver have been evaluated in more than 20 randomized controlled studies and four recent
systematic reviews and meta-analyses.47-51 Meta-conclusions from these reviews remain limited,
due to heterogeneity arising from differences in the population, design, and conduct of the
pooled studies. For example, in a study involving a pediatric burn population, Weber and
colleagues found significant reductions in CLABSI with the use of minocycline and rifampin
antimicrobial coated catheters over non-coated catheters.52 However, a prospective, doubleblinded, randomized study in adults failed to show a reduction in CLABSI with a secondgeneration CVC coated with chlorhexidine and silver sulfadiazine.53 Due to the initial acquisition
cost, variation in benefit according to patient populations, and potential concern for inducing
antimicrobial resistance, routine use of antimicrobial CVCs is not recommended.20,54 However,
in facilities where high-rates of CLABSI persist despite deployment and compliance with
comprehensive CLABSI prevention efforts, the use of antimicrobial CVCs is considered
reasonable by current HICPAC guidelines (Category IA).20
The subclavian vein as the insertion point of choice. The site of CVC placement may
influence the risk of CLABSI, owing to the differing density of bacterial skin colonization at
each entry site. In a multicenter study of 289 patients randomized to undergo venous
catheterization using either the femoral or subclavian site, CVC placement in the femoral area
was associated with a substantially greater risk of CLABSI than was subclavian insertion (20
versus 3.7 per 1,000 catheter days).55 In a recent Dutch multicenter study involving 3,750 CVCs
and 29,003 CVC days, insertion into the femoral and jugular vein was independently associated
with an increase in the risk of subsequent CLABSI.56 In a study directly comparing the
subclavian to the internal jugular and femoral sites, the subclavian site was associated with the
lowest risk of infection (0.97 versus 2.99 and 8.34 per 1,000 catheter days, respectively).24 For
this reason, whenever medically feasible, the subclavian vein is the preferred site for venous
catheterization in the current HICPAC guidelines (Category IB).6,20,57-59 However, this
recommendation remains the subject of on-going debate, as some rigorous studies have found
that the risk of CLABSI from femoral vein CVC insertion is not greater than that associated with
insertion into the subclavian or internal jugular veins.60-62
Measures To Prevent CLABSI After Central Venous Catheter Insertion

Following the insertion of a CVC, several practices may decrease the risk of developing
CLABSI. These maintenance practices are important aspects of CLABSI prevention,
especially in CVCs that remain in place for an extended period of time.
93
Disinfect hubs, needleless connectors, and injection ports prior to CVC use. Contamination
of the catheter hub due to non-sterile access technique is a recognized path for developing
CLABSI. Minimizing contamination by wiping the catheter hub with an appropriate antiseptic
specifically recommended by the device manufacturer, or swabbing the membranous septum of a
CVC with 70% alcohol have been shown to reduce both risk of catheter contamination and
incidence of CLABSI.63-65 The practice of disinfecting access sites prior to CVC use,
colloquially dubbed scrub the hub, is linked to decreases in both bacterial colonization at
access sites and rates of incident CLABSI.66-68 Educational efforts targeting providers
responsible for CVC care (such as bedside nurses) are an important component in ensuring
dissemination and compliance with this practice.69 Although current HICPAC guidelines
emphasize minimizing the risk of contamination by scrubbing the hub with an appropriate
antiseptic (Category IA),20 several in vitro studies have demonstrated that even with strict
attention to decontamination, pathogenic organisms can persist in crevices or inside CVC access
valves and/or require prolonged duration of contact with an antimicrobial to significantly
decrease the level of colonization of CVC valves.69-72 In response, innovative technologies such
as those that incorporate antimicrobial compounds in the matrix of the CVC access valve, or
devices that bathe the valve with antimicrobials are being developed and tested.70,72-74 In the
absence of significant clinical experience with these novel devices, recommendations regarding
their widespread use are not possible.
Remove nonessential CVCs. Each day with a CVC increases the risk of developing
CLABSI.75,76 Prompt removal of CVCs that are no longer warranted is thus an important practice
to reduce CLABSI. This action necessitates both awareness of CVC presence and an ongoing
risk-benefit assessment of continued central venous access. In a study tracking temporary CVC
use in hospitalized patients, Chernetsky-Tejedor and colleagues reported that patients who
underwent PICC placement for venous access paradoxically also had 5.4 concurrent days with a
peripheral intravenous line (P<0.001), and had more days in which the only justification for the
CVC was intravenous administration of antimicrobial agents (8.5 versus 1.6 days; P=0.0013).
The authors therefore concluded that a substantial proportion of CVC-days might have been
unjustified in this cohort.65 In a recent survey conducted in a European hospital, neither the
bedside nurse nor the treating physician knew why a CVC was in place for 8.3 percent of nonICU patients.77 Importantly and relatedly, routine replacement of CVCs at pre-determined time
intervals has not been shown to reduce the risk of CLABSI and is not recommended based on the
available evidence (Category IA).20
Chlorhexidine cleansing. Daily bathing of patients with a chlorhexidine-based solution in ICUor advanced care settings may lower CLABSI incidence. In a crossover study conducted in a
medical ICU, daily washing with a 2% chlorhexidine-impregnated washcloth significantly
reduced subsequent BSI compared with using soap and water (4.1 vs. 10.4 per 1,000 patientdays, P<0.05).78 A study in a surgical ICU also found that daily bathing with a 2% chlorhexidine
gluconate impregnated cloth led to significant reductions in CLABSI (12.07 vs. 3.17 CLABSIs
per 1,000 days; 73.7% rate reduction, P= 0.04).79 The benefits of chlorhexidine baths may also
extend to high-risk patients outside of ICU settings. In a quasi-experimental before and after
study of the effect of daily washing with 2% chlorhexidine solution on CLABSI incidence,
Munoz-Price and colleagues reported a 99 percent reduction in the CLABSI rate in a long-term
acute care facility.80 However, a recent retrospective study involving patients in a surgical ICU
94
suggested that the benefit from chlorhexidine bathing might not apply to all patients.81 As the
evidence base for this practice is limited and conflicting, current HICPAC guidelines cautiously
endorse the use of chlorhexidine washes (either in solution form or as a chlorhexidine
impregnated wash cloth), for daily skin cleansing as a means to prevent CLABSI with a
Category II recommendation.20 However, the level of evidence for this recommendation may
soon be upgraded, as a recent meta-analysis pooling 12 studies found significant reductions in
CLABSI risk in studies that evaluated chlorhexidine cleansing in a medical ICU setting (OR
0.44, 95% confidence interval, 0.33 to 0.59).82
CVC dressing, chlorhexidine sponges and topical antibiotic use. The type of dressing and use
of topical antibiotic ointments or creams at the catheter site may affect the risk of CLABSI. In a
meta-analysis of seven studies comparing clear dressings to gauze for CVCs, transparent
dressings were associated with greater risk of catheter tip colonization (Relative Risk [RR] 1.78,
95% confidence interval [CI] 1.30 to 2.30, P<0.05), but not CLABSI (RR 1.63, 95% CI 0.76 to
3.47).83 Another meta-analysis of randomized controlled trials comparing gauze and tape to
transparent dressings found no significant differences between dressing type and risk of
CLABSI.84 Thus, for CLABSI prevention, existing guidelines do not endorse one type of
dressing over the other and leave the choice of CVC dressing to provider/patient preference and
clinical scenario.20
The use of a chlorhexidine gluconate sponge over the site of CVC insertion has been
associated with a decrease in the frequency and cost of CLABSI. In a study involving 1,636
patients with venous and arterial catheters, Timsit and colleagues reported that chlorhexidine
gluconate sponge placement at the site of catheter insertion substantially reduced the incidence
of CLABSI (1.4 to 0.6 per 1,000 catheter days, hazard ratio 0.39, P<0.03). However, severe
contact dermatitis was observed in eight low birth-weight infants (5.3 per 1,000 catheter days),
and the potential for this adverse effect remains an important limitation in the use of
chlorhexidine gluconate sponges in this population.85 In a recent economic evaluation,
chlorhexidine-impregnated sponge use in patients with CLABSI was estimated to save $197 per
patient using a 3-day dressing change strategy vs. $83 using a 7-day standard dressing change
strategy.86 In another cost-benefit analysis, a hypothetical 400-bed hospital inserting 3,078 CVCs
annually would avoid a projected average of 35 CLABSIs, 145 local infections, and 281 ICU
days with the systematic use of a chlorhexidine-impregnated foam dressing; potential annual
hospital net savings were projected at over $895,000.87 Owing to important differences in study
design and outcomes involving primarily pediatric populations, current guidelines recommend
the use of chlorhexidine-impregnated sponge dressings only in situations where the CLABSI rate
is not decreasing despite adherence to other prevention measures (Category IB).20
The use of topical antibiotic ointment or creams at the insertion site (e.g. povidone iodine) is
recommended only for patients with hemodialysis catheters, where its use has been associated
with suppression of BSI.88,89 Interestingly, a recent prospective, non-blinded crossover study
found that chlorhexidine sponge dressings were not protective against BSI in patients with
hemodialysis catheters.90 Conversely, topical antibiotic dressings are not recommended for
CLABSI prevention in non-dialysis patients as their use may paradoxically increase fungemia
and antimicrobial resistance in this category of patients (Category IB).20,91,92
Antibiotic/anti-infective locks in high-risk patients. A catheter lock refers to the instillation
of supra-physiologic doses of an intravenous antibiotic or anti-infective solution into a catheter
95
lumen between periods of CVC access. Several studies have examined both the utility of specific
antibiotic or anti-infective agents (e.g. vancomycin, cephalosporins, taurolidine, EDTA, ethanol)
and the targeted use of antibiotic locks in high-risk patient populations. In a systematic review
and meta-analysis, vancomycin-based antibiotic locks in patients deemed high-risk for CLABSI
(planned, long-term central venous catheter duration or those with a history of prior CLABSI)
significantly decreased the risk of this outcome (RR 0.34, P=0.04).93 A more recent systematic
review also reported reductions in the risk of subsequent CLABSI using this approach as an
adjunctive treatment, specifically in patients with poor venous access where catheter salvage was
key.94 In view of concerns regarding the potential for inducing antibiotic resistance, several
novel compounds have been tested as anti-infective locks. For example, a recent study found a
solution containing minocycline and EDTA to be highly efficacious in preventing CLABSI in
patients with hemodialysis catheters.95 In patients receiving prolonged home parenteral nutrition
via a CVC, the antineoplastic compound taurolidine was found to reduce the risk of CLABSI
when used as a catheter lock in a before and after study.96 Even though several studies have
found reductions in CLABSI incidence in specific populations, generalizations beyond these
groups are difficult and not appropriate.97-100 Thus, due to important differences in study design,
type of catheter, agent used, and patient population, the use of antibiotic locks should be limited
to those who are at high baseline risk for CLABSI (Category II).20
Systemic antibiotic prophylaxis. Routine systemic antibiotic prophylaxis during or after CVC
insertion to reduce the risk of CLABSI is not recommended (Category IB).20 A recent Cochrane
meta-analysis involving patients with cancer found no convincing evidence that prophylactic
peptidoglycan administration prior to CVC insertion was associated with reduced CLABSI
incidence.101 A recent study examining the effect of prophylactic cefazolin on CLABSI
following port placement similarly found no benefit associated with antibiotic treatment.102
Institutional Initiatives To Reduce CLABSI

Educational interventions. Educational programs that emphasize appropriate indications for
CVC placement and programs that review proper procedures for catheter insertion and
maintenance have both been shown to reduce the incidence of CLABSI in various settings.103-107
Although teaching CVC insertion using simulation techniques is a growing phenomenon, a
recent systematic review found that this practice was associated with less frequent mechanical
complications, but not CLABSI.108 Reporting and monitoring for infections through a structured
infection control program is a critical component of CLABSI prevention. Consequently,
education and training regarding how to implement and assess infection control measures and
periodic reassessment of this knowledge has also been shown to reduce CLABSI.20,35,109 Despite
these important studies, a recent survey found that knowledge regarding which practices are most
associated with CLABSI prevention remains variable.110 Educational initiatives thus represent an
important area of opportunity for institutions and health systems interested in controlling
CLABSI (Category IA).20
Use of catheter checklists or bundles. A standardized approach to CVC placement that
utilizes a set of evidence-based practices represents an important innovation in CLABSI
prevention. In the Michigan Keystone ICU study, Pronovost and colleagues enrolled 103 ICUs in
67 hospitals to test whether an intervention consisting of five evidence-based practices
96
implemented at the time of CVC insertion could reduce CLABSI. Notably, these five practices
were selected because they each had strong evidence supporting their efficacy in CLABSI
reduction and the lowest barriers to implementation. These five practices were hand hygiene
prior to insertion; use of maximal sterile barrier precautions; chlorhexidine for skin antisepsis;
avoidance of the femoral site of insertion; and prompt removal of catheters when no longer
indicated. Following implementation of this intervention, the mean rate of CLABSI dropped
from 7.7 per 1,000 catheter days at baseline to 1.4 per 1,000 catheter days at 16 months across
participating sites.33 The use of these five interventions in unison has been called the checklist
or the bundle. The use of the bundle and variations thereof has been associated with a
sustained decrease in the incidence of CLABSI, not only within the U.S., but internationally as
well.38,111-116 The bundle has also been found to be cost-effective both in the U. S. and abroad,
leading to its widespread acceptance as a key strategy with which to reduce CLABSI.20,117 The
HICPAC guidelines categorize the use of bundled interventions during CVC insertion as
performance improvement initiatives and recommends this practice to reduce CLABSI (Category
IB).20
Specialized CVC insertion teams. Data from several studies suggest that CVC placement by
specialized teams dedicated to this role leads not only to greater placement skills and reduced
insertion complications, but also to reduced rates of institutional CLABSI.25,33,38,111 The use of
dedicated and trained staff ensures predictable adherence to evidence-based practices such as
hand hygiene and maximal sterile barriers. The advent of nursing-led PICC teams represents an
important transformation in the placement of CVCs in both critically ill and hospitalized patients.
Preliminary studies suggest that these teams are associated with high rates of insertion success
and low rates of mechanical complications in a variety of patient settings.118-120 However, no data
comparing the risk of CLABSI in patients who undergo PICC placement by nursing PICC teams
compared with other providers (such as hospitalists or radiologists) are currently available. The
HICPAC guidelines recommend the use of trained personnel to insert CVCs (Category IA).20
How Has CLABSI Prevention Been Implemented?

With the realization that CLABSI can be curtailed by the use of evidence-based practices,
CLABSI prevention has increasingly become an attainable goal for hospitals, health care
systems, and payors. The Michigan Keystone ICU study underscored how both technical (e.g.,
asepsis during insertion, standardized surveillance), and adaptive (e.g., buy-in from leadership, a
culture of safety) components were needed to successfully implement a CLABSI prevention
initiative.121 The identification of these two distinct, yet complementary, realms highlights how
engagement and education of staff, consistent execution of the bundle, and rigorous evaluation of
processcritical activities embodied within the CLABSI bundleare fundamental components
of CLABSI reduction.122 To ensure validity outside of Michigan, this model was replicated and
tested in Rhode Island and in the Adventist multistate health care system, where declines in
CLABSI at participating sites were also observed.122,123
Fueled by these successes, the U.S. Department of Health and Human Services prioritized
CLABSI reduction by designating it as Tier I of a comprehensive national healthcare-associated
infection prevention program. Ambitiously, the program aimed to reduce the incidence of
CLABSI by 50 percent in ICUs and specific patient populations over a period of 5 years,
primarily by encouraging the use of insertion bundles. A 2011 interim analysis found that
providers are on track with meeting this target, although continued opportunities remain for
97
patients in non-ICU settings and those receiving hemodialysis.124 In similar fashion, the Agency
for Healthcare Research and Quality (AHRQ) funded and launched an implementation program
called On the CUSP: Stop BSI. This national venture includes Federal agencies (e.g., CDC),
State organizations, and various professional societies, and aims to reduce the mean CLABSI
rate to less than 1 per 1,000 catheter days in each of the 50 United States.
What Have We Learned About CLABSI Prevention?

A decades worth of quality improvement, clinical research and policy change has led to
greater understanding of a number of pivotal aspects of prevention and control of CLABSI.
These important lessons and ongoing challenges are summarized below.
Importance of Organizational Context

CLABSI reduction efforts using bundles have been successful at some sites, but not at others.
This variable success has led to a renewed appreciation of organizational complexities (e.g., local
culture, clinical care team engagement) that influence the implementation of evidence-based
practices in health care settings. In a study that sought to answer why certain hospitals were more
likely to succeed in CLABSI reduction efforts than others, Krein and colleagues found that
themes involving structure and hierarchy within hospitals, politics and relationships between key
stakeholders, a missing sense of mission and value, and lack of commitment and passion
explained why some hospitals were not as successful at implementing CLABSI reduction
practices as others. The authors suggest that the use of externally-facilitated initiatives (e.g.,
infection prevention measures, technology-based solutions or a quality collaborative), may
provide the motivation, and sometimes resources, needed for implementation needed to
implement CLABSI prevention measures and overcome these major obstacles.125 In another
article studying the influence of context on outcomes, Dixon-Woods and colleagues examined
the Michigan Keystone ICU-initiative to develop an ex-post theory of why this quality
improvement program was so successful.126 These investigators posited that a number of
components ensured the success of the program: (a) recruitment of a large number of ICUs that
created pressure for others to join (e.g., isomorphic pressure), (b) the use of scheduled
teleconferences and meetings that created a sense of a densely networked community,
(c) reframing of CLABSI as a social problem (e.g., one that involved human action and behavior,
not a technical fix), which convinced stakeholders that they should organize to solve this issue,
(d) influencing hospital culture through checklists and integration of nursing and management,
and (e), robust measurement of outcomes as a means to enforce practice.
Similar themes emerged from a multi-ICU study involving the Department of Veterans
Affairs (VA) health care system, one of the largest integrated health care systems in the world.
Render and colleagues studied the effects of a centralized inpatient evaluation center that
supported not only bundle implementation, but also provided support by recruiting leadership,
and providing feedback, learning tools, and mentoring at VA ICUs. Although the bundle was
implemented in all ICUs, the investigators found marked declines in CLABSI specifically at sites
where the additional support tools were well received. In contrast, sites that struggled with
CLABSI reduction lacked a functional improvement team, forcing functions, or real-time
feedback systems, underscoring the importance of these factors in CLABSI reduction.127
In a national study of 1,212 health care professionals from 33 different hospitals, Flanagan
and colleagues conducted an open-ended survey and also found that poor adherence to
guidelines, lack of culture change, no impetus to change, insufficient resources, and issues
98
related to education were perceived barriers to achieving success in CLABSI improvement

programs.128 In the context of the work by Krein, Dixon-Woods, and others, these findings
highlight the importance of understanding, appreciating, and addressing contextual factors in the
quest to control CLABSI throughout the world.
Need for Accurate and Reliable Reporting

AHRQ has emphasized the reduction of CVC-associated BSI by designating it as Patient
Safety Indicator (PSI)-7 on nationally reported scorecards. Although a technical brief outlining
specifications of measurement for this PSI is publicly available,18 variations in measurement of
this indicator have led to consternation in the literature. In a criterion validity study, Zrelak and
colleagues conducted a retrospective cross-sectional study of 23 U.S. hospitals using trained
abstractors and found that among 191 cases that met PSI-7 criteria, only 104 (positive predictive
value [PPV] 54%; 95% CI, 40% to 69%) represented true CLABSI.129 In another study
examining the validity of PSI-7, Cevasco and colleagues used similar methodology and found
that only 42 of 112 reviewed cases represented true CLABSI events (PPV 38%; 95% CI, 29% to
47%).130 In both studies, coding-related issues and present-on-admission diagnoses explained a
large fraction of incorrect reporting. Inaccurate measurement is further compounded by
continued variation in public reporting of PSI-7. In a study of 14 states with mandatory CLABSI
monitoring laws, Aswani and colleagues found numerous disparities in how participating sites
selected the time span of their data collection, variably presented their infection rates, used
inconsistent methods of risk adjustment, chose which locations and care settings to report, and
demonstrated significant time lag to reporting.131 Using a standard definition of CLABSI to
retrospectively study institutional variation in reporting bloodstream infections, Lin and
colleagues found marked variability among 20 ICUs when comparing infection preventionistsreported CLABSI rates to those from a computer-generated algorithm.132 In a provocative study,
Niedner and colleagues showed that more-aggressive surveillance using stricter definitions and
written policies was associated with higher CLABSI reporting rates in 16 pediatric ICUs.133 This
variability in reporting has profound implications in pay-for-performance and benchmarking
applications that use this measure, as those most likely to accurately report CLABSI stand to be
the most penalized. This dilemma underscores the need to standardize, audit, and constantly
evaluate this system of quality measurement.
Importance of Continued Performance Improvement Efforts

Despite major strides involving knowledge generation and dissemination over the past
decade, important gaps remain in the practice of CLABSI prevention. In a cross-sectional survey
of 1,000 randomly selected physician-members of the American College of Physicians-American
Society of Internal Medicine, the reported use of maximal sterile barriers and chlorhexidine
gluconate at the time of CVC insertion remained low among internists who identified themselves
as having recently inserted a CVC.134 Similarly, around 15 percent of U.S. hospitals report
routinely changing CVCs at predetermined time intervals despite abundant evidence that this
practice should be discontinued.43,135 In an audit of staff practice and awareness of post-insertion
catheter care, Shapey and colleagues found multiple breaches involving knowledge about
dressing and catheter hub decontamination.136 Are these behaviors and practices remediable? In a
36-month followup study of the Keystone Project, zero incidents of CLABSI were found in
participating sites, despite completion of the original study. The durability of this effect suggests
that not only can behaviors be changed, but engagement, education, monitoring, and feedback
99
can sustain these behaviors beyond the intervention stage.113 Ongoing performance measurement
and process improvement must thus come to represent a fundamental facet of national and local
efforts directed towards CLABSI prevention.
Identification of New Challenges

Most BSIs related to CVCs occur not in those with long-term CVCs, but in patients with
short-term CVCs.137 A major shift in the landscape of short-term CVCs, the remarkable growth
of PICC use in hospitalized and critically ill patients, may therefore bring new challenges to
CLABSI prevention.12,76,120 Despite the rapid growth in the use of PICCs, little is known about
the indications, prevalence, and patterns of use of this device. Consequently, little is known
regarding the adherence to or appropriateness of CLABSI prevention techniques when inserting
and maintaining PICC lines. As PICCs are frequently placed in vulnerable populations such as
children and those with cancer138 and are associated with important complications, further study
of this technology and its association with CLABSI is needed.139,140 In addition, considerably
less attention has been devoted to the study and testing of best practices in maintaining long-term
CVCs, such as PICCs. As the risk of CLABSI is greatly influenced by the manner in which a
CVC is handled and treated following insertion, this knowledge gap represents an important area
for future study.

The intervening decade between the original Making Health Care Safer report18,21,22 and
this update has borne witness to a number of practices, approaches, and technologies that have
controlled and eliminated CLABSI in specific settings. Despite this progress, a number of
important policy, knowledge, and implementation gaps remain. While a CLABSI bundle that
incorporates five practices that have reasonable evidence underlying their use appears to be
successful in reducing CLABSI within ICUs, the extent to which this bundle is effective at
preventing and reducing CLABSI outside of the ICU is unknown. As the majority of CVCs are
now found in non-ICU settings, a research agenda that targets this population is necessary.
Understanding how best to assess and address the complexities of culture and behavior are
critical in this context, as these factors are likely to vary to a greater extent than ICU settings. A
summary table is located in Table 2, Chapter 10.
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Moderate-tohigh
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low-tomoderate
100
How Much do We
Know?/How Hard is It?
Not difficult
bundle)-to-moderate
(understanding
organization culture and
context)
References
1.
United States Department of Health and

Human Services.
ets/index.html#table1]. 2011;Accessed
February 7, 2012.
2.
Klevens RM, Edwards JR, Richards CL, Jr.,

et al. Estimating health care-associated
2002. Public Health Rep. 2007 MarApr;122(2):160-6. PMID 17357358.
3.
Horan TC, Andrus M, Dudeck MA.

CDC/NHSN surveillance definition of
health care-associated infection and criteria
for specific types of infections in the acute
care setting. Am J Infect Control. 2008
Jun;36(5):309-32. PMID 18538699.
4.
CDC/NHSN Surveillance Definition of

Healthcare-Associated Infection and Criteria
for Specific Types of Infections in the Acute
Care Setting.
www.cdc.gov/nhsn/PDFs/pscManual/17psc
NosInfDef_current.pdf. 2012;Accessed
February 8, 2012.
5.
Mermel LA, Allon M, Bouza E, et al.

Clinical practice guidelines for the diagnosis
and management of intravascular catheterrelated infection: 2009 Update by the
Infectious Diseases Society of America. Clin
Infect Dis. 2009 Jul 1;49(1):1-45. PMID
19489710.
6.
7.
8.
Mermel LA. Prevention of central venous

catheter-related infections: what works other
than impregnated or coated catheters? J
Hosp Infect. 2007 Jun;65 Suppl 2:30-3.
PMID 17540238.
Wenzel RP, Edmond MB. The impact of
hospital-acquired bloodstream infections.
Emerg Infect Dis. 2001 Mar-Apr;7(2):174-7.
PMID 11294700.
Climo M, Diekema D, Warren DK, et al.
Prevalence of the use of central venous
access devices within and outside of the
intensive care unit: results of a survey
among hospitals in the prevention epicenter
program of the Centers for Disease Control
and Prevention. Infect Control Hosp
Epidemiol. 2003 Dec;24(12):942-5. PMID
14700410.
101
9.
Weber DJ, Sickbert-Bennett EE, Brown V,

et al. Comparison of hospitalwide
surveillance and targeted intensive care unit
surveillance of healthcare-associated
infections. Infect Control Hosp Epidemiol.
2007 Dec;28(12):1361-6. PMID 17994516.
10.
Edwards JR, Peterson KD, Mu Y, et al.

(NHSN) report: data summary for 2006
through 2008, issued December 2009. Am J
Infect Control. 2009 Dec;37(10):783-805.
PMID 20004811.
11.
Vital signs: central line-associated blood

stream infections--United States, 2001,
2008, and 2009. MMWR Morb Mortal Wkly
Rep. 2011 Mar 4;60(8):243-8. PMID
21368740.
12.
Ajenjo MC, Morley JC, Russo AJ, et al.

Peripherally inserted central venous
catheter-associated bloodstream infections
in hospitalized adult patients. Infect Control
Hosp Epidemiol. 2011 Feb;32(2):125-30.
PMID 21460466.
13.
Stone PW, Braccia D, Larson E. Systematic

review of economic analyses of health careassociated infections. Am J Infect Control.
2005 Nov;33(9):501-9. PMID 16260325.
14.
Scott R. The direct medical costs of

healthcare-assicated infections in U.S.
hospitals and the benefits of prevention.
www.cdc.gov/HAI/pdfs/hai/Scott_CostPape
r.pdf.Accessed January 30, 2012.
15.
Roberts RR, Scott RD, 2nd, Hota B, et al.

Costs attributable to healthcare-acquired
infection in hospitalized adults and a
comparison of economic methods. Med
Care. 2010 Nov;48(11):1026-35. PMID
20940650.
16.
Tarricone R, Torbica A, Franzetti F, et al.

Hospital costs of central line-associated
bloodstream infections and costeffectiveness of closed vs. open infusion
containers. The case of Intensive Care Units
in Italy. Cost Eff Resour Alloc. 2010;8:8.
PMID 20459753.
17.
Institute For Healthcare Improvement (IHI)

Getting started kit: prevent central line
infections how-to guide.
www.ihi.org/IHI/Programs/Campaign/Centr
alLineInfection.htm. 2008;Accessed January
30, 2012.
25.

Clin Infect Dis. 2011 Jun;52(11):1283-90.
PMID 21596671.
18.
Agency for Healthcare Quality and Research

PSI, Technical Specifications.
www.qualityindicators.ahrq.gov/Downloads
/Software/SAS/V43/TechnicalSpecifications
/PSI%2007%20Central%20Venous%20Cath
eterRelated%20Blood%20Stream%20Infections
.pdf.Accessed January 30, 2012.
26.
Almuneef MA, Memish ZA, Balkhy HH, et

al. Rate, risk factors and outcomes of
catheter-related bloodstream infection in a
paediatric intensive care unit in Saudi
Arabia. J Hosp Infect. 2006 Feb;62(2):20713. PMID 16307822.
27.
Maki DG, Kluger DM, Crnich CJ. The risk

of bloodstream infection in adults with
different intravascular devices: a systematic
review of 200 published prospective studies.
Mayo Clin Proc. 2006 Sep;81(9):1159-71.
PMID 16970212.
28.
Yoshida J, Ishimaru T, Kikuchi T, et al.

Association between risk of bloodstream
infection and duration of use of totally
implantable access ports and central lines: a
24-month study. Am J Infect Control. 2011
Sep;39(7):e39-43. PMID 21652113.
29.
Bicudo D, Batista R, Furtado GH, et al. Risk

factors for catheter-related bloodstream
infection: a prospective multicenter study in
Brazilian intensive care units. Braz J Infect
Dis. 2011 Jul-Aug;15(4):328-31. PMID
21861002.
30.
Safdar N, Kluger DM, Maki DG. A review

of risk factors for catheter-related
bloodstream infection caused by
percutaneously inserted, noncuffed central
venous catheters: implications for preventive
strategies. Medicine (Baltimore). 2002
Nov;81(6):466-79. PMID 12441903.
31.
Savolainen-Kopra C, Haapakoski J, Peltola

PA, et al. Hand washing with soap and water
together with behavioural recommendations
prevents infections in common work
environment: an open cluster-randomized
trial. Trials. 2012 Jan 16;13(1):10. PMID
22243622.
32.
Bischoff WE, Reynolds TM, Sessler CN, et

al. Handwashing compliance by health care
workers: The impact of introducing an
accessible, alcohol-based hand antiseptic.
Arch Intern Med. 2000 Apr 10;160(7):101721. PMID 10761968.
19.
Association for Professionals in Infection

Control and Epidemiology. Targeting zero
healthcare associated infections.
www.apic.org/AM/CM/ContentDisplay.cfm
?ContentFileID=11707. 2008;Accessed
January 30, 2012.
20.
OGrady NP, Alexander M, Burns LA, et al.

Guidelines for the prevention of
intravascular catheter-related infections. Am
J Infect Control. 2011 May;39(4 Suppl
1):S1-34. PMID 21511081.
21.

et al. Making health care safer: a critical
analysis of patient safety practices. Evidence
report/technology assessment (Summary).
2001(43):i-x, 1-668. PMID 11510252.
22.
Saint S. Chapter 16. Prevention of

Intravascular Catheter Associated Infections.
Making Health Care Safer: A Critical
Evidence Report/Technology Assessment,
No. 43. AHRQ Publication No. 01-E058.
Quality, Rockville, MD. 2001; 149-162.
www.ahrq.gov/clinic/ptsafety/pdf/chap16.pd
f.
23.
24.
Alonso-Echanove J, Edwards JR, Richards

MJ, et al. Effect of nurse staffing and
antimicrobial-impregnated central venous
catheters on the risk for bloodstream
infections in intensive care units. Infect
Dec;24(12):916-25. PMID 14700407.
Lorente L, Henry C, Martin MM, et al.
Central venous catheter-related infection in
a prospective and observational study of
2,595 catheters. Crit Care. 2005;9(6):R6315. PMID 16280064.
102
33.

al. An intervention to decrease catheterrelated bloodstream infections in the ICU. N
Engl J Med. 2006 Dec 28;355(26):2725-32.
PMID 17192537.
41.
Hu KK, Veenstra DL, Lipsky BA, et al. Use

of maximal sterile barriers during central
venous catheter insertion: clinical and
economic outcomes. Clin Infect Dis. 2004
Nov 15;39(10):1441-5. PMID 15546079.
34.
Barrera L, Zingg W, Mendez F, et al.

Effectiveness of a hand hygiene promotion
strategy using alcohol-based handrub in 6
intensive care units in Colombia. Am J
Infect Control. 2011 Oct;39(8):633-9. PMID
21636170.
42.
35.
Zingg W, Imhof A, Maggiorini M, et al.

Impact of a prevention strategy targeting
hand hygiene and catheter care on the
incidence of catheter-related bloodstream
infections. Crit Care Med. 2009
Jul;37(7):2167-73; quiz 80. PMID
19487942.
Rosenthal VD, Maki DG, Rodrigues C, et al.

Impact of International Nosocomial
Infection Control Consortium (INICC)
strategy on central line-associated
bloodstream infection rates in the intensive
care units of 15 developing countries. Infect
Dec;31(12):1264-72. PMID 21029008.
43.
Warren DK, Yokoe DS, Climo MW, et al.

Preventing catheter-associated bloodstream
infections: a survey of policies for insertion
and care of central venous catheters from
hospitals in the prevention epicenter
program. Infect Control Hosp Epidemiol.
2006 Jan;27(1):8-13. PMID 16418980.
44.
Krein SL, Kowalski CP, Hofer TP, et al.

Preventing Hospital-Acquired Infections: A
National Survey of Practices Reported by
U.S. Hospitals in 2005 and 2009. J Gen
Intern Med. 2011 Dec 6PMID 22143455.
45.
Chaiyakunapruk N, Veenstra DL, Lipsky

BA, et al. Chlorhexidine compared with
povidone-iodine solution for vascular
catheter-site care: a meta-analysis. Ann
Intern Med. 2002 Jun 4;136(11):792-801.
PMID 12044127.
46.
Chaiyakunapruk N, Veenstra DL, Lipsky

BA, et al. Vascular catheter site care: the
clinical and economic benefits of
chlorhexidine gluconate compared with
povidone iodine. Clin Infect Dis. 2003 Sep
15;37(6):764-71. PMID 12955636.
36.

Guideline for Hand Hygiene in Health-Care
Settings: Recommendations of the
Healthcare Infection Control Practices
Advisory Committee and the
HICPAC/SHEA/APIC/IDSA Hand Hygiene
Task Force. . MMWR Morb Mortal Wkly
Rep. 2002; 2002;51(No. RR- 16):1-32.
37.
Young EM, Commiskey ML, Wilson SJ.

Translating evidence into practice to prevent
central venous catheter-associated
bloodstream infections: a systems-based
intervention. Am J Infect Control. 2006
Oct;34(8):503-6. PMID 17015155.
38.
Berenholtz SM, Pronovost PJ, Lipsett PA, et

al. Eliminating catheter-related bloodstream
infections in the intensive care unit. Crit
Care Med. 2004 Oct;32(10):2014-20. PMID
15483409.
39.
Galpern D, Guerrero A, Tu A, et al.

Effectiveness of a central line bundle
campaign on line-associated infections in
the intensive care unit. Surgery. 2008
Oct;144(4):492-5; discussion 5. PMID
18847630.
47.
Niel-Weise BS, Stijnen T, van den Broek PJ.

Anti-infective-treated central venous
catheters for total parenteral nutrition or
chemotherapy: a systematic review. J Hosp
Infect. 2008 Jun;69(2):114-23. PMID
18439717.
40.
Raad, II, Hohn DC, Gilbreath BJ, et al.

Prevention of central venous catheter-related
infections by using maximal sterile barrier
precautions during insertion. Infect Control
Hosp Epidemiol. 1994 Apr;15(4 Pt 1):231-8.
PMID 8207189.
48.
Niel-Weise BS, Stijnen T, van den Broek PJ.

Anti-infective-treated central venous
catheters: a systematic review of
randomized controlled trials. Intensive Care
Med. 2007 Dec;33(12):2058-68. PMID
17940746.
103
49.
Casey AL, Mermel LA, Nightingale P, et al.

Antimicrobial central venous catheters in
adults: a systematic review and metaanalysis. Lancet Infect Dis. 2008
Dec;8(12):763-76. PMID 19022192.
50.
Ramritu P, Halton K, Collignon P, et al. A

systematic review comparing the relative
effectiveness of antimicrobial-coated
catheters in intensive care units. Am J Infect
Control. 2008 Mar;36(2):104-17. PMID
18313512.
51.
52.
53.
54.
55.
56.
Hockenhull JC, Dwan KM, Smith GW, et al.

The clinical effectiveness of central venous
catheters treated with anti-infective agents in
preventing catheter-related bloodstream
infections: a systematic review. Crit Care
Med. 2009 Feb;37(2):702-12. PMID
19114884.
Weber JM, Sheridan RL, Fagan S, et al.
Incidence of Catheter-Associated
Bloodstream Infection After Introduction of
Minocycline and Rifampin AntimicrobialCoated Catheters in a Pediatric Burn
Population. J Burn Care Res. 2012 Jan
2PMID 22210071.
Rupp ME, Lisco SJ, Lipsett PA, et al. Effect
of a second-generation venous catheter
impregnated with chlorhexidine and silver
sulfadiazine on central catheter-related
infections: a randomized, controlled trial.
Ann Intern Med. 2005 Oct 18;143(8):57080. PMID 16230723.
Cherry-Bukowiec JR, Denchev K,
Dickinson S, et al. Prevention of catheterrelated blood stream infection: back to
basics? Surg Infect (Larchmt). 2011
Feb;12(1):27-32. PMID 21171811.
Merrer J, De Jonghe B, Golliot F, et al.
Complications of femoral and subclavian
venous catheterization in critically ill
patients: a randomized controlled trial.
JAMA. 2001 Aug 8;286(6):700-7. PMID
11495620.
van der Kooi TI, Wille JC, van Benthem
BH. Catheter application, insertion vein and
length of ICU stay prior to insertion affect
the risk of catheter-related bloodstream
infection. J Hosp Infect. 2012 Jan 11PMID
22243832.
104
57.
Hamilton HC, Foxcroft DR. Central venous

access sites for the prevention of venous
thrombosis, stenosis and infection in patients
requiring long-term intravenous therapy.
Cochrane Database Syst Rev.
2007(3):CD004084. PMID 17636746.
58.
Goetz AM, Wagener MM, Miller JM, et al.

Risk of infection due to central venous
catheters: effect of site of placement and
catheter type. Infect Control Hosp
Epidemiol. 1998 Nov;19(11):842-5. PMID
9831940.
59.
Lorente L, Jimenez A, Naranjo C, et al.

Higher incidence of catheter-related
bacteremia in jugular site with tracheostomy
than in femoral site. Infect Control Hosp
Epidemiol. 2010 Mar;31(3):311-3. PMID
20109074.
60.
Casanegra AI, Brannan S, Dadu R, et al.

Short-term femoral vein catheterization
rarely causes thrombosis or bacteremia. J
Hosp Med. 2011 Jan;6(1):33-6. PMID
20578050.
61.
Reyes JA, Habash ML, Taylor RP. Femoral

central venous catheters are not associated
with higher rates of infection in the pediatric
critical care population. Am J Infect Control.
2012 Feb;40(1):43-7. PMID 21704431.
62.
Parienti JJ, Thirion M, Megarbane B, et al.

Femoral vs jugular venous catheterization
and risk of nosocomial events in adults
requiring acute renal replacement therapy: a
randomized controlled trial. JAMA. 2008
May 28;299(20):2413-22. PMID 18505951.
63.
Cookson ST, Ihrig M, OMara EM, et al.

Increased bloodstream infection rates in
surgical patients associated with variation
from recommended use and care following
implementation of a needleless device.
Jan;19(1):23-7. PMID 9475345.
64.
Rupp ME, Sholtz LA, Jourdan DR, et al.

Outbreak of bloodstream infection
temporally associated with the use of an
intravascular needleless valve. Clin Infect
Dis. 2007 Jun 1;44(11):1408-14. PMID
17479934.
65.
Maragakis LL, Bradley KL, Song X, et al.

Increased catheter-related bloodstream
infection rates after the introduction of a
new mechanical valve intravenous access
port. Infect Control Hosp Epidemiol. 2006
Jan;27(1):67-70. PMID 16418990.
73.
Buchman AL, Spapperi J, Leopold P. A new

central venous catheter cap: decreased
microbial growth and risk for catheterrelated bloodstream infection. J Vasc
Access. 2009 Jan-Mar;10(1):11-21. PMID
19340794.
66.
Lockman JL, Heitmiller ES, Ascenzi JA, et

al. Scrub the hub! Catheter needleless port
decontamination. Anesthesiology. 2011
Apr;114(4):958. PMID 21270632.
74.
67.
Sannoh S, Clones B, Munoz J, et al. A

multimodal approach to central venous
catheter hub care can decrease catheterrelated bloodstream infection. Am J Infect
Control. 2010 Aug;38(6):424-9. PMID
20137829.
Oto J, Imanaka H, Konno M, et al. A

prospective clinical trial on prevention of
catheter contamination using the hub
protection cap for needleless injection
device. Am J Infect Control. 2011
May;39(4):309-13. PMID 20965611.
75.
Lederle FA, Parenti CM, Berskow LC, et al.

The idle intravenous catheter. Ann Intern
Med. 1992 May 1;116(9):737-8. PMID
1558345.
76.
Tejedor SC, Tong D, Stein J, et al.

Temporary central venous catheter
utilization patterns in a large tertiary care
center: tracking the idle central venous
catheter. Infect Control Hosp Epidemiol.
2012 Jan;33(1):50-7. PMID 22173522.
77.
Zingg W, Sandoz L, Inan C, et al. Hospitalwide survey of the use of central venous
catheters. J Hosp Infect. 2011
Apr;77(4):304-8. PMID 21288595.
78.
Bleasdale SC, Trick WE, Gonzalez IM, et

al. Effectiveness of chlorhexidine bathing to
reduce catheter-associated bloodstream
infections in medical intensive care unit
patients. Arch Intern Med. 2007 Oct
22;167(19):2073-9. PMID 17954801.
79.
Dixon JM, Carver RL. Daily chlorohexidine

gluconate bathing with impregnated cloths
results in statistically significant reduction in
central line-associated bloodstream
infections. Am J Infect Control. 2010
Dec;38(10):817-21. PMID 21093698.
80.
Munoz-Price LS, Hota B, Stemer A, et al.

Prevention of bloodstream infections by use
of daily chlorhexidine baths for patients at a
long-term acute care hospital. Infect Control
PMID 19751155.
81.
Popovich KJ, Hota B, Hayes R, et al. Daily

skin cleansing with chlorhexidine did not
reduce the rate of central-line associated
bloodstream infection in a surgical intensive
care unit. Intensive Care Med. 2010
May;36(5):854-8. PMID 20213074.
68.
69.
70.
71.
72.
Soothill JS, Bravery K, Ho A, et al. A fall in

bloodstream infections followed a change to
2% chlorhexidine in 70% isopropanol for
catheter connection antisepsis: a pediatric
single center before/after study on a
hemopoietic stem cell transplant ward. Am J
Infect Control. 2009 Oct;37(8):626-30.
PMID 19616869.
Simmons S, Bryson C, Porter S. Scrub the
hub: cleaning duration and reduction in
bacterial load on central venous catheters.
Crit Care Nurs Q. 2011 Jan-Mar;34(1):31-5.
PMID 21160298.
Maki DG. In vitro studies of a novel
antimicrobial luer-activated needleless
connector for prevention of catheter-related
bloodstream infection. Clin Infect Dis. 2010
Jun 15;50(12):1580-7. PMID 20455695.
Menyhay SZ, Maki DG. Disinfection of
needleless catheter connectors and access
ports with alcohol may not prevent
microbial entry: the promise of a novel
antiseptic-barrier cap. Infect Control Hosp
Epidemiol. 2006 Jan;27(1):23-7. PMID
16418982.
Menyhay SZ, Maki DG. Preventing central
venous catheter-associated bloodstream
infections: development of an antiseptic
barrier cap for needleless connectors. Am J
Infect Control. 2008 Dec;36(10):S174 e1-5.
PMID 19084153.
105
82.
OHoro JC, Silva GL, Munoz-Price LS, et

al. The Efficacy of Daily Bathing with
Chlorhexidine for Reducing HealthcareAssociated Bloodstream Infections: A Metaanalysis. Infect Control Hosp Epidemiol.
2012 Mar;33(3):257-67. PMID 22314063.
83.
Hoffmann KK, Weber DJ, Samsa GP, et al.

Transparent polyurethane film as an
intravenous catheter dressing. A metaanalysis of the infection risks. JAMA. 1992
Apr 15;267(15):2072-6. PMID 1532429.
84.
85.
86.
87.
88.
89.
90.
Gillies D, ORiordan E, Carr D, et al.

Central venous catheter dressings: a
systematic review. J Adv Nurs. 2003
Dec;44(6):623-32. PMID 14651685.
Timsit JF, Schwebel C, Bouadma L, et al.
Chlorhexidine-impregnated sponges and less
frequent dressing changes for prevention of
catheter-related infections in critically ill
adults: a randomized controlled trial. JAMA.
2009 Mar 25;301(12):1231-41. PMID
19318651.
Schwebel C, Lucet JC, Vesin A, et al.
Economic evaluation of chlorhexidineimpregnated sponges for preventing
catheter-related infections in critically ill
adults in the Dressing Study*. Crit Care
Med. 2012 Jan;40(1):11-7. PMID 21926570.
Ye X, Rupnow M, Bastide P, et al.
Economic impact of use of chlorhexidineimpregnated sponge dressing for prevention
of central line-associated infections in the
United States. Am J Infect Control. 2011
Oct;39(8):647-54. PMID 21641681.
Battistella M, Bhola C, Lok CE. Long-term
follow-up of the Hemodialysis Infection
Prevention with Polysporin Ointment
(HIPPO) Study: a quality improvement
report. Am J Kidney Dis. 2011
Mar;57(3):432-41. PMID 21216512.
Lok CE, Mokrzycki MH. Prevention and
management of catheter-related infection in
hemodialysis patients. Kidney Int. 2011
Mar;79(6):587-98. PMID 21178979.
Camins BC, Richmond AM, Dyer KL, et al.
A crossover intervention trial evaluating the
efficacy of a chlorhexidine-impregnated
sponge in reducing catheter-related
bloodstream infections among patients
undergoing hemodialysis. Infect Control
PMID 20879855.
106
91.
Zakrzewska-Bode A, Muytjens HL, Liem

KD, et al. Mupirocin resistance in
coagulase-negative staphylococci, after
topical prophylaxis for the reduction of
colonization of central venous catheters. J
Hosp Infect. 1995 Nov;31(3):189-93. PMID
8586787.
92.
Flowers RH, 3rd, Schwenzer KJ, Kopel RF,

et al. Efficacy of an attachable subcutaneous
cuff for the prevention of intravascular
catheter-related infection. A randomized,
controlled trial. JAMA. 1989 Feb
10;261(6):878-83. PMID 2492354.
93.
Safdar N, Maki DG. Use of vancomycincontaining lock or flush solutions for

prevention of bloodstream infection
associated with central venous access
devices: a meta-analysis of prospective,
randomized trials. Clin Infect Dis. 2006 Aug
15;43(4):474-84. PMID 16838237.
94.
OHoro JC, Silva GL, Safdar N. Antiinfective locks for treatment of central lineassociated bloodstream infection: a
systematic review and meta-analysis. Am J
Nephrol. 2011;34(5):415-22. PMID
21934302.
95.
Campos RP, do Nascimento MM, Chula

DC, et al. Minocycline-EDTA lock solution
prevents catheter-related bacteremia in
hemodialysis. J Am Soc Nephrol. 2011
Oct;22(10):1939-45. PMID 21852579.
96.
Toure A, Lauverjat M, Peraldi C, et al.

Taurolidine lock solution in the secondary
prevention of central venous catheterassociated bloodstream infection in home
parenteral nutrition patients. Clin Nutr. 2012
Jan 27PMID 22285029.
97.
Moran J, Sun S, Khababa I, et al. A

randomized trial comparing
gentamicin/citrate and heparin locks for
central venous catheters in maintenance
hemodialysis patients. Am J Kidney Dis.
2012 Jan;59(1):102-7. PMID 22088576.
98.
Mortazavi M, Alsaeidi S, Sobhani R, et al.

Successful prevention of tunneled, central
catheter infection by antibiotic lock therapy
using cefotaxime. J Res Med Sci. 2011
Mar;16(3):303-9. PMID 22091249.
99.
John BK, Khan MA, Speerhas R, et al.

Ethanol Lock Therapy in Reducing
Catheter-Related Bloodstream Infections in
Adult Home Parenteral Nutrition Patients:
Results of a Retrospective Study. JPEN J
Parenter Enteral Nutr. 2011 Dec 28PMID
22205580.
100.
Wales PW, Kosar C, Carricato M, et al.

Ethanol lock therapy to reduce the incidence
of catheter-related bloodstream infections in
home parenteral nutrition patients with
intestinal failure: preliminary experience. J
Pediatr Surg. 2011 May;46(5):951-6. PMID
21616259.
107.
Cohen ER, Feinglass J, Barsuk JH, et al.

Cost savings from reduced catheter-related
bloodstream infection after simulation-based
education for residents in a medical
intensive care unit. Simul Healthc. 2010
Apr;5(2):98-102. PMID 20389233.
108.
Ma IW, Brindle ME, Ronksley PE, et al.

Use of simulation-based education to
improve outcomes of central venous
catheterization: a systematic review and
meta-analysis. Acad Med. 2011
Sep;86(9):1137-47. PMID 21785310.
109.
Perez Parra A, Cruz Menarguez M, Perez

Granda MJ, et al. A simple educational
intervention to decrease incidence of central
line-associated bloodstream infection
(CLABSI) in intensive care units with low
baseline incidence of CLABSI. Infect
Sep;31(9):964-7. PMID 20658936.
101.
van de Wetering MD, van Woensel JB.

Prophylactic antibiotics for preventing early
central venous catheter Gram positive
infections in oncology patients. Cochrane
Database Syst Rev. 2007(1):CD003295.
PMID 17253487.
102.
Karanlik H, Kurul S, Saip P, et al. The role

of antibiotic prophylaxis in totally
implantable venous access device
placement: results of a single-center
prospective randomized trial. American
journal of surgery. 2011 Jul;202(1):10-5.
PMID 21601826.
110.
Koutzavekiaris I, Vouloumanou EK, Gourni

M, et al. Knowledge and practices regarding
prevention of infections associated with
central venous catheters: a survey of
intensive care unit medical and nursing staff.
Am J Infect Control. 2011 Sep;39(7):542-7.
PMID 21496955.
103.
Neumayer LA. Improving outcomes: the

importance of data monitoring and ongoing
educational interventions. Arch Surg. 2011
Mar;146(3):307. PMID 21542189.
111.
104.
Ong A, Dysert K, Herbert C, et al. Trends in

central line-associated bloodstream
infections in a trauma-surgical intensive care
unit. Arch Surg. 2011 Mar;146(3):302-7.
PMID 21422361.
Wheeler DS, Giaccone MJ, Hutchinson N,

et al. A hospital-wide quality-improvement
collaborative to reduce catheter-associated
bloodstream infections. Pediatrics. 2011
Oct;128(4):e995-e1004; quiz e-7. PMID
21930547.
112.
Kim JS, Holtom P, Vigen C. Reduction of

catheter-related bloodstream infections
through the use of a central venous line
bundle: epidemiologic and economic
consequences. Am J Infect Control. 2011
Oct;39(8):640-6. PMID 21641088.
113.
Pronovost PJ, Goeschel CA, Colantuoni E,

et al. Sustaining reductions in catheter
related bloodstream infections in Michigan
intensive care units: observational study.
BMJ. 2010;340:c309. PMID 20133365.
114.
Longmate AG, Ellis KS, Boyle L, et al.

Elimination of central-venous-catheterrelated bloodstream infections from the
intensive care unit. BMJ Qual Saf. 2011
Feb;20(2):174-80. PMID 21303772.
105.
106.
Khouli H, Jahnes K, Shapiro J, et al.

Performance of medical residents in sterile
techniques during central vein
catheterization: randomized trial of efficacy
of simulation-based training. Chest. 2011
Jan;139(1):80-7. PMID 20705795.
Leung TK, Lee CM, Tai CJ, et al. A
retrospective study on the long-term
placement of peripherally inserted central
catheters and the importance of nursing care
and education. Cancer Nurs. 2011 JanFeb;34(1):E25-30. PMID 20885304.
107
115.
116.
Chuengchitraks S, Sirithangkul S, Staworn

D, et al. Impact of new practice guideline to
prevent catheter-related blood stream
infection (CRBSI): experience at the
Pediatric Intensive Care Unit of
Phramongkutklao Hospital. J Med Assoc
Thai. 2010 Nov;93 Suppl 6:S79-83. PMID
21284138.
Palomar Martinez M, Alvarez Lerma F,
Riera Badia MA, et al. [Prevention of
bacteriema related with ICU catheters by
multifactorial intervention: a report of the
pilot study]. Med Intensiva. 2010
Dec;34(9):581-9. PMID 21041004.
117.
Halton KA, Cook D, Paterson DL, et al.

Cost-effectiveness of a central venous
catheter care bundle. PLoS One.
2010;5(9)PMID 20862246.
118.
Gamulka B, Mendoza C, Connolly B.

Evaluation of a unique, nurse-inserted,
peripherally inserted central catheter
program. Pediatrics. 2005 Jun;115(6):16026. PMID 15930222.
119.
Wang D, Amesur N, Shukla G, et al.

Peripherally inserted central catheter
placement with the sonic flashlight: initial
clinical trial by nurses. J Ultrasound Med.
2009 May;28(5):651-6. PMID 19389904.
120.
DeLemos C, Abi-Nader J, Akins PT. Use of

peripherally inserted central catheters as an
alternative to central catheters in
neurocritical care units. Crit Care Nurse.
2011 Apr;31(2):70-5. PMID 21459866.
121.
Pronovost PJ, Berenholtz SM, Needham

DM. Translating evidence into practice: a
model for large scale knowledge translation.
BMJ. 2008;337:a1714. PMID 18838424.
122.
Pronovost P. Interventions to decrease

catheter-related bloodstream infections in
the ICU: the Keystone Intensive Care Unit
Project. Am J Infect Control. 2008
Dec;36(10):S171 e1-5. PMID 19084146.
123.
DePalo VA, McNicoll L, Cornell M, et al.

The Rhode Island ICU collaborative: a
model for reducing central line-associated
bloodstream infection and ventilatorassociated pneumonia statewide. Qual Saf
Health Care. 2010 Dec;19(6):555-61. PMID
21127114.
108
124.
United States Department of Health and

Human Services: Healthcare Associated
Infection Prevention Plan.
ets/index.html#table1. 2011. Accessed
February 7, 2012.
125.
Krein SL, Damschroder LJ, Kowalski CP, et

al. The influence of organizational context
on quality improvement and patient safety
efforts in infection prevention: a multicenter qualitative study. Soc Sci Med. 2010
Nov;71(9):1692-701. PMID 20850918.
126.

program. Milbank Q. 2011 Jun;89(2):167205. PMID 21676020.
127.
Render ML, Hasselbeck R, Freyberg RW, et

al. Reduction of central line infections in
Veterans Administration intensive care
units: an observational cohort using a central
infrastructure to support learning and
improvement. BMJ Qual Saf. 2011
Aug;20(8):725-32. PMID 21460392.
128.
Flanagan ME, Welsh CA, Kiess C, et al. A

national collaborative for reducing health
care-associated infections: current
initiatives, challenges, and opportunities.
American Journal of Infection Control.
2011;39(8):685-9. PMID 21665329.
129.
Zrelak PA, Sadeghi B, Utter GH, et al.

Positive predictive value of the Agency for
Healthcare Research and Quality Patient
Safety Indicator for central line-related
bloodstream infection (selected infections
due to medical care). J Healthc Qual. 2011
Mar-Apr;33(2):29-36. PMID 21385278.
130.
Cevasco M, Borzecki AM, OBrien WJ, et

al. Validity of the AHRQ Patient Safety
Indicator central venous catheter-related
bloodstream infections. J Am Coll Surg.
2011 Jun;212(6):984-90. PMID 21489833.
131.
Aswani MS, Reagan J, Jin L, et al. Variation

in public reporting of central line-associated
bloodstream infections by state. Am J Med
Qual. 2011 Sep-Oct;26(5):387-95. PMID
21825038.
132.
Lin MY, Hota B, Khan YM, et al. Quality of

traditional surveillance for public reporting
of nosocomial bloodstream infection rates.
JAMA. 2010 Nov 10;304(18):2035-41.
PMID 21063013.
133.
Niedner MF. The harder you look, the more

you find: Catheter-associated bloodstream
infection surveillance variability. Am J
Infect Control. 2010 Oct;38(8):585-95.
PMID 20868929.
134.
Rubinson L, Wu AW, Haponik EE, et al.

Why is it that internists do not follow
guidelines for preventing intravascular
catheter infections? Infect Control Hosp
Epidemiol. 2005 Jun;26(6):525-33. PMID
16018427.
135.
Krein SL, Hofer TP, Kowalski CP, et al. Use

of central venous catheter-related
bloodstream infection prevention practices
by US hospitals. Mayo Clin Proc. 2007
Jun;82(6):672-8. PMID 17550746.
136.
Shapey IM, Foster MA, Whitehouse T, et al.

Central venous catheter-related bloodstream
infections: improving post-insertion catheter
care. J Hosp Infect. 2009 Feb;71(2):117-22.
PMID 19013680.
137.
OGrady NP, Chertow DS. Managing

bloodstream infections in patients who have
short-term central venous catheters. Cleve
Clin J Med. 2011 Jan;78(1):10-7. PMID
21199902.
138.
Barrier A, Williams DJ, Connelly M, et al.

Frequency of peripherally inserted central
catheter complications in children. Pediatr
Infect Dis J. 2011 Dec 23. PMID 22189533.
139.
Pikwer A, Akeson J, Lindgren S.

Complications associated with peripheral or
central routes for central venous
cannulation. Anaesthesia. 2012
Jan;67(1):65-71. PMID 21972789.
140.
Marnejon T, Angelo D, Abu Abdou A, et al.

Risk factors for upper extremity venous
thrombosis associated with peripherally
inserted central venous catheters. J Vasc
Access. 2012 Jan 9:0. PMID 22266584
109
Chapter 11. Ventilator-Associated Pneumonia: Brief Update

Review
Bradford D. Winters, Ph.D., M.D.; Sean M. Berenholtz, M.D., M.H.S.
Introduction
Ventilator associated pneumonia (VAP) is defined as a hospital-acquired pneumonia that
develops within 48 to 72 hours after endotracheal intubation; the diagnosis hinges on a lack of
evidence suggesting that the infection developed prior to intubation. VAP is the most common
intensive care unit (ICU)-acquired infection, accounting for 25 percent of all ICU infections and
50 percent of ICU antibiotic use. At least 250,000 VAPs occur in the United States (U.S.) each
year. This condition causes complications in 8 to 28 percent of mechanically ventilated patients
and carries a mortality risk of approximately 10 percent (range 6% to 27%), resulting in a
possible 25,000 VAP-attributable deaths every year. Patients who develop VAP stay, on average,
4 days longer in the ICU. The per-case cost of VAP is estimated to be $23,000, and the total
incremental costs to the U.S. health care system are high: $2.19 to 3.17 billion USD per year1-3
The wide range of these estimates results from the lack of universally accepted, reliable
diagnostic criteria for VAP is present. The diagnosis of VAP may be based on any of a variety of
definitions, including a surveillance definition, a clinical definition, a microbiologically
confirmed definition, or a combination of the three methods. Microbiologically confirmed
definitions also may be differentially based on blind tracheal aspirates, directed bronco-alveolar
lavage, or even protected brush specimens.2
The original Making Health Care Safer report examined four interventions related to VAP:
variation of position (semi-recumbent positioning and continuous oscillation), continuous
subglottic suctioning, selective decontamination of the gastro-intestinal tract and the use of
sucralfate. While the data in favor of semi-recumbent positioning was limited (reduced VAP but
did not change mortality), the practice was judged to be easy to implement and had essentially no
cost or adverse effects. Oscillation was less clear in its benefit secondary to poor methodological
quality of the studies. While no evidence for harm was found, there were increased costs,
estimated to be about $100/day at the time of that report. Subglottic suctioning was judged to be
a promising strategy. At the time of the report, it was infrequently used and there were only a
few studies. Harmful effects were felt to be negligible but there were incremental costs for the
specialized endotracheal tubes required for this strategy. Selective gastro-intestinal
decontamination was found to have strong benefit for reducing VAP, though cost-effectiveness
was unclear. Of the trials examined, none reported adverse events from this practice; however,
there is continued concern that this practice may have a deleterious effect on antibiotic sensitivity
in general, leading to more resistant organisms over time in individuals as well as on a
population basis. Sucralfate as a VAP prevention strategy was judged to be inconclusive.
Additionally, sucralfate is inferior compared with H-2 blockers for preventing gastro-intestinal
bleeding. Given the increased risk of mortality with gastrointestinal bleeding and the increased
costs should this complication occur, sucralfate can no longer be recommended for VAP
prevention and H-2 blockers are the preferred agent for preventing gastro-intestinal bleeding in
critically ill patients.
This updated review focuses on four strategies as well; elevation of the head of the bed,
sedation vacations, oral care with chlorhexidine and subglottic suctioning.
110
What Are the Patient Safety Practices for Preventing VentilatorAcquired Pneumonia?
We conducted a systematic review of the literature to update a 2001 review conducted for the
original report. A recent study estimates that 14,000 to 20,000 lives could be saved each year in
the U.S. if best practices to prevent VAP were universally applied to all patients on mechanical
ventilation.3 The four primary recommended practices include: elevating the head of the bed to
30 degrees, sedation vacations, oral care with chlorhexidine (CHG), and subglottic suctioning
endotracheal tubes. Ventilator bundles usually include other elements such as deep venous
thrombosis (DVT)/pulmonary embolism (PE) prophylaxis and Peptic Ulcer Disease prophylaxis,
but these procedures are designed to prevent other ventilator-associated conditions and do not
address VAP prevention specifically. In fact, Peptic Ulcer Disease prophylaxis may increase the
risk of VAP. Other VAP-specific preventive interventions may include use of closed suctioning
circuits, scheduled circuit changes and a preference for orotracheal over nasotracheal intubation.
The remainder of this section describes the evidence in support of the four primary VAP specific
practices.
Head-of-Bed Elevation
The practice of head-of-bed elevation to prevent VAP has been recommended by several
medical groups including the Canadian Critical Care Trials Group, the American Thoracic
Society and Infectious Diseases Society of America, and the Centers for Disease Control and
Prevention. This recommendation is based on early data showing that being supine was an
independent risk factor for VAP.4 Importantly, a study in 1999 by Drakulovic and colleagues5
demonstrated a reduction in VAP with patients in the semi-upright position.5
A recent systematic review by Niel-Weisse and colleagues that applied strict inclusion
criteria (randomized or quasi-randomized trial, published as a full paper and not an abstract, state
the outcome measures used and present data sufficient to calculate the risks in both groups) and
included only three of 208 potential studies, representing a total of 337 patients questioned
whether patients head elevation can be maintained continuously above 30 degrees in ICUs, and
point prevalence assessments used in many studies may overstate how often the goal is met.4 The
effect of head-of-bed elevation on the incidence of both clinically diagnosed and
microbiologically confirmed VAP was found to be non-significant (RR = 0.47, 95% CI = 0.19 to
1.17 and RR = 0.67, 95% CI =0.23 to 2.01, respectively). A second study used broncho-alveolar
lavage, and the other two used tracheal aspirates for the microbiological assessment. The third
study found no significant increase in harm (decubitus ulcers); other potential harms (such as
DVT) were not assessed. These same three trials also found no significant impact on mortality
(pooled RR = 0.90, 95% CI = 0.64 to 1.27). The data were also judged to be of low quality for
methodological reasons. Despite these findings, an evaluation of the results using an online
Delphi process recommended the practice of keeping the head of the bed elevated by greater than
30 degrees to prevent VAP (most studies had actually used 45 degrees as their target).4 The
favorable point estimates (all favored the intervention despite lack of statistical significance) and
the lack of measurable harm may have influenced this recommendation.
Sedation Vacations
The use of sedation vacations, or sedation holds, has been shown to help patients wean from
mechanical ventilation more quickly than when these techniques are not employed.6,7 Further,
sedation vacations reduce patients exposure and subsequent risk of VAP as well as several other
111
mechanical ventilation-associated complications,6,7and are, themselves, considered to be safe7.

One pre-post study that examined a sedation protocol that specified daily interruption of
sedatives in combination with spontaneous breathing trials demonstrated reduced ventilator days
and reduced length of hospital stay. Although the sedation interruption group had a higher rate of
self-extubation, the proportion of patients that required re-intubation was similar pre- and post
intervention.7 These findings suggest that sedation vacations should be part of all ventilator and
VAP prevention bundles.
Oral Care Using Chlorhexidine

Oral care using chorhexidine (CHG) to reduce VAP is based on evidence that in intubated
patients, gingival and dental plaque become rapidly colonized with bacterial overgrowth due to
loss of natural mechanical elimination and poor hygiene. This microbiological burden becomes a
source for aspiration of bacteria around the endotracheal tube cuff, resulting in pulmonary
infection. Instituting meticulous oral care can reduce this microbiological burden and the
potential for VAP. A systematic review in 20078 that included seven randomized controlled trials
(RCTs; 1,650 patients) evaluating CHG found a statistically significant reduction in the risk for
VAP using a fixed effects model (RR=0.74 95%CI=0.56-0.96). Although the effect was found to
be non-significant when a random effects model was applied, the absolute risk reduction was
slightly better (RR=0.70, 95% CI= 0.47-1.04).8 A sub-group analysis of oral care using CHG in
cardiac surgery patients did support the finding of a statistically significant reduction in the risk
for VAP (RR=0.41, 95% CI=0.17-0.98) 8
In 2008, the Canadian VAP Prevention Guidelines9 advised that oral care with CHG should
be considered for VAP prevention, and the SHEA guidelines10 recommended regular oral care
with an antiseptic solution. Although the SHEA guidelines did not specifically recommend CHG,
all three of the studies that were cited as a basis for the recommendation used CHG.10
A 2011 systematic review of the effects of CHG11 that included 12 RCTs (2,341 patients)
further supported the previous findings. The relative risk of VAP after oral care with CHG was
reduced 28 percent for all patients (RR, 0.72; 95% CI, 0.55 to 0.94); 59 percent for cardiac
surgery ICU patients (RR, 0.41; 95% CI, 0.17 to 0.98); 33 percent for trauma/surgical ICU
patients (RR, 0.67; 95% CI, 0.50-0.88); and 28 percent for mixed ICU patients (RR, 0.77; 95%
CI, 0.58-1.02 for mixed ICUs). Evidence has also shown that using a 2% solution of CHG is
superior to a 0.2% solution, which is superior to 0.12%.11
Subglottic Suctioning Endotracheal Tubes

Subglottic suctioning tubes address the tendency for nasal-oral secretions and debris to pool
above the endotracheal tube cuff and below the vocal cords. This pooling creates a rich culture
medium for micro-organisms found in the nasal-oropharynx, which leads to overgrowth and is
thought to be a major cause of VAP. Subglottic suctioning endotracheal tubes use a port or ports
just above the cuff to allow removal of this pooled material so it cannot act as a culture medium
or be aspirated. Some of the systems use a simple single suctioning port, whereas others use an
active lavage system with an inflow and outflow port to wash out the material. Our review
identified no studies that directly compared these types of subglottic suctioning tube design (or
continuous vs. intermittent suction); nevertheless, the evidence is strongly in favor of these
devices for the reduction of VAP. Among 13 RCTs (2,442 patients) identified for a recent
systematic review,12 12 of the RCTs found that subglottic suctioning reduced VAP; the pooled
risk-reduction was 0.55 (95% CI=0.46 to 0.66, p<0.00001) with no heterogeneity in the studies.
112
This practice also significantly reduced the duration of mechanical ventilation and length of stay
in the ICU, although it had no impact on ICU- or hospital mortality.
How Have Practices To Prevent Ventilator-Acquired Pneumonia

Been Implemented and What Has Been Learned?
Practices to prevent VAP are usually bundled into a care package of several elements as
described above. The package may also incorporate elements beyond the four discussed above,
including closed in-line endotracheal suctioning systems, humidification systems, and non-VAP
specific interventions such as DVT/PE prophylaxis, for which ventilated patients are at increased
risk. In a 2005 pre-post study, Resar and colleagues reported a 45 percent reduction in VAP
across 35 ICUs that used such a bundled approach in a collaborative. This particular bundle used
only sedation vacations and head-of-bed elevation as VAP-specific elements.13 Subsequent prepost studies have also found that bundled elements synergistically reduced the rate of VAP by as
much as 40 percent in both adult and pediatric patient populations.14,15
One factor that has been noted in most of these publications is the difficulty of ensuring that
all patients who qualify for the bundle and the individual elements within the bundle (e.g., for
some patientssuch as spine surgery patients with a dural tearhead-of-bed elevation may be
contraindicated,) actually receive the bundles elements consistently. The Michigan Keystone
Project addressed this quality gap through a process of developing and applying technical tools
such as checklists and ensuring their use through improvements in teamwork and the safety
climate within 112 ICUs. This pre-post study found a 71 percent risk reduction in VAP, while at
the same time demonstrating an increase in the adherence to evidence-based practices from 32
percent at baseline to 84 percent after 30 months.16 This finding suggests that a combination of
effective evidence-based bundle elements reinforced with strategies to improve teamwork and
safety can ensure that patients receive appropriate care and that outcomes improve substantially.
Others have also noted this positive effect of collaboratives on closing the quality gap. In
their evaluation of the routine use of VAP prevention practices, Krein and colleagues (2008)17
found that use of semi-recumbent positioning was much more prevalent than the use of
subglottic drainage (73% vs. 21% of hospitals that reported use of VAP practices). They also
found that use of semi-recumbent positioning was strongly influenced by participation in
collaboratives (such as the Keystone Project) and is considered primarily a responsibility of
nursing staff. In contrast, use of subglottic suctioning endotracheal tubes is not influenced very
much by collaboratives and is primarily a physician decision. It is unclear whether these
differences are secondary to the participation in collaboratives, depend on who has the primary
responsibility for decisionmaking, or both. Interestingly, the authors also noted that whereas the
prevalence of semi-recumbent positioning was dramatically higher than that of subglottic
drainage, when the effectiveness of the techniques was compared, the supportive evidence for
subglottic drainage was found to be much stronger than for semi-recumbent positioning (five
randomized studies vs. two). More recently, Krein (2011)18 reported that the prevalence of use of
VAP prevention measures was also strongly influenced by the threat of non-payment for this
hospital-acquired infection, although the use of any one bundle component for preventing VAP
varied across respondents to their survey. This finding would suggest that efforts to close the
quality gap and improve the prevalence of use of prevention practices will need to be multifactorial.
The cost-effectiveness of several VAP prevention practicesboth subglottic suctioning
endotracheal tubes and VAP bundleshas been assessed. For subglottic suctioning tubes, it is
113
estimated that 11 people need to be treated (number needed to treat) in order to prevent one
VAP. Although the cost of these endotracheal tubes is approximately $18 USD, one model of
continuous washing tubes (inflow/outflow ports with pumping system) costs about $200 USD. In
comparison, the cost of a standard endotracheal tube is approximately $1 USD. If the number
needed to treat is accurate, these special endotracheal tubes (even the most expensive versions)
are cost-effective, especially if reserved for patients likely to remain intubated for more than 48
to 72 hours (the risk for VAP in those requiring intubation less than 48- to 72 hours is considered
low). This conclusion is further supported by Hallais and colleagues,19 who compared the cost of
these tubes to the cost of VAP, using very conservative values. The authors found that averting
only three VAPs would offset the cost of the special tubes. Based on this cost analysis, any ICU
with at least 3 VAPs per year would find that switching to these tubes reduces harm as well as
costs.
Ventilator bundles have also recently been evaluated for their cost-effectiveness. A Danish
study20 retrospectively examining ventilated patients in a single ICU found that the cost of
preventing one VAP was 4451 (approximately $6,000 USD), and the cost of preventing one
death was 31792 (approximately $42,000 USD). While the cost and incidence of each VAP
varies across patient populations, the study concluded that the ventilator bundle would likely be
cost-effective in most environments.

In conclusion, of the four key practices for preventing VAP, subglottic suctioning
endotracheal tubes have strong evidence to support their ability to reduce VAP and to do it costeffectively, based on a systematic review of multiple RCTs. Strong evidence from a recent
systematic review of multiple RCTs also supports oral care using CHG. Evidence from a few
non-randomized studies supports sedation vacations directly. This evidence is of moderate
strength. The maintenance of a head-of-bed elevation of at least 30 degrees (a ubiquitous element
of VAP prevention bundles) is supported by very little evidence, yet remains part of virtually all
recommendations by U.S. quality and safety organizations. This tacit support is likely a result of
its ease and lack of evidence of harm, although the ability to effectively implement this element
consistently has been questioned. Other elements often advocated for VAP prevention but not
specifically addressed in this chapter include using antimicrobial-coated endotracheal tubes
(evidence supports effectiveness),21 closed circuit in-line suctioning systems (evidence does not
support their effectiveness)22 and humidification circuits on ventilators (evidence does not
support their benefit).23
Evidence from multiple large pre-post studies also supports the effectiveness of VAP
bundles. While the evidence for each specific VAP prevention bundle element may vary, two
principles are clear. First, VAP is most effectively reduced by the bundling of several elements
together for a potentially synergistic effect, and bundles should be developed locally based on
both institutional expertise and evidence, with ongoing evaluation of the success of the
interventions. Second, the consistent application of each of the bundle elements to all patients
who qualify for them is essential to success. The use of teamwork tools and strategies to ensure
this consistency can have a tremendous impact on closing this quality gap and improving patient
outcomes. Technical work (the bundle) needs to be supported by adaptive work (the processes
needed to apply the bundle consistently) for the best success.18 A summary table is located below
(Table 1).
114

Scope of the
Strength of
Problem targeted by Evidence for
the PSP
Effectiveness
Common/High
Moderate-tohigh
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low-tomoderate
How Much do We
Moderate/Moderate
References
9.
Vincent JL, de Souza Barros D, Cianferoni

S. Diagnosis, management and prevention of
ventilator-associated pneumonia: an update.
Drugs 2010; 70(15):1927-44.
Muscedere J, Dodek P, Keenan S, et al.

Comprehensive evidence-based clinical
practice guidelines for ventilator-associated
pneumonia: prevention. J Crit Care 2008;
23(1):126-37.
10.
Umscheid CA, Mitchell MD, Doshi JA, et

al. Estimating the proportion of healthcareassociated infections that are reasonably
preventable and the related mortality and
costs. Infect Control Hosp Epidemiol 2011;
32(2):101-14.
Coffin SE, Klompas M, Classen D, et al.

Strategies to prevent ventilator-associated
pneumonia in acute care hospitals. Infect
Control Hosp Epidemiol 2008; 29 Suppl
1:S31-40.
11.
Niel-Weise BS, Gastmeier P, Kola A, et al.

An evidence-based recommendation on bed
head elevation for mechanically ventilated
patients. Crit Care 2011; 15(2):R111.
Labeau SO, Van de Vyver K, Brusselaers N,

et al. Prevention of ventilator-associated
pneumonia with oral antiseptics: a
systematic review and meta-analysis. Lancet
Infect Dis 2011; 11(11):845-54.
12.
Muscedere J, Rewa O, McKechnie K, et al.

Subglottic secretion drainage for the
prevention of ventilator-associated
pneumonia: a systematic review and metaanalysis. Crit Care Med 2011;39(8):198591.
13.
Resar R, Pronovost P, Haraden C, et al.

Using a bundle approach to improve
ventilator care processes and reduce
ventilator-associated pneumonia. Jt Comm J
Qual Patient Saf 2005; 31(5):243-8.
14.
Bonello RS, Fletcher CE, Becker WK, et al.

An intensive care unit quality improvement
collaborative in nine Department of
Veterans Affairs hospitals: reducing
ventilator-associated pneumonia and
catheter-related bloodstream infection rates.
Jt Comm J Qual Patient Saf 2008;
34(11):639-45.
15.
Rosenthal VD, Alvarez-Moreno C, VillamilGomez W, et al. Effectiveness of a

multidimensional approach to reduce
ventilator-associated pneumonia in pediatric
intensive care units of 5 developing
countries: International Nosocomial
Infection Control Consortium findings. Am
J Infect Control 2011.
1.
Calfee DP. Crisis in hospital-acquired,

healthcare-associated infections. Annu Rev
Med 2012; 63:359-71.
2.
3.
4.
5.
6.
Drakulovic MB, Torres A, Bauer TT, et al.

Supine body position as a risk factor for
nosocomial pneumonia in mechanically
ventilated patients: a randomised trial.
Lancet 1999; 354(9193):1851-8.
Schweickert WD, Gehlbach BK, Pohlman
AS, et al. Daily interruption of sedative
infusions and complications of critical
illness in mechanically ventilated patients.
Crit Care Med 2004; 32(6):1272-6.
7.
Dries DJ, McGonigal MD, Malian MS, et al.

Protocol-driven ventilator weaning reduces
use of mechanical ventilation, rate of early
reintubation, and ventilator-associated
pneumonia. J Trauma 2004; 56(5):943-51;
discussion 951-2.
8.
Chlebicki MP, Safdar N. Topical

chlorhexidine for prevention of ventilatorassociated pneumonia: a meta-analysis. Crit
Care Med 2007; 35(2):595-602.
115
16.
Berenholtz SM, Pham JC, Thompson DA, et

al. Collaborative cohort study of an
intervention to reduce ventilator-associated
pneumonia in the intensive care unit. Infect
Control Hosp Epidemiol 2011;32(4):30514.
20.
Moller AH, Hansen L, Jensen MS, et al. A

cost-effectiveness analysis of reducing
ventilator-associated pneumonia at a Danish
ICU with ventilator bundle. J Med Econ
2012; 15(2):285-92.
17.
Krein S, Kowalski CP, Damschroder L, et

al. Preventing Ventilator-Associated
Pneumonia in the United States: A
Multicenter Mixed-Methods Study Infect
Cont and Hosp Epid 2008; 29:933-940.
21.
Kollef MH, Afessa B, Anzueto A, et al.

Silver-Coated Endotracheal Tubes and
Incidence of Ventilator-Associated
Pneumonia The NASCENT Randomized
Trial. JAMA. 2008;300:805-813.
18.
Krein S, Kowalski CP, Hofer TP, et al.

Preventing Hospital-Acquired Infections: A
National Survey of Practices Reported by
U.S. Hospitals in 2005 and 2009. J. Gen Int
Med 2011.
22.
Siempos II, Vardakas KZ, Falagas ME.

Closed tracheal suction systems for
prevention of ventilator-associated
pneumonia. Br J Anaesth. 2008;100:299306.
23.
19.
Hallais C, Merle V, Guitard PG, et al. Is

continuous subglottic suctioning costeffective for the prevention of ventilatorassociated pneumonia? Infect Control Hosp
Epidemiol 2011; 32(2):131-5.
Siempos II, Vardakas KZ, Kopterides P, et

al. Impact of passive humidification on
clinical outcomes of mechanically ventilated
patients: a meta-analysis of randomized
controlled trials. Crit Care Med.
200;35:2843-51.
116
Chapter 12. Interventions To Allow the Reuse of Single-Use

Devices: Brief Review (NEW)
Meredith Noble, M.S.
Introduction
Many hospitals choose to reprocess single-use devices (SUD)those intended by the
manufacturer to be discarded after one use--for reuse in additional patients.1 Reprocessing
includes cleaning, sterilization, and if necessary, refurbishing. Specific information on the size of
the reprocessing industry is not available.2 According to the Association of Medical Device
Reprocessors Web site, disciplines that commonly use reprocessed devices include:
cardiovascular; arthroscopic/orthopedic; general surgery; gastroenterology; laparoscopic surgery.
A wide variety of SUDs are reprocessed. Commonly reprocessed SUDs include: arthroscopic
shavers; biopsy forceps; blood pressure cuffs; clamps and dissectors; compression sleeves;
electrophysiology catheters; external fixation devices; laparoscopic scissors and forceps; opened
but unused items; orthopedic drill bits and burrs; phaco tips; pneumatic tourniquet cuffs; pulse
oximeter sensors; scissors and staplers; soft tissue ablators; trocars.1 Opened but unused items
are not technically reused but also must be reprocessed.
Using SUDs is money-saving and generally thought to be safe.1 However, SUDs are only
required to be demonstrated to the Food and Drug Administration (FDA) as safe for one use;
some manufacturers and the Medical Device Manufacturers Association contend that reusing
SUDs is unsafe because the devices are frail and cannot be adequately cleaned and resterilized.1,3
Potential risks to patients include infection, toxicity, particulate contamination, and mechanical
failure.1
Although reuse of single-use devices is common and perhaps even pervasive, little evidence
on its safety and efficacy has been published. To gather the available evidence, a literature search
of PubMed was conducted for English language articles published between January 1, 2001 and
November 2, 2011.
What Are the Practices for Assuring the Safety of Reused Devices?
Reprocessing used SUDs is subject to FDA oversight. On August 14, 2000, FDA issued a
policy on the reuse of single-use medical devices making hospitals and third-party reprocessors
subject to all the requirements of the Federal Food, Drug, and Cosmetic Acta requirement
formerly imposed only on original equipment manufacturers (OEMs).4 In response, many
hospitals that had been reprocessing SUDs in-house began using third parties to reprocess the
devices. Unused items are not subject to FDA oversight.5
The Medical Device User Fee and Modernization Act of 2002 expanded regulatory
requirements. Premarket notification submissions (510(k)s) for certain reprocessed SUDs
identified by FDA must now include validation data. Validation data include cleaning,
sterilization, and functional performance data, which confirm that each SUD will remain
substantially equivalent to a predicate device after the maximum number of times the device is
reprocessed.4 In addition, the reprocessor must be indicated with a mark or label for each
reprocessed SUD.
117
According to FDA regulations, a third-party or hospital reprocessor must comply with the
same requirements that apply to original equipment manufacturers, including:
Submitting documents for premarket notification or approval for each device and model
reprocessed
Registering as a manufacturer with FDA and listing all products
Submitting adverse event reports
Tracking devices whose failure could have serious outcomes
Correcting or removing from the market unsafe devices
Meeting manufacturing and labeling requirements.4
The FDA considers hospitals that reprocess devices as device manufacturers subject to the
requirements of the Quality System (QS) Regulation.6 However, most hospitals who use
reprocessed SUDs obtain them from third parties who perform the reprocessing.4 An estimated
95% of reprocessing in the U.S. is completed by two firms,1 Stryker Sustainability Solutions
(formerly Ascent Healthcare Solutions, Phoenix, AZ) and SteriMed Inc. (Minneapolis, MN).3
According to their Web site, Stryker claims the majority of the reprocessing market and has over
2,000 hospital members. Their service includes delivering orders and picking up used equipment
(which hospital personnel leave in marked bins), and at their facility, sorting, cleaning,
refurbishing/repairing, repackaging, and resterilizing equipment. The FDA has cleared or
approved a variety of sterilizing agents that can be used in reprocessing7 and inspects
reprocessing facilities for compliance with regulations, with steep penalties for violators.5

Reprocessing protocols in the peer-reviewed published literature vary but generally include
cleaning and sterilization. Cleaning may consist of manual or automated washing with water and
detergent or enzymatic solution. Sterilization may entail pressurized steaming (i.e., in autoclave),
ethylene oxide (especially for heat sensitive items) or gamma radiation. Quality assurance is
intended to verify sterilization success. The FDA urges the use of biological indicators to verify
that test organisms are killed.6 Chemical indicators verify that sufficient temperatures were
achieved or sterilant was present in the sterilizer in each sterilization run. Repairs and part
replacements should be made as necessary. The FDA does not require the use of particular
protocols, but may prefer standard procedures such as those recommended by the Association for
the Advancement of Medical Instrumentation.6
Reprocessing is performed in hospitals or by independent third parties at separate facilities.
Data suggest the majority of hospitals that reprocess devices use third parties.1 When such a
vendor is used, implementation for the hospital should not pose challenges. Personnel must
remember to place used devices in bins provided by the reprocessing vendor; the vendor
provides pick up, reprocesses the items, and delivers ready-to-use items. Ordering reprocessed
devices should not differ from ordering new devices.

Cleaning. Theoretically, cleaning should remove all debris and sterilization should inactivate
potentially infective viruses, bacteria, and fungi. Literature searches performed for this review
identified nine laboratory studies published in the last 10 years that tested an array of reprocessed
SUDs for microbiological contamination. Devices studied included laparoscopic instruments,8,9
118
various catheters,10-12 trocars,13 sphincterotomes,14 diathermy pencils,15 and tracheostomy

tubes.16 While most studies could not demonstrate microbial contamination after reprocessing,
four found that reprocessed SUDs were contaminated.9,11,15,16 Two of the studies also reported
damage, incomplete kits, and/or compromised functioning.9,16 Another study assessed cleaning
to remove test soils from biopsy forceps and found up to 95% of the material was removed.17
Effect on patient outcomes. The literature search spanning the last 10 years identified only one
randomized controlled study that compared new and reprocessed SUD laparoscopic instruments
used to perform cholecystectomy; the study found no significant differences in outcome.21 This
study is small (125 patients) and may be underpowered to detect rare events. A single study may
not be representative of outcomes in general; devices and protocols will vary.
A meta-analysis of nine studies that compared new and reprocessed hemodialyzers found
reuse was associated with an increased risk of hospitalization but no difference in mortality.20
The U.S. Government Accountability Office (GAO) published a report in January 2008
discussing FDA oversight of reprocessed SUDs and the available information on the potential
health risks of using reprocessed SUDs. The report concluded:
The limited number of peer-reviewed studies related to
reprocessing that we identified were insufficient to support a
comprehensive conclusion on the relative safety of reprocessed
SUDs. Despite the limitations of available data, FDAs analysis of
reported device-related adverse events does not show that
reprocessed SUDs present an elevated health risk.2
Cost savings. If using reprocessed SUDs is as safe as using new devices, saving costs on
materials would free hospital resources for other uses without compromising patient safety. Our
searches identified four cost studies published in the last 10 years. A modeled European study
found that the cost savings for reprocessing cardiac electrophysiology catheters was 33% for
ablation applications and 41% for diagnostic.18 A modeled Canadian study found savings of $0
to $739 per year per patient when hemodialyzers were reprocessed.18 Hemodialyzers are
typically reused only by the same patient. A meta-analysis of nine studies in which various
devices were used had an overall savings rate of 49%, but stipulated that the few studies
identified were of poor quality and had missing data, including adverse event cost data.19 A
meta-analysis of nine studies that compared new and reprocessed hemodialyzers found reuse was
associated with small cost savings, an increased risk of hospitalization, and no difference in
mortality.20 The Stryker Web site states that member hospitals can save 50% over purchasing
new equipment in acquisition costs, and 70%80% in operating room medical waste disposal
costs. Reusing devices should also reduce wastes for landfill.
What Methods Have Been Used To Improve These Practices?

Healthcare Risk Control (HRC) is a service ECRI Institute offers for risk managers. HRC
provides resources for a variety of patient safety issues, and specifically recommends that
hospital systems considering use of reprocessed SUDs should at a minimum, establish written
policies, procedures, and policies for such practices... [and] should be widely circulated
throughout the organization.1 They further recommend establishing a reuse committee
comprised of individuals from multiple departments, including materials management, risk
management and/or hospital legal counsel, infection control, clinical and/or biomedical
119
engineering, administration, central sterile supply, surgery, finance, and physician(s) advocating
reuse.1 A third party reprocessor should be selected based upon registration with the FDA and
compliance with FDA regulations; the types of devices to be reprocessed; and support (including
logistical support such as device pickup).1 Reprocessors usually provide template policies and
procedures to hospitals to support implementation.1

Reprocessing SUDs with appropriate quality controls should theoretically guarantee
sterilization. Less information is available on the integrity of the devices themselves after
reprocessing; the FDA recommends that reprocessors test all devices to ensure that functionality
is maintained.
Some laboratory studies in the clinical literature found that various devices were not sterile;
however, quality assurance should prevent unsterile devices from being reused. Some devices
remained unsterile after multiple attempts; use of the protocols or reuse of the particular device
warrants reconsideration in such circumstances.
Clinical literature and data on real-world use are currently not robust enough for the GAO or
independent authors to firmly conclude that reused SUDs are safe. However, one systematic
review found an increased rate of adverse events in patients treated with reused SUDs. The
protocols in the peer-reviewed literature may differ from those used by third party reprocessors.
A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Low
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low
How Much do We
A lot/Not difficult
References
1.
2.
ECRI Institute. Operating room risk

management. Vol. 2, Technology
management 11. Plymouth Meeting (PA):
ECRI Institute; 2009 Nov. Use of
reprocessed single-use medical devices.
United States Government Accountability
Office. Reprocessed single-use medical
devices: FDA oversight has increased, and
available information does not indicate that
use presents an elevated health risk. Report
to the Committee on Oversight and
Government Reform, House of
Representatives [GAO-08-147]. Washington
DC: United States Government
Accountability Office; 2008 Jan. 37 p.
www.amdr.org/documents/GAOReprocessin
gReportd08147.pdf.
120
3.
ECRI Institute. Safety and cost implications

for reusing single-use medical devices.
Plymouth Meeting (PA): ECRI Institute;
2009 May 5. 10 p. (Hotline Response).
4.
Reprocessing of single-use devices.

[internet]. Rockville (MD): Food and Drug
Administration; 2009 May 20 [accessed
2011 Nov 22]. [1 p].
www.fda.gov/MedicalDevices/DeviceRegul
ationandGuidance/ReprocessingofSingleUseDevices/default.htm.
5.
Reuse of single-use devices. [internet].

Charlotte (NC): Premier Inc.; 2011
www.premierinc.com/quality-safety/toolsservices/safety/topics/reuse/.
6.
7.
Reprocessing of single-use devices:

frequently asked questions. [internet].
Rockville (MD): Food and Drug
Administration; 2009 Aug 21 [accessed
2011 Nov 22]. [4 p].
ationandGuidance/ReprocessingofSingleUseDevices/ucm121093.htm.
FDA-cleared sterilants and high level
disinfectants with general claims for
processing reusable medical and dental
devices. [internet]. Rockville (MD): Food
and Drug Administration; 2009 Mar
[updated 2009 Apr 26]; [accessed 2011 Nov
22]. [4 p].
ationandGuidance/ReprocessingofSingleUseDevices/ucm133514.htm.
8.
Lopes Cde L, Graziano KU, Pinto Tde J.

Evaluation of single-use reprocessed
laparoscopic instrument sterilization. Rev
Lat Am Enfermagem 2011 Apr;19(2):370-7.
PMID: 21584385.
9.
Roth K, Heeg P, Reichl R. Specific hygiene

issues relating to reprocessing and reuse of
single-use devices for laparoscopic surgery.
Surg Endosc 2002 Jul;16(7):1091-7. PMID:
12165829.
10.
Lester BR, Boser NP, Miller K, et al.

Reprocessing and sterilization of single-use
electrophysiological catheters: removal of
organic carbon and protein surface residues.
J AOAC Int 2009 Jul-Aug;92(4):1165-73.
PMID: 19714986.
11.
Fedel M, Tessarolo F, Ferrari P, et al.

Functional properties and performance of
new and reprocessed coronary angioplasty
balloon catheters. J Biomed Mater Res B
Appl Biomater 2006 Aug;78(2):364-72.
PMID: 16506183.
12.
Tessarolo F, Caola I, Caciagli P, et al.

Sterility and microbiological assessment of
reused single-use cardiac electrophysiology
catheters. Infect Control Hosp Epidemiol
2006 Dec;27(12):1385-92. PMID:
17152039.
13.
dos Santos VS, Zilberstein B, Possari JF, et

al. Single-use trocar: is it possible to
reprocess it after the first use. Surg Laparosc
Endosc Percutan Tech 2008 Oct;18(5):4648. PMID: 18936667.
121
14.
Alfa MJ, Nemes R. Inadequacy of manual

cleaning for reprocessing single-use, triplelumen sphinctertomes: simulated-use testing
comparing manual with automated cleaning
methods. Am J Infect Control 2003
Jun;31(4):193-207. PMID: 12806356.
15.
Batista Neto S, Graziano KU, Padoveze

MC, et al. The sterilization efficacy of
reprocessed single use diathermy pencils.
Rev Lat Am Enfermagem 2010 JanFeb;18(1):81-6. PMID: 20428701.
16.
da Silva MV, Ribeiro Ade F, Pinto Tde J.

Safety evaluation of single-use medical
devices after submission to simulated
reutilization cycles. J AOAC Int 2005 MayJun;88(3):823-9. PMID: 16001858.
17.
Alfa MJ, Nemes R, Olson N, et al. Manual

methods are suboptimal compared with
automated methods for cleaning of singleuse biopsy forceps. Infect Control Hosp
Epidemiol 2006 Aug;27(8):841-6. PMID:
16874645.
18.
Tessarolo F, Disertori M, Caola I, et al.

Health technology assessment on
reprocessing single-use catheters for cardiac
electrophysiology: results of a three-years
study. Conf Proc IEEE Eng Med Biol Soc
2007;2007:1758-61. PMID: 18002317.
19.
Jacobs P, Polisena J, Hailey D, et al.

Economic analysis of reprocessing singleuse medical devices: a systematic literature
review. Infect Control Hosp Epidemiol 2008
Apr;29(4):297-301. PMID: 18462140.
20.
Manns BJ, Taub K, Richardson RM, et al.

To reuse or not to reuse? An economic
evaluation of hemodialyzer reuse versus
conventional single-use hemodialysis for
chronic hemodialysis patients. Int J Technol
Assess Health Care 2002 Winter;18(1):8193. PMID: 11987444.
21.
Colak T, Ersoz G, Akca T, et al. Efficacy

and safety of reuse of disposable
laparoscopic instruments in laparoscopic
cholecystectomy: a prospective randomized
study. Surg Endosc 2004 May;18(5):727-31.
PMID: 15026911.
Section C. Surgery, Anesthesia, and Perioperative

Medicine
Chapter 13. Preoperative Checklists and Anesthesia
Checklists
Jonathan R. Treadwell, Ph.D.; Scott Lucas, Ph.D., P.E.

Surgical operations greatly benefit the public health; however, they can also be directly
responsible for substantial morbidity and mortality. In industrialized countries, the rate of
perioperative death directly due to inpatient surgery has been estimated at 0.4 percent to 0.8
percent, and the rate of major complications has been estimated at 3 percent to 17 percent.1,2
Sources of these complications are numerous, including wrong-patient/procedure/site surgery,
anesthesia equipment problems, lack of availability of necessary equipment, unanticipated blood
loss, non-sterile equipment, and surgical items (e.g., sponges) left inside patients. The
complexity of most surgical procedures requires a well-coordinated team to prevent these events.
The medical community recognizes that anesthesia has reached a high level of safety;
however, with increased awareness, it is believed that the risk, particularly morbidity risk, can be
further reduced.3 As an example of increasing awareness, in June 2010, the European Board of
Anesthesiology (EBA) and the European Society of Anesthesiology (ESA) jointly adopted the
Helsinki Declaration on Patient Safety in Anaesthesiology.4 Also, the journal Health Devices
listed, Anesthesia hazards due to incomplete pre-use inspection as one of the top ten
technology hazards for 2012.5

Preoperative checklists can help prevent errors and complications related to surgery.
Checklists are often implemented within a multifactorial strategy of interventions; therefore they
usually cannot be judged alone as a patient safety practice. The World Health Organization
(WHO) Surgical Safety Checklist is a prominent example of a preoperative checklist intended to
ensure safe surgery and minimize complications; it has been translated into at least six
languages.6 Because of its prominence and importance, the majority of our review for this PSP
details the WHO checklist: its development, pilot testing, context and implementation at different
sites, and degree of adoption and diffusion around the world.
In addition to the WHO checklist, we also reviewed evidence on three other types of
checklists:
The SURPASS checklist.7-10 The checklist encompasses not only the operation itself, but
all events from admission to surgery to discharge
Checklists specifically intended to prevent wrong-site surgery. Two items on the WHO
checklist address wrong-site surgery (Has the patient confirmed his/her identity, site,
procedure, and consent? and Is the site marked?). In 2004, the Joint Commission
created the Universal Protocol for Preventing Wrong Site, Wrong Procedure, Wrong
Person Surgery.11 It comprises three sets of steps: pre-operative verification process,
marking the operative site, and a time out immediately before the operation. A
122
checklist can potentially be used to clarify the details of these three steps. The Universal
Protocol is intended to prevent wrong surgery not just in the operating room but
anywhere an invasive procedure is performed (e.g., interventional radiology unit).12
Checklists specifically intended to check anesthesia equipment. The WHO checklist also
contains a specific item about preoperative anesthesia (Is the anaesthesia machine and
medication check complete?). This single item could itself be addressed by a subchecklist. In 2008, the American Society of Anesthesiologists provided general
guidelines about items that should be checked before surgery, and institutions can
implement the guidelines to tailor the checklist to their specific equipment and clinical
settings.13
Checklists have also been developed, implemented, and assessed outside of the realm of
surgery. The Michigan ICU checklist (also referred to as the Keystone project) has been shown
to prevent central line-associated bloodstream infections (CLABSI).14,15 This program involved a
multifactorial intervention at 108 Michigan ICUs. Data showed a reduction in CLABSI from 7.7
infections per 1,000 catheter-days before the program to only 1.4 infections per 1,000 catheterdays at 16 to 18 months after program initiation follow-up. A 2001 study by Dixon-Woods and
colleagues15 proposed six reasons for this reduction, including creating a densely networked
community with strong horizontal links that exerted normative pressures on members and
harnessing data on infection rates as a disciplinary force. A recent systematic review of this
program (and other PSPs to prevent hospital-associated infections) was conducted by the Blue
Cross and Blue Shield Evidence-based Practice Center; please refer to that report for further
information about the Keystone project.16
Background Information About Preoperative Checklists

In January 2007, the WHO Patient Safety group started work on the Second Global Patient
Safety Challenge: Safe Surgery Saves Lives. This group of international experts identified four
areas of potential improvement in surgical safety: surgical site infection prevention, safe
anesthesia, safe surgical teams, and measurement of surgical services.17 Based on that work, in
early 2008, the WHO published a guideline for safe surgery.18 This guideline was used as the
basis for the WHO Surgical Safety Checklist, which was launched in June 2008.
The checklist, which was included as a Supplementary file in the original publication,6
contained 19 items in three phases with collaborative involvement of the surgeon, the anesthetist,
and the nursing team:
Before induction of anesthesia (Sign In), covering areas such as patient identification,
anesthesia equipment check, and a pulse oximetry check
Before skin incision (Time Out), covering areas such as team introductions, review of
critical steps, and antibiotic prophylaxis
Before patient leaves operating room (Sign Out), covering areas such as checking
counts of instruments, specimen labeling, and concerns for recovery
The SURPASS checklist (SURgical PAtient Safety System)7-10 is intended to address any
events that occur between patient admission and discharge. Thus, it encompasses more potential
areas of safety than the WHO checklist, which is focused only on the operating room. An
estimated 53 percent to 70 percent of surgical errors occur outside the operating room.8,19,20
Within the operating room itself, the SURPASS checklist is less specific than the WHO checklist
123
(for example, the SURPASS checklist does not specifically mention any of the following: pulse
oximetry, difficult airway, risk of blood loss (although it asks whether blood products are
available), team introductions, and anticipation of critical events).
In January 2004, the Joint Commission launched the first version of the Universal Protocol
(UP) for Preventing Wrong Site, Wrong Procedure, Wrong Person Surgery.11,21 It comprises
three sets of steps: pre-operative verification process, marking the operative site, and a time
out immediately before the operation. The preoperative verifications (of person, procedure, and
site) are supposed to occur not only in the operating room, but also (if applicable) when the
procedure is scheduled, when the patient enters the health care facility, and anytime care is
transferred between caregivers. Site marking should involve only the operative site and should be
visible before the patient is draped. The time out is to occur before incision and involve the
entire operating room team. The Universal Protocol is not a checklist,12 but it could be
implemented using one or more checklists. Both steps 1 and 3 specifically mention the potential
use of a checklist.
Anesthesia safety guidelines and standards are actively reviewed and modified globally
through organizations such as the WHO and the World Federation of Societies of
Anaesthesiologists (WFSA).22 The latest WFSA standard, which was developed as part of the
Safe Surgery Saves Lives project, recommends that an appropriate pre-list check be
performed prior to the start of each operating list and an appropriate pre-patient check be
performed prior to each anesthetic. In addition, individual anesthesia societies are developing
guidelines for pre-anesthesia checks, including the American Society of Anesthesiologists (ASA)
and the Association of Anaesthetists of Great Britain and Ireland (AAGBI). The latest U.S. preuse checkout guidelines, entitled Recommendations for Pre-Anesthesia Checkout (PAC)
Procedures, were published in 2008 by the ASA.13 These guidelines were a result of a multiyear effort by an ASA task force consisting of members from the ASA, the American
Association of Nurse Anesthetists (AANA), the American Society of Anesthesia Technicians
and Technologists (ASATT), and major anesthesia system manufacturers. The latest AAGBI
revision was published in 2004 and has been adopted by many institutions around the country.23
Similar to the WFSA guideline checklist, the full ASA and AAGBI checklists were designed to
be used at the start of the day with a subset of the full checklists performed prior to each
procedure. These societies sample checklists were developed as a basis for institutions to
develop their individual checklists.
Additional background information about preoperative checklists, including how they were
developed and modified, and the overlap between different checklists, appears in Appendix A.

No formal model exists for why preoperative checklists should reduce surgical errors, but
studies have cited several common reasons. These reasons include ensuring that all critical tasks
are carried out, encouraging a non-hierarchical team-based approach; enhancing communication;
catching near misses early, anticipating potential complications, and having technologies to
manage anticipated and unanticipated complications. With regard to anesthesia checklists,
Staender and Mahajan3 attribute the reduced anesthesia-related mortality rates to a combination
of interventions, including incident reporting, simulations, and checklists.
124

In this section, the primary issues surrounding checklists involve implementation, rather than
whether they are effective. Consequently, we briefly summarize the primary results, and the bulk
of our work appears later in detailed assessments of the implementation efforts.
World Health Organization Checklist

The 2008 WHO Surgical Safety Checklist was tested at eight sites (Prince Hamzah Hospital
in Amman, Jordan; St. Stephens Hospital in New Delhi, India; University of Washington
Medical Center in Seattle, U.S.; St. Francis Designated District Hospital in Ifakara, Tanzania;
Philippine General Hospital in Manila, Philippines; Toronto General Hospital in Toronto,
Canada; St. Marys Hospital in London, England; and Auckland City Hospital in Auckland, New
Zealand).6 These settings varied greatly in the number of beds (range 371 to 1800), the number
of operating rooms (range 3 to 39), and the income level of the country (four low, four high).
Surgical safety policies prior to implementation of the WHO Checklist also differed regarding
the use of routine intraoperative monitoring with pulse oximetry (six of eight sites), oral
confirmation of patients identity and surgical site in the operating room (only two of eight sites),
and routine administration of prophylactic antibiotics in the operating room (five of eight sites).
None of the eight sites had a standard plan for intravenous access for cases of high blood loss,
or formal team briefings preoperatively or postoperatively.
Baseline data were obtained at each site for 3 months prior to checklist introduction,
involving a total of 3,733 surgical procedures. In the subsequent 3- to 6-month period after
checklist introduction, involving 3,955 procedures, data showed decreases in patient mortality
(from 1.5% to 0.8%) and inpatient complications (from 11% to 7%). No single site was driving
the findings, as evidenced by the persistence of findings after the removal of any single site in a
sensitivity analysis. Authors found that the performance rates for six specific safety indicators
(e.g., using a pulse oximeter) also increased after checklist introduction, suggesting that the
safety indicators may have been responsible for the lower rates.
In discussing the results, authors acknowledged that the underlying reasons for the
improvements were most likely multifactorial and included explanations such as the following:
The checklist itself
A Hawthorne effect (i.e., rates may have decreased because operating room personnel
knew they were being measured). The authors argued against this possibility based on
two aspects of their data: (1) that this knowledge was in place both before and after
checklist introduction, and (2) the subset of procedures for which study personnel were
present in the operating room had the same reductions in complications as procedures
where study personnel were absent from the operating room.
The simple existence of a formal pause or preoperative briefing (which could be done
without a checklist). Such a pause is a necessary component of the checklist.
Increased uptake of safety technologies (e.g., administering antibiotics in the operating
room rather than in preoperative wards). This change could be considered a byproduct of
checklist introduction (i.e., hospitals made more antibiotics directly available in the
operating room because of the presence of an antibiotics-related item on the checklist)
A broad change in safety culture and teamwork at that site, an explanation supported by
the finding that greater increases in safety attitudes at the pilot sites were associated with
greater reductions in complications.24
125
In a subsequent, 2010, publication, Weiser and colleagues25 presented a subgroup analysis of

the 2009 NEJM publication that was focused on urgent surgery (defined as surgery required to
be performed within 24 hours of assessment in order to be beneficial). Complications dropped
from 18 percent in the pre-intervention phase to 12 percent in the post-intervention phase, and
death dropped from 3.7 percent to 1.4 percent. Also, a 2011 study by Haynes and colleagues24
reported data on the Safety Attitudes Questionnaire (SAQ) in the eight pilot sites before and after
checklist introduction. The SAQ is scored on a 1 to 5 scale, where a 5 represents the most safetyconscious attitude. Scores on the SAQ were only slightly higher in the phase after checklist
introduction than before introduction (4.01 vs. 3.91, representing an increase of only 2.5 percent
of the scale range; this small difference was nevertheless statistically significant). However, the
change in SAQ scores was associated with reduced complication rates (Pearson r=0.71), meaning
that sites with greater improvements in safety attitude tended to have greater reductions in
complications. The publication also reported that 80 percent of respondents considered the
checklist easy to use, 20 percent believed it took too long, and when respondents were asked if
they would want the checklist used if they were undergoing surgery, 93 percent said yes.
SURPASS Checklist
An empirical test of the 90-item SURPASS checklist was reported in a 2010 study by
DeVries and colleagues.7 The design was a 6-month interrupted time series with concurrent
controls; six hospitals using the checklist were matched with five other hospitals that did not, and
researchers measured the rates of surgical complications in both groups. The 11 hospitals were
distributed through the Netherlands and comprised six tertiary hospitals, three academic
hospitals, and two regional hospitals; numbers of beds per hospital ranged from 3801002. These
hospitals had already been measured for their safety performance, so the potential Hawthorne
effect is lower than it would have been in hospitals just starting to be measured. Regarding
implementation, authors stated that the SURPASS checklist involved extensive time and effort.
A random sample of cases generally revealed good compliance with the checklist (median 80%).
The 3-month period after the checklist was initiated (compared with the 3 months before)
saw numerous improvements: decreases in the percentage of patients with complications (from
15% to 11%), in-hospital mortality (1.5% to 0.8%), patient temporary disability (9.4% to 6.6%),
and reoperations (3.7% to 2.5%). No such improvements were found among the control
hospitals. Interestingly, the extent of improvement was associated with greater compliance with
the checklist: the 566 patients whose surgery involved greater checklist compliance had 7.1
complications per 100 patients, which was considerably lower than the 18.8 per 100 experienced
by the 580 patients whose surgery involved less checklist compliance. This finding provided
greater confidence that the checklist itself was the reason for the improvements. A subsequent
retrospective review of 294 medical claims10 estimated that 40 percent of deaths and 29 percent
of liability incidents might have been prevented if the SURPASS checklist had been used.
Wrong-Site Surgery Checklists

Wrong-site surgery is relative rare: Estimates for various procedures range from 1 in 13,000
procedures for wrong-site anesthesia block to 1 in 4,200 for wrong-side ureteral stents.26 A
general systematic review estimated that the overall rate was 1 to 5 per 10,000 procedures.27
Given the rarity, demonstrating a statistical reduction would require an unfeasibly large study. A
systematic review searched for literature and concluded there was no literature to substantiate
the effectiveness of the current JC [Joint Commission] Universal Protocol in decreasing the rate
126
of wrong site, wrong level surgery.27 Therefore, the preventive benefits of a checklist to prevent
wrong-site surgery, are generally assumed based on clinical expertise.
Anesthesia Equipment Checklists

In evaluating beneficial effects, the same limitations apply to anesthesia checklists as apply
to wrong-site surgery checklists. The rate of mortality associated with malfunction of anesthesia
equipment is 1:100,000, and the rate of severe morbidity ranges between 1:170 and 1:500.3
Future research may be possible to evaluate the severe morbidity rate; however, addressing the
benefits on mortality would require an unfeasibly large study.
A 2000 randomized cross-over study by Blike and Biddle28 compared the effectiveness of the
1994 hard-copy version of the FDA-approved AACR to a researcher-designed electronic
checklist. Machine faults were purposely entered into an Ohmeda Modulus II Plus Anesthesia
System. Participants using the electronic checklist were first given a researcher-drafted
philosophy of anesthesia apparatus checkout, which outlined basic strategies to reduce
anesthesia apparatus-related patient injury. They reported that the electronic checklist greatly
improved the detection of prearranged anesthesia equipment faults. For 19 of the 20 faults
studied, the electronic checklist was either equal or superior to the AACR. However, the
electronic checklist missed 30 percent of the difficult faults (e.g., breathing circuit leak). While
this percentage was better than when the AACR was used (60%), it is still substantial. Studies
like these provided the basis for revising the AACR. For additional references to the
effectiveness of the AACR, we refer the reader to the 2008 ASA guideline; most of this literature
was published prior to 2000.
Ben-Menachem and colleagues29 performed a simulation study, published in 2011 that used
the 2008 ASA guideline to measure the performance of anesthesia residents of Sheba Medical
Center (Israel). The residents were instructed to complete the ASA checklists during simulationbased scenarios, which included two pre-arranged equipment failures. The study showed that 25
of 28 participants correctly performed 70 percent or more of the items on the checklist that is
used before the first-morning case, and 27 of 30 participants correctly performed 70 percent or
more of the items on the between-case checklist. Regarding the pre-arranged equipment failures,
30 of 31 participants identified O2 supply and pressure alarms and 30 of 30 participants
recognized an abnormal capnograph waveform.

Direct harms of preoperative checklists have not been reported. In 2011, Sewell30 reported
that after WHO implementation, the rate of lower respiratory tract infections actually increased
from 2.1 percent to 2.5 percent. Whether this increase was caused by the checklist is unclear;
however the authors attributed rate reductions to the checklist, so they could also have attributed
rate increases to the checklist. In 2011, Kearns31 reported that 3 months after WHO checklist
implementation, 30 percent believed it was an inconvenience in emergency cases; however, this
percentage was lower than it had been prior to implementation of the checklist when staff were
asked hypothetically whether they believed it would be an inconvenience in emergency cases
(53 percent said it would be). In 2010, Thomassen and colleagues32 reported user experiences
with their pre-induction anesthesia checklist. In this qualitative study, focus group interviews
were conducted amongst the participating nurses and physicians. Users reported that checklist
use could divert attention from the patient and that it interfered with doctor-nurse workflow,
although the latter improved with increased use.
127

What Contexts?
World Health Organization Checklist
In 2011, one of the eight pilot sites that piloted the WHO Checklist reported checklist-related
opinions of surgical team members 18 months after checklist introduction.33 Team members
reported high levels of agreement with the questions Do you think the use of the checklist has
improved patient safety? Are you comfortable in reminding other members of the team to
carry out the checklist? If you were to undergo surgery would you want the checklist to be
used? and Do you think that use of the checklist generally has improved communication
among members of the Operating Room team? Team members generally estimated that it took
about two minutes to complete the checklist.
We identified nine reports of the implementation of the WHO checklist at other sites;
implementation details at each site appear in Table 1 in Appendix D in the section titled
Evidence Tables for Chapter 13. Eight studies used the 2008 WHO checklist as a basis, and
one did not say which version was used. Six studies modified the WHO checklist, according to
either surgical specialty (three studies) or country (three studies). Of six studies that modified the
checklist, five provided their modified checklist within the paper, and of these five, four included
all of the WHO items and one did not (they had deleted some items).
Six studies were case series, and three were before-after studies. Regarding a theory or logic
model, eight of nine provided some statements about why a checklist should work to reduce
complications (e.g., Checklists may be used to improve patient safety by ensuring that all
elements of a practice are instituted for each new clinical event).31 Six studies were conducted
in the UK, two in the U.S., and one in Finland. Four studies involved surgical specialties
(pediatrics, OB-GYN, orthopedics, and otorhinolaryngology), and the other five were general
surgery.
Only two studies reported on the pre-existing safety infrastructure: one stated that a core
group of patient safety experts was in place before checklist implementation, and another stated
that a hospital quality infrastructure had been in place for five years prior to implementation.
Two studies reported information on the pre-existing safety culture, and they both measured staff
attitudes specifically about checklist-related items. In one, some safety aspects were fairly good
(knowledge of OR-teams names and roles, the rate of recording of postoperative follow-up
instructions, and overall successful communication range from 61% to 93%); however the rate of
discussing risks was only 24 percent. In the other study, most respondents (81% to 85%)
believed that communication in the operating room could improve and that for elective surgery
the checklist would be useful) and only 31 percent already felt familiar with other operating
room team members. However, for emergency surgery, a slight majority (53%) believed that
introducing the checklist would be inconvenient. All of these opinions were hypothetical as they
were solicited before checklist introduction.
The results of the nine implementations appear in Table 2 in Appendix D in the section titled
Evidence Tables for Chapter 13. Regarding checklist training, three sites mentioned
educational sessions, three used posters in the operating room, two mentioned a hospitalwide
publicity campaign, two mentioned that training was provided (however, no details were
provided), and two either failed to mention training or stated that only limited training was
provided. Four studies mentioned a pilot testing period; these pilot tests lasted 1 to 3 months and
often involved minor modifications to the checklist. Three studies reported the degree of
128
compliance with the checklist; one simply reported 97 percent compliance, and the other two
reported improvement over time (from approximately 60% to approximately 80% in one study,
and from 85% to 95% in another study). One study reported that it took about two minutes to
complete the checklist, and that 20 percent of respondents believed it caused an unnecessary time
delay.
Feedback from surgical teams was generally positive, but support tended to be greater from
nurses and anesthetists than from surgeons. Two studies reported increases in certain attitudinal
variables such as the degree to which people felt familiar with others in the operating room, the
quality of communication, the anticipated safety of patients, and the usefulness of the checklist in
either elective or emergency cases. Behaviorally, one study reported that after 3 months, team
briefings were occurring in 77 percent of operations and time-outs in 86 percent. Another study
reported improvements in anesthetists knowledge about patients, their check of anesthesia
equipment, and staff knowledge of patient identity/procedure/site.
Reasons cited for success included good training and staff understanding, a local champion,
support from upper management, being able to modify the checklist, distribution of
responsibility, the feeling of ownership by team members, and enhanced communication and
teamwork. Barriers to implementation included general surgeon resistance to changing habits,
the belief that they were already checking those things, awkwardness of self-introductions, steep
interpersonal hierarchy, and a fear of legal responsibility if a complication occurred after they
had signed a form.
One ongoing research project, funded by the Agency for Healthcare Research and Quality, is
entitled Factors Associated with Effective Implementation of a Surgical Safety Checklist.34
This 2010-2013 project will examine implementation processes in a large group of U.S. and
international hospitals to identify factors supportive of effective implementation. Further, the
team will determine how teamwork helps explain the impact of the checklist.
The WHO Web site (www.who.int/patientsafety/safesurgery/en/) provides advice to hospitals
for implementing the checklist.35 This advice includes statements such as The key to successful
implementation is to start small. Start with a single operating room on day 1 and see how it
works. This will guide you to strategies for altering the checklist to fit your needs, as well as
identify potential barriers to adaptation.35 Other implementation advice from WHO is available
in the Frequently Asked Questions section
(www.who.int/patientsafety/safesurgery/faq_introduction/en/index.html), the 20-page Starter
Kit for Implementing the Surgical Safety Checklist
(www.who.int/patientsafety/safesurgery/testing/participate/starter_kit-sssl.pdf), and the
Checklist Adaptation Guide
(www.who.int/patientsafety/safesurgery/checklist_adaptation.pdf). Regarding checklist
modification, the Web site states, Do not hesitate to customize the checklist for your setting as
necessary, but do not remove safety steps just because you are unable to accomplish them. Also,
regarding feasibility, the WHO states that It should take no more than a minute to complete
each section of the checklist (i.e., three minutes in total). The pilot study reported that at various
sites, introduction of the checklist took only 1 week to 1 month.6
The Web site www.safesurg.org also provides additional materials relevant to the WHO
checklist. Those interested in implementing the checklist are encouraged to register with the Web
site. The Web site provides a template for the 2009 checklist (www.safesurg.org/templatechecklist.html) in Microsoft Word format (Microsoft Corporation, Redmond, Washington, U.S.),
and an implementation manual is available (www.safesurg.org/implementation-manual.html).
129
The Web site also provides a list of other institutions modified checklists
(www.safesurg.org/modified-checklists.html), where institutions can submit their modifications
of the WHO checklist to be made publically available. On October 3, 2011, the publicallyavailable list contained 79 checklists from 25 countries.
The site also provides several downloadable videos (www.safesurg.org/videos.html): one on
how to use the WHO checklist; one on how not to use it; two from the National Patient Safety
Agency in the United Kingdom; one from University Health Network Hospital in Toronto; one
from the Surgical Care and Outcomes Assessment Program in Washington State; two from Great
Ormond Street Hospital in the United Kingdom; one in French from Fattouma Bourguiba
Hospital in Monastir, Tunisia; one in Spanish from la Agencia de Calidad de Andalucia; one
Spanish translation of the WHO how-top video; and two from Auckland City Hospital in New
Zealand.
SURPASS Checklist
We performed a citation search to determine if the SURPASS checklist has yet been
attempted outside the Netherlands; however, no such attempts were identified. The SURPASS
Web site (www.surpass-checklist.nl/home.jsf?lang=en) describes an electronic version of the
checklist (called SURPASS Digital) that can be used by any web-connected computer. The
electronic version allows one to modify the checklist, although the designers of SURPASS
strongly encourage users to avoid modification (www.surpasschecklist.nl/content.jsf?pageId=FAQ&lang=en).
Wrong-Site Surgery Checklists

No implementation advice was found on the Joint Commission Web site or in other
published documents. In August 2010, the Joint Commission conduced an online survey of over
2,100 people.36 The Web site did not report how many questions were asked or the wording of
any given question. The Web site reports five findings from the survey: 1) 88 percent agreed that
their organizations could fully implement the Universal Protocol; 2) 87 percent to 92 percent
agreed that the three steps are appropriate; 3) More than 90 percent agreed that there is benefit
in using it in the operating room, ambulatory surgery, and hospital units performing invasive
procedures, but the rates of agreement of benefit were lower for ambulatory clinics and physician
offices; 4) the need to modify policies and procedures varied greatly across respondents; and 5)
no differences were found between different types of respondents (e.g., type of hospital, bed
size).
We identified four sites describing pertinent checklists (see Table 3 in Appendix D in the
section entitled Evidence Tables for Chapter 13). These sites were located in Switzerland,
Sweden, the United Kingdom, and North Carolina:
The Swiss study37 was conducted in a large anesthesiology service and focused on verifying
two key aspects: patient identity and surgical site. The protocol was developed by an
interdisciplinary team and required patient participation in the verification of identity and
surgical site (answering open-ended questions rather than closed-ended questions). Compared
with the first 3 months of implementation, the next 3 months saw better compliance in checking
patient identity (63% up to 81%) and proportion of surgical site checks performed (77% up to
93%). Compliance was stable in subsequent periods. Barriers to implementation included 1)
surgeons saying they already knew the patients or the surgical site was obvious, and 2) the
failure to include the input of all surgical services in developing the protocol.
130
The Swedish study38 involved two hospitals, each of which had a recent wrong-site surgery
incident, and a root-cause analysis suggested that a time-out procedure might help. A time-out
checklist was implemented, and one year later, a questionnaire was sent to all 704 team
members. Of the 331 responders, 93 percent expressed the belief that the checklist contributes to
increased patient safety (either without a doubt, or probably). When asked about eight
specific components of the time-out checklists, the percentage of respondents who believed the
component was very important varied widely, from a low of 14 percent for the introduction of
team members to highs of over 80 percent for patient identity, correct procedure, and correct
side.
The English study39 was conducted at a childrens hospital in which staff had incorporated an
eight-item correct-site surgery checklist into an existing preoperative checklist. Five people were
required to sign the documentation: marking surgeon, operating surgeon, ward nurse, scrub
nurse, and anesthetist. Comparing 2008 to 2006, correct completion was unimproved for four of
the eight items (ward nurse signed, operating surgeon signed, scrub nurse, signed, and operating
department practitioner signed) but was improved for the other four (mark site documented, no
mark required documented, entries legible, and marking surgeon signed).
The North Carolina study37 implemented a checklist to prevent wrong-site surgery that was
tailored to the hospitals preferences and procedures. Previously, the staff was using a
cumbersome form to document their compliance with the Universal Protocol. Champions
demonstrated the checklist during educational staff meetings, and new staff were given a primer.
Staff commented favorably that they no longer had to remember everything.
The Association of PeriOperative Registered Nurses (AORN) Comprehensive Surgical
Checklist (www.aorn.org/uploadedImages/Images/Images/comprehensive_surgical_checklist_
RGB961.jpg) was a collaborative effort between AORN, the developers of the WHO checklist,
and the Joint Commission. The Web site states that the checklist, created in April 2010, was
created to support a facilitys need to use a single checklist that includes the safety checks
outlined in the World Health Organizations (WHO) Surgical Safety Checklist, while also
meeting the safety checks within The Joint Commissions Universal Protocol in order to meet
accreditation requirements. Our searches identified no empirical studies of this checklist.
Another combined checklist (called a crosswalk) combining the WHO checklist and the
Universal Protocol was published in November 2011 by the Pennsylvania Patient Safety
Reporting System.40 This document also addresses checking preparedness for surgical fires, as
well as two intraoperative checks specifically for spinal surgery involving precise locations. Due
to the recency, no studies exist yet on this crosswalk.
Anesthesia Equipment Checklists

The ASA guidelines identify 15 items: 7 to be performed only before the first procedure of
the day, and 8 to be performed prior to each procedure. Similarly, the AAGBI guideline
recommends that 11 items be checked prior to each operating session and that 3 of these items
are to be checked again prior to each new patient procedure. These guidelines are to be
implemented by individual hospitals and tailored to their departmental needs. As stated by the
ASA on their Web site (ASA, 2011), the updated recommendations are intended to serve as
general guidelines for individual departments and practitioners to design pre-anesthesia checkout
procedures specific for the delivery systems and the needs of the local practice. Further, they
state, Guidelines are systematically developed recommendations that assist the practitioner and
patient in making decisions about health care. These recommendations may be adopted,
131
modified, or rejected according to clinical needs and constraints and are not intended to replace
local institutional policies. In addition, practice guidelines are not intended as standards or
absolute requirements, and their use cannot guarantee any specific outcome. Practice guidelines
are subject to revision as warranted by the evolution of medical knowledge, technology, and
practice. They provide basic recommendations that are supported by a synthesis and analysis of
the current literature, expert opinion, open forum commentary, and clinical feasibility data.41
The ASA encourages institutions to submit their version of the Pre-Anesthesia Checkout
(PAC) for publication on the ASA Web site (www.asahq.org/For-Members/ClinicalInformation/2008-ASA-Recommendations-for-PreAnesthesia-Checkout/SampleProcedures.aspx). Currently, sample PACs are posted for the following anesthesia system
models: (1) General Electric AESTIVA, (2) Draeger Apollo, (3) Draeger Narkmoed GS,
(4) Draeger 6000, (5) Draeger B/C/GS, and (6) Draeger Fabius GS. Eight U.S. hospitals are
currently represented on the ASA collection of sample checkouts.
As an additional international example of implementing anesthesia checklist guidelines, the
Columbian 2009 version of their Minimum Safety Standards in Anaesthesia states that
anaesthesiologists and surgeons must collaborate in completing an overall check list, which is to
include at least the items in the WHO checklist. In addition, before applying anaesthetic, the
anaesthesiologist must complete a pre-anaesthetic checklist.42
In referencing earlier implementation strategies for aviation checklists, a 2000 article by
Blike and Biddle28 propose the three Ps for successful implementation of their anesthesia
machine electronic checklist. They refer to the three Ps as a guiding philosophy, with
procedures designed to achieve the goal of the philosophy using consistent policies for
implementation. They concluded that the earlier AACR was deficient in that the associated
published checklist had no supporting philosophy.
Regarding staffing, the 2008 ASA guidelines identify particular aspects of the PAC that
could be performed by a qualified anesthesia and/or biomedical technician. However, regardless
of the level of training and support by technicians, the anesthesia care provider is ultimately
responsible for proper function of all equipment used to provide anesthesia care.13

Costs of implementing a checklist mostly involve checklist development (or checklist
modification if the WHO checklist is used), formal staff notification that use of the checklist is
expected, staff training, and additional operating room time. In 2010, Semel and colleagues43
performed a hypothetical decision analysis of checklist introduction in a U.S. hospital. The cost
of implementing the checklist was estimated using the opportunity cost of the work that would
have otherwise been performed by the three department checklist champions and the
implementation coordinator, which was an estimated $12,635 in 2008 dollars; per-use cost was
only $11. The cost of a major surgical complication was estimated at $13,372. In the base-case
analysis, checklist introduction actually saved money. Regarding time, Sewell 201130 reported
that 20 percent of staff thought the WHO checklist caused an unnecessary time delay. However,
in 2011, Taylor and colleagues33 reported that the WHO checklist took only about two minutes
on average.
With regard to operating room time, a 26-item anesthesia checklist developed in 2010 by
Thomassen and colleagues44 was completed with a median time of 88.5 seconds (n=502
patients). Additionally, when cases were compared before and after implementation, checklist
completion did not cause any significant difference in pre-induction time (25.1 vs. 24.3 minutes).
132
An additional potential cost benefit relates to reduced litigation claims. With regard to
anesthesia, comparing the period prior to the1990s to the period from 1990 to 2003, the
proportion of claims with substandard care decreased (from 39% to 22%), and payments were
made less frequently (from 58% to 42% of the time).3
Are There Any Data About Adoption and Diffusion of This Patient
Safety Practice?
On February 22, 2012, the WHOs Surgical Safety Web Map indicated that as of February 1,
2012, 4,120 hospitals had expressed interest in using the checklist and 1,790 of these hospitals
have used the checklist in at least one operating theatre (Figure 1). On the map, red crosses
represent those expressing interest, and yellow crosses represent previous/current users.
Figure 1, Chapter 13. Screenshot of adoption and diffusion of the WHO surgical safety checklist
Note: This figure is a screenshot taken on 2/22/2012 of the WHO Surgical Safety Web Map
(http://maps.cga.harvard.edu:8080/Hospital/). Red crosses represent hospitals who have expressed interest in using the WHO
Surgical Safety Checklist (as of 2/1/2012); yellow crosses represent hospitals that have used the checklist in at least one operating
theatre. Using the right-hand panel, the map can also be configured to display locations of endorsing organizations, international
endorsing organizations, pilot sites, and countries with nationwide implementation. Granted permission by the World Health
Organization.
Our searches found that a number of professional organizations have recommended adoption
of the WHO checklist (Table 1).
133
Table 1, Chapter 13. Governmental and nongovernmental organizations adopting or

recommending adoption of the WHO checklist
Organization
The Institute for
Healthcare
Improvement
(www.ihi.org)
Web site
www.ihi.org/offerings/
MembershipsNetworks/
MentorHospitalRegistry/Pages/
SurgicalSafetyChecklist.aspx
National Patient
Safety Agency
(NPSA) in the UK
www.nrls.npsa.nhs.uk/alerts/
?entryid45=59860
France
NA
Canadian Patient
Safety Institute
(CPSI)
NA
Washington State
Surgical Care and
Outcome
Assessment Program
49
(SCOAP)
www.scoap.org/
checklist/index.html
Details
In December 2008, then-president Donald Berwick
issues the Surgical Safety Checklist Challenge: to have
each hospital use the checklist in at least one
operating room by April 1, 2009. To assist facilities in
implementing the checklist, the IHI Web site provides a
list of eight mentors throughout the United States who
have already implemented the checklist. The
demographics of these eight sites are provided to
enable facilities to match themselves up with similar
mentor facilities.
NPSA mandated in February 2010 the use of the
45
checklist in all of its Trusts in England and Wales.
The NPSA Web site contains downloadable materials,
videos, and three tailored WHO surgical checklists (for
radiological interventions, cataract surgery, and
maternity cases). Also, the Surgical Checklist
Implementation Project, funded by the National Health
Service (NHS), involves four studies of implementing
the WHO checklist at 20 NHS Trusts
46
(www.safesurgery.org.uk/). The topics of the four
studies are (1) perception of the checklist and possible
barriers to use; (2) additional quantitative data on staff
perceptions of the checklist; (3) how the checklist is
actually used in operating rooms; (4) the impact of
46
checklist use on clinical outcomes.
France mandated the use of the WHO checklist in all
47
its 8000 hospitals in 2010.
The CPSI has endorsed the checklist, and checklist
implementation is now an accreditation standard for
48
Canadian hospitals.
The Canadian province of Ontario mandated use of the
47
checklist in 2011.
SCOAP stated a goal of having all of its hospitals use
the WHO checklist in every operating room by the end
of 2009 . The February 2010 SCOAP version of the
WHO checklist is available on the Web site. The Web
site also states that According to the Washington
State Hospital Association, 100% of Washington State
hospitals have either implemented a standardized
surgical checklist or are in the process of doing so.
Hospitals can also order a 2x3 foot laminated SCOAP
checklist from the Web site.
The SCHA, in the fall of 2010, planned to institute the
checklist in all the states hospitals over the next few
47
years.
Ireland and Jordan each plan to require checklist
50
implementation in all its hospitals.
The Spanish Ministry of Health and the Spanish
51
Association of Surgeons have joined the initiative.
The South Carolina

Hospital Association
(SCHA)
Ireland and Jordan
Spanish Ministry of
Health and Spanish
Association of
Surgeons
NA
134
Table 1, Chapter 13. Governmental and nongovernmental organizations adopting or

recommending adoption of the WHO checklist (continued)
Organization
Australia and New
Zealand: the Royal
Australasian College
of Surgeons, the
Australian and New
Zealand College of
Anaesthetists, the
Royal Australian and
New Zealand College
of Obstetricians and
Gynecologists, the
Australian College of
Operating Room
Nurses, and the
Australian
Commission for
Safety and Quality in
Health Care
Web site
NA
Details
These organizations developed a modified version of
52
the WHO checklist. The checklist was launched in
August of 2009 with the endorsement of national
government health departments in both countries.
The webcast event Check a Box, Save a Life was launched on October 22, 2009 to
promote the use of the WHO checklist.53 The event, run mostly by medical students, involved
182 hosting sites from 121 medical institutions and an estimated 1,400 online viewers. A
Facebook page had enrolled 111 medical students who agreed to host the event at their
institutions. At the Institute for Healthcare Improvement forum 6 weeks later, 15 case reports
were presented that detailed checklist-related projects.
In January 2010 in the UK, just before the mandatory requirement to use the WHO checklist
was instituted by the NPSA, Sivathasan and colleagues54 conducted telephone interviews with
238 hospitals in the UK (randomly selected from some 540 hospitals, therefore representing
about 44% of UK hospitals). Almost all (99%) of the hospitals had heard of the checklist, and its
use was already compulsory in 65 percent of them. In hospitals where it was not required, 81
percent used it voluntarily, and 75 percent had a plan to make it mandatory in the future.
However, some operating rooms reported partial use of the checklist, i.e., intentionally skipping
items or skipping the entire checklist because of time constraints.
In June 2009, the journal OR Manager received online data from 136 subscribers regarding
use of the WHO checklist.55 About half (48.5%) said they had implemented the checklist, and
64 percent said the checklist has improved safety in the operating room. However, 11 percent of
respondents stated that the checklist was not well accepted by surgeons, and another 63 percent
said surgeons did accept it but with reservations. Nurses were believed to have a somewhat
greater degree of acceptance, with only 2 percent not well accepted and 52 percent accepted
with reservations.
A survey in October/November 2009 of 12 oral and maxillofacial consultants in Yorkshire
England found that all were aware of the WHO checklist, but only 5/12 were actually using it.56
Ten of 12 expressed the belief that it would improve patient safety, but four of 12 said it would
not improve team communication.
Regarding the Universal Protocol, accredited hospitals are required to comply. Therefore the
diffusion of the Universal Protocol is large, by mandate. However, as stated earlier, the
Universal Protocol is not a checklist. We found no published information on how many hospitals
actually use a checklist in their efforts to comply.
135
Regarding anesthesia checklists, in 2009, Winters and colleagues57 at Johns Hopkins

University republished the AAGBI checklist and discussed resistance of physicians to adopting
anesthesia checklists in general. They cited the cases of some physicians who claimed to be
insulted, whereas others expressed doubt that a checklist will prevent a medical mistake. They
counter this argument by mentioning the complexity of modern medicine, which may
inadvertently introduce devastating risks. In 2000, Thomassen and colleagues44 of Haukeland
University Hospital in Norway developed an anesthesia checklist designed to identify preinduction deficiencies (i.e., missing equipment or inadequate preparation). The checklist was
improved over the course of 502 inductions. They reported that in 17 percent of the cases,
missing items were identified, the most critical being lack of availability of a second
laryngoscope, the introducer not having been fitted to the endotracheal tube, the endotracheal
tube cuff not having been tested, and no separate ventilation bag available. Thomassens 2010
study32 reported user experiences: Some of the senior physicians were skeptical of the usefulness
of the checklist. They concluded that the success of implementation of the checklist depends on
physician leaders having a positive attitude. The checklist itself improved confidence in
unfamiliar contexts (see Table 4, Chapter 13 in Appendix D).

Several prominent authorities in the field of patient safety have proposed checklists in an
attempt to prevent mistakes related to surgery. These checklists have been developed carefully
by experts in the field, and have evolved over time to capture only the most essential
considerations. Numerous implementation issues remain, including how to modify a given
checklist to a specific hospital setting, or to a specific anesthesia system, or to a specific surgical
staff. A recurrent theme in the literature on preoperative checklist is the explicit encouragement
of a team-based approach. Further adoption and diffusion of these checklists will depend on the
continued demonstration of effectiveness in preventing errors, checklist modifications to
improve clarity and prevent misuse, proof that the benefits are worth the added time and cost,
and flexibility to changes as needs arise. A summary table is located below (Table 2).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
High
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Low
How Much do We
A lot/Moderate
References
1.
2.
Gawande AA, Thomas EJ, Zinner MJ, et al.

The incidence and nature of surgical adverse
events in Colorado and Utah in 1992.
Surgery 1999 Jul;126(1):66-75. PMID:
10418594.
Kable AK, Gibberd RW, Spigelman AD.
Adverse events in surgical patients in
Australia. Int J Qual Health Care 2002
Aug;14(4):269-76. PMID: 12201185.
136
3.
Staender SE, Mahajan RP. Anesthesia and

patient safety: have we reached our limits.
Curr Opin Anaesthesiol 2011 Jun;24(3):34953. PMID: 21330916.
4.
Helsinki declaration on patient safety in

anesthesiology. Actual Anestesiol Reanim
2010 Jul;20(3):136-9.
5.
ECRI Institute. Guidance article: top 10

technology hazards for 2012. Health
Devices 2011 Nov;40(11):358-73.
6.
Haynes AB, Weiser TG, Berry WR, et al. A

surgical safety checklist to reduce morbidity
and mortality in a global population. N Engl
J Med 2009;360(5):491-9.
content.nejm.org/cgi/content/full/NEJMsa08
10119. PMID: 19144931
7.
de Vries EN, Prins HA, Crolla RM, et al.

Effect of a comprehensive surgical safety
system on patient outcomes. N Engl J Med
2010 Nov 11;363(20):1928-37. PMID:
21067384
8.
9.
10.
11.
de Vries EN, Hollmann MW, Smorenburg

SM, et al. Development and validation of
the SURgical PAtient Safety System
(SURPASS) checklist. Qual Saf Health Care
2009 Apr;18(2):121-6. PMID: 19342526.
de Vries EN, Dijkstra L, Smorenburg SM, et
al. The SURgical PAtient Safety System
(SURPASS) checklist optimizes timing of
antibiotic prophylaxis. Patient Saf Surg 2010
Apr 13;4(1):6. PMID: 20388204.
de Vries EN, Eikens-Jansen MP, Hamersma
AM, et al. Prevention of surgical
malpractice claims by use of a surgical
safety checklist. Ann Surg 2011
Mar;253(3):624-8. PMID: 2120959.
Facts about the universal protocol.
[internet]. Oakbrook Terrace (IL): The Joint
Commission; 2004 [accessed 2008 Jan 9]. [1
p].
www.jointcommission.org/assets/1/18/Univ
ersal%20Protocol%201%204%20111.PDF.
12.
Using surgical checklists to comply with the

universal protocol. Jt Comm Perspect
Patient Saf 2011 Jul;11(7):9-11.
13.
ASA Committee on Equipment and

Facilities. 2008 recommendations for preanesthesia checkout procedures. Park Ridge
(IL):American Society of Anesthesiologists;
2008.
14.

al. An intervention to decrease catheterrelated bloodstream infections in the ICU. N
Engl J Med 2006 Dec 28;355(26):2725-32.
PMID: 17192537.
15.

program. Milbank Q 2011 Jun;89(2):167205. PMID: 21676020.
137
16.
AHRQ Effective Healthcare Program.

Closing the quality gap: prevention of health
care-associated infections. [internet].
Research and Quality (AHRQ); 2011 Aug 3
[accessed 2011 Nov 4]. [20].
www.effectivehealthcare.ahrq.gov/index.cf
m/search-for-guides-reviews-andreports/?productid=738&pageaction=display
product#toc.
17.
Surgical safety checklist. Geneva: World

Health Organization (WHO); 2009 Jan. 1 p.
whqlibdoc.who.int/publications/2009/97892
41598590_eng_Checklist.pdf.
18.
World Alliance for Patient Safety. WHO

guidelines for safe surgery. Geneva: World
Health Organization (WHO); 2008.
19.
Greenberg CC, Regenbogen SE, Studdert

DM, et al. Patterns of communication
breakdowns resulting in injury to surgical
patients. J Am Coll Surg 2007
Apr;204(4):533-40. PMID: 17382211.
20.
Griffen FD, Stephens LS, Alexander JB, et

al. The American College of Surgeons
closed claims study: new insights for
improving care. J Am Coll Surg 2007
Apr;204(4):561-9. PMID: 17382214.
21.
Mulloy DF, Hughes RG. Chapter 36.

Wrong-site surgery: a preventable medical
error. In: Patient safety and quality: an
evidence-based handbook for nurses [AHRQ
Publication No. 08-0043]. Rockville (MD):
(AHRQ); 2008 Apr. p. 1-15.
www.ahrq.gov/qual/nurseshdbk/docs/Mullo
yD_WSS.pdf.
22.
Merry AF. Role of anesthesiologists in

WHO safe surgery programs. Int
Anesthesiol Clin 2010 Spring;48(2):137-50.
PMID: 20386233.
23.
Association of Anaesthetists. Checklist for

anaesthetic equipment 2004. London (UK):
Association of Anaesthetists of Great Britain
& Ireland; 2004. 2 p.
www.aagbi.org/sites/default/files/checklista
404.pdf.
24.
Haynes AB, Weiser TG, Berry WR, et al.

Changes in safety attitude and relationship
to decreased postoperative morbidity and
mortality following implementation of a
checklist-based surgical safety intervention.
BMJ Qual Saf 2011 Jan;20(1):102-7. PMID:
21228082.
25.
26.
34.
Weiser TG, Haynes AB, Dziekan G, et al.

Effect of a 19-item surgical safety checklist
during urgent operations in a global patient
population. Ann Surg 2010
May;251(5):976-80. PMID: 20395848.
Singer SJ. Grants awarded in FY 2010.

Factors associated with effective
implementation of a surgical safety
checklist. In: Researching implementation
and change while improving quality: grant
awards. Fact sheet. Rockville (MD): Agency
for Healthcare Research and Quality
(AHRQ); 2011 Oct. p. 3.
www.ahrq.gov/qual/qualimp.pdf.
35.
Quarterly update on preventing wrong-site

surgery [draft]. Plymouth Meeting (PA):
ECRI Institute; 2012. 13 p.
Safe surgery saves lives. [internet]. Geneva:

World Health Organization; [accessed 2011
Sep 22].
www.who.int/patientsafety/safesurgery/en/.
36.
Follow-up Survey on Universal Protocol.

Commission; 2010 Dec 14 [accessed 2011
Oct 18]. [2 p].
www.jointcommission.org/followup_survey_universal_protocol/.
37.
Garnerin P, Ares M, Huchet A, et al.

Verifying patient identity and site of
surgery: improving compliance with
protocol by audit and feedback. Qual Saf
Health Care 2008 Dec;17(6):454-8. PMID:
19064662.
38.
Nilsson L, Lindberget O, Gupta A, et al.

Implementing a pre-operative checklist to
increase patient safety: a 1-year follow-up of
personnel attitudes. Acta Anaesthesiol
Scand 2010 Feb;54(2):176-82. PMID:
19719811.
39.
Owers C, Lees E, Roberts J. Correct site

surgery - Are we up to standard? Cent Eur J
Med 2010 Aug;5(4):407-10.
40.
Possible integration of actions to satisfy

Joint Commission universal protocol and
WHO surgical safety checklist. [internet].
Harrisburg (PA): Pennsylvania Patient
Safety Authority; 2011 [accessed 2011 Nov
21]. [1 p].
www.patientsafetyauthority.org/Educational
Tools/PatientSafetyTools/PWSS/Documents
/crosswalk.pdf.
41.
2008 ASA recommendations for

preanesthesia checkout: sample procedures.
[internet]. Park Ridge (IL): American
Society of Anesthesiologists (ASA);
www.asahq.org/For-Members/ClinicalInformation/2008-ASA-Recommendationsfor-PreAnesthesia-Checkout/SampleProcedures.aspx.
27.
Devine J, Chutkan N, Norvell DC, et al.

Avoiding wrong site surgery: a systematic
review. Spine (Phila Pa 1976) 2010 Apr
20;35(9 Suppl):S28-36. PMID: 20407349.
28.
Blike G, Biddle C. Preanesthesia detection

of equipment faults by anesthesia providers
at an academic hospital: comparison of
standard practice and a new electronic
checklist. AANA J 2000 Dec;68(6):497-505.
PMID: 11272956.
29.
Ben-Menachem E, Ezri T, Ziv A, et al.

Identifying and managing technical faults in
the anesthesia machine: lessons learned
from the Israeli Board of Anesthesiologists.
Anesth Analg 2011 Apr;112(4):864-6.
PMID: 21288969.
30.
Sewell M, Adebibe M, Jayakumar P, et al.

Use of the WHO surgical safety checklist in
trauma and orthopaedic patients. Int Orthop
2011 Jun;35(6):897-901. PMID: 20730425.
31.
Kearns RJ, Uppal V, Bonner J, et al. The

introduction of a surgical safety checklist in
a tertiary referral obstetric centre. BMJ Qual
Saf 2011 Sep;20(9):818-22. PMID:
21693466.
32.
Thomassen O, Brattebo G, Heltne JK, et al.

Checklists in the operating room: help or
hurdle? A qualitative study on health
workers experiences. BMC Health Serv Res
2010;10:342. PMID: 21171967.
33.
Taylor B, Slater A, Reznick R. The surgical

safety checklist effects are sustained, and
team culture is strengthened. Surgeon 2010
Feb;8(1):1-4. PMID: 20222395.
138
42.
Ibarra P, Robledo B, Galindo M, et al.

Minimum standards 2009 for practising
anaesthesiology in Colombia: Safety
Committee. Rev Colomb Anestesiol 2009
Jul;37(3):235-53.
43.
Semel ME, Resch S, Haynes AB, et al.

Adopting a surgical safety checklist could
save money and improve the quality of care
in U.S. hospitals. Health Aff (Millwood)
2010 Sep;29(9):1593-9. PMID: 20820013.
44.
Thomassen O, Brattebo G, Softeland E, et

al. The effect of a simple checklist on
frequent pre-induction deficiencies. Acta
Anaesthesiol Scand 2010 Nov;54(10):117984. PMID: 21069898.
45.
Mahajan RP. The WHO surgical checklist.

Best Pract Res Clin Anaesthesiol 2011
Jun;25(2):161-8. PMID: 21550541.
46.
The objectives and study aim of the Surgical

Checklist Implementation Project. [internet].
London (UK): Imperial College; [accessed
2011 Oct 6]. [1 p].
www.safesurgery.org.uk/.
47.
Robbins J. Hospital checklists.

Transforming evidence-based care and
patient safety protocols into routine practice.
Crit Care Nurs Q 2011 Apr-Jun;34(2):142-9.
PMID: 21407009.
48.
Warnock GL. Of surgeons and safety

checklists. Can J Surg 2010 Dec;53(6):3646. PMID: 21092427.
49.
National push for surgical safety checklist

under way. Hosp Peer Rev 2009
May;34(5):53-7. PMID: 19453097.
50.
WHOs patient-safety checklist for surgery.

Lancet 2008 Jul 5;372(9632):1. PMID:
18603137.
51.
Grande L. [Improving safety in the

operating room reduces hospital mortality].
Cir Esp 2009 Dec;86(6):329-30. PMID:
19783243.
52.
Gough I. A surgical safety checklist for

Australia and New Zealand. ANZ J Surg
2010 Jan;80(1-2):1-2. PMID: 20575868.
53.
Henderson D, Carson-Stevens A, Bohnen J,

et al. Check a box. save a life: How student
leadership is shaking up health care and
driving a revolution in patient safety. J
Patient Saf 2010 Mar;6(1):43-7.
139
54.
Sivathasan N, Rakowski KR, Robertson BF,

et al. The World Health Organizations
Surgical Safety Checklist: should
evidence-based initiatives be enforced in
hospital policy? JRSM Short Rep
2010;1(5):40. PMID: 21103132.
55.
Patterson P. In survey, about half of ORs are

using the WHO checklist. OR Manager
2009 Sep;25(9):1, 6-8. PMID: 19824166.
56.
Abdel-Galil K. The WHO surgical safety

checklist: are we measuring up. Br J Oral
Maxillofac Surg 2010 Jul;48(5):397-8.
PMID: 20153096.
57.
Winters BD, Gurses AP, Lehmann H, et al.

Clinical review: checklists - translating
evidence into practice. Crit Care
2009;13(6):210. PMID: 20064195.
Chapter 14. Use of Report Cards and Outcome

Measurements To Improve Safety of Surgical Care: American
College of Surgeons National Quality Improvement Program
(NEW)
Melinda Maggard-Gibbons, M.D., M.S.H.S.
How Important Is the problem?

Over 40 million operative procedures are performed in the United States (U.S.) each year.1
Postoperative adverse events occur all too commonly. Mortality for complex operations in the
Medicare population ranges from 7.5 percent to as high as 17.7 percent for gastrectomy and 3.1
percent to 13.3 percent for pancreatectomy.2 Overall morbidity for three complex procedures
combined ranged from 44.3 percent to 38.9 percent. Even in less complex cases such as
colectomy (250,000 cases are performed each year), surgical site infections occur approximately
10 percent of the time.3 These adverse events increase hospitalization length and cost. A surgical
site infection is estimated to add $27,631 to the cost of a surgical stay.4 Even a simple wound
opening is estimated to cost $1,426. A urinary tract infection can add $675 or up to $2,800 if
accompanied by bacteremia.5 Following respiratory complications, length of stay increases
attributable to post-operative complications range from 3 to 11 days.6 A single case of ventilatorassociated pneumonia adds $50,000 to the baseline cost of a surgical admission.7,8 Patients who
develop postoperative pulmonary embolism/deep venous thrombosis require readmission in 44
percent of cases, with annual cost ranging from $7,594 to $16,644.6

The largest and best known intervention for measuring and reporting surgical outcomes in
the U.S. is the American College of Surgeons National Surgical Quality Improvement Project
(ACS NSQIP), now implemented at 431 sites. This multicomponent intervention grew out of
efforts initiated by Veterans Affairs (VA) Health System researchers and clinicians in the late
1980s. The original idea for VA NSQIP was to feedback data to facilities and surgeons on their
performance as a stimulus for quality improvement. To implement that original idea a number of
elements needed to be developed: methods to collect data consistently across sites, methods for
data sharing, and models for calculations observed-to-expected outcomes. Later, in VA NSQIP,
the need to bring sites together for learning and sharing across sites was recognized as necessary
to catalyze improvement and this was added to the intervention. Thus, the intervention has
changed over time in the components used to implement the original idea of feedback for
performance data. An example of observed to expected ratio reporting for ACS NSQIP is located
in Figure 1.
The current ACS NSQIP collects prospective, clinical data that are used to provide riskadjusted assessments of outcomes that are fed back to the hospitals and surgeons for comparative
purposes, with the ultimate goal of quality improvement. A benchmarked, peer-controlled
database allows hospitals to compare 30-day outcomes across hospital types. With support from
ACS NSQIP, individual sites work to design quality initiatives to achieve better outcomes and
care in the areas of need.
140
The intervention comprises five basic components. First, a trained surgical clinical reviewer
(SCR) prospectively collects data on preoperative and clinical variables and on 30-day outcomes
(outcomes are described in more detail, below). The number of variables depends on the
particular Program option chosen (which in turn depends on the needs and size of the particular
hospital) but is typically 46 or 69 for each case. A predetermined number of cases is reported,
which again depends on the Program option (also referred to as Use Option) the hospital has
elected to use. The second component is development and maintenance of models of expected
mortality and morbidity by risk and types of procedures performed. The third component is the
calculation of the observed-to-expected (O/E) 30-day mortality and morbidity ratios. Data are
then fed back to individual sites as observed-to-expected ratios of, typically, 21 morbidities, such
as wound, respiratory, central nervous system, urinary and cardiac complications, as well as
mortality. Data are provided alongside blinded national results from the other participating sites.
Sites are designated as being a high (worse than expected) or low outlier (better than expected)
for each category of morbidity and for mortality.
Figure 1, Chapter 14. Example of observed to expected ratio reporting for American College of
Surgeons national surgical quality improvement program
Figure 1 accessed from the ACS NSQIP Web site, December 2011. Reprinted by permission of American College of Surgeons
NSQIP.
Lastly, institutions then identify areas where they are a high outlier and improvement is
needed. Auditing by the ACS NSQIP staff occurs randomly and for cause, that is, site reports
many high risk patients but a low complication rate. Individual surgeon-level data are provided
to the participating hospital if they request it. While the responsibility for making changes and
addressing areas in need of improvement remains with the individual sites, the administrative
ACS NSQIP body provides support in terms of case reports, best practices, national meetings,
and monthly supportive conferences calls with the surgeon champions and surgical clinical
reviewers.

The concept that measurement and reporting of hospital outcomes can be useful for
improving quality and safety goes back more than 100 years. E.A. Codman in 1913 told the
Philadelphia Medical Society We must formulate some method of hospital report showing as
nearly as possible what are the results of the treatment obtained at different institutions. This
report must be made out and published by each hospital in a uniform manner, so that comparison
will be possible. With such a report as a starting point, those interested can begin to ask
141
questions as to the management and efficiency.9 In 1914, Codman started his own hospital in
Boston, the End-Result Hospital, to study the quality of surgical care.
Two precedents helped inform the rationale for ACS NSQIP. In 1994, in response to
concerns about high complication rates in the VA, a system to collect and report clinical
variables and outcomes across all VA sites was established (ultimately coined VA NSQIP).10,11
To compare results to non-VA hospitals or among VA sites fairly, these investigators needed to
correct for how sick patients were. Lisa Iezzoni coined the phrase, the algebra of effectiveness,
which means that outcomes are a function of patient factors, effectiveness of the care provided,
and random variation.12 Patient factors include the severity of target illness plus their
comorbidities. Thus, the VA developed a database of preoperative risk factors (severity of illness
and comorbidity) along with the database used to collect postoperative outcomes. In an attempt
to level the playing field for cross-institutional comparisons, the investigators developed riskadjustment models. The perceived success of the VA program led to its adaption for non-VA
hospitals, now known as the ACS NSQIP.
The second precedent for ACS NSQIP was the apparent success of programs like the New
York State Cardiac Surgery Reporting System (CSRS).13 In this program, which began about 20
years ago, the State Department of Health collects and distributes data from all New York State
hospitals performing coronary artery bypass grafting, with the aim of promoting accountability
and quality improvement. While originally envisioned as a confidential feedback system, public
and court pressure led to the public release of results, and the CSRS became a public reporting
system. Evaluation of the impact of the CSRS has been complicated by secular trends and the
lack of a widely-accepted comparison group, but systematic reviews of public reporting have in
general concluded that the preponderance of the evidence supports a causal relationship between
the CSRS and greater declines in cardiac surgery morbidity and mortality than would be
expected from the long-established secular trends alone.14,15 The basic concepts of the New York
State CSRS have been adopted by other states (California, Pennsylvania) and have expanded
across the U.S. through the efforts of the Society of Thoracic Surgery (STS) Registry, which
incorporated public reporting. Measurement and reporting of cardiac surgery outcomes has also
spread to England.16
The rationale for how NSQIP improves care is multifaceted. To improve care and reduce
complications, surgeons must first know the outcomes of their own procedures. The data used to
provide this feedback must be high quality and reliable, and the method of risk-adjustment must
be adequate to allay concerns about comparing apples to oranges. Lastly, the program must
establish an impetus for change: In this case, it is the comparison of care at ones own site to that
of other representative sites, that is sites with similar elements and the same risk-adjustment.
This comparison allows surgeons and hospitals to see how they compare in terms of poor
outcomes, which in turn promotes accountability and stimulates work to correct the problems.
Most sites, 59 percent of those surveyed, were unaware of their hospitals adverse event rates, let
alone how they compare to other hospitals, until after they enrolled in ACS NSQIP.4
One of the great strengths of the ACS NSQIP program is that it relies on prospective clinical
data. Administrative and claims data are limited, as they lack sufficient clinical data elements
and vary considerably in terms of quality (coder variations, subjective reporting, focus on
payment rather than on outcome reporting). A study comparing administrative and claims data
collected by the University Health System Consortium (UHC) program showed that the ACS
NSQIP identified 61 percent more total complications than were identified by UHC, including 97
percent more surgical site infections and 100 percent more urinary tract infections.17
142
Furthermore, NSQIP focuses on 30-day outcomes and is not limited to adverse events
associated with the index admission. Studies show that more than 50 percent of complications
happen after discharge. For colectomies, 45 percent of deep surgical site infections, 39 percent of
organ space infections, and 28 percent of deep venous thrombosis (DVT) occur after patients
have left the hospital.18 Identifying complications that occur outside the hospital is the
prerequisite first step to developing changes in care to help prevent those complication, which in
turn should result in reduced morbidity and mortality and save costs.19

Evaluating the effect of the measurement and reporting of outcomes is complicated by the
fact that these measurement systems have almost always resulted from policy decisions that
affect large geographic areas; thus, with no access to natural control institutions, investigators are
relegated to using time-series data within interventions sites, observational comparisons to nonintervention sites, and focused process-and-outcome evaluations seeking to explain observed
changes in outcomes over time. As alluded to earlier, the first such evaluations emanated from a
congressional mandate in 1986 for the VA to perform a National VA Surgical Risk Study
(NVASRS), with the aim of developing surgical risk-adjustment models to predict outcomes and
compare the quality of surgical care among facilities. Between 1991 and 1993, 44 VA medical
centers used clinical nurse reviewers to collect preoperative and intraoperative clinical data and
30-day outcomes on major surgical procedures. Variations in the 30-day morbidity and mortality
outcomes were identified across VA facilities.20-22 The success of this initial study led to VA
NSQIP, which was officially launched in 1994 and has provided continuous monitoring of the
outcomes of surgical care in the VA. A review of over 400,000 cases performed between 1991
and 1997 showed that 30-day mortality and morbidity rates for major surgery fell 9 percent and
30 percent, respectively.23 Reductions in one post-operative complication alone, surgical
pneumonia, are estimated to have saved the VA $9.3 million annually, and the overall reduction
in postoperative morbidity may have saved $46 billion over the lifetime of the program.
In the late 1990s, non-VA hospitals became interested in applying the VA experience to their
data reporting and quality improvement programs. A pilot study in three civilian hospitals
(University of Michigan, Emory University, University of Kentucky) showed the feasibility in
the private sector.24 Following this pilot, in 2001, the American College of Surgeons took the
lead to expand efforts to a broader group of hospitals (14 sites), and in 2004, the formal ACS
NSQIP began.25
The potential impact of participating in ACS NSQIP on complication rates and mortality has
been reported by individual hospitals and collaboratives and posted on the ACS NSQIP site.
Although reported improvements in morbidities have been large, improvements in mortality have
ranged from none to modest. However, many general and vascular surgery procedures tend to
have low 30-day mortality rates to start. Most of the reports of improvement in single institutions
or collaborative have not been published but have been presented in other venues. At the most
recent ACS NSQIP national meeting in July 2011, 20 presentations reported reductions in
morbidity following an intervention. In all these cases, ACS NSQIP data enabled the hospital or
hospitals to target an area with worse-than-expected outcomes and to intervene, with resulting
improvement. Eleven additional examples of programs that have recently begun participating in
NSQIP, have identified areas of need, and are in the process of implementing change, although
results are not yet available (see Table 1 for examples). An advantage of the annual national
143
conference is the opportunity it provides to network and collaborate on quality improvement

planning and projects.
Table 1, Chapter 14. Example of interventions and associated impact on outcomes in American
College of Surgeons national surgical quality improvement program for hospitals/collaboratives
Hospital
Hershey Medical Center,
26
Penn State
27
University of Virginia
Massachusetts General
27
Hospital
Complication
19.3% SSI in diabetics; 8%
in non-diabetics
VTE 3.4% (2008)
Intervention
Glucose control protocol
VTE risk assessment and
order set
17.6% SSI (national

average 8.1%) colorectal
resections. High BMI was
a risk factor.
Vascular surgery morbidity
O/E ratio=1.19, [99% CI
0.93 to 1.48]
Protocol for wound wicking

2
for BMI >25 kg/m , SCIP
measures, glycemic
control
physician order entry
templates, foley catheter
removal algorithm, silvercoated catheters for select
patients, identify
procedures not requiring a
catheter, educational
campaign for clinicians
Standardized orders,
proper antibiotic use,
morbidity conference
presentations, skin
preparation changes
UTI rate 7.0% vs 4.7%

(P<0.087)
Hospital A
Identified a rise in organ

space infections
Hospital B
VTE 17.6%
Risk stratification, best

practices, standardized
orders
Hospital C
Unplanned reintubation 3%
(O/E=1.56)
Ventilator >48 hrs 3.84%
(O/E=1.71)
Tracking tool, risk

assessment, improved
pulmonary toilet
intervention
Hospital D
Ventilator >48 hrs 2.24%

(O/E=1.7)
Tracking tool, standardized

orders, patient teaching
Impact
Reduction of SSI O/E 1.31
to 0.78
Reduction of VTE rate
3.4% to 0.2% (2008-2009)
Reduction of SSI from
17.6% to 11.2% (36%
reduction)
(2003-2006)
Reduction of UTI 7.0% to
1.8%.
Morbidity O/E ratio went
from 1.19 [99% CI 0.93 to
1.48] to 0.93[(99% CI 0.67
to 1.48]
(76% reduction)
(2003-2004)
Organ space infection
increase attributed to
increased leak rates and
identified surgical
technique issues and saw
improvements, but rate still
high.
(2005-2010)
VTE decreased from 17.6
to 2.3%; O/E 1.88 to 1.05
(2006-2010)
TBD
Ventilator >48 hrs 2.24% to

1.19%
(O/E=1.7 to 0.83)
(2008-2010)
Hospital E
Overall orthopedic DVT
Identified variations in DVT Reduction of overall
3.1%
prophylaxis practice,
orthopedic DVT rate 3.1%
Knee Arthroplasty DVT
surgeon specific review,
to 1.1%
rate 10.1%
standardized care
Reduction of knee
arthroplasty DVT rate
10.1% to 1.6%
(2008-2010)
Notes: DVT, deep venous thrombosis; VTE, venous thromboembolism; Hospital A-E are representative examples taken from
ACS NSQIP Data Portal Web site, accessed December, 13, 2011. Reprinted by permission from the American College of
Surgeons NSQIP.
Almost all these studies have a pre-post design, with no control groups, and therefore have
all the limitations common to studies of that design, including regression to the mean. Yet, in
aggregate, these reports consistently show that hospitals identified as high outliers in some
particular outcome that respond by implementing a targeted intervention experience a decrease in
that outcome. The magnitude of some of these decreases cannot be explained by regression to the
144
mean or confounding; for example the decrease in DVT rate from 10 percent to 1 percent in one
study or from 3.2% to 0.2% in another study.
Two published longitudinal studies have reached divergent conclusions on the effects of
reporting. The first study looked at changes over three years (2005 to 2007) in ACS NSQIPparticipating sites (N=183) for all outcomes measured and surgical specialties using riskadjustment and accounting for hospital procedure volume (Hall B, 2009).19 For the most recent
time period, 2006 to 2007, 118 hospitals were enrolled and were participating long enough to
produce clinically useful data. The authors found that 82 percent (97 of 118) of hospitals had
improved morbidity and 66 percent (78 of 118) had improved mortality. The adjusted absolute
difference in observed to expected (O/E) ratio was -0.114 for morbidity and -0.174 for mortality
(negative numbers meaning less morbidity and mortality). Similar results were seen when the
researchers accounted for institutional volume. They also found that the number of high outliers
(those with worse outcomes) decreased over time and the number of low outliers (those with
better outcomes) increased. Additionally, high outliers were more likely to improve and had
larger mean changes in outcomes. For large hospitals, it was estimated that an average of 200 to
500 complications and 12 to 36 deaths may have been avoided.19 This study was conducted by
investigators affiliated with ACS NSQIP.
The other study compared ACS NSQIP to another private sector collaborative, which was
based at the University of Michigan Medical Center.28 The data file provided by ACS NSQIP
included Michigan and non-Michigan hospitals. The Michigan Surgical Quality Collaborative
(MSQC) includes 34, largely community (68 percent) hospitals, unlike the ACS NSQIPparticipating hospitals, which are primarily academic/teaching. Sixteen MSQC hospitals were
included in the analysis, which assessed two time periods: April 2005 to March 2007 compared
with April 2007 to December 2007. Results were also compared with the 126 non-Michigan
NSQIP hospitals over the same time period. All hospitals used a similar data reporting system.
This analysis found that the MSQC hospitals had a decrease in morbidity from 10.7 percent to
9.7 percent (9% reduction, P=.002; odds ratio=.898) over the 3 years, whereas morbidity did not
change for the ACS NSQIP hospitals in either time period (12.4%; odds ratio=1.0). Mortality
rates did not change for either group of hospitals.28 This study was conducted by investigators
affiliated with MSQC. One possible explanation for the difference in results between the two
studies is that the length of time the hospitals were enrolled in the program may have been too
short to see improvements, especially in the larger hospitals, which predominated among the
non-Michigan ACS NSQIP hospitals.

Few published studies have assessed the potential and actual harms of this program. Most of
the concerns are speculative. A primary concern has always been that surgeons will avoid highrisk cases for fear of adversely affecting their observed-to-expected outcomes assessments. This
issue was raised early in the process of implementing report cards when anecdotal evidence
appeared to suggest that as the result of implementing the New York CSRS, high-risk CABG
patients were being diverted instead to the Cleveland Clinic.29 However, subsequent and more
comprehensive analyses could not document any systematic exclusion of high-risk patients from
CABG operations, and that, on the contrary, the severity of illness and comorbidities of operated
patients has increased over the years.30,31 The longitudinal ACS NSQIP study also supported this
finding, showing that the risk profile and illness severity for surgical patients has increased over
time.19 Another concern is that the outcomes for outpatient cases or for a hospital or surgeon who
145
performs a small volume of procedures might need longer follow-up, possibly more than a year,
to accurately assess quality.32 Concerns have also been raised that surgeons could alter treatment
plans or surgical options for patients based on their operative risk rather than give the patient the
option of a procedure with a potentially better long-term functional outcome. An example would
be in vascular surgery, where a high-risk patient eligible for a distal bypass would be
recommended an amputation instead.

What Contexts?
Detailed steps for hospital enrollment are provided on-line (acsnsqip.org) or can be requested
directly from ACS NSQIP. The steps are also summarized here.
The first step is to review the program requirements (information can be requested on-line)
and contact the ACS to ask questions or schedule a teleconference. The program information
needs to be presented to a surgeon champion, quality improvement personnel, and
administrators, and permission to proceed must be granted by those on-site. Budget approval
must be made for the surgical clinical reviewer and the annual administrative fee. The annual
cost is estimated to be $135,000 per year, which includes the full time employment (FTE) of the
data collector (meaning the surgical clinical reviewer).26,33 The online program application form
needs to be submitted and approved by the ACS. A Hospital Participation Agreement and
surgical clinical reviewer job description are given to the hospital and a surgical clinical reviewer
should be recruited. Experience from the program has shown that the person best qualified for
this position will have a bachelor of science degree in nursing or an advanced degree as well as
surgical clinical experience (meaning in the operating room, surgical intensive care unit, or
cardiac surgery). The candidate should ideally have additional experience with quality
improvement and the clinical hospital system. Full-time effort in the initial phases of
implementation is required.
The surgical clinical reviewer then undergoes training (online modules with a test). ACS
NSQIP assists with implementing the program on-site. The complete process of enrollment and
training ranges from a few weeks to months in duration. Six to 12 months is needed to obtain
results that are acceptable for quality assessment.
The program requirements include site administrative support, a surgeon champion, and
participation in a series of conference calls and the national ACS NSQIP meeting. Commitment
from a surgeon champion (Chief of Surgery or appointed surgeon) is needed to oversee the
program. Their involvement includes quarterly conference calls and attendance at the national
meeting. Data reporting is mandated to follow particular rules, such as accrual of particular data;
complete 30-day follow-up on participating patients, including follow-up attempts; and searches
of public death records. ACS NSQIP personnel perform audits to help maintain data quality. For
small hospitals, the effort and cost may be less than for larger facilities, depending on the volume
of cases.
ACS NSQIP has been implemented in a variety of settings including large academic
hospitals, smaller community hospitals, and statewide consortia (both large and small scale). As
the need became apparent for new program models to accommodate differing clinical needs, four
program options have evolved (Table 2). Program options vary in terms of number of variables
collected, surgical specialty, if procedures are specifically targeted, and case sampling required.
The percent of the FTE for the surgical clinical reviewer varies by program option, as the smaller
146
program can use less effort. FTE calculators are available on-line to calculate the amount
required.
Table 2, Chapter 14. Comparison of American College of Surgeons national surgical quality
improvement program use options
Elements
Classic
Essential
Small & Rural
Eligibility
Any hospital
Any hospital
Variables
Collected
69 clinical variables
(subset of Classic)
Surgical Specialty
Versions
General/Vascular
Multispecialty
General/Vascular
Multispecialty
Multispecialty
Case Volume
Requirements
General/Vascular
=1680 cases/yr (or all
if <1680)
Multispecialty=20%
case volume by
specialty
General/Vascular =40
cases/8 day cycle
Multispecialty=may be
>40 per 8 day cycle
General/Vascular
=1680 cases/yr (or all
if <1680)
Multispecialty=20%
case volume by
specialty
General/Vascular =40
cases/8 day cycle
Multispecialty=may be
>40 per 8 day cycle
Maximum=1680
cases/yr
Case Sampling
<1680 cases per year

or Rural (RUCA) data
code 7.0-10.6
(same as Essential)
Procedure
Targeted
Any hospital
Core of 46 clinical
variables +
Procedural specific
variables (same as
Essential)
General/Vascular
Multispecialty
Minimum =1680
cases/yr (depends
on # and volume)
All cases (100%)
15 Core cases per 8

day cycle
25 Procedure
targeted cases per 8
day cycle
FTE requirement
1 FTE May be more
1 FTE May be more
FTE 400 cases
1 FTE (minimum)
for Multispecialty
for Multispecialty
FTE 800 cases
May be more if
(#cases required
(#cases required
FTE 1200 cases
target more than
/1680=# FTE
/1680=# FTE
1 FTE 1680 cases
1,000 procedures
required)
required)
per year
Adapted from ACS NSQIP Information Sheet dated January 1, 2011. Reprinted by permission from the American College of
Surgeons NSQIP.
The minimum number of cases the surgical clinical reviewer will enter is 1,680 (however,
this number may be smaller or larger, depending on the program chosen). If data can be entered
automatically from the electronic medical record, then an estimated 2,000 to 2,300 cases can be
reviewed per year.
Currently, 431 sites are enrolled in ACS NSQIP, which represents roughly 10% of the almost
4,500 hospitals in the United States. A geographic map of 258 sites that had reported clinically
useful data (from the July 2011 semi-annual NSQIP report) shows the distribution within and
outside of the U.S. (Figure 2).
147
Figure 2, Chapter 14. Geographic distribution of American College of Surgeons national surgical
quality improvement program participating sites
Figure 2 provided from ACS NSQIP Web site, from July 2011 semi-annual NSQIP report, accessed December, 2011
Reprinted by permission from the American College of Surgeons NSQIP.
Of participating sites, 49 percent are teaching or academic centers. The majority of these
hospitals are high volume, as only 3 percent perform fewer than 100 cases per year, 7 percent
perform 100-299 cases per year, 43 percent perform 300-499 cases per year, and 47 percent
perform more than 500 cases per year. This skewed distribution of hospital size means that the
10 percent of hospitals participating in ACS NSQIP represent 32 percent of the procedures
performed (based on Medicare data from ACS NSQIP, personal communication with Clifford
Ko). Certain complex procedures are captured at an even higher rate, for example 57 percent of
esophagectomies and 53.4 percent of pancreatectomy cases billed to Medicare are performed at
ACS NSQIP sites (Table 3).
Table 3, Chapter 14. Percent of Medicare surgical cases covered by the national surgical quality
improvement program
Procedure
Esophagectomy
Cystectomy
Abdominal Aortic
Aneurysm Repair
Pancreatectomy
Colectomy
Proctectomy
AortoIliac bypass
LEB
Liver Resection
Hip Fracture Repair
Abdominoplasty
Lung Resection
1158
3346
MC cases
NOT in
NSQIP
875
4501
3762
3901
32444
6745
2255
12203
2465
40030
1058
16065
MC cases in
NSQIP
2033
7847
Percent
covered by
NSQIP
57.0
42.6
6448
10210
36.8
3399
103056
15767
4974
30100
2201
151140
1829
27391
7300
135500
22512
7229
42303
4666
191170
2887
43456
53.4
23.9
30.0
31.2
28.8
52.8
20.9
36.6
37.0
Total MC
cases
148
Table 3, Chapter 14. Percent of Medicare surgical cases covered by the national surgical quality
improvement program (continued)
Procedure
MC cases in
NSQIP
MC cases
NOT in
NSQIP
Total MC
cases
Percent
covered by
NSQIP
Endovascular
Abdominal Aortic
8944
17324
26268
34.0
Aneurysm Repair
Nephrectomy
9727
16375
26102
37.3
Hysterectomy
17954
45108
63062
28.5
Total Hip
56700
195528
252228
22.5
Arthroplasty
Laminectomy
60650
154858
215508
28.1
TURP
11345
42928
54273
20.9
Ventral Hernia
19360
57735
77095
25.1
Carotid
20588
59710
80298
25.6
Endarterectomy
Total Knee
72916
279642
352558
20.7
Arthroplasty
Prostatectomy
10677
18808
29485
36.2
Breast recon
455
700
1155
39.4
Appy
8802
31635
40437
21.8
Thyroid
5358
12598
17956
29.8
Gastrectomy
3782
7382
11164
33.9
Carotid stent
3648
7883
11531
31.6
Small Bowel
10784
30836
41620
25.9
Resection
Mastectomy
6417
21378
27795
23.1
Cholecystectomy
29386
117327
146713
20.0
Total
32.0
Notes: LEB=Lower extremity bypass; TURP= Transurethral resection of the prostate.
Collaboratives are a main feature of the ACS NSQIP. The collaboratives have taken many
different formsa handful of geographically close hospitals or all of the hospitals in a state
that work together as a team to implement the program and initiate quality improvement. They
also can represent a disease or patient population; thus a collaborative of hospitals need not be
geographically close. Collaboratives provide a collective voice for bargaining with potential
sources of funding. One reported approach is for the main insurer for the hospitals in the
collaborative to pay for 50 percent of the cost of the program over a set number of years.
Sometimes an option to renew the financial support is given if certain milestones are met or
improvements are shown. Some payors have judged there to be a business case for helping
support ACS NSQIP participation due to perceived cost-savings (detailed below). Table 4 shows
the current list of active collaboratives in ACS NSQIP.
Table 4, Chapter 14. List of American College of Surgeons national surgical quality improvement
program collaboratives including type, number of sites, and payor
Group
Canadian National Surgical Quality
Improvement Collaborative (CAN-NSQIP)
Connecticut Surgical Quality Coalition
(CTSQC)
Department of Defense/TRICARE
Florida Surgical Care Initiative (FSCI)
Type
# of Sites
Payor Involvement
Regional
Canadian Health Authorities
Regional
None at this time
16
Department of Defense/TRICARE
63
BlueCross BlueShield of Florida
Systemwide
Regional
149
Table 4, Chapter 14. List of American College of Surgeons national surgical quality improvement
program collaboratives including type, number of sites, and payor (continued)
Group
Fraser Health Systems (Canada)
Illinois Surgical Quality Improvement
Collaborative (ISQIC)
Kaiser Permanente Northern California
Regional NSQIP Collaborative
(KPNCRNC)
Kaiser Permanente Southern California
Regional NSQIP Collaborative
(KPNCRNC)
MaineHealth Collaborative
Mayo Clinic Surgical Quality Consortium
(MCSQC)
Northern California Surgical Quality
Collaborative (NCSQC)
Nebraska Collaborative
Oregon NSQIP Consortia
Pennsylvania NSQIP Consortia
Partners HealthCare
Type
Systemwide
# of Sites
3
Fraser Health Authority
Regional
12
None at this time
Systemwide
21
Kaiser Permanente Northern California
Systemwide
Kaiser Permanente Southern California
MaineHealth
Mayo Clinic
Regional
None at this time
Regional
Regional
Regional
Systemwide
2
8
10
BlueCross BlueShield of Nebraska

None at this time
None at this time
BlueCross BlueShield Massachusetts
Systemwide
Systemwide
Surgical Quality Action Network British

Regional
21
Columbia, Canada (SQAN)
Tennessee Surgical Quality Collaborative
Regional
10
(TSQC)
Upstate New York Surgical Quality
Regional
7
Initiative
ACS NSQIP Colectomy Collaborative
Virtual
36
ACS NSQIP Glucose Control
Virtual
4
Collaborative (Pending)
ACS NSQIP Rural Collaborative (Pending) Virtual
5
ACS NSQIP Residency Training
Virtual
TBD
Collaborative (Pending)
Indiana Collaborative (Pending)
Regional
7
Maryland Collaborative (Pending)
Regional
3
Texas Collaborative (Pending)
Regional
16
Virginia Collaborative (Pending)
Regional
11
Wisconsin Collaborative (Pending)
Regional
6
Abbreviation: TBD= to be determined
Adapted from ACS NSQIP Annual Meeting, July 2011
Reprinted by permission from the American College of Surgeons NSQIP.
Payor Involvement
BC Patient Safety & Quality Council

BlueCross BlueShield of Tennessee
Health Foundation
Excellus
None at this time
None at this time
None at this time
None at this time
None at this time
None at this time
None at this time
None at this time
None at this time
A pilot pediatric collaborative for ACS NSQIP collects data for patients under age 18.34,35
Variables have been modified to pediatric surgery practices and needs.
Henry Ford hospital recently reviewed their lessons learned after implementing ACS NSQIP
over 5 years ago.36 Their findings were summarized into 12 steps (Figure 3).
150
Figure 3, Chapter 14. The 12 steps to implement the national surgical quality improvement
program
Adapted from Velanovich V, Rubinfeld I, Patton JH Jr, Ritz J, Jordan J, Dulchavsky S. Am J Med Qual. 24(6):474-9. 2009 by
Sage Publications. Reprinted by Permission from the SAGE Publications.

The costs of participation vary depending on the program type in which the hospital enrolls.
An annual administrative fee varies by hospital size and level of participation, salary for the
surgical clinical reviewer, and sometimes additional bonus to support the effort of the surgical
champion or quality improvement team. This fee ranges from $10,000 (rural and hospitals that
perform <2000 cases/year) to $25,000 (>2,000 cases). Hospitals have opportunities to lower their
costs by participating in a collaborative or in a hospital system. The annual fee covers the 2-day
training for the surgical clinical reviewer, audits, and the semi-annual data report, as well as the
additional support provided by the ACS NSQIP in terms of materials and help with quality
improvement.
The salary for the surgical clinical reviewer, who collects data and assists with the quality
improvement, makes up the bulk of the expense of participation. Previously, the clinical reviewer
had to be a nurse, but because individuals without nursing degrees have turned out to be some of
the best surgical clinical reviewers, that requirement was dropped. As such, the full time
employment (FTE) for this person will vary, based on their experience, level of training, and the
region. For example, an FTE for an experienced person with a bachelors degree may be around
$40,000 per year, but may be somewhat less for someone with experience but without a degree,
or may be upwards of $100,000 for a registered nurse (with benefits).
Many hospitals suggest that paying the surgeon champion (such as $5,000 annually) is
helpful in increasing their interest and efforts, although, a recent survey of surgical champions
(109 respondents) found that 72.5 percent did not receive salary support to compensate their
time.38
The total cost of participating has been estimated at $135,000, which includes the full time
employment (FTE) in addition to $10,000-$25,000 annual administrative fee; however, this
would be the high end estimate for a large hospital that hires a registered nurse as the surgical
clinical reviewer.33,37 Most participating hospitals in fact pay considerably less that this estimate.
Since the overarching goal of ACS NSQIP is to reduce complications, which are costly, the
business case for participating is that the cost of the program translates into cost savings to the
151
hospital. Examples of such savings reported by NSQIP sites are shown in Table 5. Shown are
pre-post data without control groups; thus, inference of a causal relationship is limited by the
study design.
Table 5, Chapter 14. Example of reductions in complications and associated costs
Hospital
37
Surrey Memorial Hospital
Henry Ford Hospital
36
22
VA
Complication Reduction
Reduced SSI over four years: 13%,
10%, 7.5%, 7.2%
Reduced LOS by 1.54 days over 4
years for general surgery, vascular
and colorectal procedures
Surgical pneumonia alone
Savings
$2.54 million savings
$2 million annual savings (increased
billing by $2.25 million/yr as
underbillings were identified )
9.3 million in savings annually
University of Michigan Medical

7
Center
Respiratory complication
$51,409 per event. A reduction of

two such complications per year
pays participation.
Hershey Medical Center;

26
Penn State
Additional cost attributable to a

postoperative complication=$16,371.
Avoiding one postoperative

complication equals cost savings of
$9052
One cost-effectiveness analysis of participation in ACS NSQIP has been published. It

compared costs and outcomes for 2,229 general surgery and vascular surgery cases at one large
academic hospital between two pre-intervention time periods and two post-intervention times,
the first post-intervention period being the 6 months following implementation and the second
being the 12 months following implementation. The perspective was the hospitals costs. The
study found that the incremental cost of the program were $832 and $266 per patient for the two
time periods, meaning the cost per patient of the program declined after the initial 6 months of
implementation. Postoperative events also declined over time, from 17% to 13%. The
incremental cost-effectiveness ratio to avoid 1 postoperative event was $25,471 in the first 6
months, declining to $7,319 in the second time period, meaning that the longer the institution
participated in the program, the more cost-effective the program became.26

Lessons Learned From Implementation at Different Sites.
Many examples are available of successful program implementation as well as the challenges
facing different hospital types, including varieties of collaboratives. The collaboratives are
proving to be an effective approach for many hospitals, as the group can bargain for financial
support from a variety of sources, shape the program for their own specific needs, and work
together to make quality improvement changes. One example of the experience faced by a
community of hospitals in starting a small state-wide collaborative, the Tennessee Surgical
Quality Collaborative (TSQC), is detailed below. A second example is described of a group of
hospitals across a state that embarked on constructing a collaboration, the Florida Surgical Care
Initiative (FSCI).
During the recent national ACS meeting (October 2011), the Tennessee Surgical Quality
Collaborative gave a detailed presentation of how they started. In 2004, after being introduced to
the newly started ACS NSQIP at the national ACS meeting, a member of a community hospital
in Tennessee returned home and approached his hospitals CEO. One year later they signed the
contract to enroll in the program. Two other Tennessee hospitals started the application process
152
around the same time. Then at the State Chapter ACS meeting, the idea was posed to develop a
statewide collaborative and use a funding mechanism modeled after that in Michigan, where
some funding support would be provided from major payors in the area (in this case, Blue
Cross/Blue Shield). Discussions with the payor were initiated, and the surgical leaders made a
site visit to Michigan to learn more about developing a collaborative. The collaborative included
the hospitals (and the chapter), Blue Cross/Blue Shield (BC/BS), and the Tennessee Hospital
Association. Ultimately, it was decided that control of the collaborative (meaning the data)
would remain with a leadership committee that comprised four surgeons who were appointed by
the Chapter, along with two hospital CEOs and one member of the Tennessee Hospital
Association. The proposal included funding for participation of eight hospitals (three hospitals
that were currently enrolled and five new ones) estimated to be $2,550,000 for 3 years. The
money would cover half the expense of the surgical clinical reviewer at each site, and provide for
some salary support for each site surgeon champion ($5,000 each per year) and the cost of the
administrative Tennessee Center for Patient Safety (TCPS), which would house the data. By
2008, three more hospitals wanted to join, and BC/BS increased their support to include them.
This example highlights many of the key components to building a successful programsurgical
leaders taking a role, supportive administration, collaborating with other hospitals.
A new and strikingly different collaborative is underway in Florida. The story of how this
collaborative started was outlined at the national meeting. In brief, the drive to participate in
NSQIP and improve care started with the hospitals. The Florida Hospital Association (FHA) was
aware of the high surgical mortality demonstrated by the Dartmouth Atlas project (which is
dedicated to identifying and showing disparities in access and utilization of health care) in their
State. The FHA, along with the payor, BC/BS, collaborated to generate a financial incentive for
hospital participation. According to the model, the first step was to identify the surgeon
champions. In order to make the program financially viable to many of the smaller hospitals and
to reach more hospitals, the ACS NSQIP along with the FHA devised a new version of the
program that would collect only four outcomes, thus lowering costs. Currently, 64 hospitals are
participating in the Florida Surgical Care Initiative (FSCI) and participation of 39 more is
pending. This example demonstrates additional features that help encourage participation:
individuals at the State level and hospital administration taking a lead, flexible program design to
fit the needs of the collaborative, and the role of the local payor to incentivize hospital
participation.

Although no randomized trials have assessed the use of outcomes measurements and
reporting in surgery, the strength of the evidence that doing so improves operative mortality and
morbidity has to be considered moderate or even high, given the strong theoretical rationale for
why it should work, the evidence that outcome reporting has likely improved surgical outcomes
in other settings (e.g., the New York State CSRS), the numerous reports from ACS NSQIP sites
of implementation of quality and safety initiatives following identification of high outlier status,
and the ensuing, sometimes dramatic, improvements in those outcomes. A great deal of
experiential evidence exists on how to implement the ACS NSQIPit has been implemented in
more than 400 hospitalssuggesting that the program can be more widely implemented. Some
of the key components of ACS NSQIP (collecting complications data, sharing models of
observed-to-expected results, multi-site data collection systems across institutions that provide
results back to the sites for benchmarking, contexts for learning and sharing tools that appear to
153
be effective across sites) are similar to those of other successful patient safety practices such as
the Michigan Keystone ICU Project to reduce catheter-related bloodstream infections.39 Despite
ACS NSQIP and the Keystone ICU Project having started with different original interventions
(the feedback of procedure-specific surgical outcome data to surgeons and a checklist of
processes to reduce infections) the observation that the current version of the interventions
include so many similar components probably suggests something generalizable about the
implementation of certain kinds of practices across hospitals.
ACS NSQIP provides hospitals and providers with usable clinical data that are otherwise not
available to them. Currently, all hospitals use administrative data to some degree to assess
quality through the Centers for Medicare and Medicaid Services Hospital Compare program or
the Surgical Care Improvement Project (SCIP). These data lack clinical information and are
limited by the variables reported for claims. More importantly, the correlation between
administrative data and actual complications or diagnoses is inadequate. For example, urinary
tract infections are poorly reported in administrative data. Furthermore, studies show that
adherence to SCIP measures do not correlate to better outcomes. ACS NSQIP has the power to
show providers the most problematic clinical data.
The greatest benefit has been seen in the larger hospitals in the procedures with the higher
complication rates. Whether the above improvements will translate to low risk but common
procedures, such as out-patient procedures is unclear. Also, most of the early adopters have been
large academically affiliated hospitals. How successfully and widely it can be implemented at
smaller hospitals remains to be seen. A summary table is located below (Table 6).
Scope of the
Strength of
the PSP
Effectiveness
Common/High
Moderate-tohigh
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate
References
1.
Hall MJ, DeFrances CJ, Williams SN, et al.

National hospital discharge survey, 2007.
U.S. Department of Health and Human
Services. Centers for Disease Control and
Prevention. National Center for Health
Statistics. National Health Statistics Report.
Number 29, Oct, 26, 2010.
2.
Ghaferi AA, Birkmeyer JD, Dimick JB.

Hospital Volume and Failure to Rescue
With High-risk Surgery. Med Care. 2011
Dec;49(12):1076-81. PMID: 22002649.
154
3.
Englesbe MJ, Brooks L, Kubus J,

Luchtefeld M, Lynch J, Senagore A,
Eggenberger JC, Velanovich V, Campbell
DA Jr. A statewide assessment of surgical
site infection following colectomy: the role
of oral antibiotics. Ann Surg. 2010
Sep;252(3):514-9; discussion 519-20.
PMID: 20739852.
4.
NSQIP ROI calculator.

http://site.acsnsqip.org/wpcontent/themes/nsqip/extras/flex2/ROICalc.
html. December 2012.
5.
Saint S, Elmore JG, Sullivan SD, et al. The

efficacy of silver alloy coated urinary
catheters in preventing urinary tract
infection: a meta-analysis. Am J Med
1998;105:236241.
6.
Silber JH, Rosenbaum PR, Trudeau ME,

Chen W, Zhang X, Kelz RR, Mosher RE,
Even-Shoshan O. Changes in prognosis after
the first postoperative complication. Med
Care. 2005 Feb;43(2):122-31. PMID:
15655425.
14.
Fung CH, Lim YW, Mattke S, Damberg C,

Shekelle PG. Systematic review: the
evidence that publishing patient care
performance data improves quality of care.
Ann Intern Med. 2008 Jan 15;148(2):11123. Review. PubMed PMID: 18195336.
7.
Dimick JB, Chen SL, Taheri PA, Henderson

WG, Khuri SF, Campbell DA Jr. Hospital
costs associated with surgical complications:
a report from the private sector National
Surgical Quality Improvement Program. J
Am Coll Surg. 2004; 199(4):531-537.
15.
Marshall MN, Shekelle PG, Brook RH,

Leatherman S. Use of performance data to
change physician behavior. JAMA. 2000
Sep 6;284(9):1079.
16.
Westaby S, Archer N, Manning N, Adwani

S, Grebenik C, Ormerod O, Pillai R, Wilson
N. Comparison of hospital episode statistics
and central cardiac audit database in public
reporting of congenital heart surgery
mortality. BMJ. 2007 Oct
13;335(7623):759.
17.
Steinberg SM, Popa MR, Michalek JA,

Bethel MJ, Ellison EC. Comparison of risk
adjustment methodologies in surgical quality
improvement. Surgery. 2008
Oct;144(4):662-7; discussion 662-7. PMID:
18847652.
18.
Ko C. ACS NSQIP Conference and

Semiannual Report Overview. Presentation
as the 2009 ACS NSQIP National
Conference, July 2009
19.
Hall BL, Hamilton BH, Richards K,

Bilimoria KY, Cohen ME, Ko CY. Does
surgical quality improve in the American
College of Surgeons National Surgical
Quality Improvement Program: an
evaluation of all participating hospitals. Ann
Surg. 2009 Sep;250(3):363-76.
20.
Daley J, Khuri SF, Henderson W, Hur K,

Gibbs JO, Barbour G, Demakis J, Irvin G
3rd, Stremple JF, Grover F, McDonald G,
Passaro E Jr, Fabri PJ, Spencer J,
Hammermeister K, Aust JB, Oprian C. Risk
adjustment of the postoperative morbidity
rate for the comparative assessment of the
quality of surgical care: results of the
National Veterans Affairs Surgical Risk
Study. J Am Coll Surg. 1997
Oct;185(4):328-40. PMID: 9328381.
8.
Dimick JB, Weeks WB, Karia RJ, Das S,

Campbell DA Jr. Who pays for poor surgical
quality? building a business case for quality
improvement. J Am Coll Surg.
2006;202(6):933-937.
9.
Codman EA. The product of a hospital. Surg

Gynecol Obstet. 1914;18:491-496.
10.
Khuri SF, Daley J, Henderson W, Barbour

G, Lowry P, Irvin G, Gibbs J, Grover F,
Hammermeister K, Stremple JF, et al. The
National Veterans Administration Surgical
Risk Study: risk adjustment for the
comparative assessment of the quality of
surgical care. J Am Coll Surg. 1995
May;180(5):519-31. PMID: 77495
11.
Khuri SF, Daley J, Henderson W, Hur K,

Hammermeister K, Aust JB. Risk
adjustment of the postoperative mortality
Oct;185(4):315-27. PMID: 9328380.
12.
Iezzoni LI. Measuring the severity of illness

and case mix. In: Goldfield N, Nash DB,
eds. Providing Quality Care: the Challenge
to Clinicians. Philadelphia, Pa: American
College of Physicians; 1989:70-105.
13.
Hannan EL, Kilburn H Jr, Racz M, Shields

E, Chassin MR. Improving the outcomes of
coronary artery bypass surgery in New York
State. JAMA 1994;271:761 6.
155
21.

Hammermeister K, Aust JB. Risk
adjustment of the postoperative mortality
Oct;185(4):315-27. PMID: 9328380.
22.
Khuri SF, Daley J, Henderson W, Barbour

G, Lowry P, Irvin G, Gibbs J, Grover F,
Hammermeister K, Stremple JF, et al. The
National Veterans Administration Surgical
Risk Study: risk adjustment for the
comparative assessment of the quality of
surgical care. J Am Coll Surg. 1995
May;180(5):519-31. PMID: 7749526.
23.

Demakis J, Aust JB, Chong V, Fabri PJ,
Gibbs JO, Grover F, Hammermeister K,
Irvin G 3rd, McDonald G, Passaro E Jr,
Phillips L, Scamman F, Spencer J, Stremple
JF. The Department of Veterans Affairs
NSQIP: the first national, validated,
outcome-based, risk-adjusted, and peercontrolled program for the measurement and
enhancement of the quality of surgical care.
National VA Surgical Quality Improvement
Program. Ann Surg. 1998 Oct; 228 (4):491507. PMID: 9790339.
24.
25.
Fink AS, Campbell DA Jr, Mentzer RM Jr,

Henderson WG, Daley J, Bannister J, Hur
K, Khuri SF. The National Surgical Quality
Improvement Program in non-veterans
administration hospitals: initial
demonstration of feasibility. Ann Surg. 2002
Sep;236(3):344-53; discussion 353-4.
PMID: 12192321.
Khuri SF, Henderson WG, Daley J,
Jonasson O, Jones RS, Campbell DA Jr,
Fink AS, Mentzer RM Jr, Neumayer L,
Hammermeister K, Mosca C, Healey N;
Principal Investigators of the Patient Safety
in Surgery Study. Successful
implementation of the Department of
Veterans Affairs National Surgical Quality
Improvement Program in the private sector:
the Patient Safety in Surgery study. Ann
Surg. 2008 Aug;248(2):329-36. PMID:
18650645.
156
26.
Hollenbeak CS, Boltz MM, Wang L,

Schubart J, Ortenzi G, Zhu J, Dillon PW.
Cost-effectiveness of the National Surgical
Quality Improvement Program. Ann Surg.
2011 Oct;254(4):619-24. PMID: 22039608.
27.
Rowell KS, Turrentine FE, Hutter MM,

Khuri SF, Henderson WG. Use of national
surgical quality improvement program data
as a catalyst for quality improvement. J Am
Coll Surg. 2007 Jun;204(6):1293-300.
PMID: 17544087.
28.
Campbell DA Jr, Englesbe MJ, Kubus JJ,

Phillips LR, Shanley CJ, Velanovich V,
Lloyd LR, Hutton MC, Arneson WA, Share
DA. Accelerating the pace of surgical
quality improvement: the power of hospital
collaboration. Arch Surg. 2010
Oct;145(10):985-91. PMID: 20956768.
29.
Omoigui NA, Miller DP, Brown KJ, et al.

Outmigration for coronary bypass surgery in
an era of public dissemination of clinical
outcomes. Circulation 1996;93:27-33.
30.
Halm EA, Chassin MR. Why do hospital

death rates vary? N Engl J Med. 2001 Aug
30;345(9):692-4. PMID: 11547726.
31.
Hannan EL, Chassin MR. Publicly reporting

quality information. JAMA. 2005 Jun
22;293(24):2999-3000; author reply 3000-1.
PMID: 15972560.
32.
Raval MV, Hamilton BH, Ingraham AM,

Ko CY, Hall BL. The importance of
assessing both inpatient and outpatient
surgical quality. Ann Surg. 2011
Mar;253(3):611-8. PMID: 21183845.
33.
Hammermeister K. The National Surgical

Quality Improvement Program: learning
from the past and moving to the future. Am
J Surg. 2009;198:S6973.
34.
Raval MV, Dillon PW, Bruny JL, Ko CY,

Hall BL, Moss RL, Oldham KT, Richards
KE, Vinocur CD, Ziegler MM; ACS NSQIP
Pediatric Steering Committee. American
College of Surgeons National Surgical
Quality Improvement Program Pediatric: a
phase 1 report. J Am Coll Surg. 2011
Jan;212(1):1-11. PMID: 21036076.
35.
Raval MV, Dillon PW, Bruny JL, Ko CY,

Hall BL, Moss RL, Oldham KT, Richards
KE, Vinocur CD, Ziegler MM; ACS NSQIP
Pediatric Steering Committee. Pediatric
Surgical Quality Improvement Program:
feasibility of a novel, prospective
assessment of surgical outcomes. J Pediatr
Surg. 2011 Jan;46(1):115-21. PMID:
21238651.
36.
Velanovich V, Rubinfeld I, Patton JH Jr,

Ritz J, Jordan J, Dulchavsky S.
Implementation of the National Surgical
Quality Improvement Program: critical steps
to success for surgeons and hospitals. Am J
Med Qual. 2009 Nov-Dec;24(6):474-9.
PMID: 19584373.
37.
Ingraham AM, Richards KE, Hall BL, Ko

CY. Quality improvement in surgery: the
Surgical Quality Improvement Program
approach. Adv Surg. 2010;44:251-67.
PMID: 20919525.
157
38.
Raval MV, Bentrem DJ, Eskandari MK,

Ingraham AM, Hall BL, Randolph B, Ko
CY, Morton JM. The role of Surgical
Champions in the American College of
Surgeons National Surgical Quality
Improvement Program--a national survey. J
Surg Res. 2011 Mar;166(1):e15-25. PMID:
21176914.
39.
Pronovost P, Needham D, Berenholtz S,

Sinopoli D, Chu H, Cosgrove S, Sexton B,
Hyzy R, Welsh R, Roth G, Bander J, Kepros
J, Goeschel C. An intervention to decrease
catheter-related bloodstream infections in
the ICU. N Engl J Med. 2006 Dec
28;355(26):2725-32.
Chapter 15. Prevention of Surgical Items Being Left Inside

Patient: Brief Update Review
Jonathan R. Treadwell, Ph.D.
Introduction
Leaving surgical items inside patients is a rare but potentially deadly mistake. The most
common such item is a surgical sponge. Researchers at the Mayo clinic found that during the
four-year period from 2003 to 2006, the rate of retained foreign objects was 1 in every 5,500
operations,1 and 68 percent of the retained objects were sponges. The greatest subsequent risk to
the patient is infection, which can be fatal. Other risks include perforations and granulomas.
Risk factors for such incidents were explored by Gawande and colleagues (2003),2 who
examined factors surrounding 42 retained sponges and 19 retained instruments. The majority
required reoperation, and one patient died. Compared with control incidents, item retention was
more likely to occur in the context of emergency surgery, an unexpected change in surgical
procedure, high body-mass index (an x-ray is recommended), and the lack of item counts.
A review of the literature on the topic of retained surgical sponges conducted for the original
report identified only one study (a case series) that attempted to assess the use of sponge and
instrument counts to prevent retention.3 The goals of the present review were to identify
interventions implemented since the previous review and to report on studies assessing their
effectiveness. We conducted a review of the literature from 2001 to 2011 and reviewed all
studies relevant to methods used to prevent surgical items from being left inside patients during
surgical procedures.
What Are the Practices Aimed at the Prevention of Leaving Surgical

Items Inside Patients?
To prevent leaving surgical items inside patients, the Association of periOperative Registered
Nurses (AORN) recommends counting all sponges, sharps, and related miscellaneous items at
five different times: (1) before the procedure to establish a baseline, (2) before closure of a cavity
within a cavity, (3) before wound closure begins, (4) at skin closure, and (5) at the time of
permanent staff relief of either the scrub person or the circulating nurse.4 In addition, specifically
for surgical instruments, AORN recommends counting only at times 1, 3, and 5 above. AORN
also recommends all counts be documented in the intraoperative record. If a discrepancy occurs
between counts, surgical staff must search for the lost item (usually a sponge). If it is suspected
in the OR that an item was left inside the patient, a radiograph may be necessary.
For comparison, the Surgical Safety Checklist of the World Health Organization requires a
post-procedure count, but the checklist suggests neither pre-procedure counts nor intraoperative
counts.5
What Supplementary Methods Have Been Used To Improve Counts

of Surgical Items?
Three technologies can enhance the accuracy of the count, thereby further lowering
(theoretically) the risk of leaving items inside patients: (1) bar coding, (2) radiofrequency tagging
158
without unique item ID numbers (abbreviated RF), and (3) radiofrequency tagging with unique
item ID numbers (RFID). Bar coding is an established and low-cost technology, but a direct line
of sight between the bar code scanner and the items bar code label is needed in order to scan it,
and blood-soaked items may be difficult to scan accurately. The RF technologies (a penny-sized
or smaller chip implanted into the device) allow items to be detected by a specialized wand that
is waved over the patients body during and after the procedure. This scan can prevent the need
for a radiograph, which itself can increase surgical risk because of the added time on the
operating table and under anesthesia. RFID represent an advance from the simpler RF
technologies because if each item is assigned a unique ID number, then the manual count can be
checked against the RFID systems baseline count.
The FDA has cleared four products relevant to the above technologies for marketing in the
United States (U.S.):
1. Safety-Sponge System, (SurgiCount Medical, Temecula, California).6 This system
comprises bar-coded sponges.
2. RF Surgical Detection System (RF Surgical, Bellevue Washington). This technology
permits detection of devices but does not provide a count, because items do not receive
unique ID numbers. The detection wand is single use. The system can be used with
sponges, laparotomic pads, gauze, and towels, but not surgical instruments or sharps.
3. SmartSponge System (Clear Count Medical, Pittsburgh Pennsylvania). This system
assigns a unique ID for each device, so it is used for both detection and counting. As the
procedure progresses and staff remove sponges or other items, they put the items into a
specialized bucket fitted with an antenna that detects and counts the RFID items. If a
discrepancy occurs between the baseline counts and the final counts, it notifies the OR
team with auditory and visual warnings, thereby initiating a search for the lost item(s)
using a detection wand. This system also can be used with sponges, laparotomic pads,
gauze, and towels, but not surgical instruments or sharps.
4. ORLocate (Haldor, Boston, Massachusetts).7 This system also assigns a unique ID for
each device, so it is used for both detection and counting. Unlike the two systems
described above, it can be used for instruments and sharps as well as sponges and other
non-metallic items
What Have We Learned About Methods To Improve Counts of

Surgical Items?
Greenberg and colleagues (2008)8 randomized 298 patients to undergo operations involving
either manual counting (148 patients) or bar-coded sponges (150 patients). Twice as many
sponge count discrepancies were detected in the bar-coded group (24 operations) as in the
manual counting group (12 operations). This difference was mostly explained by miscounted
sponges (nine operations in the bar-code group vs. one operation in the manual counting group)
rather than by misplaced or retained sponges (17 in the bar-code group vs. nine in the manual
counting group). Interestingly, in these same operations, no difference was seen between the
groups in count discrepancies for non-bar-coded surgical instruments (11 in the bar-code group
vs. ten in the manual counting group).
A 2009 systematic review by Stawicki and colleagues9 on risks and measures to prevent
retention of surgical items that considered a variety of case reports, case series, registry reports,
and position papers concluded that the most important preventive measure is to accurately count
all the pieces of surgical gauze and surgical instruments used during an operation. Authors also
159
listed additional factors that could help minimize this type of mistake: (1) Knowledge of risk
factors, (2) Use of modern technology, (3) Improved perioperative patient processing systems.
Methods May Be Time Consuming and Present Technical Challenges

In the 2008 study by Greenberg and colleagues,8 17 incidents of technological difficulties
occurred because of the bar-code system (2.04 per 1000 sponges counted), issues that would not
have arisen with manual counting. Further, of 150 operations with bar-coded sponges, the
surgical team abandoned the bar-code system in five operations (3%) due to the extra time
required. However, the authors concluded that the use of the bar-code technology was well
tolerated by staff members. The amount of time needed to count items can potentially cause
harm to patients if other key surgical steps are missed or rushed as a result of counting.
Greenberg and colleagues)8 found that the bar-code-sponge method required more than twice as
much time to count sponges as the manual method (5.3 minutes vs. 2.4 minutes).
Other Implementation Issues May Arise

One consideration for hospitals regarding the use of RF and RFID technologies is that if only
a portion of the hospitals surgical devices are RF-enabled, confusion might result. Staff may
mistakenly assume that all devices are RF-enabled, and a post-procedural scan would miss any
non-RF-enabled device inside the patient. Thus, it is recommended that RF-adoption be all or
none.10 Also, wand technique can be important when using RF devices, because scanning too far
away from the body, or too earlythe surgeon may need to use additional tagged itemscan
fail to locate all items. Also, because adipose tissue can increase the distance between the wand
and tagged items, some items may be missed when scanning obese patients.
In the 2008 study of bar-coded sponges by Greenberg and colleagues8, a post-study survey of
41 providers found moderately high ratings for ease of us (average rating 7.3 on a 0-10 scale)
and confidence in the ability of the system to track sponges (average rating 7.5 on a 0-10 scale).
Opinions on whether the bar-code system benefitted the counting protocol were mixed but
slightly positive (on a scale from -5 to +5, the average was +1.6). Authors stated that some
providers felt that the system was especially useful in large operations with high blood loss and
many sponges, whereas others felt that the system was difficult to use in these types of
operations.8
Questions About Cost-Effectiveness

The medical and liability costs of a surgical item left inside a patient can exceed $200,000.11
In 2009, Regenbogen and colleagues11 performed a cost-effectiveness analysis of six strategies to
prevent this type of incident. In their simulation, manual counting prevented 82 percent of the
simulated incidents at a cost of $1,500. In comparison, the other five strategies performed as
follows:
Bar-coding the sponges raised the effectiveness to 97.5 percent, and the cost-perprevented-retained-sponge was $95,000.
RF-enabling the sponges (without a unique ID for each sponge) raised the effectiveness
to between 97.5 percent and 100 percent, and the cost-per-prevented-retained-sponge was
between $620,000 and $720,000.
Three radiographic strategies were dominated by the two bar code strategies with respect
to cost and effectiveness: 1. Do not count but always X-ray before closure; 2. Count, and
160
always X-ray before closure; 3. Count, and also X-ray before closure only for high-risk
operations).

To prevent leaving items (typically sponges) inside patients during surgery, manually
counting all items is widely recommended. Although several supplementary technologies exist,
their use must remain limited to that of supplementing or aiding counting. These technologies
include bar coding and radiofrequency tagging (with or without unique ID numbers). For each of
these technologies, specific institutional hurdles (e.g., cost, confusion with older non-tagged
devices, and wand technique with RF and RFID systems) must be overcome before their use can
be considered both reliable and cost effective. A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Rare/Low
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
How Much do We
Low if it simply Little

involves more
frequent manual
counting; high if
RFID is used
References
1.
Cima RR, Kollengode A, Garnatz J, et al.

Incidence and characteristics of potential
and actual retained foreign object events in
surgical patients. J Am Coll Surg 2008
Jul;207(1):80-7. PMID: 18589366.
2.
Gawande AA, Studdert DM, Orav EJ, et al.

Risk factors for retained instruments and
sponges after surgery. N Engl J Med 2003
Jan 16;348(3):229-35. PMID: 12529464.
3.
4.
Gibbs VC, Auerbach AD. The retained

surgical sponge. In: Shojania KG, Duncan
BW, McDonald KM, et al, editors. Making
health care safer: a critical analysis of
patient safety practices. Evidence Report
Technology Assessment No. 43. Rockville
(MD): Agency for Healthcare Research and
Quality (AHRQ); 2001 Jul. p. 255-8.
Recommended practices for sponge, sharp,
and instrument counts. AORN
Recommended Practices Committee.
Association of periOperative Registered
Nurses. AORN J 1999 Dec;70(6):1083-9.
PMID: 10635432.
161
5.

Safe Surgery Saves Lives Study Group. A
J Med 2009;360(5):491-9.
http://content.nejm.org/cgi/content/full/NEJ
Msa0810119. PMID: 19144931.
6.
The safety-sponge. [internet]. Irvine (CA):

Surgicount Medical, Inc; [accessed 2011
Nov 2]. [2 p].
www.surgicountmedical.com/products_safet
y.php.
7.
ORLocate OR. [internet]. Boston: Haldor

Advanced Technologies, Ltd; [accessed
2011 Nov 2]. [3 p]. www.haldortech.com/ORLocate_OR.aspx.
8.
Greenberg CC, Diaz-Flores R, Lipsitz SR, et

al. Bar-coding surgical sponges to improve
safety: a randomized controlled trial. Ann
Surg 2008 Apr;247(4):612-6. PMID:
18362623.
9.
Stawicki SP, Evans DC, Cipolla J, et al.

Retained surgical foreign bodies: a
comprehensive review of risks and
preventive strategies. Scand J Surg
2009;98(1):8-17. PMID: 19447736.
10.
ECRI Institute. Radio-frequency surgical

sponge detection: a new way to lower the
odds of leaving sponges (and similar items)
in patients. Health Devices 2008
Jul;37(7):193-202. PMID: 18771206.
11.
162
Regenbogen SE, Greenberg CC, Resch SC,

et al. Prevention of retained surgical
sponges: a decision-analytic model
predicting relative cost-effectiveness.
Surgery 2009 May;145(5):527-35.
www.ncbi.nlm.nih.gov/pmc/articles/PMC27
25304/pdf/nihms114760.pdf. PMID:
19375612.
Chapter 16. Operating Room Integration and Display

Systems: Brief Review (NEW)
Introduction
Patient monitoring is one of the central tasks in operating rooms (ORs). Because of the
increasing use of advanced technologies in surgical procedures, todays ORs are commonly
crowded with freestanding devices, support systems, and monitors. In addition to the traditional
monitors that continuously present the patients hemodynamic, respiratory, and
electrophysiological signals, many innovative devices recently introduced into the OR feature
their own platforms for data display. These devices may fall into one of four categories:1
Surgical machine-controlled applications (e.g., robotics, minimally invasive surgery,
video-endoscopic surgery, master-slave systems)
Designated diagnostic and real-time navigation devices (e.g., magnetic resonance
imaging, computed tomography, three-dimensional ultrasound)
Information technology (IT) applications generating a real-time connection between the
OR and the hospital medical record archives (e.g., picture archiving and communication
systems [PACS], videos, electronic medical records (EMRs), hospital information
systems, and other laboratory data)
Telecommunication and teleconferencing systems connecting the OR in real time with
other medical centers
The increasing use of these devices leads to a congestion of data displays in the OR, compels
OR staff members to increase the time devoted to monitoring the displays, and may divide their
attention between monitoring and other tasks.1 Meanwhile, the proliferation of freestanding
devices and displays in the OR makes coordination difficult.2 Surgeons, nurses, and
anesthesiologists have their own perioperative devices or systems on which to focus. Human
coordination of multiple electromechanical devices may lead to misunderstandings and delay
action.
As some experts have commented, what is currently lacking in most ORs is a high-level
overview of all the information that is already available in the room.3-5 The lack of integration of
patient data makes ORs inefficient, overcrowded, and less safe. When patient data are not
displayed to caregivers in an integrated fashion, OR staff have to frequently with multiple
displays to obtain updates and exert control over the various devices at their disposal.4,5
OR integration is an emerging technology that has the potential to address the long-standing
problem of segregated data display in ORs. This technology organizes and consolidates patient
data for clinicians during a surgical procedure. This chapter focuses on the latest form of the
technology, which features a centralized data display platform.
163
What Are the Practices for Integrating Operating-Room Display

Systems?
An OR is integrated if users can control the routing of audio/video (AV) signals from a
central location.3,6 These AV signals can originate from within an OR (e.g., endoscopes, wallmounted cameras) or outside an OR (e.g., PACS, AV feed from another OR). Depending on
configuration, an OR integration system may also permit centralized control of certain clinical
devices (e.g., insufflators and electrocautery units) and nonclinical equipment (e.g., lighting and
room climate controls).6,7 Some OR integration systems may allow signals to be sent outside the
room (e.g., to a conference room or to a central display used by the OR scheduling nurse) or
exchange data with an EMR system.6,7
Current-generation OR integration systems offer a range of capabilities. ECRI Institute has
identified at least 10 vendors that offer products meeting the basic definition of OR integration
(i.e., with centralized control of AV routing in the OR).3,7
Centralized display of consolidated data. OR integration technology continues to evolve as
new features are added. One of the latest developments in the field is the consolidation of realtime patient data from different devices and systems (e.g., physiologic monitors, anesthesia
systems) for display on a dashboard format single screen that can be viewed simultaneously by
all clinicians in the OR.3,7 We believe this new development represents the future of OR
integration; thus, we focus this chapter exclusively on OR integration systems that can display
consolidated data on a single screen. We exclude conventional OR integration systems that lack
the capacity to present data in a centralized, single-screen format.
Two studies published by ECRI Institute in 2008 identified at least two vendors that offered
OR integration systems featuring a centralized repository/display of consolidated data sent from
a number of monitors and devices in real time.3,7 These two systems are the OR-Dashboard
(LiveData, Inc., Cambridge, MA, U.S.) and the ICIS (integrated clinical information system)
Dashboard (Global Care Quest, Inc., Aliso Viejo, CA, U.S.). Both systems have the capacity to
interface with numerous medical devices, patient monitors, and information systems without
encountering compatibility problems. Data may be collected from patient progress logs, OR
scheduling software, real-time vital signs, anesthesiology systems, medical infusion pumps, radio
frequency identification tracking systems, PACS, EMRs, clinical laboratory systems, in-room
cameras, endoscopic systems, clinical notes and rounding lists, bidirectional video conferencing,
and audio note recording. The data can be displayed on large, wall-mounted, flat-panel screens
and on accessory monitors in the OR. Both systems allow clinicians to monitor time trends of
various waveforms, such as respiration, blood pressure, and cardiac activity.3,7,8
The central display of the OR-Dashboard changes to reflect one of four procedural stages:
case setup, time out (safety pause), intraoperative, and closing. Case setup mode displays
information such as surgical supplies and blood availability. The time out mode assists the
surgical team in verifying patient and case information. Intraoperative mode displays information
such as physiologic status, fluid status, and current readings from ventilators and infusion pumps.
Closing mode includes information on equipment counts, postanesthesia care unit assignment,
and family waiting status.3,7,8 This display arrangement is intended to improve situation
awareness during the surgery.8
Used with other products provided by LiveData, the OR-Dashboard allows
videoconferencing between the surgical team and other departments to monitor procedures
remotely or consult specialists throughout the hospital in real time.7,8 The OR-Dashboard can
164
also securely archive all case data, including video, using various device and information
protocols to permit case review after the surgery.7,8 With additional software provided by Global
Care Quest, the ICIS Dashboard provides secure access to the collected patient data in real time
through wireless mobile devices, including personal digital assistants and smart phones that
clinicians can use where a wireless connection is available.7
Contribution to operative and perioperative safety. OR integration technology with
centralized data display can potentially help improve operative and perioperative safety in
several ways. First, the technology allows easy, just-in-time access to patient information from
disparate devices or systems that is often unseen, unrealized, or unused.8,9 Increased access to
this information may improve team situation awareness (TSA, i.e., the task- and team-oriented
knowledge held by everyone in the team and the collective understanding of the unfolding
situation).8,10 TSA is one of the critical factors in OR teamwork that can affect patient safety and
quality of care. Augmented TSA can improve communication among clinical personnel and thus
help reduce the number of medical errors.8,11,12
Second, the integration of previously isolated information sources may open new
opportunities for decision support and augment vigilance.8 For example, allergy information
from the hospital information system may alert the team to not administer certain drugs to the
patient and, thus, prevent harmful drug-related adverse events. Information from the laparoscopic
insufflator can inform the team of impending asystole from insufflation. Information from the
location tracking system may help the surgical team check the accuracy of patient identity.
Integration with the order-entry system can help update the team on workflow and resource
acquisition such as pathology, radiology, and the blood bank. The order information can be
continuously displayed throughout the operative period to decrease uninformed or delayed
decisions.
Additionally, the OR integration technology provides the ability to flexibly change the source
or destination of an AV signal without requiring the cumbersome process of reconfiguring direct
links between sources and destinations each time such a change is needed.6 This ability might
decrease the risk of medical errors in the reconfiguration process. OR integration may also
generate other opportunities for improving patient safety. For example, the technology might
allow real-time, remote consultation from experts outside of the OR.6 The technology could
enhance patient data collection during the surgery and decrease stale or duplicate data. These
data can be analyzed later for the purpose of improving patient safety. The technology may also
have a positive psychological effect on clinical personnel, making them feel more comfortable
and more confident that things are going well.10
How Have Integrated Operating-Room Display Systems Been

Implemented?
We identified several sources that described issues related to the design, planning, or
installation of integrated OR systems (with or without centralized data display).5,6,14-17 In
particular, two studies provided practical guidance on the implementation of integrated ORs. One
of the studies offered step-by-step instruction for addressing the equipment and construction
needs for OR integration.15 The study outlined the technical considerations for in-room
integration, extended AV integration, and equipment control. The study also provided a detailed
list of equipment and specifications required for OR integration.
165
The second study discussed the technical issues that must be addressed when installing
integrated ORs.14 The issues included controlling the images, integrating team members, preconstruction planning, working with vendors, and managing the final project phases. It was not
feasible to provide additional details about these studies in this brief review. Readers can refer to
the original studies for detailed instructions on implementing integrated ORs.
Data about costs. According to an ECRI Institute study, as of October 3, 2007, a LiveData ORDashboard system costs about $150,000.7 Total system costs can vary widely depending on the
features that a hospital requests and the number of systems installed at a facility.6,18 Facilities
could also face significant additional costs to integrate new systems into their existing IT
infrastructure. Similar cost information for the ICIS Dashboard was not reported.7
Effect of context on effectiveness. We identified a survey of 17 surgeons and 9 scrub nurses
from a single hospital that evaluated their satisfaction after 2 years of use of integrated ORs.13
The surgeons and scrub nurses agreed that a great degree of education and a cultural change were
needed to use the system in a correct and complete way.13 However, we were not able to verify
whether the integrated ORs described in the study had the centralized data display feature. We
did not identify any other study that evaluated the effect of context on the effectiveness of an
integrated OR and centralized display systems in improving patient safety.
What Have We Learned About Integrating Operating-Room

Display Systems?
Despite all of the rationales supporting the adoption of OR integration and display systems,
published evidence to validate the effect of this technology on patient safety is rare.7 Researchers
face many practical obstacles in designing and conducting clinical trials that could deliver a
hard-and-fast measurement of that effect.3 For example, surgical patients comprise a
heterogeneous population, making it difficult to draw firm conclusions from any studies.
Additionally, because the incidence of medical errors and other adverse events is rather low
(from the statistical perspective), detecting a significant improvement in safety outcomes
typically requires a very large number of patients. Recruiting enough patients to conduct a good
study would be difficult.
Our search identified only one case report that described the experiences of a hospital in
implementing an OR integration system with centralized display (called wall of knowledge in
the review).9 The authors provided their opinion-based assessment of the system. The perceived
benefits of the system included easy access to a patients vital signs for surgeons during the
operation, improved staff handoffs, reduction of clutter in the OR, improved teaching function,
and timely data reporting. No patient safety outcomes were reported in the study.
In the survey that evaluated the satisfaction of 17 surgeons and 9 scrub nurses from 1 hospital
after 2 years of using integrated ORs,13 the clinicians agreed that integrated ORsusing a
digitalized video acquisition system, boom-mounted devices, and multiple displayscan be very
effective in increasing quality of care, reducing risk, and shortening surgery time. Scrub nurses
were particularly confident that medical device control could reduce the confusion inside the OR
and reduce the number of setting errors. However, as mentioned previously, based on the
information reported in the study, we were not able to verify whether the integrated ORs had the
centralized display feature.
166
In theory, if an OR integration system or its centralized display stop functioning

appropriately or fail entirely, clinicians in the OR could receive delayed or misleading
information about the patient. Clinical decisions based on such information could lead to patient
harm; however, our search did not identify any study that reported data on harms caused by
integrated OR centralized display systems.
Note that for this chapter, we reviewed only studies relevant to patient safety issues. We did
not review studies that focused solely on management issues (e.g., the effects of a display system
on OR efficiency or staff scheduling).

OR integration with centralized data repository/display represents the latest technology
development in the OR setting. While the technology might help improve patient safety,
evidence to demonstrate the technologys benefits in improving safety outcomes is lacking.
Given the many practical obstacles in designing and conducting empirical studies to test the
benefits of this technology, decisions on its adoption will continue to be based on rationales
rather than hard evidence in the near future. Patient safety is only one of the factors that are
considered in the decisionmaking process. Other factors, such as the technologys potential to
improve OR efficiency and productivity, need to be considered as well. As this review has
suggested, the implementation of integrated ORs with centralized data display is not inexpensive.
Decisionmakers should carefully evaluate their facilitys needs and long-term goals to determine
whether this technology is really needed and, if it is, which integration capabilities are
appropriate.6 A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Low-to-high
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate
References
1.
Bitterman N. Technologies and solutions for

data display in the operating room. J Clin
Monit Comput 2006 Jun;20(3):165-73.
PMID: 16699740.
2.
Feussner H. The operating room of the

future: a view from Europe. Semin Laparosc
Surg 2003;10(3):149-56. PMID: 14551657.
3.
ECRI Institute. How could OR integration

systems affect patient care? Health
Technology Trends 2008 Jan;20(1):1-3, 8.
4.
Bucholz RD, Laycock KA, McDurmont L.

Operating room integration and telehealth.
Acta Neurochir Suppl 2011;109:223-7.
PMID: 20960347.
167
5.
Clinical systems innovations: the operating

rooms of the future. [Internet]. Boston
(MA): Center for Integration of Medicine &
Innovative Technology; 2011 [accessed
2011 Nov 30]. [2 p].
www.cimit.org/programsoperatingroom.html.
6.
OR integration: what it is--and what it isnt.

Health Devices 2007 Sep;36(9):281-6.
PMID: 18027807.
7.
ECRI Institute. Data integration and display

systems. Plymouth Meeting (PA): ECRI
Institute; 2008 Jan 16. 4 p. (Health
Technology Forecast).
8.
Meyer M, Levine WC, Brzezinski P, et al.

Integration of hospital information systems,
operative and peri-operative information
systems, and operative equipment into a
single information display. AMIA Annu
Symp Proc 2005;1054. PMID: 16779341.
9.
Kurtz R. Wall of knowledge informs OR

team. OR Manager 2008;1-2.
10.
Parush A, Kramer C, Foster-Hunt T, et al.

Communication and team situation
awareness in the OR: Implications for
augmentative information display. J Biomed
Inform 2011 Jun;44(3):477-85. PMID:
20381642.
11.
Lai F, Spitz G, Brzezinski P. Gestalt

operating room display design for
perioperative team situation awareness. Stud
Health Technol Inform 2006;119:282-4.
PMID: 16404062.
12.
Hu PF, Xiao Y, Ho D, et al. Advanced

visualization platform for surgical operating
room coordination: distributed video board
system. Surg Innov 2006 Jun;13(2):129-35.
PMID: 17012154.
13.
Nocco U, del Torchio S. The integrated OR

Efficiency and effectiveness evaluation after
two years use, a pilot study. Int J Comput
Assist Radiol Surg 2011 Mar;6(2):175-86.
PMID: 20661656.
14.
Pinkney N. Picture perfect. Installing an

integrated operating room. Health Facil
Manage 2009 Oct;22(10):25-8. PMID:
19862903.
15.
Integrating your OR: equipment and

construction needs. Health Devices 2008
Mar;37(3):76-89. PMID: 18771218.
16.
Acevedo AL. Construction of an integrated

surgical suite in a military treatment facility.
AORN J 2009 Jan;89(1):151-9. PMID:
19121420.
17.
Swart R. Planning for the rapidly emerging

digital OR. Can Oper Room Nurs J 2005
Mar;23(1):6, 8, 32-4. PMID: 15934572.
18.
ECRI Institute. Integrating your O.R. for

less. Six cost-saving tips that can save you
thousands. Health Devices 2009
Oct;38(10):333-4. PMID: 20853766.
168
Chapter 17. Use of Beta Blockers To Prevent Perioperative

Cardiac Events: Brief Update Review
Sumant R. Ranji, M.D.; Paul G. Shekelle, M.D., Ph.D.
Introduction
Myocardial infarction and cardiovascular death are the most common complications of major
non-cardiac surgery; thus, they have long been a focus of preoperative evaluations and a target of
perioperative management strategies. Based on strong evidence linking myocardial ischemia
with postoperative myocardial events and preliminary evidence that beta-blockade blunts
electrocardiographic signs of ischemia, clinical researchers in the late 1990s began examining
the effects of perioperative beta-blocker administration on patient outcomes. The 2001 report1
reviewed the evidence up to that point regarding the effectiveness, safety, and cost-effectiveness
of this intervention. Based on the results from several well-designed clinical trials, the authors
concluded that use of beta-blockers in the perioperative period was associated with significant
reductions in patient cardiac morbidity and mortality. However, as of publication of that report,
many questions remained regarding the optimal type of beta-blocker, the patients most likely to
benefit, and the safest and most effective dosing regimen.
What Have We Learned About the Use of Beta Blockers To Reduce

the Risk of Perioperative Cardiac Events?
Since the publication of Making Health Care Safer in 2001, new studies have called for a
re-examination of the initial enthusiasm for the use of beta blockers to reduce perioperative
cardiac events. Two systematic reviews with meta-analysis and one large randomized controlled
trial (RCT) have been influential. The first systematic review/meta-analysis was published in
2005 by Devereaux and colleagues.2 This review, which scored 10 out of 11 relevant AMSTAR
domains, included 22 trials encompassing 2,437 patients. The point estimates of effect favored
patients treated with beta blockers for nearly all outcomes, but the 95% confidence intervals for
these estimates were not statistically significant. The exception was the composite outcome of
major peri-operative cardiovascular events, which included cardiovascular death, non-fatal
myocardial infarction, and non-fatal cardiac arrest, where the results significantly favored
treatment (pooled relative risk 0.44, 95% CI 0.20 to 0.97). Conversely, pooled estimates of the
risk for three adverse effects (congestive heart failure, hypotension needing treatment, and
bradycardia needing treatment), all indicated the potential for harm, with pooled relative risks of
1.27 to 2.27, the latter being for bradycardia needing treatment and being statistically significant
(95% CI 1.53 to 3.36). This review concluded that the evidence supporting the use of beta
blockers in this situation was encouraging but too unreliable to allow definitive conclusions to
be drawn.
The large RCT was the POISE (perioperative ischemic evaluation) study, published in 2008.3
In this study, 8,351 patients 45 years of age or older who were undergoing non-cardiac surgery,
and had either known vascular disease or strong risk factors were randomized to receive 100 mg
of oral extended-release metoprolol 2 to 4 hours before surgery, followed by 200 mg every day
for 30 days (patients unable to take oral medications received the comparable dose
intravenously). The primary outcome was a composite of cardiovascular death, non-fatal
169
myocardial infarction, and non-fatal cardiac arrest (in other words, the exact composite outcome
with the statistically significant effect in the earlier meta-analysis). Indeed, at 30 days, patients
receiving metoprolol had a hazard ratio (HR) for the primary outcome of 0.84 (95% CI 0.70 to
0.99), due primarily to fewer myocardial infarctions. However, patients treated with metoprolol
had a statistically significantly greater risk of stroke (HR 2.17, 95% CI 1.26 to 3.74), and, even
more alarmingly, a greater risk of all-cause death (HR 1.33, 95% CI1.03 to 1.74). The authors of
POISE concluded that the perioperative use of beta-blockers has both benefits and risks. For
example they calculated that for every 1,000 patients undergoing noncardiac surgery, the use of
extended-release metoprolol would prevent 15 patients from having a myocardial infarction and
three from undergoing cardiac revascularization, but that there would be eight extra deaths and
five extra strokes. Based on these differences in benefits and harms, and on the potential for
patients to place different values on these outcomes, the authors of POISE concluded that authors
of current guidelines advocating the use of beta blockers should reconsider their
recommendations.
The later systematic review/meta-analysis was published in 2008, and included the POISE
results.4 This review, which scored 11 out of 11 relevant domains in AMSTAR, included 33
trials, now encompassing 12,306 patients. Recalling that POISE contributed more than 8,000
patients alone, in most of the pooled analyses the POISE results contribute 75 percent or greater
weight to the pooled result. Unsurprisingly, the meta-analysis found statistically significant
benefits for treatment for the outcomes of non-fatal myocardial infarction and myocardial
ischemia, nonsignificant results for all other potential benefits, and statistically significant
adverse effects for nonfatal stroke (pooled odds ratio[OR] of 2.16), perioperative bradycardia
requiring treatment (pooled OR of 2.74), and perioperative hypotension requiring treatment
(pooled OR 1.62). The effect on mortality was adverse, but did not reach statistical significance
(pooled OR 1.20, 95% CI 0.95 to 1.51). The authors of this review concluded that evidence
does not support the use of beta blocker therapy for the prevention of perioperative clinical
outcomes in patients having non-cardiac surgery.

Evidence that has emerged since the 2001 publication of Making Health Care Safer
indicates that perioperative beta blockers1 have mixed benefits and harms and should not be
considered a patient safety practice for all patients. An observational study of more than 600,000
patients suggests that perioperative beta blockers may have more benefit in high risk than in low
risk patients.5 An observational study of more than 600,000 patients who underwent major
noncardiac surgery, which did not did not find any evidence of benefit on in-hospital mortality
for perioperative beta blockade (adjusted odds ratio = 0.99), did find a suggestion of possible
benefit in the subgroup of patients at higher risk of death due to the presence of comorbidities
(diabetes, renal insufficiency, ischemic heart disease, cerebrovascular disease) or receipt of highrisk surgery (ref 5). If these suggestive findings are confirmed in subsequent randomized clinical
trials the use of peroperative beta blockers could yet be shown to have benefits exceeding risks
for certain subgroups of patients, but this question remains a topic for clinical research.
Moreover, randomized clinical trials may yet show this intervention to have benefits exceeding
risks for some subgroups of patients undergoing noncardiac surgery, but this question remains a
topic for clinical research. A summary table is below (Table 1).
170

Scope of the
Strength of
the PSP
Effectiveness
Common/High
Evidence or
Potential for
Harmful
Unintended
Consequences
High evidence High (death,
we know
stroke,
harms may
hypotension and
equal or
bradycardia
exceed
benefits
Estimate of
Cost
Low
References
1.
Auerbach A. Chapter 25. Beta-blockers and

Reduction of Perioperative Cardiac Events.
In Making Health Care Safer: A Critical
Quality. July 2001.
2.
Devereaux PJ, Beattie WS, Choi PT, et al.

How strong is the evidence for the use of
perioperative beta blockers in non-cardiac
surgery? Systematic review and metaanalysis of randomised controlled trials.
BMJ. 2005;331(7512):313-21. PMID:
15996966.
3.
Devereaux PJ, Yang H, Yusuf S, et al.

Effects of extended-release metoprolol
succinate in patients undergoing non-cardiac
surgery (POISE trial): a randomised
controlled trial. Lancet.
2008;371(9627):1839-47. PMID: 18479744.
4.
Bangalore S, Wetterslev J, Pranesh S, et al.

Perioperative beta blockers in patients
having non-cardiac surgery: a meta-analysis.
Lancet. 2008;372(9654):1962-76. PMID:
19012955.
5.
Lindenauer PK, Pekow P, Wang K, et al.

Perioperative beta-blocker therapy and
mortality after major noncardiac surgery. N
Engl J Med. 2005;353(4):349-61. PMID:
16049209.
171
How Much do We
N/A
Chapter 18. Use of Real-Time Ultrasound Guidance During

Central Line Insertion: Brief Update Review
Paul G. Shekelle, M.D., Ph.D.; Paul Dallas, M.D.
Introduction
Central venous catheters (CVCs) have multiple indications, including parenteral nutrition,
treatment of intravascular depletion, access for vasoactive medications, hemodynamic
monitoring, intravenous access during cardiopulmonary arrest, difficult peripheral intravenous
(IV) access, and long-term IV access for medications, such as antibiotics.1,2 Although these
catheters can be life saving, they are also associated with significant risk.3 This risk is heightened
by a number of factors, including patient characteristics (e.g., morbid obesity, cachexia, or local
scarring from surgery or radiation treatment), patient setting (e.g., patients receiving mechanical
ventilation or during emergencies such as cardiac arrest), co-morbidities (e.g., bullous
emphysema or coagulopathy), the variable training and experience of the clinicians who perform
the procedure, and the method of insertion (e.g., percutaneous insertions are often performed
blind and rely on anatomic landmarks).3-5 However, protocols have been developed that use
portable ultrasound (US) devices to provide bedside imaging of the central veins during catheter
placement. The advantages associated with US-guided CVC placement include detection of
anatomic variations and exact vessel location (for example, the carotid artery is anterior to the
internal jugular vein in 3% to 9% of patients6), avoidance of central veins with pre-existing
thrombosis that may prevent successful CVC placement, and guidance of both guidewire and
catheter placement after initial needle insertion.
The original report included a review of the efficacy, safety, and cost-effectiveness of realtime US guidance on the safety of CVC insertions. This review found that, in general, US
improves the success rates and reduces the risks of CVC placement, particularly for
inexperienced clinicians and for patients in high-risk situations. The purpose of the present report
is to provide an update on the impact of US CVC insertion. We used the articles cited as
evidence in the 2001 report to create a list of search terms and then used these terms to conduct
an update search.
What Is the Practice of Using Ultrasound Guidance for Central

Venous Catheter Insertion?
As a patient safety practice, utilizing portable two-dimensional ultrasonography to guide the
insertion of CVCs (internal jugular, subclavian or femoral) can take one of two forms the
static approach, whereby a mark is placed on the skin to indicate where to insert the needle, or
the real time approach, where the needle insertion is visualized during the procedure. The
alternative to using US guidance is the landmark approach, whereby anatomic landmarks are
used to determine, to the extent possible, where the underlying vein is located. A recent 18minute video (and accompanying text7) demonstrates the use of US guidance for internal jugular
vein catheterization (www.nejm.org/doi/full/10.1056/NEJMvcm0810156#figure=preview.jpg).
172
How Has the Use of Ultrasound To Guide Central Venous Catheter

Insertion Been Implemented?
Our search identified two surveys of the use of US for CVC. In 2006, an Internet survey was
sent to members of the Society of Cardiovascular Anesthesiology; 1,494 responses were received
from 4,235 members (35%). Of these respondents, 37 percent stated they never used US for
CVC insertion, and another 30 percent almost never used it. Only 15 percent always or
almost always used US guidance.8 A survey from the United Kingdom (U.K.) asked 2000
senior members of the Association of Anesthetists of Great Britain and Ireland about their use of
US guidance; 1,455 replied, for a response rate of 73%). Of the respondents, 93 percent regularly
inserted internal jugular venous catheters as part of their practice, and 27 percent of respondents
indicated that the use of US was their first choice as a technique (50% of respondents used
surface landmarks and 30% used palpation/balloting; some respondents indicated more than
one first choice).9
Educating clinicians on the use of US for central line placement has received relatively little
attention. Studies have shown that clinical US guidance skills are improved by implementing
simulator-based training (see Chapter 38). Although several medical schools offer training in
portable ultrasonography, scant information exists on teaching US guided (USG) central line
placement to medical students.10-12 Particular specialties mandate portable US training for
residents, including procedural skills like USG central line placement, whereas others have just
begun to explore the benefits of portable US in their graduate medical education programs. In
emergency medicine residency training for instance, the first US curriculum was published in
1994.13 While no clear consensus exists regarding the need for training in USG central line
placement in emergency medicine residencies, a novel training program consisting of a brief
web-based instructional module and a practical session was effective in enhancing emergency
resident competency in USG central line placement.14 Carilion Clinic trains physicians in the use
of US using a curriculum consisting of 16 hours of didactic and hands on experience during the
first month of residency; this training covers physics, knobology (e.g., what all the knobs on
the machine are for), echocardiography, abdominal US, vascular US, and includes 2.5 hours of
procedural skills, of which USG central line placement is prominent. Physicians who are
experienced in the procedure use special models to conduct the hands-on portion of this
curriculum for groups of four to five trainees. Currently, skills assessment is done by
observation, although a competency and performance checklist is being developed. With respect
to continuing medical education, medical schools, clinics, and medical education companies
sponsor a range of activities. These activities cover hands-on USG central line placement as part
of multiday courses that concentrate on U.S. education.
What Have We Learned About the Use of Ultrasound Guidance for

Central Venous Catheter Insertion?
The most relevant meta-analysis identified was published by Hind and colleagues in 2003,
and was commissioned by the U.K. National Institute for Clinical Excellence.15 These authors
identified 18 eligible randomized trials that compared either two-dimensional US or Doppler US
with either the landmark method or the cut-down method (whereby an incision is made to
directly visualize the vein) and that measured any one of five relevant outcomes. Data for adults
and children were pooled separately, and data from 2D and from Doppler studies were also
pooled separately. For all five relevant outcomes (failed catheter placement, complication with
173
placement, failure on the first attempt, mean number of attempts to successful catheterization,
and seconds to successful catheterization), two-dimensional US had statistically significantly
better outcomes than the landmark method for internal jugular vein catheterization in adults.
More limited data in children and for subclavian and for femoral vein insertion favored the use of
two-dimensional US. Pooled results from studies of Doppler US also favored its use. No studies
directly compared two-dimensional and Doppler US. The authors made an indirect comparison
by assessing the size of the pooled effects for each compared with the landmark method. This
analysis favored the use of two-dimensional US. This review scored nine of 11 relevant
AMSTAR criteria. A companion cost-effectiveness analysis estimated the marginal cost (in
2002) for use of US in CVC to be about 10 pounds sterling (approximately $16) per procedure,
assuming the machine was used for 15 procedures each week. The base case scenario estimated
that for every 1000 patients, 90 complications would be avoided, with a net cost saving of about
2000 pounds sterling (approximately $3200).16
Since that time, randomized trials in adults have consistently supported the conclusions about
effectiveness, including patients treated in the Emergency Department,17 ventilated patients,18
critical care patients,19,20 and patients in other miscellaneous clinical settings.21,22 A new
outcomecentral venous catheter-associated bloodstream infectionhas been assessed and
found to be statistically significantly lower in one trial of US-guided catheter insertion compared
with landmark methods.19
A more recent meta-analysis included five studies that focused only on children, most of
whom were cardiac surgery patients. Although pooled point estimates favored the use of US, the
95% confidence intervals were wide and none of the results were statistically significant.23 Two
trials published since that meta-analysis, one of which compared real time to static US, both
found that two-dimensional real-time US improved some outcomes.24,25
Recent trials of US have focused less on its use in adult internal jugular vein catheterization
and more on its use in other locations and refinements of the technique, including the insertion of
hemodialysis catheters,26 the radial artery,27-29 the femoral artery,30 and even peripheral venous
catheters in difficult patients.31-36 In general, studies reported that US guidance improved
outcomes compared with techniques without US guidance. Systematic reviews of the use of US
guidance for hemodialysis catheter insertion37 and radial artery catheters38 each concluded that
the use of real-time two-dimensional US improved outcomes.
Clearly, USG central line placement education varies in undergraduate, graduate, and
continuing medical education. While educators at all levels are making inroads, greater
consistency is needed in curricula, evaluation of outcomes, and guideline development.

In 2001, Making Health Care Safer concluded that the use of US guidance for the
placement of CVCs is one of the patient safety practices with the strongest evidence. Since that
time, new evidence continues to support and strengthen this conclusion. Simulator-based training
can improve implementation of this patient safety practice. Emerging evidence suggests that
two-dimensional real-time US guidance may also be beneficial for other kinds of catheter
insertions. A summary table is located below (Table 1).
174

Scope of the
Strength of
the PSP
Effectiveness
Common/Lowtomoderate
High
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Low-tomoderate
How Much do We
A lot/Moderate
References
1.
Civetta JM, Taylor RW, Kirby RR. Critical

Care. 3rd ed. Philadelphia: LippincottRaven; 1997.
2.
3.
10.
Shoemaker W, Ayres S, Grenvik A, et al.

Textbook of Critical Care. 4th ed.
Philadelphia: WB Saunders; 2000.
Fernandez-Frackelton M, Peterson M, Lewis

RJ, et al. A bedside ultrasound curriculum
for medical students: prospective evaluation
of skill acquisition. Teach Learn Med. 2007
Winter;19(1):14-9. PMID: 17330994.
11.
Mansfield PF, Hohn DC, Fornage BD, et al.

Complications and failures of subclavianvein catheterization. N Engl J Med. 1994
Dec 29;331(26):1735-8. PMID: 7984193.
Arger PH, Schultz SM, Sehgal CM, et al.

Teaching medical students diagnostic
sonography. J Ultrasound Med. 2005
Oct;24(10):1365-9. PMID: 16179619.
12.
Rao S, van Holsbeeck L, Musial JL, et al. A

pilot study of comprehensive ultrasound
education at the Wayne State University
School of Medicine: a pioneer year review. J
Ultrasound Med. 2008 May;27(5):745-9.
PMID: 18424650.
13.
Mateer J, Plummer D, Heller M, et al.

Model curriculum for physician training in
emergency ultrasonography. Ann Emerg
Med. 1994 Jan;23(1):95-102. PMID:
8273966.
14.
Lee AC, Thompson C, Frank J, et al.

Effectiveness of a novel training program
for emergency medicine residents in
ultrasound-guided insertion of central
venous catheters. CJEM. 2009
Jul;11(4):343-8. PMID: 19594973.
15.
Hind D, Calvert N, McWilliams R, et al.

Ultrasonic locating devices for central
venous cannulation: meta-analysis. BMJ.
2003 Aug 16;327(7411):361. PMID:
12919984.
16.
Calvert N, Hind D, McWilliams R, et al.

Ultrasound for central venous cannulation:
economic evaluation of cost-effectiveness.
Anaesthesia. 2004 Nov;59(11):1116-20.
PMID: 15479322.
4.
Sznajder JI, Zveibil FR, Bitterman H, et al.

Central vein catheterization. Failure and
complication rates by three percutaneous
approaches. Arch Intern Med. 1986
Feb;146(2):259-61. PMID: 3947185.
5.
Bernard RW, Stahl WM. Subclavian vein

catheterizations: a prospective study. I. Noninfectious complications. Ann Surg. 1971
Feb;173(2):184-90. PMID: 5100094.
6.
Maecken T, Grau T. Ultrasound imaging in

vascular access. Crit Care Med. 2007
May;35(5 Suppl):S178-85. PMID:
17446777.
7.
Ortega R, Song M, Hansen CJ, et al. Videos

in clinical medicine. Ultrasound-guided
internal jugular vein cannulation. N Engl J
Med. 2010 Apr 22;362(16):e57. PMID:
20410510.
8.
9.
Bailey PL, Glance LG, Eaton MP, et al. A

survey of the use of ultrasound during
central venous catheterization. Anesth
Analg. 2007 Mar;104(3):491-7. PMID:
17312193.
McGrattan T, Duffty J, Green JS, et al. A
survey of the use of ultrasound guidance in
internal jugular venous cannulation.
Anaesthesia. 2008 Nov;63(11):1222-5.
PMID 19032257.
175
17.
Leung J, Duffy M, Finckh A. Real-time

ultrasonographically-guided internal jugular
vein catheterization in the emergency
department increases success rates and
reduces complications: a randomized,
prospective study. Ann Emerg Med. 2006
Nov;48(5):540-7. PMID: 17052555.
25.
Aouad MT, Kanazi GE, Abdallah FW, et al.

Femoral vein cannulation performed by
residents: a comparison between ultrasoundguided and landmark technique in infants
and children undergoing cardiac surgery.
Anesth Analg. 2010 Sep;111(3):724-8.
PMID: 20601450.
18.
Hayashi H, Amano M. Does ultrasound

imaging before puncture facilitate internal
jugular vein cannulation? Prospective
randomized comparison with landmarkguided puncture in ventilated patients. J
Cardiothorac Vasc Anesth. 2002
Oct;16(5):572-5. PMID: 12407608.
26.
Bansal R, Agarwal SK, Tiwari SC, et al. A

prospective randomized study to compare
ultrasound-guided with nonultrasoundguided double lumen internal jugular
catheter insertion as a temporary
hemodialysis access. Ren Fail.
2005;27(5):561-4. PMID: 16152994.
19.
Karakitsos D, Labropoulos N, De Groot E,

et al. Real-time ultrasound-guided
catheterisation of the internal jugular vein: a
prospective comparison with the landmark
technique in critical care patients. Crit Care.
2006;10(6):R162. PMID: 17112371.
27.
Levin PD, Sheinin O, Gozal Y. Use of

ultrasound guidance in the insertion of radial
artery catheters. Crit Care Med. 2003
Feb;31(2):481-4. PMID: 12576955.
28.
Ganesh A, Kaye R, Cahill AM, et al.

Evaluation of ultrasound-guided radial
artery cannulation in children. Pediatr Crit
Care Med. 2009 Jan;10(1):45-8. PMID:
19057451.
29.
Shiver S, Blaivas M, Lyon M. A prospective

comparison of ultrasound-guided and
blindly placed radial arterial catheters. Acad
Emerg Med. 2006 Dec;13(12):1275-9.
PMID: 17079789.
30.
Dudeck O, Teichgraeber U, Podrabsky P, et

al. A randomized trial assessing the value of
ultrasound-guided puncture of the femoral
artery for interventional investigations. Int J
Cardiovasc Imaging. 2004 Oct;20(5):363-8.
PMID: 15765858.
31.
Doniger SJ, Ishimine P, Fox JC, et al.

Randomized controlled trial of ultrasoundguided peripheral intravenous catheter
placement versus traditional techniques in
difficult-access pediatric patients. Pediatr
Emerg Care. 2009 Mar;25(3):154-9. PMID:
19262420.
32.
Bair AE, Rose JS, Vance CW, et al.

Ultrasound-assisted peripheral venous
access in young children: a randomized
controlled trial and pilot feasibility study.
West J Emerg Med. 2008 Nov;9(4):219-24.
PMID: 19561750.
33.
Aponte H, Acosta S, Rigamonti D, et al. The

use of ultrasound for placement of
intravenous catheters. AANA J. 2007
Jun;75(3):212-6. PMID: 17591303.
20.
21.
22.
23.
24.
Fragou M, Gravvanis A, Dimitriou V, et al.

Real-time ultrasound-guided subclavian vein
cannulation versus the landmark method in
critical care patients: a prospective
randomized study. Crit Care Med. 2011
Jul;39(7):1607-12. PMID: 21494105.
Turker G, Kaya FN, Gurbet A, et al. Internal
jugular vein cannulation: an ultrasoundguided technique versus a landmark-guided
technique. Clinics (Sao Paulo).
2009;64(10):989-92. PMID: 19841706.
Milling TJ, Jr., Rose J, Briggs WM, et al.
Randomized, controlled clinical trial of
point-of-care limited ultrasonography
assistance of central venous cannulation: the
Third Sonography Outcomes Assessment
Program (SOAP-3) Trial. Crit Care Med.
2005 Aug;33(8):1764-9. PMID: 16096454.
Sigaut S, Skhiri A, Stany I, et al. Ultrasound
guided internal jugular vein access in
children and infant: a meta-analysis of
published studies. Paediatr Anaesth. 2009
Dec;19(12):1199-206. PMID: 19863734.
Hosokawa K, Shime N, Kato Y, et al. A
randomized trial of ultrasound image-based
skin surface marking versus real-time
ultrasound-guided internal jugular vein
catheterization in infants. Anesthesiology.
2007 Nov;107(5):720-4. PMID: 18073546.
176
34.
Costantino TG, Parikh AK, Satz WA, et al.

Ultrasonography-guided peripheral
intravenous access versus traditional
approaches in patients with difficult
intravenous access. Ann Emerg Med. 2005
Nov;46(5):456-61. PMID: 16271677.
35.
Stein J, George B, River G, et al.

Ultrasonographically Guided Peripheral
Intravenous Cannulation in Emergency
Department Patients With Difficult
Intravenous Access: A Randomized Trial.
Ann Emerg Med. 2009 Jul;54(1):33-40.
PMID: WOS:000267853500008.
177
36.
Mahler SA, Wang H, Lester C, et al. Shortvs long-axis approach to ultrasound-guided

peripheral intravenous access: A prospective
randomized study. Am J Emerg Med.
2011;29(9):1194-7.
37.
Rabindranath KS, Kumar E, Shail R, et al.

Ultrasound use for the placement of
haemodialysis catheters. Cochrane Database
Syst Rev. 2011;11:CD005279. PMID:
22071820.
38.
Shiloh AL, Savel RH, Paulin LM, et al.

Ultrasound-guided catheterization of the
radial artery: a systematic review and metaanalysis of randomized controlled trials.
Chest. 2011 Mar;139(3):524-9. PMID:
20724734.
Section D. Safety Practices Aimed Primarily at

Hospitalized Elders
Chapter 19. Preventing In-Facility Falls
Isomi M. Miake-Lye, B.A.; Susanne Hempel, Ph.D.; David A. Ganz, M.D., Ph.D.; Paul G.
Shekelle, M.D., Ph.D.

The rate of falls in acute-care hospitals is estimated to range from 1.3 to 8.9 per 1,000 beddays,1 which translates into well over 1000 falls per year in a large facility. Higher rates are
reported in particular sites or wards, such as those specializing in neurology, geriatrics, and
rehabilitation. Because falls are believed to be underreported, most estimates are assumed to be
overly conservative.1 However defining what is a fall is itself a challenge, as there is
variability in the research literature and among older adults about what constitutes a fall.2,3
Authoritative bodies have definitions (e.g., the NQF defines a fall as an unplanned descent to
the floor without injury4 and WHO defines a fall as an event which results in a person coming
to rest inadvertently on the ground or floor or some lower level5)but even after accepting a
conceptual definition of a fall, there is a difference between any fall, a fall with injury, the
proportion of a population who has a fall, and the number of falls. Nevertheless, there is
widespread agreement that falls, however defined, occur frequently and can have serious
physical and psychological consequences. Between 30 percent and 50 percent of in-facility falls
are associated with reports of injuries. Hip fractures occur in 1 percent to 2 percent of falls.
Inpatient falls are also associated with increased health care utilization, including increased
length of stay and higher rates of discharge from hospitals into institutional or long-term care
facilities. In one recent analysis in three hospitals in Missouri, operational costs for patients who
have fallen with serious injuries were $13,000 higher than for control patients without falls, and
patients who have fallen had an additional 6.3 days length of stay.6 Even falls that do not cause
severe injuries can trigger a fear of falling, anxiety, distress, depression, and reduced physical
activity. Family members, caregivers, and health care professionals are also susceptible to overly
protective or emotional reactions to falls, which can also impact the patients independence and
rehabilitation.

Most in-facility fall prevention programs are multicomponent interventions. Unfortunately,
the individual components vary across each published evaluation, with the same combination of
components never being evaluated in more than one application. Therefore, in terms of
identifying and reviewing the evidence for fall prevention interventions, the best that can be done
is to describe the components most commonly included in interventions that have been
evaluated. The Prevention of Falls Network Europe (ProFaNE ) proposed a detailed
classification of fall risk assessment components (see Appendix C for the complete list),7 which
map closely to the descriptions provided in this chapter. According to a review by Oliver and
colleagues, the following were the most common components of successful interventions:
178
Post fall review: to assess potential reasons for a specific instance of a fall and to
remediate possible contributing factors
Patient education
Staff education
Footwear advice
Scheduled and supervised toileting
Medication review: to assess for use of medication(s) that can affect mental alertness and
balance (see ProFANE taxonomy for further details, Appendix C).
The most recent Cochrane review notes a striking variability in type, targeting, intensity,
and duration within the fall prevention programs and does not attempt to draw conclusions
about which components might be most effective.8 Table 1 lists all the studies in the reviews by
Cochrane and by Oliver, as well as new studies from our update search, and the components
included in the intervention.
All multicomponent interventions also included an assessment of falls risk. In about 60
percent of studies this was a formal falls risk assessment tool such as the Morse Fall Scale or
STRATIFY, and the remainder used informal or idiosyncratic or unstated methods for assessing
patients at increased risk of falls.
Other single intervention components include use or removal of bedrails, use of physical
restraints, movement alarm devices, low-low beds (beds closer to the floor), exercise or
additional physical therapy, increased observation or assistance, calcium or vitamin D, hip
protectors, and prevention of delirium (this last topic is covered in Chapter 20). Since most
reviews conclude that multi-component interventions are more effective than single components,
in this chapter we will consider only multi-component interventions. Multicomponent
interventions are also referred to in the literature as multifaceted or multifactorial interventions.
Although some authors draw distinctions between these labels, we will not do so here, and refer
to all of them as multicomponent.
179
Ang et al, 2011 9*
22%
Koh et al, 200919
Krauss et al, 200820

Oliver et al, 200221
()
()
?
Uden et al, 199924
Postfall
Review
Movement
Alarms
Medication
Review
Urine
Screening
Exercise
Toileting
Schedules
Footwear
Schwendimann et
al, 200622
Stenvall et al,
200723
Grenier-Sennelier
et al, 200216
Haines et al, 200417
Healey et al, 200418
Van der Helm et al,

200625
Cumming et al,
200813
Dykes et al, 201014*
Fonda et al, 200615
Other Interventions
Barker et al, 200910

Barry et al, 200111
Brandis, 199912
Staff
Education
Patient
Education
Bedrail
Review
Vest/Belt/Cuff
Restraint
Hip
Protectors
Alert
Wristband
Bedside Risk
Sign
References
Environment
Modified
Table 1, Chapter 19. Components of multi-factorial falls prevention trials in hospitals, 1999 to 2009
180
Low beds; interventions specific to each risk factor in model used. Used
Hendrich II Falls Risk Model
Low beds; Introduction of a computerized falls reporting and analysis
c
system Used STRATIFY falls risk assessment tool
Risk Factors assessed
Falls history and continence assessment added to standard admission
documentation / Unstated method of risk assessment
Modification of tool developed the Centre for Education and Research on
Ageing in Sydney, Australia
Tailored plan of care; computerized Fall Prevention Tool Kit (FPTK)
Used Morse Fall Scale
Low beds, volunteer observers
Used Falls Risk Assessment Scoring System
Improved assessment of mobility and self-efficacy
Unspecified method for assessing risk
Used the Peter James Centre falls risk assessment tool
Vision testing, lying and standing blood pressure
Brief falls risk factor screen
Stand by me notices to prompt staff to wait outside toilets ready to assist.
Mobility level signs at bedside
Unstated method of risk assessment
Used Morse Falls Scale
Nursing and medical checklist for remediable risk factors, content not
described and compliance poor
Used STRATIFY falls risk assessment tool
Briefly screened for falls risk using 3 items
Additional therapy and nurse staffing Routine dietary protein
supplementation Protocol driven delirium screening
No clear risk assessment instrument, but population can be assumed to all
be at elevated risk
Career education
A new formal risk assessment instrument created for the study
Identification of high risk patients on the basis of a recent fall or 4 other
criteria
Postfall
Review
Movement
Alarms
Medication
Review
Urine
Screening
Exercise
Toileting
Schedules
Other Interventions
Footwear
Staff
Education
Patient
Education
Bedrail
Review
Vest/Belt/Cuff
Restraint
Hip
Protectors
Alert
Wristband
Bedside Risk
Sign
References
Environment
Modified
Table 1, Chapter 19. Components of multi-factorial falls prevention trials in hospitals, 1999 to 2009 (continued)
Vassallo et al,
Medical Review/Used Downton fall risk assessment

200426
Von Renteln-Kruse
0.5%
Bedside commodes
and Krause, 200727
Used STRATIFY falls risk assessment tool
Table adapted from Oliver 1
* New studies added from update search
yes = component included within the intervention; (yes) = component planned but not implemented; ? = component implied but not explicit; = intervention discouraged use of this component;
= intervention encouraged use of this component.
a
(yes) indicates intervention in design but not applied in practice (e.g., environmental hazards identified but not addressed). ? indicates that the article implies, but does not specify, that an intervention
was included. For bedrails and body restraints, indicates the intervention was to discourage their use, indicates the intervention aimed to encourage their use, while yes indicates either direction
not described or a neutral risk versus benefit review was required.
b
Where interventions are described that would be considered very standard practice for control as well as intervention (e.g., call bell left in reach, walking aids provided as appropriate), these are not
listed.
c
This potentially confounded the findings as this changed the method of collecting outcome data on falls at the same time as the intervention was introduced.
Reprinted from Clin Geriatr Med. 26(4), Oliver D, Healey F, Haines TP., Preventing falls and fall-related injuries in hospitals, 645-92, 2009 with permission from Elsevier
181

None of the controlled trials of fall prevention programs explicitly articulate the conceptual
framework for their intervention. However, underlying each is the stated or implied
understanding that falls have a multifactorial etiology and that attention to multiple risk factors
will be more effective than an intervention that targets any single risk factor. A fall is usually the
result of interactions between patient-specific risk factors and the physical environment. Patientspecific risk factors include patient age (particularly age over 85, sometimes called the oldest
old), male sex, a history of a recent fall, muscle weakness, behavioral disturbance, urinary
incontinence or frequency, certain medications, and postural hypotension or syncope.
Environmental causes include poor lighting; trip hazards (such as uneven flooring or small
objects on floor); suboptimal chair heights; and staff availability, attitude, and skills. Given the
multifactorial nature of falls, a patient safety practice designed to assess and remediate multiple
factors is believed to be more likely to be effective. Indeed, the list of successful components in
multi-component fall prevention interventions matches well with this list of patient and
environmental contributors to falls. We identified one published logic model for why individual
fall prevention components should work (Figure 1). For example, a bed alarm detects patient
movements, which can allow a faster response to patients and reduce falls. Similarly, use of a
visible sign or identification bracelet increases awareness of falls and at-risk patients and inform
necessary responses, which in turn should reduce falls.
The second underlying assumption of most fall prevention programs in the published
literature is that fall risk assessment is primarily a nursing function, but that insufficient attention
is currently paid to this task due to other demands for nursing time, and that some method of
reminder, checklist, or similar tool can be effective to ensure the assessment of fall risk.
182
Figure 1, Chapter 19. Multi-systemic fall prevention model
(a) Firm mattresses; low beds; appropriate chair heights and depths for easy transfer; chairs with arm rests; and secured handrails throughout the movement of a patient. (b) Nonslip surfaces in floors/bathtubs; shower seats; grab bars next to the toilet/bathtub; toilet seats that allow easy transfer; door magnets that hold doors in the open position; and arm
rests next to the toilet.
*An intervention or a factor whose efficacy was NOT tested as a single factor in any healthcare setting. **An intervention or a factor whose efficacy was tested as a single factor in
other healthcare settings but NOT specifically in a hospital setting. ***An intervention or factor whose efficacy was tested in a hospital setting.
Figure taken from Choi et al, 201128
Choi YS, Lawler E, Boenecke CA, et al. Developing a multi-systemic fall prevention model, incorporating the physical environment, the care process and technology: a systematic
review. J Adv Nurs. 2011. Permission granted by John Wiley & Sons, Inc.
183

The primary sources of evidence about multi-component in-facility fall prevention programs
are three systematic reviews: a 2008 review from the Cochrane Collaboration by Cameron and
colleagues,8 a review by Coussement and colleagues also published in 2008,29 and a review by
Oliver and colleagues originally published in 2006,30 which was updated in 2010 as a narrative
review.1 All three reviews scored well on the AMSTAR criteria for systematic reviews (11/11,
10/11, and 10/11 respectively).31 The Cochrane review searched a number of databases through
November 2008 for randomized trials to assess the effectiveness of falls reduction interventions
for older adults in nursing care facilities and hospitals.8 Of the 41 trials they included, 11 were
conducted in hospital settings, of which four addressed multifactorial interventions. The review
by Coussement identified four studies, three of which were included in the Cochrane review.29
The Oliver and colleagues review also searched multiple databases for relevant literature through
January 2005.30 This reviews objective was to evaluate the evidence for fall prevention
strategies in care homes and hospitals, with an additional focus on the effect of dementia and
cognitive impairment on fall risk. Broader inclusion standards than the Cochrane review led to
the inclusion of 43 trials, case-control studies, and observational cohort studies. Thirteen of these
studies addressed multicomponent inpatient interventions. The updated narrative review focused
directly on inpatient fall prevention and discussed 17 multifactorial studies spanning 1999-2009,
which include the four trials found by the Cochrane group.1
The three reviews reached similar conclusions. The Oliver and Cochrane reviews found that
multi-component in-facility fall prevention programs result in statistically and clinically
significant reductions in rates of falls (see Table 2). The Cochrane pooled analysis of four fall
prevention programs in 6,478 participants found a 31 percent decrease in the rate of falling
(pooled rate ratio [RR]0.69 (95% CI, 0.49 to 0.96) and a 27 percent decrease in the incidence of
falls among three trials involving 4,824 participants (RR 0.73; 95% CI, 0.56 to 0.96).8 The
Coussement review found a similar pooled rate ratio as the Oliver review; however, this effect
was not quite statistically significant.29 Principal results from the Oliver meta-analysis are
reproduced below (see Figure 2).30 The other systematic reviews and meta-analyses identified in
the Oliver update review were surprisingly consistent (p. 679) and support the argument that
multi-factorial interventions reduce fall rates more effectively does than any single intervention
in acute care settings.1
Table 2, Chapter 19. Meta-analytic estimate of the effect of multicomponent fall intervention
programs on inpatient fall rates
Meta-Analysis (First Author)
8
Cameron, 2010
29
Coussement, 2008
30
Oliver, 2007
Number of Included Studies

4
4
12
184
Pooled Rate Ratio

0.69 (95% CI 0.49 0.96)
0.82 (95% CI 0.65 - 1.03)
0.82 (95% CI 0.68 1.00)
Figure 2, Chapter 19. Meta-analysis from Oliver et al. 2006 for multifaceted interventions in
30
hospital falls (random effects model)
Reproduced from Strategies to prevent falls and fractures in hospitals and care homes and effect of cognitive impairment:
systematic review and meta-analyses. Oliver D, Connelly JB, Victor CR, et al. 334(7584):82. 2007 with permission from BMJ
Publishing Group Ltd.
The Cochrane and Oliver reviews were supplemented with an update search (described
below) and an additional search by Hempel and colleagues (discussed in more detail later),
which addressed the prevention of inpatient falls. After using 15 existing reviews and reports to
identify pertinent sources, which included the two reviews in this chapter, Hempel then searched
multiple databases for relevant literature. The search covered January 2005 to August 2011 and
included randomized controlled trials, non-randomized trials, and before-after studies in Englishlanguage publications that addressed falls in the hospital setting. Details of the search strategy
are in Appendix C.
In the update search, we focused on studies with large sample sizes (at least N=1,000), that
assessed multi-component interventions in acute-care hospitals, in the general population or
older adult population. We were looking for pivotal studies, as defined by Shojania and
colleagues (see Methods, Chapter 2 p.ES-4) that could provide a signal when an existing
systematic review is out of date.32 We identified two new relevant studies, both of which showed
statistically significant improvements in intervention groups when compared with controls, and
which we discuss briefly here. A third study is reviewed because of its unique design. Data for
all studies included in the Oliver review, the Cochrane review, and our update search are in an
evidence table in Appendix D. Table 3 provides an abbreviated description of each study.
185
Table 3, Chapter 19. Abridged evidence tables, adapted from Oliver and colleagues
Author, year
Ang et al, 2011
9
*
Barker et al,
200910
Barry et al,
200111
Brandis, 199912
Study
design
RCT
Before/After
Before/After
Before/After
Cumming et al,
200813
Dykes et al,
2010 14*
Cluster RCT
Fonda et al,
200615
GrenierSennelier et al,
200216
Haines et al,
200417
Before/After
Cluster RCT
Before/After
RCT
Healey et al,
200418
Koh et al,
200919
Krauss et al,
200820
Oliver et al,
200221
Schwendimann
et al, 200622
Cluster RCT
Stenvall et al,
200723
RCT
Uden et al,
199924
Van der Helm
et al, 200625
Before/After
Cluster RCT
Before/After
Before/After
Before/After
Before/After
Outcomes+
1822 patients.
Quality
Score**
25
271,095 patients
16
SFI
All patients admitted to 95

beds for 3 years
All patients admitted to 500
beds for 2 years
3999 patients
15
SFI
11
NFF
27
NFF
All patients admitted or

transferred to units over 6
month study period
3961 patients
27
SFF
20
SFF
All admitted patients over 4

years
11
SFF
626 patients
26
SFF
3386 patients
26
NFF
All admissions during 1.5

years
All admissions over 18
months
3200 patients admitted
annually; data over 2 years
34,972 admissions
14
NFF
18
NFF
NGF
15
NFF
199
25
SFF
379 patients
12
NGF
2670 patients
11
NGF
Setting
Participants
8 medical wards; acute

care; Singapore
Small; acute care;
Australia
Small; long-stay and
rehab; Ireland
Acute, Australia
24 wards; acute and
rehab; Australia
8 units; medical; urban
U.S.
4 wards; elderly acute
and rehab; Australia
400 bed; rehab; France
3 wards; subacute
rehab and elderly;
Australia
8 wards; acute and
rehab; 3 hospitals; UK
2 hospitals; acute;
Singapore
General medicine;
acute academic hospital
Elderly medical unit;
acute hospital; UK
300 bed; internal
medicine, geriatric and
surgical; Switzerland
3 wards; orthogeriatric,
geriatric, orthopedic;
Sweden
Geriatric dept; acute
hospital; Sweden
Internal med ward and
neurology ward; acute
hospital; Netherlands
3 wards; rehab; UK
SFF
Vassallo et al,
Cohort Study
825 patients
25
NFF
200426
Von RentelnBefore/After
Elderly acute and rehab 7254 patients
17
SFF
Kruse et al,
wards; Germany
200727
*New studies added from update search
** Downs and Black Quality Score,33 evaluated by the authors
+
SFF= significantly fewer falls; SFI=significantly fewer injuries; NFF= nonsignificantly fewer falls; NGF= nonsignificantly
greater falls
Reprinted from Clin Geriatr Med. 26(4), Oliver D, Healey F, Haines TP., Preventing falls and fall-related injuries in hospitals,
645-92, 2009 with permission from Elsevier.
186
Dykes and colleagues compared the fall rates of four intervention units to matched control
units in four urban United States hospitals over a 6-month period.14 Control units received usual
care, which included fall risk assessments, signage for high-risk patients, patient education as
needed, and manual documentation in patient records. The intervention group tested the Fall
Prevention Tool Kit (FPTK), which was developed by the study team. The FPTK is a health
information technology application that includes a risk assessment and tailored signage, patient
education, and plan of care components. The FPTK is integrated with, and seeks to enhance,
existing workflow and communication patterns. Adjusted fall rates in the intervention units (3.15
per 1,000 patient days [95% CI, 2.54 to 3.90]) were significantly lower than in control units
(4.18 per 1,000 patient days [95% CI, 3.45 to 5.06]), with a particularly strong impact among
patients aged 65 or older (rate difference of 2.08 per 1,000 patient days [95% CI: 0.61 to 3.56]).
This study was judged to have a low risk of bias using the criteria of the Effective Practice and
Organizational Organisation of Care (EPOC) Cochrane Group (score of 8 of 9 components). 34
In the second study, Ang and colleagues 9 randomized patients in eight medical wards of an
acute-care hospital in Singapore over a 9-month interval. They used an assessment tool to match
high-risk patients with appropriate interventions, in addition to a tailored educational session, in
the intervention group. Both the intervention and control groups in this study received usual care,
which included environmental modifications, review of medications and fall history, and
educational sessions. The proportion of patients with at least one fall in the intervention group
was 0.4 percent (95% CI, 0.2 to 1.1) while in the control group this was 1.5 percent (95% CI, 0.9
to 2.6) for a relative risk reduction of 0.29 (95% CI, 0.1 to 0.87). Using the EPOC criteria, this
study was judged to be at low risk of bias (score of 8 of 9 components).34
One additional study was identified and is noted here because of its unique design. The study
by van Gaal and colleagues evaluated a program that targeted three patient safety practices
(pressure ulcers, urinary tract infections, and falls prevention) simultaneously and found an
overall positive effect on the development of any adverse event, a composite measure of pressure
ulcers, urinary tract infections, and falls.35,36 The study was not powered to assess falls
separately, yet it is worth noting that the point estimate for the relative risk reduction in falls was
0.69, which is within the range of results reported in other studies and meta-analyses. The value
of this study is the demonstration of simultaneous improvements in several intervention targets.
Thus, new large controlled trials continue to support the conclusion of existing meta-analyses
that multifactorial falls prevention programs are effective in reducing inpatient fall rates.

Most trials of fall prevention programs have not reported any harms. The Cochrane review
reported none.8 It is not clear whether the possibility of harms was explicitly assessed in these
trials. However, concern exists that some falls prevention interventions may lead to harms. The
review by Oliver and colleagues detailed a number of potential harms, including an increased use
of restraints or sedating medications. However, Oliver and colleagues also note so little empiric
evidence on adverse effects of fall prevention activities on other clinical activities has been
incorporated into clinical trials that one has very little with which to substantiate or refute these
concerns.1
187

What Contexts?
The ways in which falls prevention programs have been implemented and a description of
contexts are lacking in most reports. The limited evidence available is summarized below.
Structural Organizational Characteristics

Fall prevention programs have been implemented in both acute-care hospitals and nursing
homes. For this report, we focused on inpatient interventions, with a mix of acute-care,
rehabilitation, long-term care, and geriatric wards and facilities represented. All but two of the
studies came from outside the United States: five from Australia, three from the United
Kingdom, two each from Sweden and Singapore, and one each from France, Switzerland, the
Netherlands, and Germany. Six studies mentioned having an academic affiliation or being a
teaching hospital. Of the 15 studies that reported the size of the setting, three were under 100
beds, five were between 100 and 500 beds, and two were over 500 beds. Three other studies
described size using alternative measures: 24 wards in 12 hospitals, a staff of 641, and 2300
inpatients annually. Thus, falls prevention programs have been successfully implemented in
hospitals of varying size, location, and academic/teaching status.
No studies reported on financial concerns (e.g., how patients care or the interventions were
financed), although one U.S. study mentioned the potential impact of reimbursement on the
emphasis on falls prevention.14 Since some countries where these studies have been conducted
have national health insurance, this context may be less applicable, and therefore not reported.
Existing Infrastructure
Five studies reported on the existing quality and safety infrastructure. Here we describe this
infrastructure in terms of factors that may affect implementation of a patient safety practice,
which could include presence of electronic health records or prior experience with quality
improvement or patient safety practices. The five studies included text that captured this concept;
of these, four described their usual fall prevention care. The fifth study provided a more explicit
statement, namely, prior to this study none of the wards carried out specific fall assessments or
interventions, and investigations such as lying and standing blood pressure or ophthalmology
referral occurred on an ad hoc basis. There was no specialist falls clinic or other falls service
available at this hospital.18 Another explanation was less explicit, and was embedded in the
authors explanation of the intervention, which noted that the two control wards continued with
the regular fall prevention policy used at the hospital (i.e., daily assessment of fall risk, review of
fall prevention with the patient and/or their family, use of fall prevention signage, and
implementation of other prevention strategies as needed).20 Two other reports of randomized
controlled trials discussed usual care in a similar fashion when contrasting it with the
intervention.9,14 These descriptions illustrate the potential diversity that may exist in the
control sites in terms of usual care.
In addition to a description of the current fall prevention care, a second type of infrastructure
description addressed an inadequate information system, reporting that the existing information
system was not useful for producing data that we could use to analyze the causes of falls.16 A
further example of this type of explanation is presented by Dykes and colleagues, who suggest
that including hospitals with diverse clinical information and documentation systems enhanced
the [intervention] generalizability.14 The remaining studies do not mention existing quality and
safety infrastructure.
188
Consequently, a dearth of data exists regarding the infrastructure needed to support fall
prevention programs or how the effectiveness of implementation may vary as a result of
infrastructure differences.
External Factors
Although a few studies briefly mentioned patient safety culture, teamwork, or leadership,
only four studies presented expanded explanations that merited mention. Grenier-Sennelier16 use
a framework from Shortell and colleagues37,38 to analyze safety on the unit level, teamwork at
both the organizational and unit level, and leadership on the organizational and unit level.
Stenvall discusses teamwork at the unit level in Table 2 of their article (See Appendix D).23 Koh
discusses leadership on the organizational and unit level: Successful implementation is
mediated by strong leadership and environmental support, which are integral to building positive
attitudes among nurses, ensuring that the sociocultural environment is conducive to the process
of change. In our study, the multifaceted strategy targeting barriers to change exemplified the
commitment of the leadership and environmental support.19 (p. 429) Van der Helm made
multiple observations addressing leadership on both the organizational and unit level:
Although the clinical ward management underlined the importance of implementing the
guideline at the outset of the project, the actual support given was too weak to be
effective. Some managers expressed doubt about the projects chances for success to the
project leader, stating that implementation had already failed before. Ward staff often
regarded improvement activities as unwanted additional work that hindered daily
operations. The two senior nurses often displayed a delegating rather than a directive
management style, for example, in terms of ensuring that the risk assessment tool was
completed or all incidents reported. (p.157)
nurses told us that the medical center did not take the falls problem seriously, which
therefore undermined their own motivation to contribute to the projects success. (p.158)
A measure in the Questionnaire Regarding Knowledge of the Guideline and Attitude
Toward Implementation, There is enough support from the management for guideline
implementation scored 44% to 53%.25
Implementation
The most commonly reported implementation details were patient characteristics (17 studies)
and an initial plan, or what was going to be done in the intervention (17 studies). Slightly less
often (14 studies), studies reported the intended roles of project staff, or by whom the intended
plan components were to be completed. The majority of studies reported the recipients of any
training component (15 studies), with slightly fewer reporting the type of training or giving a
description of the training (12 studies), and even fewer studies reporting the length of training (5
studies).
Another characteristic that distinguished studies was who conducted the risk assessments and
performed the interventions. In the reviews by Oliver and colleagues and the Cochrane group,
among the 17 studies of inpatient fall prevention programs, the risk assessments were performed
by the existing ward staff in 15 and by research staff in two. In 15 studies, the intervention was
performed by the ward staff: seven involved the nursing staff only, seven were multiprofessional,
and two involved physical therapy. In both of the new studies, clinicians or nurses from the
wards performed the risk assessments. The study with nurse risk assessments had research team
189
nurses provide the intervention, whereas the other study relied mainly on ward nurses, although
reference was made to clinicians more generally.
Thirteen studies provided the tools or materials used in the program implementation.
Whereas eight reported on adherence or fidelity to the designed initiative, only five described
how and why the plan evolved. Adherence or fidelity was most often characterized in a
qualitative statement, as with Brandis: The strategies implemented had high acceptance by
staff it is suggested that the higher reductions occurred in areas where the multidisciplinary
team enthusiastically embraced the project.12 An example from a less positive characterization
comes from Cumming: The lack of effect was evident in both wards and occurred despite the
planned nursing and physiotherapy interventions being successfully implemented.13 Dykes and
colleagues provided a strong example of adherence reporting, where protocol adherence was
measured by the completion of components in both control (81%) and intervention wards (94%).
Measures of adoption and reach were usually provided in the form of a flow chart: Six studies
presented these data for providers, and eight presented the data for patients.
For additional information on implementation, we used our update search and sought
suggestions of additional studies from experts. All of these studies had pre-post designs or were a
time series. Six were post-study evaluations of of falls implementations that reported a great deal
of detail about the potential reasons for effectiveness or lack thereof. Nine of the eleven studies
assessed implementation at only one or two facilities. Four of the studies did not report beneficial
effects of the fall prevention program and the article highlighted potential implementation factors
that might account for the lack of success. One study explicitly assessed the effect of some
contextual factors on intervention success across 34 facilities.39 One study explicitly assessed
sustainability. Details of these studies are presented in Appendix D.
We used five of the implementation articles to develop themes regarding effective
implementation and then reviewed all articles for these themes. The following are the most
consistently supported themes:
Leadership support is critical, both at the facility level and at the unit level (e.g.
clinical champions).
Engagement of front line clinical staff in the design of the intervention helps ensure
that it will mesh with existing clinical procedures.
Multidisciplinary committees guided or oversaw most interventions
developed/guided/overseen by
Pilot testing the intervention helps identify potential problems with implementation
Informational technology systems capable of providing data about falls can facilitate
evaluations of the causes, compliance with the intervention components, and (in one
case) be a crucial facilitator of the intervention.
Changing the prevailing attitude that falls are inevitable and nothing can be done
about them is required to get buy-in to the goals of the intervention
Education and training of clinical staff is necessary to help ensure compliance does not
diminish.
Table 4, below, presents textual support from the implementation articles for five of the
seven themes (pilot testing and information technology systems are not presented due to space
limitations).
190
Table 4, Chapter 19. Implementation themes highlighted in implementation studies

Author/Year
Leadership
Support
Frontline
Engagement
Multidisciplinary
Committees
Pilot Testing
Browne et al.,
40
2004
--
--
Falls Committee;
quarterly meetings
Once the tool

was developed, it
was piloted and
validated. The
results were
presented to the
MHS Falls
Committee, who
gave permission
for automated
implementation
system-wide.
191
Information
Technology
Systems
the redesign of
an adult
inpatient falls
program using a
computerized
information
systemthe
tool provides an
accurate
assessment of
the fall risk of
each patient.
Indicators are
embedded into
routine
assessment
documentation,
eliminating
added chargting
time. The
program allows
tailored
interventions for
specific patient
risks.
Attitude Change
Education and
Training
--
Nurses were
taught about the
redesigned falls
program by fall
and restraint
fairs that
coincided with its
implementation.
Results of
Intervention and
Implementation
Successful
Table 4, Chapter 19. Implementation themes highlighted in implementation studies (continued)

Author/Year
Leadership
Support
Frontline
Engagement
Multidisciplinary
Committees
Pilot Testing
Capan et al.,
41
2007
A unit
champion was
selected to act
as a staff
resource
who was
respected as a
mentor and
passionate
about patient
safety
Staff involved
in choosing
equipment
A pilot test of the

new tool was
conducted in a
medical/neurolog
y unit with a high
fall incidence
rate. The original
plan to roll the
tool out one unit
at a time was
modified to an
immediate
hospital-wide
implementation
after the success
of the pilot
program.
Dempsey,
42
2004
--
Raised
concern over
nurses power
to induce
change
the hospital
quality council
chartered a
multidisciplinary
falls prevention
task force. The
team included
nurses, nursing
management, a
physician/geriatrici
an, nursing
educators, a
psychiatric clinical
specialist , risk
management staff,
performance
improvement/mea
surement staff,
and
representatives
from physical
therapy and
pharmacy.
--
A tool was
developed and
tested for interrater reliability in
a pilot study
when five nurses
of different
experience levels
assessed the
same patient.
On the basis of
the results of the
research project,
the Falls
Prevention
Programme
became standard
practice for
medical
patients
192
Information
Technology
Systems
--
Attitude Change
Education and
Training
Nurses were reluctant

to impose the
interventions [but]
they came to
recognize the
importance of each
step As the staff
began using the
interventions falls
began to decline
The research
team educated
the staff about
falls and the
importance of
fall prevention,
including
background
information on
falls and how the
new tool was to
be used. 95% of
staff completed
the education
prior to the
implementation
of the tool.
--
In the pilot study.a

number of nurses
expressed the belief
that falls were
inevitable and that
there was nothing that
could be done to
change this. Although
the study
demonstrated that it
was possible to reduce
the rate of patient falls,
the remarks of the
nurses support the
suggestionthat the
successful reduction of
patient falls lay in the
attitude of the nurses
themselves.
The Falls
Prevention
programme
consisted of an
assessment tool,
an alert graphic,
and education
(patient and
staff) Staff
education
commenced at
the introduction
of the study and
continued
intermittently
though formal
and informal
means.
Results of
Intervention and
Implementation
Successful
Mixed results,
initial success
followed by
deterioration over
five years.

Author/Year
Leadership
Support
Frontline
Engagement
Multidisciplinary
Committees
Pilot Testing
Gutierrez,
43
2008
Identify clinical
champions;
leadership on
unit agreed to
send a nurse to
the EvidenceBased Practice
Institute
Leadership
formed a team
to address falls
issue, team
was led by a
senior vice
president,
information
was presented
to leadership
throughout
project
project
design
included
soliciting staff
and physician
feedback
--
--
--
The fall team

meets regularly,
with in-depth
analysis at
regular
intervals
Multiple tools
were tested
before the
redesign team
developed their
own, which was
also tested.
Kolin et al.,
44
2010
193
Information
Technology
Systems
--
Attitude Change
Education and
Training
--
Yes, one key

component was
a brief elevator
speech for
engaging and
educating staff
Currently, the
team is are
working on an
interface to
connect the
system
electronic
medical record
with the event
reporting
system. The
system had a
combination of
paper
documentation
and electronic
record sites,
which had
separate
program roll out.
Implementation
means changing the
way nurses think
about falls accepting
that all patients are at
risk.
Comprehensive
nursing
education was
conducted
Results of
Intervention and
Implementation
Successful
Successful

Author/Year
Leadership
Support
Frontline
Engagement
Multidisciplinary
Committees
Pilot Testing
McCollam,
45
1995
Nursing
Administration
involved in full
implementation
--
Research in
Practice
Committee
oversaw the
project
Senior
leadership
support helps
remove
organizational
barriers to
change and
provides
resources
needed to
implement
change The
four sites that
reported
spreading
changes to
other facilities
also indicated
that leadership
was a major
success
factor.
--
teamwork skills
are an important
component of
sustained
success
Interdisciplinary or
multidisciplinary
falls team was a
core component of
all four high
performing sites.
Problems
identified during
the pilot included
inconsistent and
incomplete
reassessment,
identification of
secondary
diagnoses, and
score
consistencies
between shifts.
Adjustments
were made for
full
implementation.
--
Neily, 2005
39
194
Information
Technology
Systems
--
Attitude Change
Education and
Training
Compliance for care

plans and
interventions lagged
behind risk
assessment, which
could be due to
skepticism about the
program. Some
nurses may question
the instruments
findings or not believe
the problem serious
enough to address.
Training
sessions were
conducted for
nursing; video
tape was shown
about tool;
understanding
checked using
evaluation
--
--
--
Results of
Intervention and
Implementation
Successful
Successful

Author/Year
Leadership
Support
Frontline
Engagement
Multidisciplinary
Committees
Pilot Testing
OConnell,
46
2001
--
--
Team of
researchers and
clinicians
No pilot test was

conducted.
Rauch et al.,
47
2009
Leadership
hired a
consulting
team. All levels
of leadership
were engaged
and accepted
ownership of
the project. A
champion was
identified in
each unit.
It is
imperative to
obtain
frontline staff
input and
feedback to
ensure that
successful
change
management
occurs in the
clinical arena
If there are
any words of
advice here,
they would
be: never
change a
program
without
directly
involving and
getting buy-in
from those it
immediately
affects.
The Fall Team,

multidisciplinary in
nature and
inclusive of
managers and
frontline staff
[were involved in
all phases of the
project] Weekly
teleconferences
during
implementation;
monthly fall team
meetings after
implementation
During the 30 day

pilot, staff were
routinely
questioned and
encouraged to
provide feedback
on elements
working well and
elements that
were failing
Changes were
made as
neededthe pilot
was
extendedto
ensure a solid
process before
total hospital rollout.
195
Information
Technology
Systems
--
--
Attitude Change
Education and
Training
Risk assessment tool

difficulties may have
undermined staff
confidence and the
program may have
lost some of its
significance. Staff felt
that they were already
doing everything they
could, and this
program did not add
anything
--
--
educational
needs were
identified and
sessions were
scheduled
[including] an
introduction of
the assessment
tool and proper
utilization
Results of
Intervention and
Implementation
Unsuccessful
Successful

Author/Year
Leadership
Support
Frontline
Engagement
Multidisciplinary
Committees
Pilot Testing
SeminGoossens,
48
2003
Attempt to
involve medical
chiefs and
nurse
managers
could have
promoted
implementation
In our case,
efforts to reach
and involve the
people higher
in the hierarchy
such as the
Medical Chiefs
and nursing
managers were
not successful.
A project team
was formed
consisting of 9
nurses in various
positions, a clinical
epidemiologist,
and a consultant
for quality
improvement
projects.
After a 3 month
pilot, the
guidelines were
finalized.
Weinberg et
49
al., 2011
Hospital
leadership
initiated effort
and prioritized
fall prevention
We did not
believe in a
top-down
strategy and
so we
involved the
nurses in
rewriting and
implementing
the guideline.
Authors would
have tried to
get more buyin from floor
nurses if
given another
try, but they
did receive
feedback and
modify the
intervention
accordingly.
--
Committee was
formed by
leadership and
attendance was
mandated;
monthly fall
reviews were
attended by unit
managers, staff
involved in patient
care, and fall
prevention
initiative co-chairs
The Fall
Prevention
Initiative was
rolled out
incrementally,
using continuous
quality
improvement
methods
196
Information
Technology
Systems
--
Attitude Change
Education and
Training
Nursesfrequently
stated that it was
simply impossible to
prevent patients from
falling. Falling was
recurrently considered
to be an inevitable part
of aging,
hospitalization, and
illness, and therefore
seen as an
unavoidable accident,
rather than something
predictable and often
preventable.
Dissemination of
the guideline,
including large
posters.
--
Transforming the
culture was integral to
implementation;
emphasis placed on
building a just culture
and having a
constructive,
nonpunitive forum for
discussion
Yes
Results of
Intervention and
Implementation
Unsuccessful
Successful

The Cochrane review found no economic evaluations of the falls prevention programs that
met inclusion criteria.8 The review by Oliver and colleagues estimated the cost for specific
combinations of components in terms of environment and equipment and in terms of staff.
Fourteen of 17 trials were considered low cost in terms of equipment and environment
(meaning some equipment costs like slippers, hip protectors, or alarms for a limited proportion of
patients), and 14 of 17 were considered as nil, meaning none or inconsequential, for extra staff
FTE.

The study by Neily and colleagues was the only one identified that explicitly assessed the
effect of context on effectiveness. Across 34 Veterans Affairs health centers, a mix of acute care
and long-term care facilities, leadership support was cited as one of the strongest factors for
success. At 1-year followup, high-performing sites reported greater agreement with questions
assessing leadership support, teamwork skills, and useful information systems than lowperforming sites.39

Inpatient multicomponent programs have been shown to be effective at reducing falls. The
strength of evidence is high.
The effects of context have not been as well studied; however multicomponent interventions
have been effective in hospitals that vary in size, location, and teaching status.
An assessment for themes in eleven implementation studies found the following to be most
consistently supported:
Leadership support is critical, both at the facility level and at the unit level (e.g. clinical
champions).
Engagement of front line clinical staff in the design of the intervention helps ensure that
it will mesh with existing clinical procedures.
Most interventions were developed/guided/overseen by multidisciplinary committees
A pilot test of the intervention helps identify potential problems with implementation
An informational technology system capable of providing data about falls can facilitate
evaluations of the causes and compliance with the intervention components, and (in one
case) can be a crucial facilitator of the intervention.
Changing the prevailing attitude that falls are inevitable and nothing can be done
about them is required to get buy-in to the goals of the intervention
Adequate time for education and training of clinical staff is necessary to help ensure
compliance does not diminish.
By January 2013, AHRQ intends to make available a list of tool kits for inpatient fall
prevention programs. A summary table is located below (Table 5).
197

Scope of the
Strength of
Problem Targeted by Evidence For
the PSP
Effectiveness
Common/Low
High
Evidence or
Potential for
Harmful
Unintended
Consequences
Moderate
(increased use of
restraints and/or
sedation)
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate
References
1.
Oliver D, Healey F, Haines TP. Preventing

falls and fall-related injuries in hospitals.
Clin Geriatr Med. 2010 Nov;26(4):645-92.
PMID 20934615.
2.
Zecevic AA, Salmoni AW, Speechley M, et

al. Defining a fall and reasons for falling:
comparisons among the views of seniors,
health care providers, and the research
literature. Gerontologist. 2006
Jun;46(3):367-76. PMID 16731875.
3.
Schwenk M, Lauenroth A, Stock C, et al.

Definitions and methods of measuring and
reporting on injurious falls in randomised
controlled fall prevention trials: a systematic
review. BMC Med Res Methodol.
2012;12:50. PMID 22510239.
4.
National Database of Nursing Quality

Indicators. Guidelines for Data Collection
on the American Nurses Associations
National Quality Forum Endorsed Measures:
Nursing Care Hours per Patient Day, Skill
Mix, Falls, Falls with Injury. March 2012.
5.
World Health Organization. Violence and

Injury Prevention: Falls. 2012.
www.who.int/violence_injury_prevention/ot
her_injury/falls/en. Accessed July 25, 2012.
6.
Wong CA, Recktenwald AJ, Jones ML, et

al. The cost of serious fall-related injuries at
three Midwestern hospitals. Jt Comm J Qual
Patient Saf. Feb;37(2):81-7. PMID
21939135.
7.
Lamb SE, Hauer K, Becker C. Manual for

the fall prevention classification system.
2007.
www.profane.eu.org/documents/Falls_Taxo
nomy.pdf.
198
8.
Cameron ID, Murray GR, Gillespie LD, et

al. Interventions for preventing falls in older
people in nursing care facilities and
hospitals. Cochrane Database of Systematic
Reviews. 2010;1(Art No.: CD005465).
9.
Ang E, Mordiffi SZ, Wong HB. Evaluating

the use of a targeted multiple intervention
strategy in reducing patient falls in an acute
care hospital: a randomized controlled trial.
J Adv Nurs. 2011 Sep;67(9):1984-92. PMID
21507049.
10.
Barker A, Kamar J, Morton A, et al.

Bridging the gap between research and
practice: review of a targeted hospital
inpatient fall prevention programme. Qual
Saf Health Care. 2009 Dec;18(6):467-72.
PMID 19955459.
11.
Barry E, Laffoy M, Matthews E, et al.

Preventing accidental falls among older
people in long stay units. Ir Med J. 2001
Jun;94(6):172, 4-6. PMID 11495234.
12.
Brandis S. A collaborative occupational

therapy and nursing approach to falls
prevention in hospital inpatients. J Qual Clin
Pract. 1999 Dec;19(4):215-20. PMID
10619149.
13.
Cumming RG, Sherrington C, Lord SR, et

al. Cluster randomised trial of a targeted
multifactorial intervention to prevent falls
among older people in hospital. BMJ. 2008
Apr 5;336(7647):758-60. PMID 18332052.
14.
Dykes PC, Carroll DL, Hurley A, et al. Fall

prevention in acute care hospitals: a
randomized trial. JAMA. Nov
3;304(17):1912-8. PMID 21045097.
15.
Fonda D, Cook J, Sandler V, et al. Sustained

reduction in serious fall-related injuries in
older people in hospital. Med J Aust. 2006
Apr 17;184(8):379-82. PMID 16618235.
16.
Grenier-Sennelier C, Lombard I, JenyLoeper C, et al. Designing adverse event

prevention programs using quality
management methods: the case of falls in
hospital. Int J Qual Health Care. 2002
Oct;14(5):419-26. PMID 12389808.
26.
Vassallo M, Vignaraja R, Sharma JC, et al.

The effect of changing practice on fall
prevention in a rehabilitative hospital: the
Hospital Injury Prevention Study. J Am
Geriatr Soc. 2004 Mar;52(3):335-9. PMID
14962145.
17.
Haines TP, Bennell KL, Osborne RH, et al.

Effectiveness of targeted falls prevention
programme in subacute hospital setting:
randomised controlled trial. BMJ. 2004 Mar
20;328(7441):676. PMID 15031238.
27.
18.
Healey F, Monro A, Cockram A, et al.

Using targeted risk factor reduction to
prevent falls in older in-patients: a
randomised controlled trial. Age Ageing.
2004 Jul;33(4):390-5. PMID 15151914.
von Renteln-Kruse W, Krause T. Incidence

of in-hospital falls in geriatric patients
before and after the introduction of an
interdisciplinary team-based fall-prevention
intervention. J Am Geriatr Soc. 2007
Dec;55(12):2068-74. PMID 17971140.
28.
Choi YS, Lawler E, Boenecke CA, et al.

Developing a multi-systemic fall prevention
model, incorporating the physical
environment, the care process and
technology: a systematic review. J Adv
Nurs. 2011 May 6.
29.
Coussement J, De Paepe L, Schwendimann

R, et al. Interventions for preventing falls in
acute- and chronic-care hospitals: a
systematic review and meta-analysis. J Am
Geriatr Soc. 2008 Jan;56(1):29-36. PMID
18031484.
30.
Oliver D, Connelly JB, Victor CR, et al.

Strategies to prevent falls and fractures in
hospitals and care homes and effect of
cognitive impairment: systematic review and
meta-analyses. BMJ. 2007 Jan
13;334(7584):82. PMID 17158580.
31.

Methodol. 2007;7:10. PMID 17302989.
32.

Med. 2007 Aug 21;147(4):224-33. PMID
17638714.
33.
Downs SH, Black N. The feasibility of

creating a checklist for the assessment of the
methodological quality both of randomised
and non-randomised studies of health care
interventions. J Epidemiol Community
Health. 1998 Jun;52(6):377-84. PMID
9764259.
34.
Chapter 8: Assessing risk of bias in included

studies: Table 8.5.c. In: Higgins JPT,
Altman DG, eds. Cochrane Handbook of
Systematic Reviews of Interventions Vol.
5.0.0. The Cochrane Collaboration; 2008.
19.
20.
Koh SL, Hafizah N, Lee JY, et al. Impact of

a fall prevention programme in acute
hospital settings in Singapore. Singapore
Med J. 2009 Apr;50(4):425-32. PMID
19421690.
Krauss MJ, Tutlam N, Costantinou E, et al.
Intervention to prevent falls on the medical
service in a teaching hospital. Infect Control
Hosp Epidemiol. 2008 Jun;29(6):539-45.
PMID 18476777.
21.
Oliver D, Martin F, Seed P. Preventing

patient falls. Age Ageing. 2002
Jan;31(1):75-6. PMID 11850313.
22.
Schwendimann R, Buhler H, De Geest S, et

al. Falls and consequent injuries in
hospitalized patients: effects of an
interdisciplinary falls prevention program.
BMC Health Serv Res. 2006;6:69. PMID
16759386.
23.
Stenvall M, Olofsson B, Lundstrom M, et al.

A multidisciplinary, multifactorial
intervention program reduces postoperative
falls and injuries after femoral neck fracture.
Osteoporos Int. 2007 Feb;18(2):167-75.
PMID 17061151.
24.
Uden G, Ehnfors M, Sjostrom K. Use of

initial risk assessment and recording as the
main nursing intervention in identifying risk
of falls. J Adv Nurs. 1999 Jan;29(1):145-52.
PMID 10064293.
25.
van der Helm J, Goossens A, Bossuyt P.

When implementation fails: the case of a
nursing guideline for fall prevention. Jt
Comm J Qual Patient Saf. 2006
Mar;32(3):152-60. PMID 16617946.
199
35.
36.
37.
38.
39.
40.
van Gaal BG, Schoonhoven L, Mintjes JA,

et al. Fewer adverse events as a result of the
SAFE or SORRY? programme in hospitals
and nursing homes. part i: primary outcome
of a cluster randomised trial. Int J Nurs Stud.
2011 Sep;48(9):1040-8. PMID 21419411.
van Gaal BG, Schoonhoven L, Hulscher
ME, et al. The design of the SAFE or
SORRY? study: a cluster randomised trial
on the development and testing of an
evidence based inpatient safety program for
the prevention of adverse events. BMC
Health Serv Res. 2009;9:58. PMID
19338655.
Shortell SM, OBrien JL, Carman JM, et al.
Assessing the impact of continuous quality
improvement/total quality management:
concept versus implementation. Health Serv
Res. 1995 Jun;30(2):377-401. PMID
7782222.
Gillies GL, Reynolds JH, Shortell SM, et al.
Implementing continuous quality
improvement. In: Kimberly J, Minvielle E,
eds. The Quality Imperative. Measurement
and Management of Quality in Healthcare.
London: Imperial College Press; 2001.
Neily J, Howard K, Quigley P, et al. Oneyear follow-up after a collaborative
breakthrough series on reducing falls and
fall-related injuries. Jt Comm J Qual Patient
Saf. 2005 May;31(5):275-85. PMID
15960018.
Browne JA, Covington BG, Davila Y. Using
information technology to assist in redesign
of a fall prevention program. J Nurs Care
Qual. 2004;19(3):218-25.
200
41.
Capan K, Lynch B. A hospital fall

assessment and intervention project. JCOM.
2007;14(3):155-60.
42.
Dempsey J. Falls prevention revisited: a call

for a new approach. J Clin Nurs. 2004
May;13(4):479-85. PMID 15086634.
43.
Gutierrez F, Smith K. Reducing falls in a

Definitive Observation Unit: an evidencebased practice institute consortium project.
Crit Care Nurs Q. 2008 Apr-Jun;31(2):12739. PMID 18360143.
44.
Kolin MM, Minnier T, Hale KM, et al. Fall

initiatives: redesigning best practice. J Nurs
Adm. Sep;40(9):384-91. PMID 20798621.
45.
McCollam ME. Evaluation and

implementation of a research-based falls
assessment innovation. Nurs Clin North Am.
1995 Sep;30(3):507-14. PMID 7567575.
46.
OConnell B, Myers H. A failed fall

prevention study in an acute care setting:
lessons from the swamp. Int J Nurs Pract.
2001 Apr;7(2):126-30. PMID 11811315.
47.
Rauch K, Balascio J, Gilbert P. Excellence

in action: developing and implementing a
fall prevention program. J Healthc Qual.
2009 Jan-Feb;31(1):36-42. PMID 19343900.
48.
Semin-Goossens A, van der Helm JM,

Bossuyt PM. A failed model-based attempt
to implement an evidence-based nursing
guideline for fall prevention. J Nurs Care
Qual. 2003 Jul-Sep;18(3):217-25. PMID
12856906.
49.
Weinberg J, Proske D, Szerszen A, et al. An

inpatient fall prevention initiative in a
tertiary care hospital. Jt Comm J Qual
Patient Saf. Jul;37(7):317-25. PMID
21819030.
Chapter 20. Preventing In-Facility Delirium


Delirium (also known as acute confusional state) refers to an acute decline in attention and
cognition that constitutes a serious problem for older hospitalized patients and long-term care
residents. Estimated hospital occurrence rates have ranged from 14% to 56% and vary depending
upon reason for hospitalization (e.g., urgent surgery, intensive care, general medical admission)
and the patients risk of developing delirium.1 Development of delirium is associated with an
increased risk of mortality, postoperative complications, longer hospital and intensive care unit
stays, and functional decline.1,2 In addition, delirium presents a significant burden in terms of
short and long-term health care costs. A study of 841 patients (age 70 years) admitted to nonintensive care general medical units over a three year period at Yale-New Haven hospital found
that costs per day were more than 2.5 times higher for patients with delirium compared with
those without delirium. The total cost estimates associated with delirium ranged from $16,303 to
$64,421 per patient, which the authors extrapolated to national costs ranging from $38 billion to
$152 billion each year.1 As these cost estimates were based on data from 1995-1998, the costs of
delirium today would be even higher. Accordingly, prevention of delirium is extremely
important both for improving patient outcomes and for lowering health care costs.

Several delirium prevention programs are multifactorial bundles of interventions. In general,
the components in the bundle vary across each published evaluation, and the same bundle is
rarely evaluated in more than one application (see Appendix D, Table 2). Therefore, the best that
can be done is to describe the components most commonly included in bundles that have been
found to reduce incident delirium. Based on our review (described later), we identified the
following as the most common components of successful bundles:
Anesthetic protocols
Assessment of bowel/bladder functions
Early mobilization
Extra nutrition
Geriatric consultation
Hydration
Medication review
Pain management
Prevention and treatment of medical complications
Sleep enhancement
Staff education
Supplemental oxygen
Therapeutic cognitive activities/orientation
Vision and hearing protocols
201
Additional components have been reported in successful multifactorial bundles. An

intervention used in a Swedish university hospital for patients with hip fracture included
increased physiological monitoring, avoidance of delays in transfer through different areas of the
hospital, daily delirium screening, and avoidance of polypharmacy (as well as several
components from the bolded list, including extra nutrition, IV fluid supplementation, pain
management, and perioperative/anesthetic period protocols).3 A multifactorial intervention used
at another Swedish university hospital for patients with hip fracture included treatment of sleep
apnea, prevention and treatment of decubitus ulcers, and measurement of blood pressure along
with components from the bolded list, although it is not clear that all of these components were
specifically designed to prevent delirium.4
The Hospital Elder Life Program (HELP) or a modified version of HELP was the most
frequently-evaluated multifactorial intervention, appearing in three studies.5-7 This program
typically consists of six components, including orientation, therapeutic activities, vision and
hearing protocols, sleep enhancement, and early mobilization. Two studies (one U.S., one
Australian) used proactive geriatric consultation with targeted recommendations (several from
the bolded list) based on a structured protocol.8,9
Components that have been used as single interventions include: medical therapy (anesthetics
or other drugs believed to lower the risk of delirium), hydration, and music therapy (see Table 3).
The overwhelming majority of single interventions consisted of some type of medication; this
included Dexmedetomidine for post-operative anesthesia (two studies), Rivastigmine (two
studies), Propofol (two studies), Olanzapine (one study), Ketamine (one study), Melatonin (one
study), Risperidone (one study), Haloperidol (one study), Donepezil (one study), and Diazepam
plus Flunitrazepam plus Pethidine drip infusion (one study).

Evidence from risk-factor studies suggests that delirium has a multifactorial etiology. Our
literature review identified 55 studies of factors associated with delirium occurrence that met
inclusion criteria (see Appendix D, Table 1 for individual study data). Collectively, these studies
found significant associations between several factors and occurrence of delirium. However, no
two studies evaluated the exact same set of factors or found the same combination of significant
factors associated with delirium. The risk of bias was moderate in 31 studies and high in
24 studies.
Age was the most commonly evaluated factor, assessed in 34 studies. Twenty studies
(58.8%) found a significant association between older age and delirium occurrence, including the
two largest studies that evaluated data from more than a million patients recorded in large
databases (most of the other studies included between 40-500 patients). These large studies had a
high risk of bias due to retrospective design, identification of delirium from ICD-9 codes, and
inclusion of prevalent as well as incident cases of delirium in the same analysis, but smaller
studies with a moderate risk of bias supported the findings. Since many studies exclusively
enrolled older patients (age >65 or >70 years), it may have been more difficult to demonstrate an
association in some of these studies (due to restriction of range), which may partially explain the
inconsistent findings in the evidence base. Another potential explanation is that some studies
may have lacked adequate power to find statistical significance, although this was clearly not the
case in all studies that did not have a significant finding. Cognitive impairment or dementia was
evaluated in 26 studies; 22 studies (84.6%) found a significant association between this factor
202
and incidence of delirium. Depression was evaluated in 10 studies, but only four (40%) found a
significant association with delirium occurrence.
Other patient-specific risk factors that showed a significant association with delirium in more
than one study include male gender, multiple medications, comorbidities (e.g., diabetes),
pneumonia, various anesthetics, neuropsychiatric drugs (e.g., benzodiazepines), anticholinergics,
blood transfusions, abnormal serum chemistry (e.g., urea levels, creatinine levels),
apolipoprotein E4 (APOE4), atrial fibrillation, heavy alcohol intake, volume depletion
(dehydration), oxygen levels, complications, restraints (rendering patients immobile) and visual
impairment. Several studies evaluated patients undergoing specific surgical procedures (e.g., hip
repair or replacement, cardiac surgery); some of these studies focused on surgery-specific risk
factors (e.g., blood transfusions, intraoperative anesthesia) and evaluated few non-surgical
factors.
Given the multifactorial nature of delirium, a patient safety practice designed to assess and
remediate multiple factors is believed to be more likely to be effective. Indeed, the list of
components in successful delirium prevention bundles targets several factors identified in this list
of patient and environmental contributors to delirium. For example, the Hospital Elder Life
Program (HELP) specifically targets six risk factors for delirium: cognitive impairment, visual
impairment, hearing impairment, sleep deprivation, immobility, and dehydration.7 Of this list,
only hearing impairment was not identified as a risk factor by the studies in our evidence base,
but this may be because only one of those studies even evaluated it as a possible risk factor.

To assess the effects of delirium prevention interventions, we performed a systematic review
of six databases (including Medline and CINAHL) from 1999 to 2011 from which we got 587
titles of which 85 were reviewed in detail. From this we identified 31 studies that met inclusion
criteria for addressing this question. Fifteen studies evaluated the efficacy of multicomponent
interventions (see Appendix D, Table 2), and the remaining 16 studies evaluated single
interventions (see Appendix D, Table 3). Most of these studies reported the incidence of delirium
following intervention compared with a control arm of usual care treated concurrently or during
a period immediately prior to adoption of the new intervention. Some studies of medical therapy
used an alternative medical therapy as the comparative arm. Since very few studies used the
same intervention, comparison group, study design and/or patient population, meta-analyses
were not performed for the majority of interventions.
Multicomponent Interventions
Hospital Inpatient Care
Of the multicomponent intervention studies, two used HELP and a third used a modification
of HELP. One was a controlled before-and-after study with a concurrent control group consisting
of patients from usual care units7,10; this study had a moderate risk of bias. The remaining two
studies were before-and-after studies where the usual care group consisted of patients treated
prior to implementation of HELP (historical control)5,6; these studies had a high risk of bias. All
three of these studies found a significant reduction in incident delirium after implementation of
HELP compared with usual care. Although the findings of the studies were consistent, the
average risk of bias was high mainly due to lack of randomization and blinding.
203
Two studies used proactive geriatric consultation with targeted recommendations based on a
structured protocol for patients with hip fracture. One was a single-blind RCT with usual care
control,8 while the other was a before-after study with a historical usual care control.9 Both
studies reported a significant reduction of incident delirium for the geriatric consult group
compared with the usual care group; however, the RCT findings were no longer statistically
significant after adjustment for baseline imbalances. The risk of bias was high and moderate for
the respective studies.
Of the remaining multicomponent studies, all but one reported a significant reduction in
delirium by at least one measure in the intervention group versus the control group. The
exception was a study of a system-wide quality improvement project.11 A study of nursefacilitated family participation reported significantly fewer patients with a diagnosis of delirium
(defined by a score 4 on the Intensive Care Delirium Screening Checklist [ICDSC]) in the
intervention group, but also reported no significant between-group difference in mean scores;
this study placed more emphasis on the latter measure.12 Overall, the findings are consistent with
the findings from studies of the HELP intervention, although the risk of bias was high again due
to lack of randomization and blinding.
Long-Term Care
The single study set in a nursing home setting reported that homes randomized to use
pharmacist-led geriatric risk assessment medguide (GRAM) reports and automated medication
monitoring plans had a significant reduction in potential delirium onset among newly-admitted
residents compared with homes randomized to usual care.13 However, it is unclear how much of
this is due to delirium prevention or resolution of new-onset delirium.
The majority of the evidence suggests that multicomponent interventions are effective in
preventing onset of delirium in at-risk patients. However, these studies do not address the
question of which particular components within a program provide the most benefit.
Single Interventions
The majority of the single-intervention studies also found a significant reduction in delirium
incidence for the study interventions, but roughly one-third (five studies) did not find a
significant reduction. Unlike the multicomponent evidence base, almost all of the singleintervention studies were RCTs. However, few studies used the same medication or comparison
treatment in the same patient population, making it difficult to determine consistency of findings
for most of these interventions.
Hospital Inpatient Care

Sedatives/anesthetics. The sedative Dexmedetomidine was compared with other post-operative
anesthetics in two studies of patients who underwent cardiac surgery. One found a significant
reduction in post-operative delirium for the Dexmedetomidine group compared with patients
receiving Propofol or Midazolam.14 This study had a high risk of bias. The other study did not
find a significant reduction in post-operative delirium for Dexmedetomidine compared with
Morphine, although the study was underpowered to detect a small difference between groups.15
This study did find a significant reduction in duration of delirium for those receiving
Dexmedetomidine; the risk of bias was moderate. Because these two studies used different
comparison groups, the consistency of the findings cannot be determined.
204
A study of light versus deep Propofol sedation during spinal anesthesia for hip repair found
that patients receiving light Propofol sedation (measured by the bispectral index [BIS]) had a
significantly lower rate of postoperative delirium.16 The risk of bias was moderate. It is unclear
how this compares to the amount of Propofol used in the study comparing Dexmedetomidine and
Propofol, which reported the amount in g/kg/minute.14
Patients undergoing cardiac surgery had significantly lower rates of postoperative delirium
after receiving Ketamine (an NMDA receptor antagonist) during anesthetic induction compared
with placebo.17 The risk of bias was moderate, and the findings should be confirmed by other
studies.
Patients undergoing joint surgery had significantly lower incidence of delirium after
receiving fascia iliac block prophylaxis via Bupivicaine (a local anesthetic) compared with
placebo.18 The risk of bias was moderate, and the findings should be confirmed by other studies.
Acetylcholinesterase inhibitors. One study of patients undergoing cardiac surgery did not find a
significant between-group difference for those receiving Rivastigmine compared with those
receiving placebo.19 This study was judged to have a low risk of bias. Patients undergoing joint
surgery who received a different acetylcholinesterase inhibitor (Donepezil) did not show a
significant reduction in postoperative delirium compared with those receiving placebo.20 This
study had a moderate risk of bias. Pooling of these two studies findings resulted in a relative risk
of 1.11 (95% CI 0.69 to 1.79); the confidence interval was too imprecise to rule out the
competing possibilities that acetylcholinesterase inhibitors are ineffective or might confer a
benefit.
Atypical antipsychotics. Patients undergoing cardiac surgery had significantly lower rates of
postoperative delirium after receiving Risperidone compared with placebo in one RCT with a
moderate risk of bias.21 Patients undergoing joint surgery who received Olanzapine had a
significant reduction in postoperative delirium compared with placebo in another RCT with low
risk of bias.22 Both studies showed a substantial reduction with almost identical risk ratios; the
combined summary relative risk is 0.35 (95% CI 0.25 to 0.50, P<0.0001). The findings are
therefore consistent and precise for this drug class.
Typical antipsychotics. Patients undergoing joint surgery who received Haloperidol did not
show a significant reduction in postoperative delirium compared with those receiving placebo.23
The risk of bias was moderate and the findings were imprecise, therefore requiring confirmation
from additional studies.
Melatonin. Patients undergoing joint surgery had significantly lower incidence of delirium after
receiving Melatonin compared with placebo.18,22,24 The risk of bias was moderate, and the
findings should be confirmed by other studies.
Benzodiazepines. One RCT found that postoperative delirium was significantly lower in
gastrointestinal surgery patients who received the benzodiazepines Diazepam + flunitrazepam as
a drip infusion in addition to a pethidine drip infusion for the first 3 days compared with those
who did not receive these infusions.25 The risk of bias was high.
205
Music therapy. In two RCTs conducted by the same authors at the same hospital, patients
undergoing hip or knee surgery had significantly lower rates of acute confusion after receiving
music therapy compared with those receiving usual care.26,27 Both studies had a high risk of bias,
in part because they employed an unvalidated delirium assessment method, and should be
repeated at other hospitals for confirmation of the results.
Long-Term Care
Hydration therapy. A quasi-randomized study comparing 8 weeks of hydration therapy to usual
care for delirium prevention among residents of four nursing homes (hydration or control was
randomized by nursing home) did not find a significant difference between intervention or
control homes in episodes of acute confusion.28 The risk of bias was high.
Acetylcholinesterase inhibitors. A study using Rivastigmine daily for two years in patients with
vascular dementia found that the Rivastigmine group had significantly fewer episodes of
delirium than those taking cardioaspirin (the control group).29 This was the only study in the
entire evidence base that exclusively enrolled patients with dementia, and it was judged to have a
high risk of bias. Although these were ambulatory outpatients, they were judged to be similar to
patient populations in long-term care settings.

Most trials of delirium prevention programs have not reported any harms. However, it is not
clear whether or not the possibility of harms was explicitly assessed in all of these trials. One
study of a multicomponent intervention based on a structured quality improvement model
reported four unexpected minor events (rectal or feeding tube removal) but no significant
complications (and no significant difference compared with usual care).30 Two other
multicomponent studies reported no significant differences in complications between
intervention and usual care groups.3,13 Seven out of 16 studies on single interventions reported
information on adverse events; all seven studies evaluated a variety of medical therapies
(medications or anesthesia). Three of these studies reported no significant between-group
difference in adverse event rates.16,19,21 One study of Dexmedetomidine versus Morphine for
patients after cardiac surgery found that bradycardia occurred significantly more often in the
Dexmedetomidine group, while systolic hypotension occurred significantly more often in the
Morphine group.15 Another study reported that patients who received Olanzapine had
significantly more severe and longer-lasting delirium than patients who received placebo,
although incidence of delirium was significantly lower in the Olanzapine group.22 One study of
melatonin reported that 2/61 patients had side effects of nightmares or hallucinations, while no
patients who received placebo had side effects.24 The remaining study reported no complications
associated with fascia iliac block prophylaxis other than local hematomas at the injection site,
which resolved spontaneously.18
206

What Contexts?
Literature searches identified 15 studies of multicomponent delirium prevention programs
that met inclusion criteria (see Table 2). The limited information on how these programs have
been implemented and in what contexts is summarized below.

Multicomponent delirium prevention programs have been successfully implemented in both
acute care hospitals (14 studies) and in nursing homes (1 study). Five of the acute care hospital
studies were conducted in the United States, three in the United Kingdom, three in Sweden, and
one each in Australia, Spain, and Taiwan. Ten studies were from academically-affiliated urban
hospitals, two studies were conducted in urban hospitals that were not described as teaching
hospitals, and the remaining two studies were set in community hospitals (in one study the
participating community hospitals were part of a larger Health System). No studies have been
reported from rural hospitals. The single study of nursing homes was conducted in the U.S..
Existing infrastructure. Only one study reported minimal information on patient safety culture
at the organizational level; the authors stated merely that SHS [Summa Health System]
maintains a strong commitment to patient safety and quality.11
External factors. External factors or motivators were not mentioned in any delirium study.
Implementation. All multicomponent intervention studies provided at least minimal information
concerning teamwork and/or leadership at the level of the unit where the intervention was
implemented. Eleven of 15 studies specifically identified the study leaders, while 14/15 studies
identified the team members by job status (e.g., nurses, geriatricians) or at least stated that all
staff in the intervention ward or unit were part of the team. All of these studies reported
multidisciplinary teamwork that included clinical experts, nurses, and other staff (e.g., physical
therapists, volunteers). One study reported minimal information on teamwork or leadership at the
hospital level.11
Seven studies described multi-professional implementation, one had the intervention
performed by the ward staff, one involved ward staff plus physical therapists (at home visits),
one involved ward staff plus ambulance workers, one involved unit staff plus volunteers, one
involved the nursing staff only, one involved nursing staff plus consultant pharmacists, one
involved nurses assisting family members with the intervention, and one involved elder life
specialists plus volunteers.
Twelve studies reported on staff education/training if this was part of the intervention, and
seven studies reported the individual(s) responsible for implementation. Most of these studies
reported that all staff involved in the implementation underwent some type of education or
training. Ten studies reported the type of training, and only four studies reported the length of
training.
Four studies reported a change in the implementation process due to local tailoring or an
iterative process. Only one study (Rubin et al. 2011) reported that internal incentives were used
to promote implementation.5 Allen et al. (2011) published the only study that provided a table
summarizing an actual implementation instrument (a scorecard used to track process and
outcome variables).11
207
Fourteen studies outlined the intended intervention and the general sequence in which the
components were implemented; only 11 studies included enough detail to determine the roles of
the various team members. However, this was generally a description of how the intervention
was supposed to be implemented; most studies did not describe any modifications or failures of
adherence that might have occurred during the actual implementation. Only one study actually
measured adherence to targeted recommendations, reporting an adherence rate of 77% regarding
implementation of a geriatric consultants recommendations for patients after hip fracture repair.8
Twelve studies reported patient characteristics.
Although implementation of multicomponent delirium prevention programs has not been
well-described in most studies, a few themes seem sufficiently constant to report here:
A multidisciplinary team comprising clinical experts, nurses and additional staff is
helpful for implementation of a complex intervention
fall.

Two studies in the evidence base reported information on costs or cost savings associated
with multicomponent delirium prevention programs. Rizzo et al. calculated the total intervention
costs of HELP over a three year period (1995-1998) at Yale-New Haven hospital as $257,385
(personnel plus equipment costs). In a cost-effectiveness analysis, they found that the
intervention was cost-effective for patients at intermediate risk of delirium but not for patients at
high risk of delirium (lack of effectiveness and higher overall costs). However, these findings
may be due to inadequate power based on their sample size of higher risk patients.31 Rubin et al.
calculated that implementation of HELP at their hospital led to estimated cost savings of over
$2 million per year from prevention of delirium cases. In addition, there was over $2.2 million
per year of estimated revenue generated by shorter hospital stays for patients without delirium.5
What Is Known About the Effect of Context on Outcomes?

Only two studies reported on the effect of context on outcomes. One study of an educational
package for medical and nursing staff reported that it was effective at preventing delirium in
hospitalized men but not in women.32,33 A study of proactive geriatric consultation with target
recommendations based on a structured protocol for patients with hip fracture reported a trend
toward more effectiveness among patients without pre-fracture dementia or activities of daily
living (ADL) impairment, but the differences were not statistically significant.8
One study assessed the somewhat related concept of patient adherence and its effect on
outcomes of a multifactorial intervention (HELP). Based on a composite adherence score for the
three components assigned to all patients (orientation, mobility, and therapeutic activities),
increased adherence scores were associated with a reduction in delirium incidence rates
(OR 0.69, 95% CI 0.56 to 0.87).7

Evidence from multiple studies suggests that a variety of factors may contribute to a hospital
patients risk for developing delirium; cognitive deficit was most consistently shown to be a risk
208
factor in these studies. All but two studies were judged to have high risk of bias, and these
exceptions were judged to have a moderate risk of bias. The majority of the evidence suggests
that most multicomponent interventions are effective in preventing onset of delirium in at-risk
patients in a hospital setting (Strength of Evidence: Moderate). In general, successful delirium
prevention programs involved a multidisciplinary team of clinical experts, nurses, and other staff
(e.g., physical therapists, volunteers) and included protocols for early mobilization of patients,
volume repletion (for hydration and electrolyte balance), and addressing visual or hearing
deficits; a few programs included elimination of unnecessary medications.
Other components reported in more than one study included staff education, geriatric
consultation, therapeutic cognitive activities/orientation, extra nutrition, sleep enhancement, pain
management, anesthetic protocols, supplemental oxygen, assessment of bowel/bladder functions,
and prevention and treatment of medical complications. However, these studies do not address
the question of which particular components within a program provide the most benefit.
There was not enough evidence to evaluate the benefit of delirium prevention programs in
long-term care settings.
Although implementation of multicomponent delirium prevention programs has not been
well-described in most studies, a few themes seem sufficiently constant to report here:
A multidisciplinary team comprising clinical experts, nurses and additional staff is
helpful for implementation of a complex intervention
fall.
Although several of the single-intervention studies also found a significant reduction in
delirium incidence for the study interventions, few studies used the same medication or
comparison treatment in the same patient population; this makes it difficult to determine
consistency of findings for most of these interventions. For atypical antipsychotics, two RCTs
with a low to moderate risk of bias evaluating different drugs within this class showed consistent
and precise findings of reduction in postoperative delirium among surgical patients (Strength of
Evidence: Moderate). Although two RCTs reported a significant reduction in acute confusion for
patients receiving music therapy, these studies were conducted at the same institution by the
same authors and used an unvalidated delirium assessment method. Therefore, the evidence is
insufficient for a conclusion regarding music therapy. Two RCTs had inconclusive findings
(even when pooled) regarding the efficacy of acetylcholinesterase inhibitors, rendering the
strength of evidence insufficient.
Most of the remaining treatments (or treatment comparisons) were represented by only one
study with a moderate or high risk of bias; we judged the evidence about these treatments to be
insufficient.
Future comparative effectiveness studies with standardized protocols are needed, particularly
to identify which components in multicomponent interventions are most effective for delirium
prevention. Identification of the most effective bundle of components might encourage hospitals
to adopt a more standardized approach to delirium prevention. Additional RCTs are also needed
to determine which single-component medical therapies or drug classes are truly beneficial for
patients at risk of delirium. A summary table is located below (Table 1).
209

Scope of the
Strength of
the PSP
Effectiveness
Common/Low
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate
References
1.
2.
3.
4.
9.
Wong DM, Niam T, Bruce JJ, et al. Quality

project to prevent delirium after hip fracture.
Aust J Ageing 2005;24(3):174-7.
10.
Rudolph JL, Marcantonio ER. Postoperative

Delirium: acute change with long-term
implications. Anesth Analg 2011
May;112(5):1202-11.
Inouye SK, Bogardus ST Jr, Charpentier

PA, et al. A multicomponent intervention to
prevent delirium in hospitalized older
patients. N Engl J Med 1999 Mar
4;340(9):669-76. PMID: 10053175
11.
Bjorkelund KB, Hommel A, Thorngren KG,

et al. Reducing delirium in elderly patients
with hip fracture: a multi-factorial
intervention study. Acta Anaesthesiol Scand
2010 Jul;54(6):678-88. PMID: 20236093.
Allen KR, Fosnight SM, Wilford R, et al.

Implementation of a system-wide quality
improvement project to prevent delirium in
hospitalized patients. J Clin Outcomes
Manag 2011 Jun;18(6):253-8.
12.
Lundstrom M, Olofsson B, Stenvall M, et al.

Postoperative delirium in old patients with
femoral neck fracture: a randomized
intervention study. Aging Clin Exp Res
2007 Jun;19(3):178-86. PMID: 17607084.
Black P, Boore JR, Parahoo K. The effect of

nurse-facilitated family participation in the
psychological care of the critically ill
patient. J Adv Nurs 2011 May;67(5):1091101. PMID: 21214624
13.
Lapane KL, Hughes CM, Daiello LA, et al.

Effect of a pharmacist-led multicomponent
intervention focusing on the medication
monitoring phase to prevent potential
adverse drug events in nursing homes. J Am
Geriatr Soc 2011 Jul;59(7):1238-45. PMID:
21649623.
14.
Maldonado JR, Wysong A, van der Starre

PJ, et al. Dexmedetomidine and the
reduction of postoperative delirium after
cardiac surgery. Psychosomatics 2009 MayJun;50(3):206-17. PMID: 19567759.
15.
Shehabi Y, Grant P, Wolfenden H, et al.

Prevalence of delirium with
dexmedetomidine compared with morphine
based therapy after cardiac surgery: a
randomized controlled trial
(DEXmedetomidine COmpared to
Morphine-DEXCOM Study).
Anesthesiology 2009 Nov;111(5):1075-84.
PMID: 19786862.
Leslie DL, Marcantonio ER, Zhang Y, et al.

One-year health care costs associated with
delirium in the elderly population. Arch
Intern Med 2008 Jan 14;168(1):27-32.
PMID: 18195192.
5.
Rubin FH, Neal K, Fenlon K, et al.

Sustainability and scalability of the hospital
elder life program at a community hospital. J
Am Geriatr Soc 2011 Feb;59(2):359-65.
PMID: 21314654.
6.
Chen CC, Lin MT, Tien YW, et al. Modified

hospital elder life program: effects on
abdominal surgery patients. J Am Coll Surg
2011 Aug;213(2):245-52. PMID: 21641835.
7.
Inouye SK, Bogardus ST Jr, Williams CS, et

al. The role of adherence on the
effectiveness of nonpharmacologic
interventions: evidence from the delirium
prevention trial. Arch Intern Med 2003 Apr
28;163(8):958-64. PMID: 12719206.
8.
Marcantonio ER, Flacker JM, Wright RJ, et

al. Reducing delirium after hip fracture: a
randomized trial. J Am Geriatr Soc 2001
May;49(5):516-22. PMID: 11380742.
210
16.
17.
18.
Sieber FE, Zakriya KJ, Gottschalk A, et al.

Sedation depth during spinal anesthesia and
the development of postoperative delirium
in elderly patients undergoing hip fracture
repair.[Erratum appears in Mayo Clin Proc.
2010 Apr;85(4):400 Note: Dosage error in
article text. Mayo Clin Proc 2010
Jan;85(1):18-26. PMID: 20042557.
Hudetz JA, Patterson KM, Iqbal Z, et al.
Ketamine attenuates delirium after cardiac
surgery with cardiopulmonary bypass.
J Cardiothorac Vasc Anesth 2009
Oct;23(5):651-7. PMID: 19231245.
Mouzopoulos G, Vasiliadis G, Lasanianos
N, et al. Fascia iliaca block prophylaxis for
hip fracture patients at risk for delirium:
a randomized placebo-controlled study.
J Orthop Traumatol 2009;10(3):127-33.
19.
Gamberini M, Bolliger D, Lurati Buse GA,

et al. Rivastigmine for the prevention of
postoperative delirium in elderly patients
undergoing elective cardiac surgery-a randomized controlled trial. Crit Care Med
2009 May;37(5):1762-8. PMID: 19325490.
20.
Liptzin B, Laki A, Garb JL, et al. Donepezil

in the prevention and treatment of postsurgical delirium. Am J Geriatr Psychiatry
2005 Dec;13(12):1100-6. PMID: 16319303.
21.
22.
23.
24.
Prakanrattana U, Prapaitrakool S. Efficacy

of risperidone for prevention of
postoperative delirium in cardiac surgery.
Anaesth Intensive Care 2007 Oct;35(5):7149. PMID: 17933157.
Larsen KA, Kelly SE, Stern TA, et al.
Administration of olanzapine to prevent
postoperative delirium in elderly jointreplacement patients: a randomized,
controlled trial. Psychosomatics 2010 SepOct;51(5):409-18. PMID: 20833940.
Kalisvaart KJ, de Jonghe JF, Bogaards MJ,
et al. Haloperidol prophylaxis for elderly
hip-surgery patients at risk for delirium: a
randomized placebo-controlled study. J Am
Geriatr Soc 2005 Oct;53(10):1658-66.
PMID: 16181163.
Al-Aama T, Brymer C, Gutmanis I, et al.
Melatonin decreases delirium in elderly
patients: a randomized, placebo-controlled
trial. Int J Geriatr Psychiatry 2011
Jul;26(7):687-94. PMID: 20845391.
211
25.
Aizawa K, Kanai T, Saikawa Y, et al. A

novel approach to the prevention of
postoperative delirium in the elderly after
gastrointestinal surgery. Surg Today
2002;32(4):310-4. PMID: 12027195.
26.
McCaffrey R, Locsin R. The effect of music

on pain and acute confusion in older adults
undergoing hip and knee surgery. Holist
Nurs Pract 2006 Sep-Oct;20(5):218-24; quiz
225-6. PMID: 16974175.
27.

listening on acute confusion and delirium in
elders undergoing elective hip and knee
surgery. J Clin Nurs 2004;13(s2):91.
28.
Mentes JC, Culp K. Reducing hydrationlinked events in nursing home residents.

Clin Nurs Res 2003 Aug;12(3):210-25;
discussion 226-8. PMID: 12918647.
29.
Moretti R, Torre P, Antonello RM, et al.

Cholinesterase inhibition as a possible
therapy for delirium in vascular dementia: a
controlled, open 24-month study of 246
patients. Am J Alzheimers Dis Other Demen
2004 Nov-Dec;19(6):333-9. PMID:
15633941.
30.
Needham DM, Korupolu R, Zanni JM, et al.

Early physical medicine and rehabilitation
for patients with acute respiratory failure:
a quality improvement project. Arch Phys
Med Rehabil 2010 Apr;91(4):536-42.
PMID: 20382284.
31.
Rizzo JA, Bogardus ST Jr, Leo-Summers L,

et al. Multicomponent targeted intervention
to prevent delirium in hospitalized older
patients: what is the economic value. Med
Care 2001 Jul;39(7):740-52. PMID:
11458138.
32.
Tabet N, Hudson S, Sweeney V, et al.

An educational intervention can prevent
delirium on acute medical wards. Age
Ageing 2005 Mar;34(2):152-6. PMID:
15713859.
33.
Tabet N, Stewart R, Hudson S, et al. Male

gender influences response to an educational
package for delirium prevention among
older people: a stratified analysis. Int J
Geriatr Psychiatry 2006 May;21(5):493-7.
PMID: 16676296.
Section E. General Clinical Topics

Chapter 21. Preventing In-Facility Pressure Ulcers

Pressure ulcers (PUs) are preventable, but PU rates continue to escalate alarmingly fast. In
fact, between 1995 and 2008, the incidence of PUs increased by as much as 80%.1 Estimates of
incidence are high for both acute and long-term care patients. Current estimates indicate that 2.5
million patients will develop a PU, and 60,000 U.S. patients will die from complications related
to hospital-acquired PUs.2 A 2009 National Center for Health Statistics (NCHS) Data Brief
reported about 11% of nursing home residents had PUs (in 2004), stage 2 being the most
common.3
Preventing PUs is important not only to protect patients from harm, but also to reduce costs
of care. Estimates suggest that PU treatment costs could be as high as $11 billion annually.4
Patients with PU-related morbidity need more care and resources and have longer inpatient stays.
In some cases, late-stage PUs lead to life-threatening infections. Because of the ever-increasing
number of obese, diabetic, and elderly patients, PU rates are predicted to continue to rise. To
gather available information about the effectiveness of PU prevention programs, we searched
CINAHL, the Cochrane Library, EMBASE, MEDLINE, and PreMEDLINE from 1981 to 2011,
in addition to searching gray literature. We identified 454 abstracts from which 87 full-text
articles were reviewed in more detail, yielding 47 articles contributing data to this review.

Sources for patient safety practices to prevent pressure ulcers included evidence- and
consensus-based guidelines, how-to guides from national organizations, and comprehensive
frameworks from well-recognized wound organizations.
A national campaign by Advancing Excellence in Americas Nursing Homes provides an
implementation guide that includes efficient, consistent, evidence-based approaches to address
the prevention and minimization of pressure ulcers.5 The coalition recommends that nursing
homes seeking to identify areas for improvement in processes and practices should verify
whether the homes current policies and protocols are consistent with current evidence-based
approaches, (i.e., new National Pressure Ulcer Advisory Panel [NPUAP] guidelines). In 2008,
the Joint Commission included healthcare-associated PU prevention as a National Patient Safety
Goal (NPSG) for long-term care. Elements of performance for this NPSG include using a
validated risk assessment tool, reassessing PU risk at intervals defined within the organization,
and educating staff on how to identify risk for and prevent PUs.6 The Institute for Healthcare
Improvement describes six key evidence-based care components in its How-to Guide: Prevent
Pressure Ulcers. Essential elements include making risk and skin assessments (upon admission
and daily), managing moisture, optimizing nutrition and hydration, and minimizing pressure.7
A search of the National Guideline Clearinghouse identified recommendations to prevent
and manage PUs by many well-respected organizations including the Wound, Ostomy and
Continence Nurses Society (WOCN).8 Preventive interventions based on Level B evidence
212
include scheduling regular repositioning and turning for bed- and chair-bound individuals and
using pressure redistribution surfaces in the operating room for high-risk individuals.
We identified two frameworks for Patient Safety Practices that appear to thoroughly embody
components used in PU prevention initiatives today (see Table 1 and Table 2 following). A
recent systematic review9 describes the Indiana State Department of Healths classification of PU
initiative components, as follows:
Organization components include team makeup, policies and procedures, ongoing quality
evaluation processes, educating staff, utilizing skin champions, and the development
and system-wide communication of the written care plan
PU prevention components include risk and skin assessment, moisture management,
nutrition and hydration optimization and pressure management
Care coordination components include the establishment of regular meetings to facilitate
communication, collegiality, and learning
The ABCDE of Pressure Ulcer Incidence Reduction Initiatives was outlined at the 12th NPUAP
Biennial Conference held on February 2011. The initiatives were described as: administrative
support backed by support at the patient care level; bundling care practices and having an
identifiable theme; creating a culture of change, commitment, and communication;
documentation of PU prevention practices must be visible; and education is essential.10
213
Table 1, Chapter 21. Components of pressure ulcer prevention trials in U.S. hospitals, 2000 to 2011
Study
Implement Review
Protocol
Wound
Care
Products
Upgrade
Automated
Systems
Integrate
New
Reporting
Education/ Risk
Skin
MultiAudit and
Training
Assessment Champion disciplinary Feedback
Tool
Team
Lynch and
Vickery
11
2010
Young et al.
12
2010
Bales et al.
b,d13
2009
Chicano and
Droishagen
b14
2009
Walsh et al.
15
2009
Dibsie L.
16
2008
McInerney J.
17
2008
Ballard et al.
18
2007
Catania et al.
19
2007
LeMaster K.
b,c20
2007
Courtney et
21
al. 2006
Gibbons et al.
d22
2006
Hiser et al.
23
2006
X
X
X
X
X
X
X
X
X
X
214
Shading = reported a significant reduction in pressure ulcer rates

a
Audit only.
b
Reduced prevalence/incidence to zero.
c
Describes role of CNS as a direct consultant.
d
Describes use of incentives.
X
X
X
X
Lyder et al.
24
2004
Stier et al.
25
2004
X
X
a
Table 2, Chapter 21. Components of pressure ulcer prevention trials in long term care, 2000 to 2011
Study
Implement Upgrade
Integrate Education/ New
Use of
Skin
MultiProtocol
Training
Champion disciplinary
Automated New
Assessment Outside
Systems
Reporting
Tool
Team
Consultants
*
Horn et al.
26
2010
Rantz et al.
b27
2010
Milne et al.
d28
2009
Tippet A.
c,d29
2009
Rosen et
c30
al. 2006
Abel et al.
31
2005
Ryden et
32
al. 2000
Rantz et al.
a33
2001
Audit and
Feedback
X
X
X
X
X
X
Shading = reported a significant reduction in pressure ulcer rates

*Examples include advanced practice nurses,28 physician wound consultant,29 and state quality improvement program staff.27,31 Services included
identifying team leaders/multidisciplinary teams,26 streamlining documentation,26 educating staff,29 providing evidence-based tools (i.e., assessment
cards),31 team leadership and technical assistance29
a
Randomized controlled trials.
b
Nonrandomized controlled trial.
c
Describes use of incentives.
d
Reviewed wound care products.
215

Age, immobility, incontinence, inadequate nutrition, sensory deficiency, multiple
comorbidities, circulatory abnormalities, and dehydration are a handful of the more than 100
factors that have been identified as placing adults at risk for developing PUs.2,34 In addition to
having many risk factors, PUs can develop very quickly. PUs have been documented as
developing in just 1 hour.12
However, despite the many risk factors and the quick development of PUs, they can be
successfully prevented with several strategies. Improvements in care processes (e.g., skin
assessments) and patient outcomes (e.g., incidence, length of stay) have resulted from singlecomponent PU prevention initiatives such as a turn-team nursing program,35 an educational
intervention,36 and a tracking system.37 A recent systematic review concluded that using support
surfaces, repositioning the patient, optimizing nutritional status, and moisturizing sacral skin are
appropriate strategies for preventing PUs.4
Use of intervention bundles has also been effective in eliminating PUs.38 The concept of
bundling care practices is credited to the Institute for Healthcare Improvement. Defined as a
structured way of improving processes of care and patient outcomes, a bundle typically includes
three to five evidence-based practices that when performed collectively and reliably, have been
proven to improve patient outcomes.39 Although successful in and of themselves, bundling care
practices is only one of several important components listed in the ABCDE of Pressure Ulcer
Incidence Reduction Initiatives.
During the NPUAP keynote address, former president Elizabeth Ayello mentions the
importance of a partnership between administration and bedside caregivers. She describes the
administrations role in making PU prevention a priority by providing adequate resources and
infrastructure and listening to staff about how to implement best practices.10 Jankowski and
Nadzam (2011) concur on the importance of facilities administration in their quest to identify
gaps, barriers, and solutions in implementing PU prevention programs.9 Additionally, NPUAP
lists creating a culture of change, commitment and communication as paramount to reduction
initiatives. Training and communication among turn-team members, the enterostomal staff, and
clinical nursing directors was critical for the success of one initiative.35
Both single and multicomponent programs described the importance of adding
documentation and education into PU prevention initiatives. Challenges encountered during the
implementation of a tracking system included manpower resources and documentation.37
Jankowski and Nadzam (2011) reported that major barriers to protocol implementation were
related to documentation. They indicated that although every hospital had a written PU protocol
and used the Braden Scale for Predicting Pressure Sore Risk, none of the hospitals routinely
included the risk scores or PU prevention care plans in shift-to-shift reports, RNs-NA reports,
RN-physician communications, or other handoffs between hospital staff (e.g., staff nurse to
transporter).9 A 2009 ECRI Institute risk analysis on PUs40 recommends that to establish
mechanisms of effective communication between facilities, include the following for all transfer
documentation: (1) standardized location of information; (2) current risk assessment; (3) skin and
observed wound assessment; and (4) current interventions (if applicable).

We limited our research to studies implementing multicomponent initiatives in acute (k = 15)
and long-term care settings (k = 8) in the U.S. from 2000 to the present (see evidence tables in
216
appendix). Study designs were mostly time series assessments of changes before, during, and
after implementation of the intervention. Other study designs included randomized controlled
(k=2)27,33 and controlled before-and-after (k=1).32 A majority of the studies focused on universal
prevention (all risk levels); one focused on high-risk patients.26 Pressure ulcers were the primary
focus of 20 studies and part of a comprehensive approach in three.27,32,33
The review group agreed not to assess risk of bias or rate the strength of evidence for those
reviews primarily focused on implementation. Therefore, in this section, we briefly summarize
the primary results; subsequently, we provide detailed assessments of the implementation efforts.
Acute Care
Evidence presented below on PU prevention programs implemented in the acute care setting
is based primarily on one systematic review (Soban 2011) of nurse-focused quality improvement
(QI) initiatives.41 Of the 39 studies included in the review, 12 met our inclusion criteria.
Additionally, we discuss three other studies that were published since the Soban review was
completed.
The Soban review41 had three objectives: describe the intervention strategies used, describe
the process and outcome measures reported, and examine the interventions effects on outcomes.
Study findings were categorized as processes of care (e.g., staging of acquired stage 2 PUs)
and/or patient outcomes (e.g., PU incidence). Eleven studies reported patient outcomes; only one
study24 reported both. Because the review included limited data, we accessed information
directly from the studies.
First, we examine intervention effects on several processes of care reported in one
multihospital QI collaboration overseen by the Connecticut Quality Improvement Organization,
Qualidigm. In 2004, Lyder et al.24 reported a 2-year follow-up on 14 measures, seven of which
were process of care measures. Four plan-do-study-act (PDSA) improvement cycles
implemented at 17 hospitals resulted in significant increases in identifying high-risk patients
(20.3% vs. 35.3%, p<0.001); repositioning of bed-bound patients every 2 hours or every hour in
chair-bound patients (50.9 vs. 56.9, p = 0.01); use of nutritional consults in malnourished
patients (34.3% vs. 48.6%, p<0.001); and staging of acquired stage 2 or greater PUs (22.4% vs.
44.2%, p<0.01). No statistically significant findings were reported for the remaining processes of
care (including staging of acquired stage 1 PUs) or hospital-acquired incidence rate (baseline vs.
follow-up; 20.6 vs. 20.8, p=0.90).
Five of the 11 remaining studies reported in the Soban review conducted nurse-focused
initiatives facility- or system-wide. A majority of studies reported on prevalence or incidence
measures; both types of measures have been described as useful in assessing and improving
patient care Catania indicates that a declining incidence of PUs would indicate that a prevention
program is working to decrease new PU cases, while a declining prevalence indicates that the
treatment strategy was also having an impact on the duration of PUs.19 Several studies reported
that initiatives reduced prevalence or incidence to zero.
After a 10-month implementation of the SKIN (Surfaces, Keep the Patient Turning,
Incontinence Management, Nutrition) bundle, Gibbons et al.22 reported a 90% reduction (5.7% to
0.448%) in prevalence at the Nations largest Catholic and nonprofit health system.22 Similar
reductions were reported at a 548-bed, two-hospital system in Southwest Florida:17 a 5-year trend
analysis indicated a significant reduction in PU prevalence (overall [-81%] and ulcers located on
the heel [-90%]) after proactive assessment and management of at-risk patients. One study
reported zero PU prevalence and incidence after 1 year of a nurse-focused initiative at a 300-bed
217
community hospital.13 Other benefits listed were optimal patient care and avoiding the cost of
treating stage 3 or 4 ulcers.
In 2006, Courtney (SOS program/Six Sigma methodology) reported that one 710-licensed
bed, multisite, not-for-profit hospital reduced the PU incident rate from 9.4% to 1.8% over 3
years. Incidence was reduced by 6.3% after only 1 quarter.21 Two years earlier, Stier et al.25
reported reducing incidence by more than 50% at a 5,600 bed nonprofit health care system.
Significant improvements were also reported from initiatives implemented in patient care
units, with two studies reporting zero prevalence postimplementation. In 2006, a two-unit
intensive care unit (ICU) significantly reduced PU prevalence (34% to 8%), noting that National
Database of Nursing Quality Indicators (NDNQI) benchmark data were instrumental in helping
our unit focus on PU prevention, ultimately leading to improved patient outcomes.18 Catania et
al.19 reported reducing PU prevalence by more than 50% due to implementing the Pressure Ulcer
Prevention Protocol Interventions in five in-patient units at one cancer hospital. From September
2004 (baseline) to June 2006 (postimplementation), the percentage of patients with all types of
PUs and with hospital-acquired pressure ulcers (HAPUs) was reduced to 4% and 2%,
respectively. NDNQI benchmarks at the time were 12.65% for all ulcers and 6.84% for HAPUs.
In a 2007 study, preimplementation PU prevalence rates for a pulmonary and oncology unit
were 9% and 12%, respectively. LeMaster20 reported that nurse-focused QI initiatives reduced
prevalence in these two hospital units to zero. Rates were reduced from 9.2% to 6.6% in five
units in one Florida hospital.23 One medical ICU, which reportedly had the highest HAPU
prevalence (average of 29.6%) among the participating units, reduced prevalence to zero.
Dibsie et al.16 reported a facility-wide reduction in percentage of patients with stage 2 or
greater HAPUs (4.2% vs. 3.2%) in four adult critical care units (54 beds) at two U.S. hospitals.
Rates for the surgical intensive care unit, however, did not improve over time (6.1% pre- and
postimplementation). A 23-month initiative (Chicano 2007) at a 25-bed intermediate care unit
reduced incidence.14
Benefits described in three separate studies (Walsh et al., Young et al., and Lynch and
Vickery) included a reduction in prevalence and incidence and improvements in processes of
care. In 2009, Walsh et al. reported a reduced PU prevalence from 12.8% to 0.6% from an 18month initiative. Walsh also reported increased focused communication among patient
caregivers; once improvements were noticed, clinicians behavior and clinical processes
improved.15 Young et al. reported several successes, including streamlining online policies (from
7 to 1) and reduction in time spent documenting skin care. Young also reported clinically
relevant reductions in development of nosocomial PUs.12
In one year, PU rates were reduced by 82.8% (2.8% to 0.48%) at one rehabilitation hospital.
Lynch and Vickery reported that streamlining documentation increased timely and accurate
completion of documentation from 60% to 90% in 90 days. This facility also increased
patient/family involvement in patients care by providing an educational brochure and reviewing
interventions on admission.11
Long-Term Care
A total of eight studies met our inclusion criteria in their evaluation of multicomponent
programs to prevent PUs in long-term care facilities. Study duration ranged from 6 months to 6
years. Among these eight programs, four reported significant reduction in PU incidence or
prevalence rates.
218
In 2010, Horn et al. reported on three main outcome measures.26 First, they considered
prevalence of PUs using Centers for Medicare & Medicaid Services (CMS) quality measures
(QMs). Based on data from seven facilities, they report that the CMS high-risk PU QM
decreased from 13.0% (baseline) to 8.7% (12 months postimplementation). A second outcome of
interest was the number of in-house acquired PUs. The average number of in-house PUs (all
stages) per facility was reduced by 62% (12.1 [baseline] to 4.6 [postimplementation]). Lastly,
Horn et al. reported a 53.2% reduction in the average number of certified nursing assistant
(CNA) documentation forms and a mid-90% completeness rate of CNA documentation.
Tippet et al. reported that nosocomial PU ulcers were eliminated after 6 months.29 They also
reported an 86% reduction in average incidence: 5.19% (preinitiative) vs. 0.73% (postinitiative);
p<0.0001. By the end of the fourth year, both incidence and prevalence were reduced by 99%.
In 2006, Rosen et al. reported a significant reduction in PU incidence.30 The percentages of
patients identified as high-risk were 22.3% and 28.0% at the baseline and intervention periods,
respectively. Significant reductions in PU incidence were reported for stage 1 and beyond
(P<0.001) and stage 2 and beyond (p<0.05). However, these improvements were not sustained
during the postintervention periods when no weekly reports (indicating completion of training)
were provided; no targets or goals were established; and no financial incentives were offered to
staff.
One RCT was conducted in three privately owned facilities in the midwestern United
States.32 This 6-month study evaluated the effectiveness of advanced practice nurses (APNs) to
successfully implement scientifically based protocols for PUs and other clinical problems. The
APNs delivered treatment in two facilities; in the third facility, patients received usual care. At
6 months, the percent of APN-treated patients with PUs was significantly reduced (19.8% vs.
3.5%; p=0.04). The percentage of patients with PUs for the usual care group was also reduced,
although not significantly.
Four studies conducted in long-term care facilities reported no significant findings from
primary analysis. At 12 months, Rantz et al.27 reported relative improvements in high-risk PU
scores (negative indicating improvement) were -53%, -12%, -5%, and +435% for Group 1, 2, 3,
4, respectively. At 24 months, relative improvement was -3%, -8%, +59%, and +105% for these
same groups, respectively.
Abel et al. describe results from a 2-year study conducted in 20 sites.31 They report a
significant improvement for 8 of 12 PU-related quality indicators; however, only a trend toward
a lower incidence of facility-acquired PUs (x2 MH = 3.66, p = 0.06) was observed. Facilities also
fell short in two other key measurements: proportion of high-risk residents with facility-acquired
PUs whose care plan intervention reflects treatment orders and proportion of skilled nursing
facilities that have a wound protocol.
Milne et al.28 reported several successes. The facility-acquired PU prevalence rate at baseline
was 41%. PUs were reduced to <3% on two units due to increased monitoring of modified nasal
cannula (pulmonary unit) and increased attentiveness to heel offloading, support surfaces, and
proper positioning (SCI/trauma unit). Of the 396 charts reviewed, fewer than 1% had missing
data. A review of 45 patient charts showed that wound teams consistently determined staging
and wound etiology in more than 90% of cases. The facility-acquired PU rate was reduced by
37% within 1 year postimplementation.
In 2001, Rantz et al.33 randomly assigned 87 facilities to receive workshop plus feedback
reports (Group 1); workshop, feedback reports, and clinical consult (Group 2); and control
219
(Group 3). Primary analysis indicated no statistically significant findings for prevalence of stage
1-4 PUs or prevalence of stage 1-4 PUs (low-risk residents).

There have been no reported harms of the PSP.

What Contexts?
We examined studies of multicomponent PU prevention programs for information on
contexts. The description of contexts is limited: most reports contain information on certain
contextual factors but lack information on other factors. The limited evidence available is
summarized below.
Use of a Model or Theory

Of the 15 studies conducted in acute care facilities, only four programs described a model or
theory as the basis of their implementation strategy. The PDSA framework was a methodology
used in 17 hospitals in Connecticut.24 The four PDSA improvement cycles involve identifying
the problem and designing an intervention (Plan), implementing change (Do), evaluating the
collected data (Study), and implementing what was learned (Act). Lyder et al. indicated that
these processes should be developed rapidly to sustain momentum in changing behaviors,
procedures, and policies as quickly as possible. Although described as being used on a
hospital-wide scale, this framework may also be applied on a smaller scale such as a hospital
unit.
Courtney et al. integrated Six Sigma methodologies into treatment processes developed for a
multisite, not-for-profit facility. Described as a data-driven quality strategy for improving
processes, DMAIC consists of five interconnected steps: (1) defining the problem, (2) measuring
the performance; (3) analyzing the data, (4) improving the process, and (5) controlling change.42
Young12 stressed the importance of empowering staff at point of care, which suggests a model
of shared governance where decisions are made at the point of service. A shared governance
model was also employed at an Illinois-based intermediate care unit.14 Use of quality council
members or a self-managing work team has been described as a second-generation shared
governance model. According to Nursingworld.com, a self-managing work team such as the
quality council members are jointly responsible for achieving goals, lead themselves, and thus
have authority and control over the work and access to information.43
Of the eight long-term care studies, one study27 referenced effectiveness of similar
components in a previous study;33 another described using the failure mode and effects analysis
(FMEA) developed by the U.S. military.28 FMEA is defined as a systematic process for
identifying potential design process failures before they occur to eliminate them or minimize
risk. The basis for one program was Havelocks (1974) model of effective research utilization.
This model is described as integrating knowledgeable resource individuals as links between
relevant sources of scientific knowledge and the user (e.g., staff).32
External Factors Motivating Attention to Pressure Ulcer Prevention

Most studies in acute care facilities reported feeling pressure from impending changes in
CMS reimbursement. Specifically, subsequent to passage of the Deficit Reduction Act of 2005,
220
CMS will no longer allow higher DRG (diagnosis-related group) payments for patients with
stage 3 and 4 HAPUs. Catania reported that one dedicated cancer hospital was responding to the
identification of two stage-4 PUs and evidence from the NDNQI survey that the prevalence of
PUs in the hospital exceeded the national benchmark by close to 50%.19 A 25-bed intermediatecare unit indicated that identification of high prevalence rates, nursing peer reviews, chart audits,
and unit observations played an important role in the hospitals response.14 Lynch mentioned that
the facility experienced an upward trend in PUs on two units.11
One 528-bed nonprofit facility, at which prevalence of HAPUs was lower than national
norms, set out to eliminate HAPUs completely.22 In 2006, Courtney et al.21 described the
emergence of new guidelines from the American Nurses Association and the Agency for
Healthcare Research and Quality as showing a revitalized interest in preventing and treating
PUs. Additionally, studies using the Nursing Care Quality Initiative guidelines revealed high
prevalence of PUs (13%) and lack of documentation and management. Two critical incidents
(not specified), concerns within individual units, and inconsistent documentation were listed as
external motivators by Dibsie et al.16 Additionally, the frequency with which concerns and
incidents were discussed, but went unreported within the internal reporting system were of
concern. Young reports stakeholder commitment to improve patient outcomes and a goal to be
recognized as a quality provider of patient services.12
External factors influencing the staff at one 151-bed Midwest skilled facility were a G-level
citation (a deficiency judged to cause actual harm to residents) and State survey deficiencies.29
This facility recorded PU prevalence and average incidence rates as high as 25% and 23.9%,
respectively. One facility reported receiving multiple citations from the Department of Health.30
Abel et al.31 indicated that the 20 participating facilities were identified from 143 Medicarecertified skilled nursing facilities with high rates of PUs despite a high volume of residents
receiving preventive care.

Organizational characteristics described in all studies included financial and academic status,
location, and size. Bed range for acute care studies ranged from 18 to 800. Settings included a
community hospital, a multisite academic medical center, and a cancer hospital. Implementation
studies included as few as two units and as many as 17 hospitals.18,24
Long-term care programs were conducted in not-for-profit facilities;26,30 a privately owned
facility,32 a Midwest-skilled nursing facility,29 and a mix of for-profit, not-for-profit, and
governmental facilities.27,33 Other studies were conducted in Medicare-certified skilled nursing
facilities31 and in a 108-bed long-term acute care facility.28
Two studies published in 201026,27 were conducted in seven and three states, respectively; 20
studies were conducted in one state. The number of facilities included in each study ranged from
1 to 87. Bed size ranged from 1 to 60 in one study33 and from 44 to 432 in another.26 Only the
highest-risk units (three maximum) participated in one program.26 Five studies indicated prior
experience with QI or presence of electronic medical record (EMR).26-28,30,33 Two studies
reported on organizational complexity.30,32 Of the 23 included studies, only two studies described
patient characteristics.29,32
Teamwork/Leadership
Although a majority of studies utilized a multidisciplinary team, skin champions were
described as key team members. Studies set in acute care settings described use of certified
221
wound ostomy continence nurses (CWOCNs),13,18,23 staff registered nurses or patient care
technicians,21 clinical nurse specialists (CNSs),19 and a collaboration between CNSs and wound
ostomy nurses20 in this role. Of eight studies set in long-term care settings, skin champions were
designated in five studies;26,28,29,31,32 in two studies, advanced practice nurses served in this
role.28,32
Three studies included lengthy descriptions of leadership within their facilities. Stier et al.
described leadership support to multidisciplinary teams at a 5,600-bed nonprofit New Yorkbased health care system.25 Teams consisting of clinical experts from 18 facilities convened to
openly discuss the various risk assessment tools and facility protocols in place. Multidisciplinary
teams agreed to develop a uniform policy, skin care formulary, and specialty bed contract.
System leadership (e.g., nurse executives, quality management directors, and senior physicians)
provided support to the team at both the system and facility level vis-a-vis resources, ensured
staff orientation and education, maintained quality control programs, and continually assessed
actions to improve performance through system-wide care committee meetings.
Dibsie16 described the importance of teamwork and leadership at a multisite academic
medical center. Discussions on serious skin-related issues were held with unit nursing
management, immediate senior nursing management, and selected peers. Discussions later
involved a larger group of managers and clinical specialists after it became evident that the
issues crossed many areas and could be better handled by the group together. When necessary,
senior management stepped in to stress the importance of resolving issues related to
preventionthroughout the organization.
Young et al. described a change in leadership at a 540-bed acute care facility in Indiana.
Clinicians were initially wary of managements intent for clinician involvement. Their
hesitation was attributed to past experiences when some clinicians joined the task force to attain
required activities relating to clinical advancement or in response to a managers request. As a
result of mandates that the new skin team be clinician-led, the majority of the original task force
members leftThe few remaining committee members were charged with selecting new task
force members who could serve as unit champions.12
Patient Safety Culture

Several studies provided a glimpse of the patient safety culture that existed before programs
were implemented. Staff at a 528-bed nonprofit hospital believed that PUs were unavoidable in
complex, critically ill patients. At this facility, chart reviews of 30 patients who developed PUs
indicated that 87% of the time a nutritional consult had been ordered, but nutritional
recommendations were only followed 35% of the time.22 McInerney17 reported a high prevalence
of PUs at a two hospital system; greater than 50% of ulcers were located on the heel. Further
review revealed that physician and nurse reluctance to use a rigid boot was the root of the
problem.
Lyder et al. reported that most hospitals participating in a multihospital QI collaboration
did not believe that PU prevention was a huge priority.24 In 2010, Lynch11 discussed several
process issues at a 166-bed acute rehabilitation facility. A review of 2007 data indicated many
misidentified PUs at admission, incomplete and inconclusive skin assessments, incorrect staging,
and inconsistent documentation of interventions.
At the unit level, staff at one medical ICU also believed that PUs were inevitable among
seriously ill patients.23 Staff at a two-unit ICU did not believe the prevalence data, stating the
higher acuity patients were more likely to develop skin breakdown.18 Accountability,
222
knowledge deficit, and communication deficits were identified as root causes of reported high
incidence/prevalence at one teaching hospital. Incomplete initial and ongoing skin assessments,
inconsistent implementation of prevention interventions, and lack of coordination among staff
were cited as examples of preimplementation safety culture.21 Lastly, analysis of survey results
at one 25-bed intermediate care unit revealed that admission documentation did not identify
patients with an increased risk for developing PUs.14
In the long-term care setting, five studies reported minimal information on patient safety
culture at the organizational level.26,28-31 One study included information at the unit level.28
Abel31 reported that facilities were plagued by inadequate assessments and data omissions
associated with risk. Milne indicated that one facility had above-average PU prevalence, used a
faulty EMR that was inconsistently used by clinicians, and had deficient documentation of risk
assessment.28
Implementation Tools
Below, we describe examples of unique tools that were used for audit and feedback,
education and training, monitoring progress, identifying specific groups of patients at risk, and
streamlining products and processes in more than 20 PU prevention initiatives. For a complete
listing of implementation tools, see evidence tables in the appendix.
Audit and Feedback

Audit and feedback were mentioned as key elements in initiatives implemented in long-term
care studies. In one study, facilitators provided direct feedback to CNAs regarding data
inconsistencies by unit and by shift to help track progress.26 Real-time management feedback in
Rosen et al.30 consisted of a prominently displayed graphic thermometer tracking weekly PU
incidence, and positive ($10 reward) or negative reinforcement (termination). Weekly informal
feedback by nursing supervisors,29 formal weekly walk-rounds11 and frequent patient positioning
audits were also used during implementation.29 See below for more detailed descriptions
included in acute care studies:
Identification of skin breakdown must be reported within the electronic system, and
weekly surveillance summaries need to be shared with administration.14
Feedback was provided during weekly SKIN operations meeting where unit leadership
reported compliance with the SKIN bundles and related issues.22
The team provided clinical staff with consistent and frequent feedback about the results
of prevalence studies for their specific units so they could benchmark their results over
time. This immediate and ongoing feedback helped engage staff members in their
program and allowed them to take credit for the improved clinical outcomes. To reinforce
the positive changes, medical ICU staff members were given a certificate for the Most
Improved Unit.23
While providing feedback to nursing staff, the CNSs attempted to balance compliments
for a job well done with recommendations for improvement.20
Education and Training

The majority off the 23 studies reported including some form of education or training.
Training was reported as mandatory in four studies.12,13,29,30 Only one study reported on duration
of training (40 minutes).30 Unique tools used in education and training sessions are described as
follows:
223
Guest speakers discussed new concepts and educated physicians about the CWOCNs
role, level of clinical expertise, and best-practice wound care interventions.23
During 30-minute mandatory continuing education sessions, participants sit on bedpans
that act as a reminder that PUs can occur within as brief a time as 1 hour.12 Compensation
was provided to staff participating on their own time. Educational content was tailored
for RNs, LPNs, secretaries, and CNAs. A post-test survey measured effectiveness of
presentation.
Staff education included critical thinking using case studies and role-playing exercises.19
Staff participated in skin-care training using an interactive video.30
Identifying Specific Groups of Patients at Risk
McInerney et al.17 reported that using computerized charting and order entry helped
identify specific groups of patients at risk for developing PUs. A consult with a specially
trained nurse is automatically generated when patients are identified as high/very high
risk. Consults are also generated for two other patient groups considered high risk
(e.g., patients placed on a ventilator and patients receiving hemodialysis).
Monitoring Progress
To monitor progress at a large 528-bed hospital, a SKIN Bundle Compliance Tool was
developed. Nutrition-related items include noting completion of a nutrition consult,
orders written, and orders carried out.22
One rehabilitation hospital posted report cards on every unit allowing staff to track
progress against other units and unit goals.11
Streamlining Products and Processes
A large nonprofit health care system streamlined a skin product line, trimming it from
100 products to 24. Standardizing products controlled costs and reduced training.25
Four critical care units conducted extensive in-service education on a new wound care
product line and made vendor support available. Vendor clinical experts were available to
educate staff on new products. Dibsie recommends informing clinicians when
modifications are made to the skin protocol or product line.16
Nursing leadership, nursing staff and those from other disciplines (e.g., nutritionists,
respiratory therapists) compared current policies and procedures to clinical practice
guidelines. The Director of Informatics facilitated revisions12 of seven existing policies
into one.
Barriers
Reported barriers to implementing PU prevention programs included expansion of the
initiative to a larger scale,16 unmotivated staff,13,14,19 staff turnover,26,31-33 staff resistance,24,31
limited resources,12 inconsistent documentation,13,20,31 difficulties in exporting data,26and
miscommunication between electronic systems.20 One facility also faced increased PU rates.28
Examples of barriers described and ultimate solutions are as follows:
224
Acute Care
Dibsie16 reported that expanding initiatives from a 20-bed critical care unit to all nursing
units in two sites provided an extra challenge. Obstacles included the coordination of a
skin committee, difficulty in coordinating schedules, and keeping abreast of new
equipment that can contribute to PU development.
Chicano14 reported a challenge motivating staff who were relatively uninvolved in
planning and implementing initiatives. Perseverance of council members encouraged the
staff to finally accept the concept of shared governance and acknowledge that it can
positively influence patient care.
During the early stages of one initiative, staff members were not sufficiently committed.
Although a QI analysis indicated that two stage 4 PUs were the result of inconsistencies
in or lack of documentation, staff awareness and assessment, staff members denied that
the PUs were due to poor nursing care. Catania indicated that overcoming it took
additional education, mentoring, and support at the unit level.19
As reported by Lyder et al.,24 hospitals identified as a major barrier the view that PU
prevention was a nursing issue. The medical staff was reportedly the most resistive when
asked to play a role in PU prevention. Due to this mindset, considerable time was spent
re-educating various disciplines about their roles in PU prevention.
Young et al. reported that clinicians complained of time constraints, insufficient
computer resources, and competing goals. To address these concerns, clinicians were
allocated 4 paid hours to carry out responsibilities related to PU prevention, and web
access to library resources was added to the organizations intranet.12
Bales et al.13 reported unmotivated staff and lack of proper reporting and documentation.
Monthly to quarterly campaigns were launched to maintain staff motivation. Nursing
units that had zero HAPUs were recognized and awarded during campaigns.
LeMaster20 reported Braden scores (a scale for predicting PU risk) were not documented
at 100% according to policy. Patients were missed because of a communication failure
between two different electronic documentation systems. To address this, the facility
transitioned to a single, universal electronic record system.
Long-Term Care
Horn et al. reported difficulties at one of 11 facilities in exporting data elements.26 As a

result, the facility could create only one of four possible clinical decision-making reports.
Additional issues concerned the preparation of documentsspecifically forms needing
the residents study ID number and faxing forms for report generation. Staff turnover,
especially for director of nursing, also seemed to slow program momentum. To overcome
these barriers, suggestions included adding new CNA documentation processes into
orientation programs, phasing in use of documentation, and developing a strong
multidisciplinary team to lead improvement efforts. Ryden et al.32 reported barriers to
their implementation program included high turnover (range of 11% to 45%) of
unlicensed staff.
Barriers faced by 20 facilities in Texas included incomplete admission assessments, staff
reverting to previously unsuccessful practices, and inappropriate use of monitoring
systems. Staff resistance, staff turnover, and variations in new staff orientation also
affected program implementation. Monthly visits by a State Quality Improvement
225
Organization (TMF Health Quality Institute) and improving performance may have
helped resolve these issues.31
Barriers reported by Milne et al.28 included a climb in PU rates once strict monitoring of
processes was leveled off from weekly to monthly for 1 month. To overcome this barrier,
the wound team increased presence on the units, monitored charts more closely, provided
feedback to staff, and scheduled biweekly prevalence rounds.
Rantz 200133 indicated that staff turnover, especially the nurse coordinator, cancellations
of site visits at the last minute, and teams mired in the MDS [minimum data set]
assessment process and coding issues impeded initiatives. To address some of these
issues, multiple nurses were assigned responsibility for processes, accomplishments were
posted, and a quality manager was placed on staff to support care delivery.
Sustainability
Several acute and long-term care facilities reported sustainability of PU prevention programs.
Conducting quarterly prevalence studies and continually monitoring all HAPUs were key to
sustaining improvements at a large nonprofit health system. A focus on skin pigmentation and
the development of a skin fragility assessment tool were the most recent strategies introduced.22
McInerney17 indicated that publicizing improvements in PU rates will keep staff focused on
prevention efforts.
One 710-bed multisite facility reported that the overall culture change at the medical center
remains a work in progress. Therefore, PU incidence continues to be a measurement used in
organization-wide scorecards and staff bonus programs. Measuring performance routinely has
since become a priority facility-wide and at sister facilities.21 LeMaster20 reported a request by
staff members for items to provide both visual and auditory clues to sustain improvements at
their facility. Dibsie et al.16 reported staff members continued to meet on a monthly basis to
discuss skin issues, participate in quarterly prevalence data collection, and learn from the
medical centers expert WOCNs as well as from one another.
Walsh indicates the importance of maintaining gains, including keeping current regarding
initiatives for improved patient safety, changes in regulatory mandates, and changes in EBP
[evidence-based practices]. At a 540-bed acute care facility, RNs and LPNs must demonstrate
competency annually; monthly updates provided via intranet to staff include product changes.12
One rehabilitation hospital printed quarterly newsletters and attached them to paychecks. The
newsletters described findings, results, and new initiatives in PU management.11 According to
Bales et al.,13 sustainability requires awareness of key management skills and priorities, such as
strong leadership, involvement of staff in decision-making, and a desire to foster
interdisciplinary relationships.
Eleven long-term care facilities across seven states stated that managing the manual data
collection, faxing forms to the project office and creating clinical reports for distribution back to
the facilities on a weekly basis could not be maintained over the long term. By date of study
publication, more than 70 additional facilities were participating in the On-Time program.26 A
wound care coordinator position29 and a wound care committee32 were established to help sustain
program gains in two long-term care programs. Lastly, one 108-bed LTC facility noted that two
wound coordinators provided additional education. Monthly review of documentation and the
presence of multiple PU prevention interventions on units also helped to sustain improvements.28
226
Lessons Learned
Several programs implemented facility-wide reported key lessons learned. Recommendations
from one large nonprofit health system include the following: (1) use current leadership to
support staff; (2) disregard a spike in reported skin breakdown, which is probably due to staffs
increased awareness, education and reporting; (3) make staff accountable for success of
initiatives; (4) do not assume that the knowledge base between disciplines is equal; (5) stay on
task and celebrate successes.22
Seventeen hospitals reportedly liked the idea that PDSA cycles could be completed using a
small number of cases to identify improvement areas and implement potential interventions at a
reduced scale before implementing programs at full scale, saving time and human resources.24
Lessons learned in Courtney et al. include the following: (1) identify and involve the process
owner (unit manager) and S.O.S. (Safe our Skin) unit champion early in the project; (2) develop
a detailed training plan to include delegation and team work; (3) communicate the changes to
everyone; (4) streamline the process to make it as easy as possible; (5) celebrate success, give
recognition, spread the success; (6) define roles and responsibilities; (7) realize that you cannot
fix everything at one time; (8) support and commitment from senior leadership are critical to the
success and help to sustain the gains.21
Abel et al. reported lessons learned by 20 Medicare-certified skilled nursing homes while
working collaboratively with TMF, a State quality improvement organization.31 A TMF
representative provided the following recommendations:
Strive to transition from quality assurance to QI, moving from defect detection to defect
prevention, while making continuous improvement in the process of care delivery.
Incorporate interventions designed to address barriers to preventive care while sustaining
existing processes that have proven effective.
Publicly recognize front line staff successes.
Share data often.
Implement change on one unit/hall at a time to ensure staff buy-in.
Accept failure in the systems as an opportunity to improve.
Operationalize systems that use continued measurement to monitor and improve
performance that is reported to be below a designated threshold.
Allow staff to maintain a level of autonomy during design of the intervention.
Ensure accountability in following agreed-upon process changes.
Ensure consistency in a formalized staff training/orientation related to documentation
requirements.
Include staff that directly affect the process in the intervention design.44
Are There Any Data About Costs of Pressure Ulcer Prevention

Programs?
Five studies included information on costs of PU prevention programs: two in acute care
facilities and three in long-term care facilities. In 2008, McInerney17 indicated millions of dollars
in cost savings due to the significant reductions (-81%) in PU prevalence at a two-hospital
system in Naples, Florida. A conservative analysis (assuming savings of $3,000 per case) found
that this 548-bed system saved approximately $11.5 million annually as a result of the
program.
227
A Midwest 710-bed teaching hospital estimated additional cost per case (in 2001) was
$3,037, and the additional overall cost related to PU development was $4,877,000. Based on
these figures, a reduction in the number of HAPUs by 50% to 5% would reduce overall costs by
$2,438,000.00. This amount would not preclude the anticipated significant improvements in
patient satisfaction, length of stay/quality, staff awareness of skin integrity issues, and risk
management issues.21
The most recently reported costs in a long-term care setting were from 2010 in a four-group
comparison study that assessed the value of a bedside electronic health record PU initiative.27
Total costs for the 3-year evaluation for the groups of facilities implementing technology
increased $15.11 (12.5%) for Group 1 and $16.89 (9.6%) for Group 2, while those for the
comparison groups did not. Rantz et al. attributed cost increases to the cost of technology,
including maintenance and support and ongoing staff training to use the EMR system.27
In 2009, a 151-bed Midwest skilled facility described cost savings 4 years after program
implementation. The authors estimated cost savings per PU/per month was $1,617; monthly
savings, $10,187; and yearly savings greater than $122,000.29 Lastly, in 2006, Rosen et al. stated
based on a mean cost of $2700 to treat a single stage II PU,45 reducing the incidence of ulcers
by approximately 15 over 12 weeks would yield savings of approximately $40,000. Less than
$10,000 was distributed as financial incentives.
What Is Known About the Effect of Context on Outcomes?

Several studies directly commented on the effect of context on outcomes, and all studies
specifically mentioned the influence of staff on implementation. Dibsie16 reported that the
changes in the climate and practice related to skin care and prevention of breakdown are the
direct result of nursing taking ownership of their practice with the support of nursing leaders at
all levels.
Lyder indicated that focusing pressure ulcer prediction and prevention programs on the
nursing staff is limited. Effective pressure ulcer prevention requires a multidisciplinary effort.
The PDSA model assists hospitals in working in multidisciplinary teams and places the onus for
improvement on the team rather than on a particular discipline.24 According to Lyder, hospitals
found that the most sustainable interventions were those that were institutionalized. For example,
two hospitals changed their hospital mattresses to pressure-relieving mattresses. However,
interventions that are most dependent on sufficient staffing were more difficult to sustain (for
example, ensuring that every resident is turned every 2 hours).24
Bales et al.13 reported that the hospitals managerial style encouraged staff involvement in
decision-making about developing a program, and leadership gave strong support to the program
and promoted it to both other leaders and hospital staff. Chicano14 reported that commitment
and diligence from the QI team and from the members of the staffs self-governance councils
played a significant factor in achieving our goal of reducing HAPU prevalence in our
intermediate care unit.
Contextual factors that Horn et al.26 identified as key to their success were resident
participation, use of a multidisciplinary team, and integration of all clinical reports. The facility
with the highest reduction in PUs (-82.4%) had 100% involvement of residents. In addition, this
facility incorporated all 4 weekly clinical reports into care planning. Two facilities with the
lowest reduction in PUs did not involve a multidisciplinary team. This study included only the
highest risk units at each facility.
228
Abel et al. reported that higher QI scoring and greater improvement in scores equaled lower
PU incidence rates. He indicated that 10 facilities with the highest quality indicator scores at remeasurement showed a trend toward lower PU incidence rates than the 10 facilities with the
lowest QI scores at re-measurement (-4.6% vs. -2.6%) (S = 125.5, p = 0.07). In addition, the 10
facilities with the greatest improvement in QI scores had significantly lower PU incidence rates
than the 10 facilities with the least improvement in QI scores (-6% vs. -1.2%) (S = 131.0,
P = 0.03). Later, Abel noted that study results demonstrated a relationship between PU incidence
rates and improvement in QI scores, suggesting a relationship between the process of care
measures and patient outcome (i.e., PU incidence). He adds that improved performance was
primarily noted in measures that were less dependent on staffing, i.e., use of support surfaces and
risk assessment tools rather than patient turning.
Milne et al.28 reported on revisions of care in two units with greatly improved PUs. On a
pulmonary focused unit, PUs were observed at the ear/scalp junction of 25% of patients. With
the adaption of the nasal cannula, PU rates on the ear/scalp junction were reduced within
2 months to less than 3%. On a separate unit, 33.8% of SCI and trauma patients were suffering
from sacral and heel sores due to immobility and sensory deficit. Due to several focused
initiatives, including pressure redistribution support surfaces, PU rates dropped to 2.9%, a 30.9%
reduction.
Rantz 200133 observed that intensive consultation contributed to a greater improvement in
MDS QI measures for several outcome measures, including two focused on PUs.

A review of the evidence indicates that a variety of multicomponent initiatives have been
implemented in U.S. acute and long-term care settings to prevent PUs. Improvements (many
significant) were reported both from comprehensive initiatives (targeting several patient safety
concerns) and from those primarily focused on reducing PUs. Successful prevention initiatives
were reported regardless of setting (system-, facility-, or unit-wide) or number of components
integrated in the initiative. However, evidence indicated that the majority of successful initiatives
integrated several core components. Key to improving these measures were the simplification
and standardization of pressure-ulcer-specific interventions and documentation, involvement of
multidisciplinary teams and leadership, designated skin champions, ongoing education, and
sustained audit and feedback for promoting both accountability and recognizing successes.
Several studies commented on the influence of context on outcomes. Useful information was
also included on topics such as barriers, solutions to barriers, and issues surrounding
sustainability. Key lessons learned include keeping both leadership and staff accountable for the
initiatives success, streamlining processes, learning from front-line staff, implementing change
one unit at a time, providing feedback, and celebrating successes.
Of the 23 studies included in this review, eight studies (evenly split between acute and longterm care settings) reported on both processes of care and patient outcomes. A majority of these
studies reported improvements in processes of care and corresponding improvements in patient
outcomes. We did, however, identify two studies that reported significant improvement in
several processes of care with little or no improvement in patient outcomes. Future research
should focus on the specific preventive measures or processes of care undertaken to better
understand their influence on patient outcomes. We also encourage clinicians to report findings
regardless of success level and to report strategies to sustain momentum of preventive programs.
A summary table is located below (Table 3).
229

Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate
References
1.
Lyder CH, Ayello EA. Annual checkup: the

CMS pressure ulcer present-on-admission
indicator. Adv Skin Wound Care 2009
Oct;22(10):476-84. PMID: 20026923.
2.
Duncan KD. Preventing pressure ulcers: the

goal is zero. Jt Comm J Qual Patient Saf
2007 Oct;33(10):605-10. PMID: 18030862
3.
Park-Lee E, Caffrey C. Pressure ulcers

among nursing home residents: United
States, 2004. NCHS Data Brief 2009
Feb;(14):1-8.
www.cdc.gov/nchs/data/databriefs/db14.pdf.
PMID: 19389326.
4.
Reddy M, Gill SS, Rochon PA. Preventing

pressure ulcers: a systematic review. JAMA
2006 Aug 23;296(8):974-84. PMID:
16926357.
5.
Advancing Excellence Steering Committee.

Implementation guide: goal 4: reducing high
risk pressure ulcers. NH Quality Campaign;
24 p.
www.nhqualitycampaign.org/files/impguide
s/4_PressureUlcer_TAW_Guide.pdf.
6.
Joint Commission. Accreditation program:

long term care. National patient safety goals.
Oakbrook Terrace (IL): Joint Commission;
2011 Jan 1. 13 p.
www.jointcommission.org/assets/1/6/2011_
NPSGs_LTC.pdf.
7.
How-to guide: prevent pressure ulcers.

Cambridge (MA): Institute for Healthcare
Improvement (IHI); 2011. 19
www.ihi.org/knowledge/Knowledge%20Ce
nter%20Assets/Tools%20-%20HowtoGuidePreventPressureUlcers_0a06721b7d9b-42f6-8d4053347fa21b65/HowtoGuidePreventPressure
Ulcers.pdf.
230
8.
Wound, Ostomy, and Continence Nurses

Society (WOCN). Guideline for prevention
and management of pressure ulcers. Mount
Laurel (NJ): Wound, Ostomy, and
Continence Nurses Society (WOCN); 2010
Jun 1. 96 p. (WOCN clinical practice
guideline; no. 2).
9.
Jankowski IM, Nadzam DM. Identifying

gaps, barriers, and solutions in
implementing pressure ulcer prevention
programs. Jt Comm J Qual Patient Saf 2011
Jun;37(6):253-64. PMID: 21706985.
10.
Ayello AE. Keynote: successful pressure

ulcer initiatives to reduce pressure ulcer
incidence [slide set]. In: 12th NPUAP
Biennial Conference; 2011 Feb; National
Pressure Ulcer Advisory Panel; 2011 Feb.
www.npuap.org/PDF%20FINAL%20Hando
ut%20Feb%2025%20Ayello%20PRessure%
20Ulcer%20Initiatives%20NPUAP%202011
%20Las%20Vegas.pdf.
11.
Lynch S, Vickery P. Steps to reducing

hospital-acquired pressure ulcers. Nursing
2010 Nov;40(11):61-2. PMID: 20975436.
12.
Young J, Ernsting M, Kehoe A, Holmes K.

Results of a clinician-led evidence-based
task force initiative relating to pressure ulcer
risk assessment and prevention. J Wound
Ostomy Continence Nurs 2010 SepOct;37(5):495-503. PMID: 20736858.
13.
Bales I, Padwojski A. Reaching for the

moon: achieving zero pressure ulcer
prevalence. J Wound Care 2009
Apr;18(4):137-144. PMID: 19349933
14.
Chicano SG, Drolshagen C. Reducing

hospital-acquired pressure ulcers. J Wound
Ostomy Continence Nurs 2009 JanFeb;36(1):45-50. PMID: 19155823.
15.
Walsh NS, Blanck AW, Barrett KL.

Pressure ulcer management in the acute care
setting: a response to regulatory mandates. J
Wound Ostomy Continence Nurs 2009 JulAug;36(4):385-8. PMID: 19609158.
16.
Dibsie LG. Implementing evidence-based

practice to prevent skin breakdown. Crit
Care Nurs Q 2008 Apr-Jun;31(2):140-9.
PMID: 18360144.
17.
McInerney JA. Reducing hospital-acquired

pressure ulcer prevalence through a focused
prevention program. Adv Skin Wound Care
2008 Feb;21(2):75-8. PMID: 18349734.
18.
Ballard N, McCombs A, Deboor S, Strachan

J, Johnson M, Smith MJ, et al. How our ICU
decreased the rate of hospital-acquired
pressure ulcers. J Nurs Care Qual 2008 JanMar;23(1):92-6. PMID: 18281882.
19.
Catania K, Huang C, James P, Madison M,

Moran M, Ohr M. Wound wise: PUPPI: the
Pressure Ulcer Prevention Protocol
Interventions. Am J Nurs 2007
Apr;107(4):44-52; quiz 53. PMID:
17413732.
20.
LeMaster KM. Reducing incidence and

prevalence of hospital-acquired pressure
ulcers at Genesis Medical Center. Jt Comm J
Qual Patient Saf 2007 Oct;33(10):611-6,
585. PMID: 18030863.
21.
Courtney BA, Ruppman JB, Cooper HM.

Save our skin: initiative cuts pressure ulcer
incidence in half. Nurs Manage 2006
Apr;37(4):36, 38, 40 passim. PMID:
16603946.
22.
Gibbons W, Shanks HT, Kleinhelter P,

Jones P. Eliminating facility-acquired
pressure ulcers at Ascension Health. Jt
Comm J Qual Patient Saf 2006
Sep;32(9):488-96. PMID: 17987872.
23.
Hiser B, Rochette J, Philbin S, Lowerhouse

N, Terburgh C, Pietsch C. Implementing a
pressure ulcer prevention program and
enhancing the role of the CWOCN: impact
on outcomes. Ostomy Wound Manage 2006
Feb;52(2):48-59. PMID: 16464994.
24.
Lyder CH, Grady J, Mathur D, Petrillo MK,

Meehan TP. Preventing pressure ulcers in
Connecticut hospitals by using the plan-dostudy-act model of quality improvement. Jt
Comm J Qual Saf 2004 Apr;30(4):205-14.
PMID: 15085786.
231
25.
Stier L, Dlugacz YD, OConnor LJ, Eichorn

AM, White M, Fitzpatrick J. Reinforcing
organizationwide pressure ulcer reduction
on high-risk geriatric inpatient units.
Outcomes Manag 2004 Jan-Mar;8(1):28-32.
PMID: 14740581.
26.
Horn SD, Sharkey SS, Hudak S, Gassaway

J, James R, Spector W. Pressure ulcer
prevention in long-term-care facilities: a
pilot study implementing standardized nurse
aide documentation and feedback reports.
Adv Skin Wound Care 2010 Mar;23(3):12031. PMID: 20177165.
27.
Rantz MJ, Hicks L, Petroski GF, Madsen

RW, Alexander G, Galambos C, et al. Cost,
staffing and quality impact of bedside
electronic medical record (EMR) in nursing
homes. J Am Med Dir Assoc 2010
Sep;11(7):485-93. PMID: 20816336.
28.
Milne CT, Trigilia D, Houle TL, Delong S,

Rosenblum D. Reducing pressure ulcer
prevalence rates in the long-term acute care
setting. Ostomy Wound Manage 2009
Apr;55(4):50-9. PMID: 19387096.
29.
Tippet AW. Reducing the incidence of

pressure ulcers in nursing home residents: a
prospective 6-year evaluation. Ostomy
Wound Manage 2009 Nov 1;55(11):52-8.
PMID: 19934464.
30.
Rosen J, Mittal V, Degenholtz H, Castle N,

Mulsant BH, Hulland S, et al. Ability,
incentives, and management feedback:
organizational change to reduce pressure
ulcers in a nursing home. J Am Med Dir
Assoc 2006 Mar;7(3):141-6. PMID:
16503306.
31.
Abel RL, Warren K, Bean G, Gabbard B,

Lyder CH, Bing M, et al. Quality
improvement in nursing homes in Texas:
results from a pressure ulcer prevention
project. J Am Med Dir Assoc 2005 MayJun;6(3):181-8. PMID: 15894247.
32.
Ryden MB, Snyder M, Gross CR, Savik K,

Pearson V, Krichbaum K, et al. Value-added
outcomes: the use of advanced practice
nurses in long-term care facilities.
Gerontologist 2000 Dec;40(6):654-62.
PMID: 11131082.
33.
34.
35.
Rantz MJ, Popejoy L, Petroski GF, Madsen

RW, Mehr DR, Zwygart-Stauffacher M, et
al. Randomized clinical trial of a quality
improvement intervention in nursing homes.
Gerontologist 2001 Aug;41(4):525-38.
PMID: 11490051.
Lyder CH, Ayello EA. Pressure ulcers: a
patient safety issue. In: Patient safety and
quality: an evidence-based handbook for
nurses. Rockville (MD): Agency for
Healthcare Research and Quality (AHRQ);
2008 Apr. p. 1-33.
www.ahrq.gov/qual/nurseshdbk/docs/Lyder
C_PUPSI.pdf.
Hobbs BK. Reducing the incidence of
pressure ulcers: implementation of a turnteam nursing program. J Gerontol Nurs 2004
Nov;30(11):46-51. PMID: 15575191.
36.
Uzun O, Aylaz R, Karadag E. Prospective

study: reducing pressure ulcers in intensive
care units at a Turkish medical center. J
Wound Ostomy Continence Nurs 2009 JulAug;36(4):404-11. PMID: 19609161.
37.
Zaratkiewicz S, Whitney JD, Lowe JR, et al.

Development and implementation of a
hospital-acquired pressure ulcer incidence
tracking system and algorithm. J Healthc
Qual 2010 Nov-Dec;32(6):44-51. PMID:
20946425.
38.
Gray-Siracusa K, Schrier L. Use of an

intervention bundle to eliminate pressure
ulcers in critical care. J Nurs Care Qual
2011 Jul-Sep;26(3):216-25. PMID:
21278597.
39.
What is a bundle? [internet]. Cambridge

(MA): Institute for Healthcare Improvement
(IHI); [accessed 2011 Nov 18]. [3 p].
www.ihi.org/knowledge/Pages/Improvement
Stories/WhatIsaBundle.aspx.
232
40.
Healthcare risk control: pressure ulcers. Vol.

3, Risk analysis. Plymouth Meeting (PA):
ECRI Institute; 2011 May. 21 p. (Nursing;
no. 4).
41.
Soban LM, Hempel S, Munjas BA, Miles J,

Rubenstein LV. Preventing pressure ulcers
in hospitals: a systematic review of nursefocused quality improvement interventions.
Jt Comm J Qual Patient Saf 2011
Jun;37(6):245-52. PMID: 21706984.
42.
What is DMAIC? [internet]. Chicago (IL):

iSixSigma; [accessed 2011 Nov 10]. [1 p].
www.isixsigma.com/index.php?option=com
_k2&view=item&layout=item&id=1483&It
emid=369.
43.
Anthony MK. Shared governance models:

the theory, practice, and evidence. Online J
Issues Nurs 2004 Jan 31;9(1):7.
www.nursingworld.org/MainMenuCategorie
s/ANAMarketplace/ANAPeriodicals/OJIN/
TableofContents/Volume92004/No1Jan04/S
haredGovernanceModels.html?css=print.
PMID: 14998357.
44.
Warren K, Bean G, Abel B, Gabbard B,

Bing M. Improvement report: reducing
pressure ulcer rates in nursing homes.
[internet]. Cambridge (MA): Institute for
Healthcare Improvement; 2004 Dec
[accessed 2011 Jun 15]. [3 p].
www.ihi.org/IHI/Topics/PatientSafety/Safet
yGeneral/ImprovementStories/ReducingPres
sureUlcerRatesinNursingHomes.htm.
45.
Xakellis GC, Frantz R. The cost of healing

pressure ulcers across multiple health care
settings. Adv Wound Care 1996 NovDec;9(6):18-22. PMID: 9069752.
Chapter 22. Inpatient Intensive Glucose Control Strategies

To Reduce Death and Infection (NEW)
Devan Kansagara, M.D., M.C.R., FACP

Hyperglycemia is a common finding among medical and surgical inpatients with and without
known diabetes.1,2 Moreover, several observational studies have found an association between
inpatient hyperglycemia and poor outcomes in patients undergoing general and cardiac
surgery,3,4 and in patients with a variety of conditions including myocardial infarction, stroke,
and trauma.5-7 Hyperglycemia could contribute to these poor health outcomes by causing
inflammation, oxidative stress, poor immune function, endothelial dysfunction, and tissue
ischemia.8 Given the mechanistic and observed association between hyperglycemia and poor
outcomes in hospitalized patients, significant interest has developed in using intensive insulin
therapy (IIT) to control blood glucose in a variety of inpatient subpopulations. However, overaggressive use of insulin can result in dangerously low levels of blood glucose, which also can be
harmful to patients. This chapter reports the results of a systematic review conducted in 20102011 on the use of IIT to control blood glucose among inpatients as well as the findings of
studies released subsequent to the searches we conducted for that review.

The evidence evaluating the balance of benefits and harms from the use of IIT to control
blood glucose in hospitalized patients is detailed in the following sections. Overall, trials have
not consistently found that use of IIT to lower blood glucose to normal levels (i.e., 80-110
mg/dL) improves health outcomes, whereas aggressive IIT approaches are clearly associated
with high rates of hypoglycemia. Nevertheless, many organizations continue to recommend
moderate blood glucose control (e.g., 140200 mg/dL) because of the association of high blood
glucose with infection, poor wound healing, dehydration, and other complications. Given the
uncertainty about the benefits of IIT, the remaining concerns that untreated hyperglycemia is
harmful, and the hypoglycemia risks associated with IIT, the patient safety practice of interest is
the implementation of inpatient glycemic control strategies that minimize the risk of
hypoglycemia.

Initially encouraging trial data in critically ill patients spurred some organizations to
recommend tight glycemic control strategies be implemented in hospitals across a variety of
settings.8,9 In one of the key trials fueling these recommendations, 1,548 patients in a single
surgical intensive care unit (SICU) were randomized to either an intensive insulin regimen, with
a goal glucose range of 80-110 mg/dl, or a conventional insulin regimen with a goal glucose
range of 180-200 mg/dl.10 The trial was terminated early after finding all-cause ICU mortality
was significantly lower in the IIT group (4.6% vs. 8%, relative risk [RR] 0.58, 95% CI, 0.38 to
0.78) (Table 1). The short-term mortality benefit was limited to the subgroup of patients who
required 5 or more days of ICU care (10.6% vs. 20.2%, p = 0.005); long-term mortality did not
differ between the two groups.11
233
Table 1, Chapter 22. Large trials (n > 500) evaluating the health outcome effects of intensive insulin therapy
Patient Population;
Implementation Context Glucose Target, Inpatient BG Mortality T v C (RR, 95% CI)
Diabetes Mellitus (%);
T v C (mg/dL)
Achieved,
Single or Multicenter;
T v C (mg/dL)
Country;
Study Quality
SICU
Insulin protocol was
80-110
103 v 153 ICU mortality 4.6 v 8% (p=0.005
13
developed and use overseen
v 180-200
(p<0.001)
unadjusted)
by study investigators.
Single center
RR 0.42 (95% CI 0.22-0.62);
10
Hospital mortality: 7.2 v 10.9%
Belgium
Fair
(p=0.01)
RR 0.66; 95% CI 0.48-0.92
Neurosurgical ICU
Efforts made to limit nursing 80-110 v 180-200
92 v 143
6-month mortality: 74.0 v 72.0%
NR
turnover. New nursing staff
(p<0.001)
(p=0.82)
Single center
worked with experienced
58
Italy
staff.
Fair
MICU
Study conducted in a
80-110 v
111 v 153
ICU mortality: 24.2 v 26.8%
16
hospital that had already
180-200
(p<0.001)
(p=0.31)
conducted similar IIT study in
Single center
59
SICU patients. Authors note
(p=0.33)
Belgium
RR 0.93; 95% CI 0.81-1.08
Fair
the nurse:bed ratio of 2.5
90d mortality: 35.9 v 37.7%
was not changed for study.
(p=0.53)
MICU
No details provided
80-110
112 v 151
28d mortality: 24.7 v 26%
30
v 180-200
(p<0.001)
(p=0.74)
Multicenter
RR 0.95, 95% CI 0.70-1.28
21
Germany
Fair
(p=0.31)
MICU/SICU
Characteristics from each
80-110
117 v 144
17 T, 22 C (p=0.031)
study site were reported.
v 140-180
(p<0.001)
(p=0.41)
Multicenter
Median nurse:bed ratio was
60
(p=0.11)
Europe
2. ICUs ranged widely in
Fair
size, patient volume, and
(p=0.14)
number of glucometers per
ICU.
234
Hypoglycemia
Definition
(mg/dL),
rate ,T v C,
RR (95%CI)
<40,
5 v 0.76%,
RR 6.65 (2.8315.62)
Other Reported
Outcomes*
TvC
Renal replacement
4.8 v 8.2% (p=0.007)
Sepsis
4.2 v 7.8% (p=0.0003)
<50, 93.8 v
62.8%, p<0.001
Sepsis 2.9 v 3.3%

(p=NS)
Long-term disability:
40.2 v 41.1% (p=0.98)
<40, 18.7 v 3.1%
Infection 0.7 vs 0.8%

(p=NS)
Renal replacement 20.8
v 22.7% (p=0.50)
<40, 17 v 4.1% Renal replacement 27.5

RR 4.11 (95% CI
v 22.5% (p=0.001)
2.21-7.63)
< 40, 8.7 v 2.7%
Renal replacement
(patient days) 519 v 523
(p=0.75)
Table 1, Chapter 22. Large trials (n > 500) evaluating the health outcome effects of intensive insulin therapy (continued)
Patient Population;
Implementation Context Glucose Target,
T v C (mg/dL)
Country;
Study Quality
MICU/SICU
24/7 ICU coverage by
80-110
32 T, 48 C (p<0.001)
intensivists. Protocol
v 180-200
designed by multidisciplinary
Single center
19
team at study site.
Saudi Arabia
Fair
Physicians and nurses
attended training sessions
before and during study.
MICU/SICU
Three month staff training 80-110 v 180-200
13 T, 12 C (p=NS)
period before study.
Single center
54
Colombia
Fair
MICU/SICU
Pre-trial pilot studies carried 80-108 v <180
20
out to test/improve insulin
13
Multicenter International protocol. Final computerized
Fair
insulin protocol algorithm
accessible to study sites
through a central Web site.
No clear explicit training prior
to study.
Acute MI
No details provided
126-198 v NR
39
Multicenter CCU
61
Sweden
Fair
Acute MI
No details provided
group 1 and 2:
77 established DM; 23
126-180
new DM
group 3: NR
of < 1y
62
Multicenter Europe
Poor
Inpatient BG
Achieved,
T v C (mg/dL)
115 v 171
(p<0.001)
120 v 149
(p,0.001)
115 v 144
(p<0.001)
24 hours:
T: 172.8 (59.4)
C: 210.6 (73.8)
p < .001
Mortality T v C (RR, 95% CI)
Hypoglycemia
Other Reported
Definition
Outcomes*
(mg/dL),
TvC
rate ,T v C,
RR (95%CI)
< 40, 28.6 v
Renal replacement
(p=0.70)
3.1%, p < 0.001 11.7 v 12.1% (p=0.89)
RR 1.09 (0.70 -1.72)
Sepsis 36.9 v 40.9%
(p=0.35)
(p=0.19)
RR 0.84 (0.64 -1.09)
ICU mortality: 33.1 v 31.2%; RR <40, 8.3 v 0.8%
1.06 (0.82-1.37)
28d mortality: 36.6 v 32.4%;
RR 1.1 (0.85-1.42)
(p=0.17)
RR 1.07 (0.97-1.18)
(p=0.02)
RR 1.14 (1.02-1.28)
<40, 6.8 v 0.5%

OR 14.7 (9.025.9)
<54, 15.0 v 0%
3 month mortality:
(p < .001)
12.4% v 15.6%, p = NS
1 year mortality:
18.6% v 26.1 %,
RR 0.69; 95% CI 0.49-0.96
24 hours:
Adjusted 2-year mortality:
< 54, Gr 1 v Gr2
group 1: 163.8 Group 1 v 3 = 1.19 (0.86 - 1.64) v Gr3: 12.7 v 9.6
v 1.0
(54.0), group 2: Group 2 v 3 = 1.23 (0.89 - 1.69)
163.8 (50.4),
group 3: 180.0
(64.8)
p = .0001
235
Infection 27.2 v 33.2%

(p=NS)
Renal replacement
10.8 v 13% (p=0.45)
Renal replacement
15.4 v 14.5% (p=0.34)
Sepsis
12.8 v 12.4% (p=0.57)
Table 1, Chapter 22. Large trials (n > 500) evaluating the health outcome effects of intensive insulin therapy (continued)
Patient Population;
Implementation Context Glucose Target,
T v C (mg/dL)
Country;
Study Quality
Stroke
Conducted as a pragmatic 72-126 v <306
17
trial as part of routine clinical
63
care. No clear explicit
Mutlicenter Britain
Poor
training prior to study.
Inpatient BG
Achieved,
T v C (mg/dL)
Hypoglycemia
Other Reported
Definition
Outcomes*
(mg/dL),
TvC
rate ,T v C,
RR (95%CI)
24 hour mean
90-day mortality:
< 72 for > 30
difference I v C
30.0% v 27.3%,
mins, 15.7,
(95% CI): 10.3 OR (95% CI) = 1.14 (0.86-1.51) control group rate
90-day severe disability:
NR
(4.9 - 15.5), p <
35.1% v 36.0%,
.0001
OR (95% CI) = 0.96 (0.70-1.32)
Notes: Abbreviations: BG = Blood glucose; d = day; CCU = coronary care unit ; ICU = intensive care unit; MICU = medical intensive care unit; SICU = surgical intensive care
unit; C = comparator; DM = diabetes mellitus; NR = not reported; NS = not statistically significant; RR = relative risk; T = treatment
Other reported outcomes include renal replacement, infection, cardiovascular events, and long-term disability.
Quality was assessed using criteria from the U.S. Preventive Services Task Force.
SI unit conversion for glucose: 1 mg/dL x 0.0555 = 1 mmol/L.
* Infection includes wound infection, urinary tract infection, or pneumonia; or a combination of these.
Morning blood glucose.
Average of blood glucose measurements, not otherwise specified.
236
Mortality T v C (RR, 95% CI)
However, over the last decade, IIT trials have failed to replicate these results consistently.
Our recent meta-analysis of 21 randomized controlled trials (RCTs), including a total of 14,768
inpatients, found no effect of IIT on short-term mortality (RR 1.00, 95% CI 0.94 to 1.07) (see
Figure 1), with remarkably little heterogeneity among studies (I2 = 0.0%, p=0.463).12 The body
of evidence is strongest in ICU settings.
Figure 1, Chapter 22. Short-term mortality in studies of intensive insulin therapy, by inpatient
setting and condition
Figure taken from Kansagara et al., 201112

Kansagara D, Fu R, Freeman M, Wolf F, Helfand M. Intensive insulin therapy in hospitalized patients: a systematic review. Ann
Intern Med. Feb 15 2011; 154. Permission granted from the American College of Physicians (Annals of Internal Medicine is the
original source of the material).
Table 2 summarizes the main characteristics and results from the largest trials (n > 500
patients). The NICE-SUGAR trial, with 6,104 medical intensive care unit (MICU) and SICU
patients, was by far the largest and likely provides the most generalizable results, given its size,
multicenter design, and the broad ICU population included.13 Patients randomized to a lower
blood glucose target (80-110 mg/dL) had higher 90-day mortality than those assigned a higher
blood glucose target (140-180 mg/dL), with approximately one excess death for every 39
patients treated with the more intensive protocol (n = 6,022 with 90-day outcomes reported; RR
1.14; 95% CI 1.02 to 1.28).
237
Morbidity outcomes have also been assessed. Trials of IIT failed to demonstrate consistent
benefits in reducing renal failure or length of stay. The effects of IIT on infection are mixed. The
2001 Van den Berghe SICU trial did find IIT reduced the incidence of sepsis,10 whereas the
NICE-SUGAR trial did not.13 Trials also reported a variety of other infection-related outcomes,
including wound infections, pneumonia, and urinary tract infection. A pooled analysis of these
trials found a nonsignificant reduction in infection, though the results were quite heterogenous
(RR 0.68; 95% CI 0.36 to1.30, I2 = 56.3%, P=0.033).12
Several trials have reported results since January 2010 (the end of the search period of our
systematic review). Two trials found no benefit of IIT on neurologic or mortality outcomes in
patients with traumatic brain injury or stroke.14,15 One trial did find postoperative IIT reduced
wound infections in diabetic patients who had undergone partial gastrectomy (7.6% v 18.4%, p =
0.03), but the trial had several methodologic flaws.16 Finally, the recent RABBIT 2 trial was
among the first to compare the effects of a subcutaneous basal-bolus insulin regimen with
sliding-scale insulin on health outcomes.17 In this trial, 211 noncritically ill general surgery
patients were randomized to either a basal-bolus insulin regimen using insulin glargine and mealtime glulisine or a sliding-scale insulin regimen (whereby fixed doses of insulin are given in
response to specific glucose readings), with a goal glucose target in both groups of 100-140
mg/dL. The basal-bolus group achieved significantly better glycemic control and a lower
incidence of wound infections (2.9% vs. 10.3%, p = 0.05), but several methodologic issues,
including poor outcome ascertainment methods and no blinding of outcome assessors, threaten
the validity of results.

The main harm of IIT is hypoglycemia. Insulin has a narrow therapeutic window: Underuse
may fail to resolve potentially risky hyperglycemia, whereas overly aggressive insulin use can
lead to severe hypoglycemia. Nearly all 31 trials in our recent systematic review reported that IIT
was associated with excess risk of hypoglycemia.12 Our meta-analysis of 10 trials found that IIT
was associated with a six-fold increased risk of severe hypoglycemia, defined as glucose <40
mg/dL (RR 6.00; 95% CI 4.06 to 8.87; I2 = 57.9%; p<0.001) (Figure 2).12 The few trials
published since we did the searches for our review corroborated these findings.14-16
238
Figure 2, Chapter 22. Risk for hypoglycemia in studies of intensive insulin therapy in various
inpatient settings
Figure taken from Kansagara, 201112

Kansagara D, Fu R, Freeman M, Wolf F, Helfand M. Intensive insulin therapy in hospitalized patients: a systematic review. Ann
Intern Med. Feb 15 2011; 154. Permission granted from the American College of Physicians (Annals of Internal Medicine is the
original source of the material).
The consequences of hypoglycemia in hospitalized patients may be severe. Several MICU

studies found excess risk-adjusted mortality and/or extended length of stay among patients
experiencing one or more episodes of severe hypoglycemia.18-22 However, these studies reported
few in-hospital adverse effects of hypoglycemia during IIT, though many critically ill patients in
these studies were sedated, which limits the completeness of neurologic assessment.
How Has the Patient Safety Practice Been implemented, and in

What Contexts?
The safety of IIT may depend on intervention and implementation characteristics of the IIT
protocols. In addition to the IIT trials reviewed above, we reviewed an additional 40 insulin
protocol studies that did not report health outcomes in order to better understand how
intervention and implementation characteristics of protocols impact safety. These protocol
studies differed in terms of patient characteristics, target glucose ranges, the time required to
achieve the target glucose levels, the definition and incidence of hypoglycemia, the rationale or
algorithm used for adjusting the insulin rates, the methods used to assess effectiveness, and the
methods of glucose monitoring.12 Nevertheless, some themes emerge from a review of these
protocol studies.
The vast majority of studies evaluated intravenous insulin infusions; fewer examined the
effects of subcutaneous insulin protocols. Protocols were most widely tested in ICU populations
(both surgical and medical); few studies occurred in general medical or surgical ward settings.
Most of the studies were single-center studies of insulin infusion protocols iteratively developed
by a local group of providers. The rate of hypoglycemia in these studies was, in general, lower
than that seen in IIT trials evaluating health outcomes.
239
Glucose Targets
Not surprisingly, we found that protocols targeting higher blood glucose ranges were
generally associated with lower rates of hypoglycemia. This observation echoes findings from
the trial literature in which insulin infusions were used to target strict (80-110 mg/dL) or
moderate (140-200 mg/dL) glucose goals. The rates of severe hypoglycemia were substantially
higher in the strict glucose target groups.
We also found protocols that achieved mean blood glucose <120 mg/dL were not
consistently safe, even when clinicians used relatively sophisticated computerized
algorithms.23,24 Two observational studies evaluated the safety of phasing in progressively
stricter glucose targets over time.25,26 One of these was a large single-center retrospective study
evaluating the effects of an increasingly aggressive IIT policy in the ICU.25 The authors found a
nearly four-fold increase in the incidence of hypoglycemia as the institution moved from no
insulin protocol to IIT with a target of 80-130 mg/dL and finally to a target of 80-110 mg/dL.25
The infusion protocol details were not available. A second study of a relatively simple infusion
protocol reported a doubling of the rate of severe hypoglycemia as the glucose target moved
from 120-150 mg/dL to 80-110 mg/dL, although the overall rate of severel hypoglycemia
remained less than 5 percent.26
Insulin Dosing Factors

The factors used to guide insulin dosing may also be important. Some protocols use only
current blood glucose levels to guide dosing, whereas others attempt to anticipate insulin needs
based on measures of insulin sensitivity. For example, one recent observational study examined
an insulin-resistance-guided protocol in which cardiac surgery patients were assigned to one of
six insulin resistance categories.27 The insulin dosing adjustments depended on the category to
which each patient was assigned. Hourly arterial blood glucose monitoring was used along with
changes in medications and patient condition to determine subsequent changes to insulin
resistance category. The authors found that such an approach reduced the rates of both
hypoglycemia and hyperglycemia. The SPRINT protocol used a similarly complex approach and
also found reductions in hyperglycemia and hypoglycemia in critically ill patients.28
Other reviews speculate that better protocols incorporate bolus insulin doses, account for the
direction and rate of glucose change, and make allowances for off-protocol adjustments.
However, these conclusions are not based on direct comparisons of protocols.29,30
Computerized Protocols
Most protocols studied have used a paper-based algorithm to guide nurses in making insulin
dose adjustments and timing glucose measurements. In recent years, computer-based algorithms
have become available and, in 2010, the first RCT comparing these algorithms to a paper-based
algorithm was reported.31 This multicenter trial of 153 MICU patients compared a paper-based
algorithm to the computerized Glucommander system, which directed insulin dosing and glucose
monitoring timing using glucose measurements at the patients bedside. The glucose target was
80-120 mg/dL. Patients in the computer-algorithm group had fewer instances of marked
hyperglycemia (glucose > 200 mg/dL, 11.7% vs. 25%, p = 0.05, but a similar rate of severe
hypoglycemia (< 40 mg/dL:, 3.9% vs. 5.6%, p = NS). An older observational study of the same
computerized system had reported similar results.32 However, a recent small observational study
found only minor improvements in glucose control using a computerized protocol and no change
in hypoglycemia rates.33 This study also found that using the computerized protocol led to more
240
frequent glucose testing and insulin dose adjustments. Finally, another recent observational study
found that fewer dosing errors occurred with a computer-based protocol than with a paper-based
protocol.34
Continuous Glucose Monitoring

Recently, studies have also examined the use of continuous glucose monitoring devices,
although none have tested their use outside of small, single-center populations. In a single-center
RCT of 124 MICU patients, a subcutaneous continuous glucose monitoring strategy did not
improve glycemic control but did reduce the rate of severe hypoglycemia (1.6% vs. 11.5%, p =
0.03).35 One observational study evaluated a closed-loop glycemic control device, which
continuously monitors glucose and automatically delivers insulin and glucose, and found no
hypoglycemic events.36
Glucose Monitoring Site

Capillary blood glucose is the most common source for glucose monitoring. However, it has
several notable limitations to its accuracy, which, in turn, affect the safety of IIT. Capillary blood
sampling is less dependable than arterial sampling in critically ill patients for several possible
reasons, including low perfusion pressure, use of vasopressors, and low pH.37-40 The rate of
agreement between capillary and whole blood samples is particularly low in the hypoglycemic
range.41,42 Capillary blood testing also tends to overestimate glucose levels in anemic patients,
which could lead to overaggressive use of insulin to achieve tight glucose control. One recent
study found high rates of measurement error in patients with hematocrit less than 34 percent; the
investigators suggested a mathematical correction factor, but it has yet to be tested on a wide
scale.43
Nutrition
Most insulin protocols neither coordinate insulin dosing with patient nutrition nor provide
detailed nutritional guidance. In one RCT, 337 critically ill patients were randomized to either a
carbohydrate-restrictive strategy or an insulin infusion regimen targeted to blood glucose levels
of 80-120 mg/dL.44 Although the glucose level achieved in the carbohydrate-restrictive group
was higher than in the infusion group (144 vs. 134 mg/dL, p = 0.03), the difference was modest
and the rate of hypoglycemia was substantially lower in the carbohydrate-restrictive group (3.5%
vs. 16%, p < 0.01). The results suggest that more-intensive nutritional strategies may be a
promising adjunct, or alternative to, IIT. An observational study of the SPRINT protocol, which
directly prescribes both insulin dosing and dietary intake, found that it improved glycemic
control and reduced the risk of severe hypoglycemia.28
Sliding Scale Insulin

Although most trials evaluating health outcomes of IIT have used insulin infusions to achieve
blood glucose control, subcutaneous insulin is more often used in real-world settings, especially
in general ward patients. Subcutaneous sliding scale insulin (SSI) regimens have several
theoretical disadvantages when used as the sole method for inpatient glycemic control. For that
reason, various researchers have called for a reduction in the widespread use of SSI
approaches.45,46
Very few controlled trials have compared SSI with basal-bolus subcutaneous insulin
regimens in which both long-acting and meal-time insulin are provided. The multicenter
241
RABBIT 2 surgery trial17 and several small, single-center trials in general medical47-49 and
gastric bypass populations50 found that basal-bolus regimens were more effective in lowering
blood glucose than SSI, although both strategies had similar rates of hypoglycemia. The
RABBIT 2 surgery trial is the only one to have reported the effects of basal-bolus insulin on
health outcomes.17

No studies evaluating the cost-effectiveness of IIT incorporate findings from trials reported
within the past few years. Several earlier studies suggest that IIT is cost-effective, but these
studies relied on findings from older studies that had found cost reductions from shorter length of
stay and lower risk of costly complications such as infections. However, as noted above, such
benefits have not been replicated consistently in more recent trials.
The incremental impact of IIT on resource utilization is unclear. Some studies have suggested
that costs are relatively low. For instance, the cost attributable to IIT in cardiac surgery patients
in one center were estimated to be $138 per patient.51 Because nurses are typically responsible
for glucose monitoring and insulin adjustments, implementing IIT protocols might require more
nursing time and effort. One large ICU cohort study found that ITT implementation did not
change either nurse:patient ratios or nursing hours.52 By contrast, a multi-center ICU study using
more detailed time-in-motion observations found that the nurse-led intensive glucose monitoring
and insulin dosing adjustments were burdensome and costly. The authors estimated that nursing
personnel could spend up to 2 hours on IIT-related activities for a given patient per 24-hour
period; this level of effort totaled $182,488 for nurses salaries and about $58,500 for supplies in
cost over 1 year.53

All IIT strategies involve frequent insulin dose adjustments and glucose monitoring; these
tasks are usually performed by nursing staff and guided by a protocol. Above, we detailed the
protocol characteristics that may affect safety, but, because of the frequent human input required,
careful implementation strategies and training may also be important to execute IIT safely. Most
large IIT trials (see Table 1) did not provide much detail about the clinical context within which
they had implemented their IIT interventions. Only two trials specified explicit nursing training
before the start of the study.19,54
The high rates of hypoglycemia in recent multicenter IIT trials, such as NICE-SUGAR, may
provide some information about implementation of IIT protocols.13 Some have argued that, in
general, the protocols used in IIT trials are difficult to implement safely across multiple centers
because of the lack of specific instructions, the simplicity of inputs used to guide insulin dosing,
and the relative lack of clinical expertise that nurses would gain with IIT in each study site.55
Van den Berghe and colleagues used a simple infusion protocol in their 2001 SICU study, but
theirs was a single-center study in which the investigators were practicing clinicians and the
nursing staff had more opportunity to develop clinical expertise with the protocol because all
patients were treated in one setting.10
In our review of observational studies and smaller IIT trials, we could indirectly glean similar
lessons about implementation. Furnary and colleagues acknowledged the importance of local
physician champions, an iterative process, and nursing buy-in to the successful implementation
of their IIT protocol in cardiac surgery patients.56 They gradually lowered glucose targets from
200 mg/dL to 100-150 mg/dL over 15 years. The rate of deep sternal wound infections dropped
242
with use of IIT, although theirs was an uncontrolled study, and they did not report overall
hypoglycemia rates. In contrast, another institution went from no glycemic control policy to a
normal glucose target over 5 years in a large population of critically ill patients (n = 10,456);
these investigators reported markedly increased rates of hypoglycemia and a trend to increased
mortality.25 These results may well reflect the difficulty with broad, rapid implementation of
aggressive glucose control practices.

The use of IIT to achieve very tight blood glucose control does not reduce short-term
mortality in MICU patients (high strength of evidence) or SICU patients (moderate strength of
evidence). It increases the risk of severe hypoglycemia in all settings (high strength of evidence).
The lack of consistent benefit from very tight blood glucose control and the increased risk of
hypoglycemia has led to recommendations for a moderate blood glucose target of 140-200
mg/dL in ICU populations.57 However, it is unknown whether implementation of IIT protocols
targeted to moderate blood glucose levels (140-200 mg/dL) with low rates of hypoglycemia
improves health outcomes. Despite the lack of evidence to support a specific blood glucose
target, many organizations continue to recommend moderate blood glucose control in inpatients
because of the association of high blood glucose with infection, poor wound healing,
dehydration, and other complications. Although glycemic control protocols remain an important
part of quality inpatient care, the lack of clear and consistent evidence of benefit underscores that
minimization of hypoglycemia is of paramount importance in the implementation of any
glycemic control protocol.
Based on data from a review of insulin protocols and of trials evaluating the health outcome
effects of IIT, the following emerge as important issues to consider when implementing IIT
protocols:
The glucose target is important. Glucose targets in the normal range (80-110 mg/dL)
markedly increase the risk of severe hypoglycemia and do not improve health outcomes.
Higher glucose targets (e.g., 140-200 mg/dL) can be safely achieved with careful IIT
implementation.
The clinical factors used to guide insulin dosing are important. Very simple protocols
based only on current and past glucose levels may be difficult to replicate safely across
institutions. Protocols incorporating some estimate of a patients insulin sensitivity may
be safer and more effective than those that ignore these factors.
Newer technologies such as continuous glucose monitoring and computerized protocols
may improve glycemic control. However, the evidence base is limited. Whether these
technologies reduce hypoglycemia rates remains unclear. The cost of such technology has
not been adequately assessed.
In critically ill patients, capillary blood glucose can be markedly inaccurate, particularly
in the hypoglycemic range. Clinicians should exercise caution using capillary blood
glucose measurements in these patients and in patients with anemia.
IIT protocols should be coupled to patient nutrition whenever possible because failure to
modify IIT dosing in response to discontinuities in nutritional intake can increase the risk
of hypoglycemia. Additionally, some nutritional interventions may, themselves, be
effective in reducing the risk of hypoglycemia.
In surgical patients, weight-based subcutaneous insulin protocols using both basal and
bolus insulin may reduce infection rates more than sliding-scale-only insulin. The
243
comparative effects of different subcutaneous insulin regimens has not been well studied
in non-surgical populations.
IIT is a complex endeavor requiring buy-in from nurses and physicians. Implementation
of IIT in a given setting is likely best done iteratively, with multidisciplinary involvement
and training, and using real-time data to inform continuous quality improvement of the
process.

Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
it doesnt help
Evidence or
Potential for
Harmful
Unintended
Consequences
High
(hypoglycemia)
Estimate of
Cost
Low-tomoderate
How Much do We
N/A
References
1.
Umpierrez GE, Isaacs SD, Bazargan N, You

X, Thaler LM, Kitabchi AE.
Hyperglycemia: an independent marker of
in-hospital mortality in patients with
undiagnosed diabetes.[see comment]. J Clin
Endocrinol Metab. Mar 2002;87(3):978-982.
2.
Levetan CS, Passaro M, Jablonski K, Kass

M, Ratner RE. Unrecognized diabetes
among hospitalized patients. Diabetes Care.
Feb 1998;21(2):246-249.
3.
Furnary AP, Gao G, Grunkemeier GL, et al.

Continuous insulin infusion reduces
mortality in patients with diabetes
undergoing coronary artery bypass
grafting.[see comment]. J Thorac
Cardiovasc Surg. May 2003;125(5):10071021.
4.
5.
Pomposelli JJ, Baxter JK, 3rd, Babineau TJ,

et al. Early postoperative glucose control
predicts nosocomial infection rate in
diabetic patients. JPEN J Parenter Enteral
Nutr. Mar-Apr 1998;22(2):77-81.
Bochicchio GV, Sung J, Joshi M, et al.
Persistent hyperglycemia is predictive of
outcome in critically ill trauma
patients.[erratum appears in J Trauma. 2005
Nov;59(5):1277-8]. J Trauma. May
2005;58(5):921-924.
244
6.
Capes SE, Hunt D, Malmberg K, Gerstein

HC. Stress hyperglycaemia and increased
risk of death after myocardial infarction in
patients with and without diabetes: a
systematic overview.[see comment]. Lancet.
Mar 4 2000;355(9206):773-778.
7.
Capes SE, Hunt D, Malmberg K, Pathak P,

Gerstein HC. Stress hyperglycemia and
prognosis of stroke in nondiabetic and
diabetic patients: a systematic overview.[see
comment]. Stroke. Oct 2001;32(10):24262432.
8.
Clement S, Braithwaite SS, Magee MF, et

al. Management of diabetes and
hyperglycemia in hospitals.[erratum appears
in Diabetes Care. 2004 Mar;27(3):856 Note:
Hirsh, Irl B [corrected to Hirsch, Irl B];
dosage error in text]. Diabetes Care. Feb
2004;27(2):553-591.
9.
Institute for Health Care Improvement.

Establish a glycemic control policy in your
ICU.
www.ihi.org/IHI/Topics/CriticalCare/Intensi
veCare/Changes/IndividualChanges/Establis
haGlycemicControlPolicyinYourICU.htm
2008.
10.
van den Berghe G, Wouters P, Weekers F, et

al. Intensive insulin therapy in the critically
ill patients. N Engl J Med. Nov 8
2001;345(19):1359-1367.
11.
Ingels C, Debaveye Y, Milants I, et al. Strict

blood glucose control with insulin during
intensive care after cardiac surgery: impact
on 4-years survival, dependency on medical
care, and quality-of-life. Eur Heart J. Nov
2006;27(22):2716-2724.
20.
Devos P, Preiser JC, Melot C. Impact of

tight glucose control by intensive insulin
therapy on ICU mortality and the rate of
hypoglycaemia: final results of the
GLUCONTROL study. Intensive Care Med.
2007;33(Supplement 2):S189.
12.
Kansagara D, Fu R, Freeman M, Wolf F,

Helfand M. Intensive insulin therapy in
hospitalized patients: a systematic review.
Ann Intern Med. Feb 15 2011;154(4):268282.
21.
Brunkhorst FM, Engel C, Bloos F, et al.

Intensive insulin therapy and pentastarch
resuscitation in severe sepsis. The New
England journal of medicine. Jan
2008;358(2):125-139.
13.
NICE-SUGAR Study, Finfer S, Chittock

DR, et al. Intensive versus conventional
glucose control in critically ill patients. N
Engl J Med. Mar 26 2009;360(13):12831297.
22.
Hermanides J, Bosman RJ, Vriesendorp TM,

et al. Hypoglycemia is associated with
intensive care unit mortality. Crit Care Med.
Jun 2010;38(6):1430-1434.
23.
14.
Green DM, OPhelan KH, Bassin SL, Chang

CWJ, Stern TS, Asai SM. Intensive versus
conventional insulin therapy in critically ill
neurologic patients. Neurocrit Care. Dec
2010;13(3):299-306.
15.
Coester A, Neumann CR, Schmidt MI.

Intensive insulin therapy in severe traumatic
brain injury: a randomized trial. J Trauma.
Apr 2010;68(4):904-911.
Dortch MJ, Mowery NT, Ozdas A, et al. A

computerized insulin infusion titration
protocol improves glucose control with less
hypoglycemia compared to a manual
titration protocol in a trauma intensive care
unit. JPEN J Parenter Enteral Nutr. Jan-Feb
2008;32(1):18-27.
24.
Toschlog EA, Newton C, Allen N, et al.

Morbidity reduction in critically ill trauma
patients through use of a computerized
insulin infusion protocol: a preliminary
study. J Trauma. Jun 2007;62(6):1370-1375;
discussion 1375-1376.
25.
Treggiari MM, Karir V, Yanez ND, Weiss

NS, Daniel S, Deem SA. Intensive insulin
therapy and mortality in critically ill
patients. Crit Care. 2008;12(1):R29.
26.
Taylor BE, Schallom ME, Sona CS, et al.

Efficacy and safety of an insulin infusion
protocol in a surgical ICU. J Am Coll Surg.
Jan 2006;202(1):1-9.
27.
Caddell KA, Komanapalli CB, Slater MS, et

al. Patient-specific insulin-resistance-guided
infusion improves glycemic control in
cardiac surgery. Ann Thorac Surg. Dec
2010;90(6):1818-1823.
28.
Chase JG, Shaw G, Le Compte A, et al.

Implementation and evaluation of the
SPRINT protocol for tight glycaemic control
in critically ill patients: a clinical practice
change. Critical Care. 2008;12(2).
16.
17.
18.
19.
Cao S-G, Ren J-A, Shen B, Chen D, Zhou

Y-B, Li J-S. Intensive versus conventional
insulin therapy in type 2 diabetes patients
undergoing D2 gastrectomy for gastric
cancer: a randomized controlled trial. World
J Surg. Jan 2011;35(1):85-92.
Umpierrez GE, Smiley D, Jacobs S, et al.
Randomized study of basal-bolus insulin
therapy in the inpatient management of
patients with type 2 diabetes undergoing
general surgery (RABBIT 2 surgery).
Diabetes Care. Feb 2011;34(2):256-261.
Krinsley JS, Krinsley JS. Association
between hyperglycemia and increased
hospital mortality in a heterogeneous
population of critically ill patients.[see
comment]. Mayo Clin Proc. Dec
2003;78(12):1471-1478.
Arabi YM, Dabbagh OC, Tamim HM, et al.
Intensive versus conventional insulin
therapy: a randomized controlled trial in
medical and surgical critically ill patients.
Crit Care Med. Dec 2008;36(12):3190-3197.
245
29.
Meijering S, Corstjens AM, Tulleken JE,

Meertens JHJM, Zijlstra JG, Ligtenberg
JJM. Towards a feasible algorithm for tight
glycaemic control in critically ill patients: a
systematic review of the literature. Crit
Care. Feb 2006;10(1):R19.
38.
Critchell CD, Savarese V, Callahan A,

Aboud C, Jabbour S, Marik P. Accuracy of
bedside capillary blood glucose
measurements in critically ill patients.
Intensive Care Med. Dec 2007;33(12):20792084.
30.
Wilson M, Weinreb J, Hoo GW, Wilson M,

Weinreb J, Hoo GWS. Intensive insulin
therapy in critical care: a review of 12
protocols. Diabetes Care. Apr
2007;30(4):1005-1011.
39.
Tanvetyanon T, Walkenstein MD, Marra A.

Inaccurate glucose determination by
fingerstick in a patient with peripheral
arterial disease. Ann Intern Med. Nov 5
2002;137(9):W1.
31.
Newton CA, Smiley D, Bode BW, et al. A

comparison study of continuous insulin
infusion protocols in the medical intensive
care unit: computer-guided vs. standard
column-based algorithms. J Hosp Med. Oct
2010;5(8):432-437.
40.
Juneja D, Pandey R, Singh O. Comparison

between arterial and capillary blood glucose
monitoring in patients with shock. European
Journal of Internal Medicine. Jun
2011;22(3):241-244.
41.
32.
Juneja R, Roudebush C, Kumar N, et al.

Utilization of a computerized intravenous
insulin infusion program to control blood
glucose in the intensive care unit. Diabetes
Technol Ther. Jun 2007;9(3):232-240.
Kanji S, Singh A, Tierney M, Meggison H,

McIntyre L, Hebert PC. Standardization of
intravenous insulin therapy improves the
efficiency and safety of blood glucose
control in critically ill adults. Intensive Care
Med. May 2004;30(5):804-810.
33.
Barletta JF, McAllen KJ, Eriksson EA, et al.

The effect of a computer-assisted insulin
protocol on glycemic control in a surgical
intensive care unit. Diabetes Technol Ther.
Apr 2011;13(4):495-500.
42.
Desachy A, Vuagnat AC, Ghazali AD, et al.

Accuracy of bedside glucometry in critically
ill patients: influence of clinical
characteristics and perfusion index. Mayo
Clin Proc. Apr 2008;83(4):400-405.
34.
Lee A, Faddoul B, Sowan A, Johnson KL,

Silver KD, Vaidya V. Computerisation of a
paper-based intravenous insulin protocol
reduces errors in a prospective crossover
simulated tight glycaemic control study.
Intensive Crit Care Nurs. Jun
2010;26(3):161-168.
43.
Pidcoke HF, Wade CE, Mann EA, et al.

Anemia causes hypoglycemia in intensive
care unit patients due to error in singlechannel glucometers: methods of reducing
patient risk. Crit Care Med. Feb
2010;38(2):471-476.
44.
35.
Holzinger U, Warszawska J, Kitzberger R,

et al. Real-time continuous glucose
monitoring in critically ill patients: a
prospective randomized trial. Diabetes Care.
Mar 2010;33(3):467-472.
de Azevedo JRA, de Araujo LO, da Silva

WS, de Azevedo RP. A carbohydraterestrictive strategy is safer and as efficient as
intensive insulin therapy in critically ill
patients. J Crit Care. Mar 2010;25(1):84-89.
45.
36.
Yatabe T, Yamazaki R, Kitagawa H, et al.

The evaluation of the ability of closed-loop
glycemic control device to maintain the
blood glucose concentration in intensive
care unit patients. Crit Care Med. Mar
2011;39(3):575-578.
Metchick LN, Petit WA, Jr., Inzucchi SE.

Inpatient management of diabetes
mellitus.[see comment]. Am J Med. Sep
2002;113(4):317-323.
46.
Cheung NW, Chipps DR. Sliding scale

insulin: will the false idol finally fall? Intern
Med J. Sep 2010;40(9):662-664.
47.
Umpierrez GE, Smiley D, Zisman A, et al.

patients with type 2 diabetes (RABBIT 2
trial).[see comment]. Diabetes Care. Sep
2007;30(9):2181-2186.
37.
Atkin SH, Dasmahapatra A, Jaker MA,

Chorost MI, Reddy S. Fingerstick glucose
determination in shock. Ann Intern Med. Jun
15 1991;114(12):1020-1024.
246
48.
Dickerson LM, Ye X, Sack JL, Hueston WJ.

Glycemic control in medical inpatients with
type 2 diabetes mellitus receiving sliding
scale insulin regimens versus routine
diabetes medications: a multicenter
randomized controlled trial. Ann Fam Med.
May-Jun 2003;1(1):29-35.
56.
Furnary AP, Zerr KJ, Grunkemeier GL,

Starr A. Continuous intravenous insulin
infusion reduces the incidence of deep
sternal wound infection in diabetic patients
after cardiac surgical procedures.[see
comment]. Ann Thorac Surg. Feb
1999;67(2):352-360; discussion 360-352.
49.
Korytkowski MT, Salata RJ, Koerbel GL, et

al. Insulin therapy and glycemic control in
hospitalized patients with diabetes during
enteral nutrition therapy: a randomized
controlled clinical trial. Diabetes Care. Apr
2009;32(4):594-596.
57.
50.
Datta S, Qaadir A, Villanueva G, Baldwin

D. Once-daily insulin glargine versus 6-hour
sliding scale regular insulin for control of
hyperglycemia after a bariatric surgical
procedure: a randomized clinical trial.
Endocr Pract. May-Jun 2007;13(3):225-231.
Qaseem A, Humphrey LL, Chou R, Snow

V, Shekelle P, Clinical Guidelines
Committee of the American College of P.
Use of intensive insulin therapy for the
management of glycemic control in
hospitalized patients: a clinical practice
guideline from the American College of
Physicians. Ann Intern Med. Feb 15
2011;154(4):260-267.
58.
Bilotta F, Caramia R, Paoloni FP, et al.

Safety and efficacy of intensive insulin
therapy in critical neurosurgical patients.
Anesthesiology. Mar 2009;110(3):611-619.
59.
Van den Berghe G, Wilmer A, Hermans G,

et al. Intensive insulin therapy in the medical
ICU. N Engl J Med. Feb 2 2006;354(5):449461.
60.
Preiser JC, Devos P, Ruiz-Santana S, et al.

A prospective randomised multi-centre
controlled trial on tight glucose control by
intensive insulin therapy in adult intensive
care units: the Glucontrol study. Intensive
Care Med. Oct 2009;35(10):1738-1748.
61.
Malmberg K, Ryden L, Efendic S, et al.

Randomized trial of insulin-glucose infusion
followed by subcutaneous insulin treatment
in diabetic patients with acute myocardial
infarction (DIGAMI study): effects on
mortality at 1 year. J Am Coll Cardiol. Jul
1995;26(1):57-65.
62.
Malmberg K, Ryden L, Wedel H, et al.

Intense metabolic control by means of
insulin in patients with diabetes mellitus and
acute myocardial infarction (DIGAMI 2):
effects on mortality and morbidity. Eur
Heart J. Apr 2005;26(7):650-661.
63.
Gray CS, Hildreth AJ, Sandercock PA, et al.

Glucose-potassium-insulin infusions in the
management of post-stroke hyperglycaemia:
the UK Glucose Insulin in Stroke Trial
(GIST-UK). Lancet neurol. May
2007;6(5):397-406.
51.
Furnary AP, Wu Y, Bookin SO. Effect of

hyperglycemia and continuous intravenous
insulin infusions on outcomes of cardiac
surgical procedures: the Portland Diabetic
Project. Endocr Pract. Mar-Apr 2004;10
Suppl 2:21-33.
52.
Krinsley JS. Effect of an intensive glucose

management protocol on the mortality of
critically ill adult patients.[see
comment][erratum appears in Mayo Clin
Proc. 2005 Aug;80(8):1101]. Mayo Clin
Proc. Aug 2004;79(8):992-1000.
53.
Aragon D. Evaluation of nursing work effort

and perceptions about blood glucose testing
in tight glycemic control. Am J Crit Care.
Jul 2006;15(4):370-377.
54.
De La Rosa GD, Donado JH, Restrepo AH,

et al. Strict glycaemic control in patients
hospitalised in a mixed medical and surgical
intensive care unit: a randomised clinical
trial. Critical Care. 2008;12(5):R120.
55.
Van den Berghe G, Schetz M, Vlasselaers

D, et al. Clinical review: Intensive insulin
therapy in critically ill patients: NICESUGAR or Leuven blood glucose target? J
Clin Endocrinol Metab. Sep
2009;94(9):3163-3170.
247
Chapter 23. Interventions To Prevent Contrast-Induced Acute

Kidney Injury
Sumant R. Ranji, M.D.; Stephanie Rennke, M.D.; Yimdriuska Magan, B.S.; Erika Moseson,
M.D.; Robert M. Wachter, M.D.

Over 70 million computed tomography (CT) scans are performed yearly in the United
States,1 approximately half of which use iodinated radiocontrast media, and over 2 million
patients undergo other studies using radiocontrast media such as coronary angiograms.2 Contrastinduced acute kidney injury (CI-AKI) is one of the major risks of procedures using radiocontrast
media. CI-AKI is generally defined by laboratory criteria: biochemical CI-AKI is usually defined
as an increase in serum creatinine of 25%, or an absolute increase of 0.5 mg/dl, within 2-5 days
after receiving contrast.3 A prospective study4 found that the incidence of CI-AKI by this
definition was 7.7% in patients with impaired baseline kidney function (defined as an estimated
glomerular filtration rate of less than 60 mL/min/ 1.73 m2), ranging from 6.5% in patients
undergoing CT scans to 13.2% in patients undergoing non-coronary angiography.
Risk factors for CI-AKI include chronic kidney disease (CKD) of any cause, especially in
diabetic patients. Other risk factors include intravascular volume depletion and disease states
associated with decreased effective circulating volume and renal perfusion, such as congestive
congestive heart failure (CHF) and liver failure, and concominant use of nephrotoxic
medications, particularly non-steroidal anti-inflammatory drugs (NSAIDs).5 Procedural risk
factors also play a role, with larger volumes of contrast media, intra-arterial contrast
administration (such as in coronary angiography), and use of high-osmolarity contrast media all
independently associated with elevated risk for CI-AKI. Patients with normal baseline kidney
function have minimal risk of CI-AKI.
Although biochemical CI-AKI is commonly documented, the link between laboratory
abnormalities and clinical outcomes is controversial. Several studies have shown an independent
link between CI-AKI diagnosis in hospitalized patients and subsequent increases in length of
stay,6 progression to end-stage renal disease,7 and short- and long-term mortality.8 However,
causality is difficult to determine despite the presence of this association, because many factors
that predispose to CI-AKI (especially CHF and CKD) also are associated with adverse clinical
outcomes independent of CI-AKI development. In addition, AKI of any cause is associated with
worsened short- and long-term outcomes in hospitalized patients.9 In prospective studies, CIAKI has been found as an asymptomatic laboratory abnormality in the vast majority of patients.
Only 1 of 660 patients in a 2008 study by Weisbord et al.4 required kidney dialysis after
receiving contrast.

The standard of care to prevent CI-AKI includes several widely accepted, evidence-based
interventions:
Intravascular volume expansion with intravenous normal saline10
Limiting the volume of contrast administered
248
Avoidance of high-osmolar contrast media in patients with impaired baseline renal

function11
Stopping nephrotoxic medications, especially NSAIDs
Published guidelines from the American College of Radiology, the European Society of
Radiology,12 and the Canadian Association of Radiology13 all recommend the above measures.
The original review of this topic for Making Health Care Safer (2001) also recommended
volume expansion with normal saline and avoidance of high-osmolar contrast. The 2009
American College of Cardiology/American Heart Associated guidelines for percutaneous
coronary interventions also recommend avoidance of high-osmolar contrast media.2
In addition to standard care, several interventions have been widely studied to prevent CIAKI. These practices are the focus of this review:
Volume expansion with intravenous sodium bicarbonate
Administration of n-acetylcysteine
Use of iso-osmolar (instead of low- or high-osmolar) contrast media
Prophylactic renal replacement therapy (dialysis)
Administration of HMG CoA-reductase inhibitors (statins)

The pathophysiology of CI-AKI is complex and incompletely understood.3 Intravascular
contrast administration is thought to induce renal vasoconstriction, which may lead to medullary
ischemia, particularly in the presence of intravascular volume depletion or other medications that
may cause afferent renal artery vasoconstriction such as NSAIDs. Contrast media, particularly
older high-osmolar media, may be directly toxic to the renal tubules. Finally, some component of
renal damage is thought to be mediated by generation of reactive oxygen species (free
radicals). Because patients suspected of suffering CI-AKI rarely undergo kidney biopsy for
definitive diagnosis, the relative contribution of these mechanisms is unclear. As a result, the
mechanisms by which the proposed PSPs prevent CI-AKI are also somewhat speculative.
Opportunities for improving CI-AKI prevention definitely exist, as studies show that
appropriate and proven prophylactic interventions are not universally applied. Studies have
found that volume expansion is used in only 40% of at-risk patients undergoing coronary
angiography 4 and 60% of patients undergoing computed tomography.14 In the latter study, only
7% of patients had nephrotoxic medications discontinued.

We designed a structured literature search with the assistance of a medical librarian to
identify studies of interventions to prevent CI-AKI. Searching PubMed identified 193
randomized controlled trials and 53 meta-analyses of various interventions to prevent CI-AKI
published in the past 10 years. (Searching of the Cochrane Controlled Trials Registry and the
Cochrane Database of Systematic Reviews did not identify any additional trials.) In contrast, the
original Making Health Care Safer report published in 2001 identified only 10 RCTs and 1 metaanalysis.
Based on the expansion in this literature, we opted to conduct a systematic meta-review of
the meta-analyses of CI-AKI prevention published since January 1, 2007. We chose this
inclusion date based on prior literature demonstrating that the results of systematic reviews are
249
generally not stable by 5 years after publication.15 The revised search identified 32 studies, of
which 20 were confirmed to be meta-analyses after full-text review (the others were largely
narrative reviews). These 20 meta-analyses evaluated the effectiveness of 5 distinct interventions
for preventing CI-AKI:
Hydration with intravenous sodium bicarbonate (N=11)16-26
Administration of oral N-acetylcysteine (NAC, N=3)27-29
Use of iso-osmolar radiocontrast media (N=3)30-32
Prophylactic renal replacement therapy (RRT, N=1)33
Administration of HMG CoA-reductase inhibitors (statins, N=1)34
In addition, one study35 evaluated the combination of NAC and bicarbonate in preventing CIAKI compared with NAC alone.
We followed the methodology previously outlined by Whitlock36 for incorporating
previously published systematic reviews into a new review. Each identified review was
evaluated for quality using the AMSTAR checklist37, and information was extracted on the
interventions and outcomes assessed, the study populations (including the types of radiologic
studies for which contrast media was used) and sample size, the definition of CI-AKI used, and
the overall conclusions of the review (Table 1).
Hydration With Intravenous Sodium Bicarbonate

We identified a total of 11 meta-analyses published since 200716-26 comparing sodium
bicarbonate hydration to volume expansion with normal saline. These meta-analyses all used the
same definition of CI-AKI (a 25% increase in the serum creatinine level, or an absolute increase
of > 0.5 mg/dl, within 2-5 days of the procedure).
The review with the most recent inclusion date19 completed its search through February
2009, and identified a total of 18 published and unpublished trials. This meta-analysis was
methodologically sound, scoring 11 (of a possible 11) on the AMSTAR scale, and overall found
a slight benefit for bicarbonate compared with saline volume expansion in preventing CI-AKI by
the laboratory definition (pooled OR 0.66, 95% CI 0.45-0.95). There was no reported
improvement in the need for renal replacement therapy or mortality. This seemingly positive
result was tempered by numerous caveats. The authors noted significant heterogeneity across
included trials, found evidence for publication bias, and considered the quality of included trials
to be low. Therefore, the authors concluded only a limited recommendation can be made in
favour of sodium bicarbonate.
Another meta-analysis with a slightly earlier study inclusion date of December 200825
actually included more trials (N=23, including 14 unpublished trials). This meta-analysis also
scored 11 on the AMSTAR scale. The pooled trial results found evidence for a slight benefit for
bicarbonate compared with saline volume expansion in preventing laboratory-defined CI-AKI
(pooled relative risk 0.62, 95% CI 0.45 to 0.86). However, the authors performed a metaregression analysis and found that bicarbonate was effective only in smaller, poor-quality trials.
Larger, higher-quality trials generally found neutral results. This meta-analysis, which appears to
be the most comprehensive study of bicarbonate prophylaxis for CI-AKI, concludes that the
effectiveness of sodium bicarbonate treatment to prevent contrast-induced nephropathy remains
unclear.
The other 9 meta-analyses identified in our search did not include any other trials (published
or unpublished) that were not included in the 2 meta-analyses discussed above. Significant
250
heterogeneity was found in all 11 meta-analyses, and all of the meta-analyses that included
unpublished studies found evidence of publication bias.
Therefore, we conclude that sodium bicarbonate therapy appears to offer only marginal
benefit at best over routine saline volume expansion, and the primary literature suffers from
significant limitations. Routine bicarbonate administration cannot be recommended to prevent
CI-AKI.
Administration of Oral N-Acetylcysteine

The role of N-acetylcysteine in CI-AKI prevention has been quite thoroughly studied. We
identified 3 meta-analyses published since 200727-29, but prior to 2007 an additional 12 metaanalyses and 2 meta-reviews had already been published. Limitations in the prior literatureand
the meta-analyses of this literaturehave been well documented; in fact, a 2006 meta-review38
described the plethora of NAC trials and meta-analyses as a case study in the pitfalls of the
evolution of evidence. No consensus on the effectiveness of NAC existed as of 2007, as the
existing meta-analyses produced differing results.
The most recent meta-analysis of interventions included randomized controlled trials
published through February 200829 and evaluated only studies of high-dose NAC protocols
(defined as administration of >1,200mg/day of oral NAC or a single periprocedural dose of >
600mg) compared with saline volume expansion. This high-quality meta-analysis (AMSTAR
score of 11) found that high-dose NAC protocols were effective in preventing biochemically
defined CI-AKI (random effect odds ratio 0.52; 95% CI, 0.34 to 0.78) in a trial population
predominantly composed of patients undergoing coronary angiography. This meta-analysis did
not extract or report information on clinical outcomes. However, a large RCT39 that was
published after this review and also used a high-dose NAC protocol did not find any reduction in
biochemical CI-AKI, need for hemodialysis, or mortality in patients undergoing coronary
angiography. This study enrolled 2,308 patients, whereas the 16 RCTs included in the metaanalysis in total enrolled only 1,677 patients.
Another earlier meta-analysis that included trials published through March 200628 identified
26 trials of NAC, using different dosing regimens ranging from 400 mg/day to 1,200 mg/day.
This meta-analysis did find evidence for a significant reduction in biochemically defined CIAKI. However, there was significant unresolved heterogeneity in this study. The meta-analysis
published by Gonzales et al.27, which included all but 6 of the same studies, noted that evidence
of benefit was confined to a small group of relatively low-quality studies which showed very
large relative benefits from NAC. These studies were also performed and published earlier than
subsequent larger, higher-quality trials that reported negative results.
Based on these findings, we conclude that routine use of NAC at any dose does not appear to
convincingly reduce the incidence of CI-AKI. As with bicarbonate infusion, there is no evidence
that NAC administration decreases the incidence of clinically meaningful outcomes such as the
need for renal replacement therapy.
Use of Iso-Osmolar Contrast Media

There are three types of iodinated radiocontrast media: high-osmolar, low-osmolar, and isoosmolar. High-osmolar contrast is little used due to its nephrotoxic effects, and low-osmolar
contrast media has become the standard of care. So-called iso-osmolar contrast has an even
lower osmolality than low-osmolar contrast, and 3 meta-analyses30-32 have evaluated the
renoprotective effect of the iso-osmolar contrast medium iodixanol compared with low-osmolar
251
contrast media (LOCM, of which there are several agents). The most recent and largest metaanalysis32 identified 36 randomized controlled trials published before December 2009. This
meta-analysis was high quality, scoring 11 on the AMSTAR scale. It did not find a statistically
significant reduction in biochemical CI-AKI for iso-osmolar contrast compared with all LOCM
agents (pooled OR 0.77, 95% CI 0.56 to 1.06). However, a subgroup analysis did find that isoosmolar contrast was associated with a reduction in CI-AKI in studies comparing iodixanol to
one specific low-osmolar agent, iohexol (pooled OR 0.25, 95% CI 0.11-0.55, N=10 trials). This
finding was also noted in the other two meta-analyses of this question.30,31 None of the metaanalyses evaluated the effect of iso-osmolar contrast media on clinical outcomes.
Other than this advantage of iodixanol over the specific agent iohexol, there is therefore no
convincing evidence supporting the routine use of iso-osmolar contrast. The 2009 ACC/AHA
guidelines for percutaneous coronary intervention2 recommend use of iso-osmolar contrast or use
of LOCM other than iohexol. This is a change from the 2007 guidelines, which specifically
recommended use of iso-osmolar agents.
Prophylactic Renal Replacement Therapy

One meta-analysis33 analyzed 9 RCTs evaluating the effectiveness of prophylactic renal
replacement therapy (RRT) on prevention of biochemically defined CI-AKI, need for long-term
RRT, and mortality. The patients included in the individual studies uniformly had baseline
kidney dysfunction (at least stage 3 chronic kidney dysfunction, with baseline serum creatinines
ranging from 1.5 to 4.2 across the studies). Overall, prophylactic RRT was not associated with
decreased biochemical CI-AKI or the need for long-term hemodialysis. The authors did find a
statistically significant reduction in mortality associated with prophylactic RRT (RR 0.33, 95%
CI 0.11 to 0.77), but the significance of this finding is quite questionable given the lack of effect
on the primary outcome. The authors speculated that the mortality benefit might instead
represent a general benefit of RRT in critically ill patients with AKI.
Administration of Statins
One recent meta-analysis34 identified 6 small RCTs evaluating the effect of statins on
biochemical CI-AKI. There was no overall beneficial effect of statins on prevention of CI-AKI.
Coadministration of Bicarbonate and N-Acetylcysteine

One meta-analysis35 identified 10 RCTs that studied the effectiveness of combination of
bicarbonate and NAC compared with NAC alone. The authors reported a reduction in
biochemical CI-AKI with combination therapy, but the result did not reach statistical
significance (pooled RR 0.65, 95% CI 0.40 to 1.05), nor did combination therapy reduce the
incidence of renal failure requiring dialysis.

The individual interventions that have been evaluated to prevent CI-AKI are generally
considered low risk. Bicarbonate and NAC are not associated with a significant risk of clinically
relevant adverse effects, and likewise, iso-osmolar contrast media do not have a unique side
effect profile compared with other routinely used radiocontrast agents. The exception is renal
replacement therapy, which requires placement of large bore central venous access, exposing
patients to complications of this procedure including hemorrhage, pneumothorax, or central lineassociated bloodstream infections.
252
One potential harm is that administration of intravenous fluids may increase the risk of
clinically significant congestive heart failure (CHF) in patients with a known diagnosis of CHF.
However, the largest meta-analysis of intravenous bicarbonate administration did not find an
increased incidence of symptomatic CHF.25

What Contexts?
Interventions to prevent CI-AKI have been studied in patients with a range of risk factors for
CI-AKI, and have included patients with no preexisting renal dysfunction as well as those with
chronic kidney disease. Studies have also assessed patients undergoing a variety of radiologic
procedures, including those associated with a higher risk of CI-AKI such as coronary
angiography. Within specific interventions, there are a range of specific protocols used for
administering prophylactic medications. However, across all the meta-analyses of this subject, no
unique subgroup of patients has been identified that benefits from any specific intervention.
At the health care system level, some steps have been taken to implement protocols to
minimize the risk of CI-AKI. Brown et al.40 conducted a mixed-methods study of CI-AKI
prevention practices at 10 centers enrolled in the Northern New England Cardiovascular Disease
Study Group PCI Registry. The incidence of biochemically defined CI-AKI varied widely across
sites, ranging from 1.9% to 10% even after adjustment for covariates. The two centers with the
lowest CI-AKI rates both had strong clinical leadership that prioritized CI-AKI prevention and
utilized standardized protocols for volume administration, NAC administration, and minimizing
the time that patients were NPO prior to procedures. Interestingly, one of these centers used
normal saline and the other bicarbonate for volume administration, indicating that the choice of
fluid likely matters less than ensuring that patients receive adequate volume prior to the
procedure.
Are There Any Data About Costs of the Patient Safety Practice?
We did not identify any formal cost-effectiveness analyses of the various modalities
proposed to prevent CI-AKI published since 2007. Interventions such as bicarbonate and NAC
are low cost, whereas iso-osmolar contrast media (IOCM) is more costly than standard LOCM.
One cost-effectiveness analysis demonstrated that IOCM is cost-effective compared with
LOCM,41 but this analysis was based on earlier, more favorable estimates of the benefits of
IOCM that have not been borne out in subsequent trials or meta-analyses. We also identified one
cost-effectiveness analysis of prophylactic RRT published in 2006,42 which found that
prophylactic RRT might be cost-effective only in a subset of patients with stage 4 chronic kidney
disease. This analysis was also based on favorable treatment estimates that have not been
confirmed in formal systematic reviews.

There is no definitive evidence that any single intervention to prevent CI-AKI is more
effective in specific patient populations (e.g., patients with more advanced chronic kidney
disease) or undergoing specific radiologic procedures (e.g., patients undergoing intra-arterial
contrast procedures such as coronary angiography versus patients undergoing procedures
requiring intravenous contrast. Health care system factors have not been studied as an effect
modifier for specific CI-AKI preventive interventions.
253

We identified 20 meta-analyses testing various interventions to prevent CI-AKI. However,
despite this intensive research, we were unable to identify any unique interventions that clearly
are effective at preventing either biochemical CI-AKI or clinically relevant outcomes such as
renal failure requiring hemodialysis. Moreover, even the significance of biochemical evidence of
kidney injury after contrast is debated, and some experts question the importance of this as a
proxy measure or target for intervention. At this point, it appears that standard therapy, most
importantly volume administration with intravenous normal saline prior to procedures, is the
most efficacious method of preventing CI-AKI. Use of standardized CI-AKI prevention
protocols that emphasize volume administration may be associated with a lower risk of CI-AKI
in patients undergoing coronary angiography. A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Low
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Low
How Much do We
References
1.
2.
Mettler FA, Bhargavan M, Faulkner K, et al.

Radiologic and nuclear medicine studies in
the United States and worldwide: frequency,
radiation dose, and comparison with other
radiation sources--1950-2007. Radiology
2009;253:520-31.
Kushner FG, Hand M, Smith SC, Jr., et al.
2009 focused updates: ACC/AHA
guidelines for the management of patients
with ST-elevation myocardial infarction
(updating the 2004 guideline and 2007
focused update) and ACC/AHA/SCAI
guidelines on percutaneous coronary
intervention (updating the 2005 guideline
and 2007 focused update) a report of the
American College of Cardiology
Foundation/American Heart Association
Task Force on Practice Guidelines. J Am
Coll Cardiol 2009;54:2205-41.
3.
Weisbord SD, Palevsky PM. Strategies for

the prevention of contrast-induced acute
kidney injury. Curr Opin Nephrol
Hypertens;19:539-49.
4.
Weisbord SD, Mor MK, Resnick AL, et al.

Prevention, incidence, and outcomes of
contrast-induced acute kidney injury. Arch
Intern Med 2008;168:1325-32.
254
5.
Weisbord SD, Palevsky PM. Contrastinduced acute kidney injury: short- and
long-term implications. Semin
Nephrol;31:300-9.
6.
Bartholomew BA, Harjai KJ, Dukkipati S, et

al. Impact of nephropathy after percutaneous
coronary intervention and a method for risk
stratification. Am J Cardiol 2004;93:1515-9.
7.
James MT, Ghali WA, Knudtson ML, et al.

Associations between acute kidney injury
and cardiovascular and renal outcomes after
coronary angiography. Circulation
2011;123:409-16.
8.
Brown JR, Malenka DJ, DeVries JT, et al.

Transient and persistent renal dysfunction
are predictors of survival after percutaneous
coronary intervention: insights from the
Dartmouth Dynamic Registry. Catheter
Cardiovasc Interv 2008;72:347-54.
9.
Levy EM, Viscoli CM, Horwitz RI. The

effect of acute renal failure on mortality. A
cohort analysis. JAMA 1996;275:1489-94.
10.
Mueller C, Buerkle G, Buettner HJ, et al.

Prevention of contrast media-associated
nephropathy: randomized comparison of 2
hydration regimens in 1620 patients
undergoing coronary angioplasty. Arch
Intern Med 2002;162:329-36.
11.

Wachter RM, Markowitz AJ. Making health
care safer: a critical analysis of patient
safety practices. Evid Rep Technol Assess
(Summ) 2001:i-x, 1-668.
20.
Joannidis M, Schmid M, Wiedermann CJ.

Prevention of contrast media-induced
nephropathy by isotonic sodium
bicarbonate: a meta-analysis. Wien Klin
Wochenschr 2008;120:742-8.
12.
Stacul F, van der Molen AJ, Reimer P, et al.

Contrast induced nephropathy: updated
ESUR Contrast Media Safety Committee
guidelines. Eur Radiol;21:2527-41.
21.
13.
Benko A, Fraser-Hill M, Magner P, et al.

Canadian Association of Radiologists:
consensus guidelines for the prevention of
contrast-induced nephropathy. Can Assoc
Radiol J 2007;58:79-87.
Kanbay M, Covic A, Coca SG, Turgut F,

Akcay A, Parikh CR. Sodium bicarbonate
for the prevention of contrast-induced
nephropathy: a meta-analysis of 17
randomized trials. Int Urol Nephrol
2009;41:617-27.
22.
Meier P, Ko DT, Tamura A, Tamhane U,

Gurm HS. Sodium bicarbonate-based
hydration prevents contrast-induced
nephropathy: a meta-analysis. BMC Med
2009;7:23.
23.
Navaneethan SD, Singh S, Appasamy S,

Wing RE, Sehgal AR. Sodium bicarbonate
therapy for prevention of contrast-induced
nephropathy: a systematic review and metaanalysis. Am J Kidney Dis 2009;53:617-27.
14.
Fishman EK, Reddan D. What are

radiologists doing to prevent contrastinduced nephropathy (CIN) compared with
measures supported by current evidence? A
survey of European radiologists on CIN
associated with computed tomography. Acta
Radiol 2008;49:310-20.
15.
Shojania KG, Sampson M, Ansari MT, Ji J,

Doucette S, Moher D. How quickly do
systematic reviews go out of date? A
survival analysis. Ann Intern Med
2007;147:224-33.
24.
Trivedi H, Nadella R, Szabo A. Hydration

with sodium bicarbonate for the prevention
of contrast-induced nephropathy: a metaanalysis of randomized controlled trials.
Clin Nephrol 2010;74:288-96.
16.
Brar SS, Hiremath S, Dangas G, Mehran R,

Brar SK, Leon MB. Sodium bicarbonate for
the prevention of contrast induced-acute
kidney injury: a systematic review and metaanalysis. Clin J Am Soc Nephrol
2009;4:1584-92.
25.
Zoungas S, Ninomiya T, Huxley R, et al.

Systematic review: sodium bicarbonate
treatment regimens for the prevention of
contrast-induced nephropathy. Ann Intern
Med 2009;151:631-8.
26.
17.
Ho KM, Morgan DJ. Use of isotonic sodium

bicarbonate to prevent radiocontrast
nephropathy in patients with mild preexisting renal impairment: a meta-analysis.
Anaesth Intensive Care 2008;36:646-53.
Kunadian V, Zaman A, Spyridopoulos I,

Qiu W. Sodium bicarbonate for the
prevention of contrast induced nephropathy:
a meta-analysis of published clinical trials.
Eur J Radiol 2011;79:48-55.
27.
18.
Hogan SE, LAllier P, Chetcuti S, et al.

Current role of sodium bicarbonate-based
preprocedural hydration for the prevention
of contrast-induced acute kidney injury: a
meta-analysis. Am Heart J 2008;156:414-21.
Gonzales DA, Norsworthy KJ, Kern SJ, et

al. A meta-analysis of N-acetylcysteine in
contrast-induced nephrotoxicity:
unsupervised clustering to resolve
heterogeneity. BMC Med 2007;5:32.
28.
19.
Hoste EA, De Waele JJ, Gevaert SA,

Uchino S, Kellum JA. Sodium bicarbonate
for prevention of contrast-induced acute
kidney injury: a systematic review and metaanalysis. Nephrol Dial Transplant
2010;25:747-58.
Kelly AM, Dwamena B, Cronin P, Bernstein

SJ, Carlos RC. Meta-analysis: effectiveness
of drugs for preventing contrast-induced
nephropathy. Ann Intern Med
2008;148:284-94.
29.
Trivedi H, Daram S, Szabo A, Bartorelli

AL, Marenzi G. High-dose N-acetylcysteine
nephropathy. Am J Med 2009;122:874 e915.
255
30.
Heinrich MC, Haberle L, Muller V, Bautz

W, Uder M. Nephrotoxicity of iso-osmolar
iodixanol compared with nonionic lowosmolar contrast media: meta-analysis of
randomized controlled trials. Radiology
2009;250:68-86.
39.
Acetylcysteine for prevention of renal

outcomes in patients undergoing coronary
and peripheral vascular angiography: main
results from the randomized Acetylcysteine
for Contrast-induced nephropathy Trial
(ACT). Circulation;124:1250-9.
31.
Reed M, Meier P, Tamhane UU, Welch KB,

Moscucci M, Gurm HS. The relative renal
safety of iodixanol compared with lowosmolar contrast media: a meta-analysis of
randomized controlled trials. JACC
40.
Brown JR, McCullough PA, Splaine ME, et

al. How do centres begin the process to
prevent contrast-induced acute kidney
injury: a report from a new regional
collaborative. BMJ Qual Saf.
41.
32.
From AM, Al Badarin FJ, McDonald FS,

Bartholmai BJ, Cha SS, Rihal CS. Iodixanol
versus low-osmolar contrast media for
meta-analysis of randomized, controlled
trials. Circ Cardiovasc Interv 2010;3:351-8.
Aspelin P, Aubry P, Fransson SG, Strasser

R, Willenbrock R, Lundkvist J. Costeffectiveness of iodixanol in patients at high
risk of contrast-induced nephropathy. Am
Heart J 2005;149:298-303.
42.
Klarenbach SW, Pannu N, Tonelli MA,

Manns BJ. Cost-effectiveness of
hemofiltration to prevent contrast
nephropathy in patients with chronic kidney
disease. Crit Care Med 2006;34:1044-51.
33.
Song K, Jiang S, Shi Y, Shen H, Shi X, Jing

D. Renal replacement therapy for prevention
meta-analysis of randomized controlled
trials. Am J Nephrol 2010;32:497-504.
34.
Zhang T, Shen LH, Hu LH, He B. Statins

nephropathy: a systematic review and metaanalysis. Am J Nephrol 2011;33:344-51.
35.
Brown JR, Block CA, Malenka DJ,

OConnor GT, Schoolwerth AC, Thompson
CA. Sodium bicarbonate plus Nacetylcysteine prophylaxis: a meta-analysis.
JACC Cardiovasc Interv 2009;2:1116-24.
36.
Whitlock EP, Lin JS, Chou R, Shekelle P,

Robinson KA. Using existing systematic
reviews in complex systematic reviews. Ann
Intern Med 2008;148:776-82.
37.

Methodol 2007;7:10.
38.
Bagshaw SM, McAlister FA, Manns BJ,

Ghali WA. Acetylcysteine in the prevention
of contrast-induced nephropathy: a case
study of the pitfalls in the evolution of
evidence. Arch Intern Med 2006;166:161-6.
256
Chapter 24. Rapid-Response Systems (NEW)

Bradford D. Winters, M.D., Ph.D.; Sallie Weaver, Ph.D.; Sydney Dy, M.D., M.Sc.

General ward patients often experience unrecognized deterioration in their clinical status that
may progress to cardio-respiratory arrest. Such cardio-respiratory arrests are known to carry a
poor prognosis for hospitalized patients. Mortality for in-hospital arrest is as high as 80%. One
study, examining patient data prior to an arrest event, found that clear signs and symptoms
heralding arrest often exist in these patients for many hours prior to the arrest (median time6
hours) yet are unrecognized and/or unappreciated. In addition, an average of two visits by health
care staff occurred during those median 6 hours of developing instability without apparent
recognition of the patients condition or any intervention.1
Rapid response systems (RRSs) were developed by clinicians as a way to improve
recognition of deterioration (this portion is called the Afferent Limb) and provide a critical care
team to respond to those deteriorations (the Efferent Limb), in order to improve outcomes such
as reducing the incidence of cardio-respiratory arrest and hospital mortality. RRSs have been
implemented in many hospitals to remedy the failure of our current system model (intermittent
vital signs) to monitor general ward patients adequately, to recognize the signs and symptoms of
deterioration, to rescue such patients, to deliver optimal care rapidly in patients who develop
signs or symptoms of clinical deterioration; and to escalate care and triage appropriately.2

At the 3rd International Medical Emergency Team (MET) conference, the disparate
nomenclature for this intervention was codified to bring all the terms under one umbrella term:
the Rapid Response System or RRS. An RRS includes a multidisciplinary team, most frequently
consisting of intensive care unit (ICU)-trained personnel who are available 24 hours per day, 7
days per week to evaluate patients not in the ICU who develop signs or symptoms of clinical
deterioration. RRSs include Medical Emergency Teams (METs, which includes a physician),
Rapid Response Teams (RRTs, which do not include a physician), and Critical Care Outreach
Teams (CCOT, which provide specific follow-up care for patients discharged from an intensive
care unit to a general ward, and may also include as part of the intervention, the ability to
respond to deteriorating ward patients that may or may not have been in the ICU previously).
The response team is referred to as the Efferent Limb and the system of tracking and recognizing
deterioration and activating the Efferent Limb is referred to as the Afferent Limb.
Rapid Response Systems aim to improve the safety of hospital-ward patients whose
condition is deteriorating. These systems are based on identification of patients at risk, early
notification of an identified set of responders, rapid intervention by the response team, and
ongoing evaluation of the systems performance and hospital-wide processes of care.2 Similar
types of systems exist for acute myocardial infarction (AMI)/cardiac stenting emergencies (Heart
Attack Teams or HATs), cerebrovascular accident (CVA) (Brain Attack Teams or BATS), and
other specialty issues such as hyperkalemia. However, these are different programs, with
different structures and effectiveness, designed to address very specific disease states. In
257
contrast, RRSs are non-specific and address a panoply of conditions. Therefore, we do not
include the disease-specific systems (BATs and HATs etc.) in this review.
A Rapid Response System generally has four components:
Criteria for notifying the response team and a system for activating it (the Afferent
Limb). The criteria usually include vital signs (single trigger criteria or more complicated
algorithms including aggregate and weighted early warning scores). However, in some
cases a clinician or family member might initiate activation, based on clinical judgment
and concern even though specific activation criteria are not met (e.g., heart rate >130).
The response team the Efferent Limb. Refers to personnel and equipment (can be led by
a critical-care physician, other physician, or by an nurse or respiratory therapist). Team
composition varies based on local needs and human resources.
Feedback loop to collect and analyze event data and quality improvement.
Administrative component, coordinating resources, staff, equipment, and education.
Jones et al2 also cites importance of support of leadership and administration, use of criteria
that are not too complicated (argues for simple vital signs triggers as opposed to complicated
early warning scores), education of the personnel on the team regarding the criteria (including
possibly simulation training), and involvement of physicians who can facilitate ICU transfers and
end-of-life planning. In a narrative review of data from the MERIT trial (the only multi-center
cluster randomized trial of RRS) and subsequent data, Jones et al. also note that RRSs exhibit a
dose response curve, where utilization rates (number of RRS activations) positively correlate
with reduction in the incidence of cardiac arrest. The authors found that a utilization rate of
approximately 17 RRS calls/1000 patient admissions is required to reduce the incidence of
cardio-respiratory arrest by 1/1000 admissions. Given this relationship, many hospitals have
sought to increase utilization of their RRSs to realize improvements in outcome.

That RRSs should be able to improve patient outcomes has strong face validity. These
outcomes include the incidence of cardio-respiratory arrests and unexpected mortality. All but a
small number of cardio-respiratory arrests have clear antecedents indicating that the patient is
deteriorating, yet these signs and symptoms of deterioration are not recognized or recognition is
delayed. In usual care, even when recognition of deterioration occurs, the process of responding
to that patient runs into a range of barriers, including a culture of medicine that is not patientcentered (i.e., concepts of patient ownership, autonomy, respect for authority and the chain of
command) and imbalances in the need (patient) to resource (available physicians, nurses,
respiratory therapy, monitoring etc.) ratio. These combined problems of poor recognition and/or
poor response create the opportunity for intervention. RRSs have been the primary intervention
of choice for the last decade to address the problems of poor recognition (afferent limb) and poor
response (efferent limb).
The afferent limb defines the parameters that indicate deterioration and democratizes that
knowledge to all clinicians. It also often allows for bedside clinicians (primarily nurses) to
trigger the Efferent Limb, even in cases where individual thresholds are not met but the bedside
clinician has a sense that something is not right. Since these signs often exist for hours before a
crisis actually occurs, improving the recognition process through defined criteria and
democratization of knowledge should lead to earlier recognition and hopefully intervention
before the patient becomes too unstable to be rescued. Providing a critical care response team
258
that can be directly triggered should also help to circumvent the delays that typically occur in
summoning a physician or higher level expertise. Together, these two elements (afferent and
efferent limbs) should catch treatable problems early before they are life-threatening. Finally, the
feedback component should help make clinicians aware of the need and benefits of using the
RRS, the quality improvement component should ensure improvement or maintenance over time,
and the administrative component should ensure that adequate resources are available to respond
to patient rescue needs.

RRSs were not addressed as a topic in Making Health Care Safer. RRS have mostly been
implemented and evaluated since 2000, although a small number of hospitals such as Dandenong
Hospital in Australia and University of Pittsburgh in the U.S established them in the mid1990s.1,3
For this review, a total of 2177 unique abstracts were captured by the search strategy. Of
these, 1,982 were excluded during the abstract screening phase. A total of 174 additional articles
were excluded at the article screening phase. Twenty one articles in total met the inclusion
criteria for this systematic review. Twenty articles met the inclusion criteria for intervention
studies evaluating the effectiveness of rapid response systems and 15 articles met the inclusion
criteria for intervention studies evaluating the implementation of rapid response systems.
We identified seven systematic reviews of RRSs: The one high-quality review is described
below. A second review addressed implementation, and we discuss it in that section. We
excluded two reviews from 2007 that contained many fewer publications than reviews published
in 2009 or later.4,5 We also excluded three additional reviews with low AMSTAR criteria scores
(5-6/11); they generally cover the same literature and time period, and report similar findings.6
The highest-quality systematic review and only meta-analysis7 (AMSTAR criteria score
10/11) identified 18 studies from 17 publications through November 2008, involving nearly 1.3
million hospital admissions. The meta-analysis concluded that, among adults, implementation of
an RRS was associated with a statistically significant reduction in cardiopulmonary arrest outside
the intensive care unit (ICU) (relative risk [RR], 0.66; 95% confidence interval [CI], 0.54 to
0.80) but not with lower hospital mortality (RR, 0.96). In children, implementation of an RRT
was associated with statistically significant reductions in both cardiopulmonary arrest outside the
ICU (RR, 0.62; 95% CI, 0.46 to 0.84) and hospital mortality (RR, 0.79; 95% CI, 0.63 to 0.98).
The review assessed studies as high quality if they adjusted for confounding and for time trends
by using either concurrent control groups or an interrupted time series design. Studies were rated
as fair quality if they adjusted only for confounding. Five studies were rated high quality, two as
fair quality, and the rest were rated as low quality.
This review identified two cluster-randomized, controlled trials (RCTs) but treated one in
their meta-analysis as a concurrent cohort controlled study (the MERIT Study) and the other as a
before-after historically controlled trial (Priestley, 2004 which used 3 different methodologies in
their analysis one of which was a before-after control). A key finding was that the major
multicenter RCT (the MERIT study) did not show an effect in the main analysis. However, in
further analysis, the change in arrest rate was exactly as expected given the utilization rates, and
exposure to the intervention was well below that which is necessary to realize a significant
change. The implication was that the implementation of RRSs may be critical to their success.
Additionally, in the MERIT trial, the intervention hospitals did see a statistically significant
improvement compared with their baseline period (before/after historical control), but the control
259
hospitals demonstrated essentially the same before/after improvement as the intervention

hospitals. The end result was no difference between intervention and control hospitals. Reasons
for the lack of difference may include other systems changes that improved care or decreased
mortality, or the Hawthorne effect since the intervention could not be blinded. Post-hoc analysis
did show that control hospitals increased their code team calls for non-code events, suggesting
that they engaged in RRS-like activities using their existing cardiac arrest teams.8
We identified 20 additional effectiveness studies that met our inclusion criteria published
since this systematic review. None were randomized trials or had a concurrent control group, and
only one study included multiple centers. Three studies were in pediatric hospitals. Most
occurred in the United States, Australia, or Canada, with only a few in Europe or Asia; most
studies were conducted in teaching hospitals. Almost no studies included any information on
context, and no studies reported a theoretical or logic model. The number of included hospital
admissions or discharges during the study periods ranged from 2426 to 277,717.
Most studies reported the main outcomes of total hospital or non-ICU cardiac arrests and
total hospital mortality; some studies also reported variations on these outcomes, such as
unexpected or non-DNR cardiac arrests or mortality. Of those studies that reported results and
statistics on total hospital (or non-intensive care unit) mortality, 8/14 (57%) reported statistically
significantly decreased mortality in the period after the RRS was implemented; one study
reported decreased mortality only on the medical (not the surgical) service (the study had
separate RRSs for the two services). Two studies that also reported non-DNR or unexpected
death rates in addition to in-hospital mortality also found a significant decrease in those
outcomes.
Of the studies that reported the outcome of cardiac arrest, 9/14 (64%) reported a significant
decrease after implementation of the RRS. One study reported unexpected cardiac arrest and
found no significant change; one study reported unplanned intubations and found no significant
change. Finally, of the 13 studies that reported outcomes and statistical testing for both cardiac
arrest and for mortality, 4 (31%) found different results for these 2 outcomes: 2 found significant
results for mortality but not for cardiac arrest, and 2 found significant results for cardiac arrest
but not mortality.
The overall strength of evidence for this topic was low. Risk of bias was high for all studies
due to study design issuesthere were no studies using any type of randomization since the
multi-institution MERIT study published in 2005; almost all studies were pre-post, with no
interrupted time series or concurrent controls. Few studies reported or accounted for differences
in patient populations over time or reported characteristics of providers in the two time periods.
Few studies reported or attempted to control for secular trends over time that could have
impacted mortality or cardiac arrest rates. The one study that did account for secular trends over
time in these outcomes found that, after adjusting for them, the changes in mortality and cardiac
arrest rate were no longer statistically significant. No studies reported on or accounted for other
safety initiatives in the hospital that might have also contributed to trends in decreasing mortality
or cardiac arrests.
No studies conducted blinded outcome assessment; although mortality is an objective
outcome, the other key outcome measured, incidence of cardiac arrest, can be defined in a
number of different ways (e.g., calling the code team vs. documented use of cardiac
compressions, stopped breathing, etc.) and is subject to bias, as are some of the other variations
in outcomes reported in some studies (e.g., unexpected mortality vs. total mortality, which
required retrospective, implicit assessment of medical records). Ideally, studies should report
260
cardio-respiratory arrest (codes) rates outside of the ICU and Emergency Departments since
these patient populations are not part of the exposure group (RRSs do not respond to these
locations), yet often hospital-wide rates were reported. One study9 included ICU arrests in their
analysis, concluding there was no effect, though data presented on their non-ICU code rate
showed a statistically significant difference. Cardiac arrest rates are also affected by changes in
patient casemix over time and the frequency of do-not-resuscitate orders and terminal illness,
which most studies did not account for.
Most studies reported in-hospital mortality. Only one reported longer-term mortality (such as
180-day mortality) reflecting patient survival more accurately. Most other outcomes reported,
such as the cardiac arrest rate, unanticipated intensive care unit admissions, or other health care
utilization measures are also indirect outcomes. In terms of precision, we did not identify any
additional studies that would have been assessed as high-quality in the 2009 meta-analysis7 all
would have been fair or poor quality. Evidence for association of RRSs with lower in-hospital
mortality was not strong. A summary table is located below (Table 1).
Table 1, Chapter 24. RRS summary table: effectiveness
Author, Year Description of PSP
Study Design Outcomes: Benefits*
Anwar ul,
PICU physicians (Pediatric MET)
Pre-post
Cardiac arrest: Y
10
2010
11
Bader, 2009 Nurse led, with Critical care outreach component
Pre-post
Mortality (non ICU): NR
(proactive rounding on ICU-discharged patients
Cardiac arrest: Y
Benson,
Advanced practice nurses (APN) with intensivists
Pre-post
Mortality: Y
12
2008
and other disciplines involved as needed
Cardiac arrest: NR
Campello,
ICU physician and ICU nurse
Pre-post
Mortality: Y
13
2009
Cardiac arrest: Y
9
Chan, 2008 Respiratory therapist and 2 ICU nurses (RRT model) Pre-post
Mortality: N
Cardiac arrest (hospital-wide): N
Cardiac arrest (non-ICU): Y
Gerdik,
RRT (specifics not described) including option for
Pre-post
Mortality: N
14
2010
patient and family activation
Cardiac arrest: NR
Hanson,
PICU fellow, resident, nurse and respiratory therapy Pre-post
Mortality: N
15
2009
Cardiac arrest (ward): N
16
Hatler, 2009 ICU nurse and respiratory therapy (RRT model)
Pre-post
Cardiac arrest: NR
Konrad,
17
2010
Kotsakis,
18
2011
Laurens,
19
2011
Lighthall,
20
2010
ICU nurse and ICU physician
Pre-post
Peds ICU attending and/or fellow, respiratory

therapists and ICU nurse, family activation
MET: anesthesiologist, medical house officer and
ICU/ED nurse
MET: ICU fellow, anesthesiologist nurse,
pharmacist, respiratory therapist
Pre-post
MedinaMET (no specifics given)

21
Rivera, 2011
Rothberg,
Hospitalist-led MET -critical care nurse, respiratory
22
2011
therapist, intravenous therapist, physician
Santamaria, MET: ICU registrar, general medical registrar and
23
2010
the ICU nurse
24
Sarani, 2011 2 METs surgery, medicine; critical care nurse,
pharmacy, respiratory therapy, resident, ICU
attending /fellow
25
Scott, 2009 ICU nurse and respiratory therapy (RRT model)
26
Shah, 2011 Critical care nurse and respiratory therapist (RRT
model)
261
Pre-post
Pre-post
Pre-post
Pre-post
Pre-post
Pre-post
Pre-post
Pre-post
Mortality (adjusted total): Y

Cardiac arrest: Y
Mortality (hospital): N
Cardiac arrest: N
Mortality: Y
Cardiac arrest: Y
Mortality (all): Y
Mortality (non-DNR): Y
Cardiac arrest: Y
Mortality: N
Cardiac arrest: N
Mortality (overall hospital): N
Cardiac arrest: Y
Mortality (unexpected): Y
Cardiac arrest: Y
Mortality: Y (Medical service only)
Cardiac arrest: Y
Cardiac arrest: NR
Mortality (In-hospital): Y**
Cardiac arrests: N
Table 1, Chapter 24. RRS summary table: effectiveness (continued)

Author, Year
Snyder,
27
2009
Tibballs,
28
2009
Description of PSP
MET: critical care physician and nurses
Study Design Outcomes: Benefits*

Pre-post
Mortality: N
Unplanned intubations: N
MET: hospitals resuscitation officer, RN
Pre-post
Mortality (in-hospital): Y
coordinating position, ICU physician and RN, ED
Mortality (unexpected): Y
physician and RN
Cardiac arrest (unexpected): N
*Overall results statistically significant Yes, No, or NR (Not reported no statistics reported)
** Significant in early time period but not later

Potential harms include decrease in the skills of ward staff due to dependence on the RRS,
inappropriate patient care for other patients (decreased responsibility or responsiveness of the
usual team), staff conflict, and diversion of critical care staff from usual care in the ICU.2
Unexpected beneficial consequences include improvements in the frequency and quality of endof-life discussions with patients and their families.
Despite several papers discussing these potential harms and unexpected consequences,
neither the high-quality systematic review7 nor any of the additional studies we identified
reported any harms or unexpected consequences.

What Contexts?
External factors. The need for programs such as RRSs is part of the Joint Commissions
National Patient Safety Goals (Goal #16): organizations should select a suitable method that
enables health care staff members to directly request additional assistance from a specially
trained individual(s) when the patients condition appears to be worsening. While this goal does
not specifically state RRSs as the correct strategy for meeting the goal, RRS have been the near
exclusive response to this requirement. RRSs are also included as one of several interventions in
the Institute for Healthcare Improvements 100K and 5 Million Lives Campaigns
(www.ihi.org/ihi/programs/campaign).
Structural organizational characteristics. In the high-quality review, of the 12 studies that
reported academic status, 10 were in academic centers and 1 multicenter study included
academic and community hospitals. Studies were mainly from Australia and the United States; 2
were in England and 1 was in Canada.7
Teamwork/leadership/patient safety, management tools. While the systematic reviews of
RRSs we identified and reviewed did not address issues such as teamwork and leadership several
papers did so individually.
Jones et al29 analyzed the literature for the implementation issues of factors impacting nurses
use of Medical Emergency Teams. Five major themes emerged: education on the MET,
expertise, support by medical and nursing staff, nurses familiarity with and advocacy for the
patient, and nurses workload.
Rapid response systems have been implemented in a variety of contexts (different countries,
and hospital and patient characteristics) and have varied in their composition, activation criteria,
and implementation process. In term of composition, the RRS studies reviewed might include
physicians, nurses, respiratory therapists, and other staff with different training or based in
262
different settings (intensive care unit, emergency room), as well as different management,
administrative staff, or quality oversight involvement. The majority of studies utilized
interdisciplinary teams comprised of at least one physician and one nurse. However, several
studies examined alternative RRS configurations. For example, two studies examined systems
that leveraged nurse leaders or nurse liaisons as primary first responders.12,30 Implementation
processes varied widely, often guided by the Institute on Healthcare Improvement (IHI)
suggestions or using IHI materials. Education and promotion of the new service was often a
factor, although actual staff training (such as simulation training) was uncommon. A variety of
different objective criteria were used for calling the team, and some interventions depended on
nurses clinical judgment; a few studies also developed and promoted a system for family or
patient initiation of the team.
Fifteen studies met our inclusion criteria for studies of the implementation processes
surrounding Rapid Response Systems. Eleven of these studies used quantitative methods,
primarily for evaluating the impact of a change in the implementation process for an RRS
program, and four used primarily qualitative methods such as interviews or focus groups of staff
regarding RSS implementation issues.31-33 The majority of implementation studies were
conducted in academic hospitals; however two studies specifically detailed implementation
efforts within community hospitals.31,34 Another study also examined the effects of separating
the overall emergency response system into two teams with different activation criteria and
processes in order to increase utilization.35 Results indicated significant increases in utilization
(15.7calls/1000 admissions vs. 24.7 admissions/1000admissions, p < .0001) after changes were
implemented.
Activation criteria and reasons for activation were focal study topics related to RRS
implementation. Several studies included subjective activation criteria (e.g., staff were worried
that a patient was at risk for an adverse event) in addition to traditional activation triggers based
on vital sign abnormalities.8,35 For example, one study that examined data from the MERIT trial
found that MET hospitals were 35 times more likely to activate their emergency response team
based upon this worried criteria compared with control hospitals (14.1% of activations vs.
0.4% of activations, p < .001).8
Descriptions of themes in the implementation processes included the categories of
technology and tools, staff and training, and barriers and facilitators. In terms of technology and
tools, no studies reported use of technology (such as computerized alerts) in RRS
implementation. Tools mentioned included changing activation criteria, triggers, or activation
methods, including one study changing to mandatory activation based on alert criteria; and
review of events, feedback, and rewards. In terms of staff and education, several implementation
studies brought on new staff, such as a nurse educator or liaison. Most studies indicated that
implementation processes explicitly included educational activities; however, these varied in the
degree to which they were strictly information-based (e.g., emails, meetings) or included
dedicated training and practice opportunities for either RRS members or staff. The majority of
studies also explicitly noted that on-the-job cognitive aids such as posters with activation criteria
or badge cards listing activation criteria were included. Finally, barriers and facilitators
mentioned included knowledge of activation criteria and other knowledge and attitudes about the
RRS; communication, teamwork, and lack of criticism for calling the team; perceptions about the
teams helpfulness to nurses and patients; and the importance and role of RRS champions.
One study specifically examined MET processes over time with the maturation of the MET
(and therefore potentially higher team skill level and more acceptance from ward staff). The
263
study found that the proportion of patients with delayed MET activation was significantly lower
(40.3% vs. 22%, p <.001) and that the proportion of patients with unplanned ICU admissions
was lower in a later cohort (31.3% vs. 17.5%, p < .001). A summary table is located below
(Table 2).
Table 2, Chapter 24. RRS summary table: implementation studies
Author, year
Adelstein,
36
2011
Buist, 2007
30
Calzavacca,
37
2010
Chen, 2010
Cretikos,
38
2007
Donaldson,
33
2009
Foraida,
39
2003 ,
DeVita,
3
2004
Genardi,
34
2008
Main Study objective

To assess if new strategies could improve
the time to delivery of MET
To assess impact of change programs

(education for nurses and housestaff)
To assess impact of maturation of an RRS
on the failure to rescue rate (recognition of
deterioration) and associated outcomes
To compare reasons for calling
emergency help between hospitals with a
MET and those without
To assess process components of MET
implementation correlated with utilization
To identify factors associated with

successful implementation across
hospitals- qualitative
To determine if specific educational and
feedback interventions would increase
MET utilization
To revitalize existing RRT and improve
code reductions
Jones,
35
2006
To assess whether systems changes in

existing MET would increase utilization
rate
Jones,
40
2006
Jones,
41
2010
To assess education program to increase

utilization of existing MET
To determine if mandatory MET activation
improves outcomes compared with
elective
To determine nurses perceptions of RRS
impact on practice and what constitutes a
successful RRS qualitative
Shapiro,
32
2010
264
Implementation Themes
Tools: centralized activation system, review of all
events, automatic escalation to code team if MET did
not respond within 30 min
Staff/training: nurse educator for training and
compliance
Staff/training: nurse liaison, development and
education
Barriers/facilitators: Maturation of system over time
Barriers/facilitators: worry about the patient, effect of

teaching hospital, metropolitan hospital, patient
location and time of activation
Barriers/facilitators: knowledge of activation criteria,
understanding of MET purpose, perceptions of
readiness for change, overall attitude to MET
program
Barriers/facilitators: Extra resources, rapid transfer,
communication enhancement, one stop shopping
(single team assessment), strength of adoption
Tools: immediate review of all stat sequential paging
events, feedback to those involved in delaying MET
activation, creating better objective alert criteria,
dissemination and education for those new criteria.
Tools: rewards program (recognition of effort),
improved documentation, alter alert criteria, increase
access to RRT, change to centralized paging
Staff/education: education, support for nurses, critical
thinking skills, ensure competencies
Tools: Method of activation (changing activation
methods to separate the teams), triggers (changing
alert criteria for calling MET)
Staff/training: team composition (separation of unified
code/MET into separate teams with separate
activations), re-education on purpose of MET, criteria,
and the changes
Staff/training: education, improved communication,
on-the-job aids (e.g., posters, observational charts)
Tools: conversion from elective MET activation to
mandatory based on alert criteria
Barriers/facilitators: Nurse enthusiasm about teams;
clarity about when to call team; concerns about being
reprimanded for calling team; institutional and
individual inertia; concerns about who would care for
other patients during a call
Table 2, Chapter 24. RRS summary table: implementation studies (continued)

Author, year
42
Soo, 2009
Main Study objective

To evaluate major features of the patient
safety practice champion role
Williams,
31
2011
To clarify nurse perceptions of RRS

qualitative
Barriers/facilitators: Both executive and managerial
champions were important; were skilled
communicators, well-respected and familiar with
institutional culture. Champions were educators,
advocated for RRT, built relationships, and navigated
boundaries between professions/units.
Barriers/facilitators: advantages of RRT to nurses
(develops skills, autonomy, resource and way to
circumvent unit problems), perceived benefits for
patients; degree of teamwork with RRT; RRT skills;
concerns about activating an RRT

This was not evaluated in the high-quality systematic review by Chan in 20107 or in any of
the additional effectiveness or implementation articles that we reviewed.

The high-quality meta-analysis concluded that RRSs were associated with significantly
reduced hospital mortality in pediatric but not in adult populations.7 Effectiveness appeared high
in earlier studies, but less in later studies. In our update, however, we found the opposite to be
true. We found that the most recent studies are more likely to demonstrate positive results for
mortality. In fact, there were 7 studies in a row, starting with Kenward in 2004 and continuing to
Chan in 2008, where the point estimate of effect doesnt go below 0.95. After Chan 2008, all
point estimates are < 0.95. Potential explanations for this include maturation of the intervention
and improved implementation strategies that may have led to improved results within and across
institutions.
We did not find any studies evaluating the impact of context on effectiveness. One study that
had two separate MET teams for the two groups showed an impact in a medical, but not a
surgical population.24

In summary, a previous high-quality meta-analysis of 18 studies published from 1990
through November 2008 found that although RRSs were associated with a significant reduction
in rates of cardiopulmonary arrest outside the intensive care unit, there was a significant
reduction in mortality only in pediatric studies (not in studies in adults). Our update identified an
additional 20 studies, none of which was high quality, and the strength of evidence in those
studies for the impact of RRSs on in-hospital mortality in both adult and pediatric populations
was low.
Tools mentioned in qualitative and quantitative implementation studies included changing
activation criteria, triggers, or activation methods, but technology was not mentioned. In terms of
staff and education, themes included bringing on new staff or educational activities, but efforts
were mainly focused on information rather than training. Finally, barriers and facilitators
mentioned included knowledge and attitudes about the RRS; communication, teamwork, and
lack of criticism for calling the team; and perceptions about the helpfulness of the team for
nurses and patients. Studies included little information about context, and we found no evidence
about how context impacted effectiveness or implementation.
265
Despite their strong face validity, RRSs have exhibited mixed results in the literature. There
are several potential explanations for this--none mutually exclusive. The afferent limb can
provide clear definitions to identify which patients are likely deteriorating and can educate staff
on those definitions. However, activation triggers were originally developed through clinical
chart review of patients who had arrested or been transferred to the ICU, and subsequent
attempts to improve upon this model have not generated a better approach. Studies of aggregate
scores, weighted scores, and single parameter triggers have not demonstrated clear superiority of
one over another.2 Confounding this approach is the way that vital signs, which constitute most
of the data for afferent limb systems, are collected. On general wards, vital signs are, at best,
collected every 4 hours and more typically every 6 or 8 hours, leaving ample time for
deterioration to develop unrecognized. The fidelity with which vital signs are collected and
recorded is also known to be poor,1 amplifying the problem. Finally, vital signs are not the only
variable predicting risk of deterioration. Weighted and aggregated scores try to address this
issue, but the interconnectedness of these changes is complex and varies with specific
populations.
There are also a number of issues with the implementation of the efferent limb (the RRS
team). Optimal team composition is unknown, including the structure (including a physician or
not, level or training, and overall team composition), and whether the RRS should be unified
with the code team or be separate not only in function but in personnel. Hospitals are reluctant to
fund free standing RRSs whose only responsibility is to attend to deteriorating patients and/or
arrests. As a result, RRS team members need to leave other duties (often caring for critically ill
patients in the ICU) to respond. This may limit the available resources they can bring to the ward
patient and risks harm to the patients they have stepped away from. Restricted financial
resources may also impact the RRSs ability to selfaudit and evaluate code events and
unanticipated ICU transfers that occur outside an RRS intervention. As a consequence, the RRS
cannot make appropriate assessments in order to improve systematically. Efforts to improve
utilization may likewise suffer, especially given evidence that utilization (dose) matters, that
utilization can be improved with changes in implementation strategies, and that many programs
have low utilization rates. Utilization of RRSs is reported to be low often because of issues with
the culture of safety, including reluctance on the part of the ward staff to activate the team.
Finally, there are a number of issues regarding how outcomes in RRS studies are measured.
Cardiac arrests and hospital mortality can be affected by many other factors such as patient
characteristics and other aspects of care, including trends over time in reducing hospital mortality
and length of stay and in caring for more terminally ill patients outside the hospital setting.
Additionally, several metrics commonly used to evaluate RRSs count patients who are not
exposed to the intervention (i.e., total hospital mortality), potentially affecting the results.
Unfortunately, using metrics such as preventable general ward-only mortality is more difficult
and potentially introduces bias (chart review to determine preventability of a death).
In summary, RRSs are clearly associated with decreased rates of cardiopulmonary arrest, but
the question of whether RRSs as currently defined and implemented affects mortality is unclear.
Insufficient evidence exists on the impact of context, different implementation strategies, or RRS
structure. RRSs are not likely to realize their full potential for improving outcomes without
accurate, more frequent (possibly even continuous) and integrated patient specific data to inform
the afferent limb, an understanding of what team structure and training works best, greater
commitment to fully support RRSs so they can carry out all necessary functions unencumbered,
266
a greater focus on patient-centered care and patient safety, and improved measurement and
reporting. A summary table is located below (Table 3).
Scope of the
Strength of
the PSP
Effectiveness
Common/High
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate
References
1.
Bristow PJ, Hillman KM, Chey T et al. Med

J Aust: Rates of in-hospital arrests, deaths
and intensive care admissions: the effect of a
medical emergency team. 2000; 173:236-40.
2.
Jones DA, DeVita MA, Bellomo R. Rapidresponse teams. N Engl J Med 2011;
365(2):139-46.
3.
DeVita MA, Braithwaite RS, Mahidhara R,

Stuart S, Foraida M, Simmons RL. Quality
& Safety in Health Care: Use of medical
emergency team responses to reduce
hospital cardiopulmonary arrests. 2004; 2514.
4.
Winters BD, Pham JC, Hunt EA, Guallar E,

Berenholtz S, Pronovost PJ. Crit Care Med:
Rapid response systems: a systematic
review. 2007; 35:1238-43.
5.
McGaughey J, Alderdice F, Fowler R,

Kapila A, Mayhew A, Moutray M.
Cochrane Database Syst Rev: Outreach and
Early Warning Systems (EWS) for the
prevention of intensive care admission and
death of critically ill adult patients on
general hospital wards. 2007; CD005529.
6.
Massey D, Aitken LM, Chaboyer W. J Clin

Nurs: Literature review: do rapid response
systems reduce the incidence of major
adverse events in the deteriorating ward
patient? 2010; 19:3260-73.
7.
Chan PS, Jain R, Nallmothu BK, Berg RA,

Sasson C. Arch Intern Med: Rapid Response
Teams: A Systematic Review and Metaanalysis. 2010; 170:18-26.
8.
Chen J, Bellomo R, Hillman K, Flabouris A,

Finfer S. J Crit Care: Triggers for
emergency team activation: a multicenter
assessment. 2010; 25:359 e1-7.
267
9.
Chan PS, Khalid A, Longmore LS, Berg

RA, Kosiborod M, Spertus JA. JAMA:
Hospital-wide code rates and mortality
before and after implementation of a rapid
response team. 2008; 300:2506-13.
10.
Anwar ul H, Saleem AF, Zaidi S, Haider

SR. Indian J Pediatr: Experience of pediatric
rapid response team in a tertiary care
hospital in Pakistan. 2010; 77:273-6.
11.
Bader MK, Neal B, Johnson L et al. Jt

Comm J Qual Patient Saf: Rescue me:
saving the vulnerable non-ICU patient
population. 2009; 35:199-205.
12.
Benson L, Mitchell C, Link M, Carlson G,

Fisher J. Jt Comm J Qual Patient Saf: Using
an advanced practice nursing model for a
rapid response team. 2008; 34:743-7.
13.
Campello G, Granja C, Carvalho F, Dias C,

Azevedo LF, Costa-Pereira A. Crit Care
Med: Immediate and long-term impact of
medical emergency teams on cardiac arrest
prevalence and mortality: a plea for periodic
basic life-support training programs. 2009;
37:3054-61.
14.
Gerdik C, Vallish RO, Miles K, Godwin SA,

Wludyka PS, Panni MK. Resuscitation:
Successful implementation of a family and
patient activated rapid response team in an
adult level 1 trauma center. 2010; 81:167681.
15.
Hanson CC, Randolph GD, Erickson JA et

al. Qual Saf Health Care: A reduction in
cardiac arrests and duration of clinical
instability after implementation of a
paediatric rapid response system. 2009;
18:500-4.
16.
Hatler C, Mast D, Bedker D et al. Medsurg

Nurs: Implementing a rapid response team
to decrease emergencies outside the ICU:
one hospitals experience. 2009; 18:84-90,
126.
17.
Konrad D, Jaderling G, Bell M, Granath F,

Ekbom A, Martling CR. Intensive Care
Med: Reducing in-hospital cardiac arrests
and hospital mortality by introducing a
medical emergency team. 2010; 36:100-6.
18.
Kotsakis A, Lobos AT, Parshuram C et al.

Pediatrics: Implementation of a Multicenter
Rapid Response System in Pediatric
Academic Hospitals Is Effective. 2011;
128:72-8.
19.
Laurens N, Dwyer T. Resuscitation: The

impact of medical emergency teams on ICU
admission rates, cardiopulmonary arrests
and mortality in a regional hospital. 2011;
82:707-12.
20.
Lighthall GK, Parast LM, Rapoport L,

Wagner TH. Anesth Analg: Introduction of
a rapid response system at a United States
veterans affairs hospital reduced cardiac
arrests. 2010; 111:679-86.
21.
Medina-Rivera B, Campos-Santiago Z,
Palacios AT, Rodriguez-Cintron W. Critical
Care and Shock: The effect of the medical
emergency team on unexpected cardiac
arrest and death at the VA Caribbean
Healthcare System: A retrospective study.
2010; 13:98-105.
22.
Rothberg MB, Belforti R, Fitzgerald J,

Friderici J, Keyes M. Journal of Hospital
Medicine: Four years experience with a
hospitalist-led medical emergency team: An
interrupted time series. 2011.
23.
Santamaria J, Tobin A, Holmes J. Crit Care

Med: Changing cardiac arrest and hospital
mortality rates through a medical emergency
team takes time and constant review. 2010;
38:445-50.
24.
Sarani B, Palilonis E, Sonnad S et al.

Resuscitation: Clinical emergencies and
outcomes in patients admitted to a surgical
versus medical service. 2011; 82:415-8.
25.
Scott SS, Elliott S. Crit Care Nurse:

Implementation of a rapid response team: a
success story. 2009; 29:66-75; quiz 76.
268
26.
Shah SK, Cardenas VJJr, Kuo YF, Sharma

G. Chest: Rapid response team in an
academic institution: does it make a
difference? 2011; 139:1361-7.
27.
Snyder CW, Patel RD, Roberson EP, Hawn

MT. American Surgeon: Unplanned
intubation after surgery: Risk factors,
prognosis, and medical emergency team
effects. 2009; 75:834-8.
28.
Tibballs J, Kinney S. Pediatric Critical Care

Medicine: Reduction of hospital mortality
and of preventable cardiac arrest and death
on introduction of a pediatric medical
emergency team. 2009; 10:30612+423+424+425.
29.
Jones L, King L, Wilson C. J Clin Nurs: A

literature review: factors that impact on
nurses effective use of the Medical
Emergency Team (MET). 2009; 18:3379-90.
30.
Buist M, Harrison J, Abaloz E, Van Dyke S.

BMJ: Six year audit of cardiac arrests and
medical emergency team calls in an
Australian outer metropolitan teaching
hospital. 2007; 335:1210-2.
31.
Williams DJ, Newman A, Jones C, Woodard

B. J Nurs Care Qual: Nurses perceptions of
how rapid response teams affect the nurse,
team, and system. 2011; 26:265-72.
32.
Shapiro SE, Donaldson NE, Scott MB. Am J

Nurs: Rapid response teams seen through
the eyes of the nurse. 2010; 110:28-34; quiz
35-6.
33.
Donaldson N, Shapiro S, Scott M, Foley M,

Spetz J. J Nurs Adm: Leading successful
rapid response teams: A multisite
implementation evaluation. 2009; 39:17681.
34.
Genardi ME, Cronin SN, Thomas L. Dimens

Crit Care Nurs: Revitalizing an established
rapid response team. 2008; 27:104-9.
35.
Jones DA, Mitra B, Barbetti J, Choate K,

Leong T, Bellomo R. Anaesth Intensive
Care: Increasing the use of an existing
medical emergency team in a teaching
hospital. 2006; 34:731-5.
36.
Adelstein BA, Piza MA, Nayyar V,

Mudaliar Y, Klineberg PL, Rubin G. J Crit
Care: Rapid response systems: A
prospective study of response times. 2011.
37.
Calzavacca P, Licari E, Tee A et al.

Resuscitation: The impact of Rapid
Response System on delayed emergency
team activation patient characteristics and
outcomes--a follow-up study. 2010; 81:31-5.
38.
Cretikos MA, Chen J, Hillman KM,

Bellomo R, Finfer SR, Flabouris A. Crit
Care Resusc: The effectiveness of
implementation of the medical emergency
team (MET) system and factors associated
with use during the MERIT study. 2007;
9:206-12.
39.
Foraida MI, DeVita MA, Braithwaite RS,

Stuart SA, Brooks MM, Simmons RL. J Crit
Care: Improving the utilization of medical
crisis teams (Condition C) at an urban
tertiary care hospital. 2003; 18:87-94.
40.
Jones D, Bates S, Warrillow S et al. Intern

Med J: Effect of an education programme on
the utilization of a medical emergency team
in a teaching hospital. 2006; 36:231-6.
41.
Jones CM, Bleyer AJ, Petree B. Joint

Commission journal on quality and patient
safety / Joint Commission Resources:
Evolution of a rapid response system from
voluntary to mandatory activation. 2010;
36:266-70, 241.
42.
Soo S, Berta W, Baker GR. Role of

champions in the implementation of patient
safety practice change. Healthc Q. 2009;12
Spec No Patient:123-8.
269
Chapter 25. Medication Reconciliation Supported by Clinical

Pharmacists (NEW)
Lisha Lo, M.P.H.; Janice Kwan, M.D.; Olavo A. Fernandes, B.Sc.Phm., Pharm.D.; Kaveh G.
Shojania, M.D.
We have specified support by clinical pharmacists in the title for this review, because the
evidence for the clinical impact of medication reconciliation exclusively involves interventions
in which pharmacists play a key role. We regarded this specification as important since
accreditation standards to implement medication reconciliation do not require involvement by
pharmacists. Thus, medication reconciliation as implemented in many hospitals may not achieve
the same impacts reported in the literature that led to this required practice.

Transitions in care, such as admission and discharge from an acute care hospital or changes
in setting within a hospital, place patients at risk for errors due to poor communication and
inadvertent information loss. Unintended medication discrepancies represent one well-studied
category of such patient safety problems related to information loss at transitions of care.
When patients are admitted to or discharged from a hospital, treating physicians may
intentionally make changes to patients preadmission medication regimens. However, they may
also make changes unintentionally (e.g., as the result of not being aware of the full list of
preadmission medications or having inaccurate information on the most recent doses). Published
studies suggest that 4054% of patients experience unintentional medication discrepancies upon
admission to acute care hospitals.1 Slightly higher rates of unintentional discrepancies may occur
during internal hospital transfers (e.g., intensive care unit to ward), and at least 40% of patients
experience discrepancies at hospital discharge.2-5 A recent large observational study using
population data from Ontario, Canada showed that 187,912 patients admitted to a hospital were
at significantly increased risk for unintentional discontinuation of chronic, evidence-based
therapies as compared with control patients not admitted to a hospital (n = 208,468).6 Admission
to an intensive care unit carried an even greater risk of unintentional discontinuation of these
medications.
Not all unintended discrepancies carry substantial risks for harm. In a systematic review of
unintended discrepancies at hospital admission, only five of 22 studies estimated the clinical
importance of errors in the medication history. The proportion of all discrepancies likely to cause
clinical problems, as estimated by these five studies, ranged widely from 11% to 59%.1 The two
common categories of unintended discrepancies that contribute to clinical risk are omissions
key prescription and non-prescription medications are inappropriately not started or continued
(range: 4656% of all discrepancies)and commissionsmedications that patients have
discontinued are inadvertently re-started.1,2,5

Medication reconciliation is the proposed formal, systematic strategy to overcome
medication information communication challenges and reduce unintended medication
discrepancies that occur at transitions in care (Figure 1). Ideally, health care providers from
270
different professions (physicians, nurses, pharmacists) work together and with patients (and their
families) to ensure the accurate and consistent communication of medication information across
transitions in care.
Figure 1, Chapter 25. Overview of medication reconciliation
Adapted from Pharmacy Practice 2009;25(6):26 with permission
The World Health Organization (WHO) has prioritized medication reconciliation as one
of five top patient safety strategies, within the Action on Patient Safety: High 5s.7 National
campaigns targeting the reduction of preventable patient adverse events such as the Institute for
Healthcare Improvements 100,000 Lives Campaign in the United States (U.S.) as well as the
Canadian Patient Safety Institutes Safer Healthcare Now! have championed medication
reconciliation as one of a few core interventions. Furthermore, accreditation authorities such as
The Joint Commission in the U.S. and Accreditation Canada made medication reconciliation best
practices a mandatory requirement for various health care settings. However, of note, The Joint
Commission no longer formally scores medication reconciliation during accreditation surveys,
although the latter loosely remains part of the National Patient Safety Goal to Maintain and
communicate accurate patient medication information.8
The Best Possible Medication History (BPMH) constitutes the cornerstone for medication
reconciliation. The BPMH is more comprehensive than a routine primary medication history, as
it involves (1) a systematic process for interviewing the patient/family; and (2) a review of at
least one other reliable source of information (e.g., review of a central medication database,
inspection of medication vials, or contact with the community pharmacy) to obtain and verify
patient medications (prescribed and non-prescribed).9
271
Some may argue that ambulatory patients face greater risks from medication problems than
do patients in a protected hospital setting. Studies of ambulatory reconciliation have begun to
appear.10,11 However, most studies of medication reconciliation still focus on the hospital setting,
which remains the focus of this review.

One previous systematic review12 has summarized the evidence on inpatient medication
reconciliation, but this review did not include quantitative synthesis. We sought to quantify the
impact of medication reconciliation on unintentional discrepancies with the potential for harm
(clinically significant discrepancies) and hospital utilization following discharge, as assessed
by unplanned emergency visits and readmission to hospital.
To evaluate these impacts of medication reconciliation, we searched major bibliographic
databases (MEDLINE, Embase, Cochrane CENTRAL) and scanned article reference lists.
Appendix C, Section A and B, present the search strategy, article flow, and methods. Eligible
studies reported emergency department visits and hospitalizations within 30 days of discharge or
evaluated the severity of clinical significance of unintentional discrepancies. We included
randomized controlled trials, before-after evaluations, and post-intervention studies.
All included 18 studies reporting 20 medication reconciliation interventions came from
hospitals in the United States or Canada (Table 1). We identified studies with interventions
related to medication reconciliation from other countries, but all met pre-specified reasons for
exclusion, such as not clearly distinguishing intended from unintended medication
discrepancies13-15 and assessment of clinical severity performed solely by personnel conducting
medication reconciliation.16,17
It is notable that all but three interventions involved pharmacists playing a major role, which
does not reflect routine practice, nor is it a requirement in the accreditation standard in either the
U.S. or Canada. Some of the studies also involved additional enhancements beyond medication
reconciliation itself (Table 2), such as the creation of a single place in the electronic medical
record (EMR) to enter and update the preadmission medication history, or functionality in the
EMR to facilitate creation of the pre-admission medication history. These characteristics
probably also differentiate medication reconciliation as reported in the literature from routine
practice.
significant unintended discrepancies and emergency department visits and hospitalizations
within 30 days of discharge
Study
Setting
Study Design
Transition
Targeted
Coffey,
200918
Pediatric ward in
academic medical
center in Canada
Admission to
hospital
Cornish,
20052
Medical ward in
academic medical
center in Canada
Prospective
post-intervention
study (272
patients)
Prospective
post-intervention
study (151
patients)
Admission to
hospital
272
Additional
Interventions
Beyond Medication
Reconciliation
None
Outcome
None
Clinically significant
unintentional
discrepancies
unintentional
discrepancies
significant unintended discrepancies and emergency department visits and hospitalizations within
30 days of discharge (continued)
Study
Setting
Study Design
Transition
Targeted
Gleason,
200419
Surgical and
medical wards in
U.S. academic
medical center
Admission to
hospital
Gleason,
201020
Medical ward in
U.S. academic
medical center
Kripalani,
201221
Medical and
cardiology wards
in two U.S.
academic medical
centers
Medical and
cardiology wards
in two U.S.
academic medical
centers
Post-intervention
study (204
patients, 12 adult
medical-surgical
units)
Prospective
post-intervention
study (651
patients)
RCT (428
patients)
Kripalani,
201221
Lee, 20105
Inpatient wards
and critical care
units in two
academic medical
centers in Canada
Pippins,
200822
Medical wards in
two U.S. academic
medical centers
Stone,
201023
Pediatric ward in
U.S. academic
medical center
Vira,
200624
Acute care units in

urban community
hospital in Canada
Wong,
20084
Medical ward in
academic medical
center in Canada
RCT (423
patients)
Prospective
post-intervention
study (129
patients, 10
patient care
units)
Prospective
post-intervention
study (180
patients, 7
medical teams)
Prospective
post-intervention
study (23
patients on 2
medical teams)
Retrospective
post-intervention
study (60
patients)
Prospective
post-intervention
study (150
patients)
Additional
Interventions
Beyond Medication
Reconciliation
None
Outcome
unintentional
discrepancies
Admission to
hospital
None
unintentional
discrepancies
At time of
enrollment in
study, discharge
home, and inhospital transfer
At time of
enrollment in
study, discharge
home, and inhospital transfer
Discharge counseling
unintentional
discrepancies
Pharmacist
intervention including
in-patient pharmacist
counseling, lowliteracy adherence
aids, and postdischarge phone call
None
unintentional
discrepancies
Discharge home
None
unintentional
discrepancies
Admission to
hospital
None
unintentional
discrepancies
Admission to
hospital;
discharge home
None
unintentional
discrepancies
Discharge home
None
unintentional
discrepancies
In-hospital
transfer
273
unintentional
discrepancies
Study
Setting
Study Design
Transition
Targeted
Schnipper,
200925
Medical wards in
two U.S. academic
medical centers
RCT (162
patients, 7
medical teams)
Admission to
hospital;
discharge home
Dedhia,
200926
Additional
Interventions
Beyond Medication
Reconciliation
None
Medical wards in
U.S. academic
medical center,
community
teaching hospital,
and urban
community
hospital
Medical ward in
U.S. academic
medical center
Prospective
before-after
study (185
patients)
Discharge home
Safe STEPS
intervention including
admission
assessment,
communication with
PCP, multidisciplinary
discharge meeting
RCT (373
patients)
Discharge home
Nurse discharge
advocates created
after-hospital care
plan, and postdischarge phone call
Koehler,
200928
Medical ward in
U.S. academic
medical center
RCT (21
Admission to
patients, 2
hospital,
hospitaldischarge home
medicine groups)
Counseling by
registered nurse
Koehler,
200928
Medical ward in
U.S. academic
medical center
RCT (20
Admission to
patients, 2
hospital,
hospitaldischarge home
medicine groups)
Kramer,
200729
Medical ward in
U.S. community
teaching hospital
Prospective
before-after
study (136
patients)
Admission to
hospital;
discharge home
Supplemental elderly
care bundle:
counseling by
pharmacist, postdischarge phone call,
discharge letter to
PCP
None
Schnipper,
200630
Medical ward in
U.S. academic
medical center
RCT (92
patients, 4
medical teams)
Discharge home
Jack,
200927
274
None
Outcome
unintentional
discrepancies
Emergency
department visits
and
hospitalizations
within 30 days of
discharge
Emergency
department visits
and
hospitalizations
within 30 days of
discharge
Emergency
department visits
and
hospitalizations
within 30 days of
discharge
Emergency
department visits
and
hospitalizations
within 30 days of
discharge
Emergency
department visits
and
hospitalizations
within 30 days of
discharge
Emergency
department visits
and
hospitalizations
within 30 days of
discharge
Emergency
department visits
and
hospitalizations
within 30 days of
discharge
Study
Setting
Study Design
Transition
Targeted
Additional
Outcome
Interventions
Beyond Medication
Reconciliation
Showalter, All admitted
Emergency
Retrospective
Discharge home Standardized
201131
mandatory electronic
department visits
patients through
before-after
discharge instructions and
emergency
study (17,516
department in U.S. patients)
document with
hospitalizations
academic medical
embedded
within 30 days of
computerized
discharge
center
medication
reconciliation
Walker,
Medical ward in
Prospective
Discharge home Pharmacist-facilitated Emergency
200932
U.S. academic
quasidischarge program
department visits
center
experimental
including counseling,
and
study (358
provision of
hospitalizations
patients, 2
medication
within 30 days of
medical teams
reconciliation list to
discharge
and 1 hospitalist
PCP, and postservice)
discharge phone call
Abbreviations: RCT, randomized control trial; PCP, primary care physician; Safe STEPS, Safe and Successful Transition of
Elderly Patients Study.
275
Table 2, Chapter 25. Key features of the 12 included medication reconciliation interventions
Study
Intervention
Coffey,
200918
Medication
reconciliation
by pharmacy
student
Medication
reconciliation
by pharmacist,
pharmacy
student, or
medical
student
Medication
reconciliation
by pharmacist
Cornish,
20052
Gleason,
200419
Selection for
More
Complex
Patients
No
4
medications
No
Medication
History
BPMH
performed by
pharmacy
student
BPMH
performed by
pharmacist,
pharmacy
student, or
medical
student
Structured
history
performed by
pharmacist or
PharmD
student
Structured
history
performed by
pharmacist
History
performed by
pharmacist
Electronic- or
Paper-Based
Medication
Reconciliation
Paper*
Institutional
Informatics
Functionality
CPOE*
Medication
Reconciliation
Became Order
Process
No*
Paper*
Limited*
No*
Paper*
Limited*
No*
Electronic*
EMR, CPOE*
No*
Electronic (at
one site)
EMR, CPOE,
Preadmission
Medication List
Builder
(embedded at
one site)
EMR, CPOE,
Preadmission
Medication List
Builder
(embedded at
one site)
EMR, CPOE
(partial)*
Yes (at one

site)
Not reported
Not reported
Not reported
Gleason,
201020
Medication
reconciliation
by pharmacist
No
Kripalani,
201221
Medication
reconciliation
by physician
and nurse
No
Kripalani,
201221
Medication
reconciliation
by pharmacist
with
pharmacist
intervention
Medication
reconciliation
by pharmacist
Medication
reconciliation
by pharmacist
Medication
reconciliation
by pharmacist
No
History
performed by
pharmacist
Electronic (at
one site)
No
BPMH
performed by
pharmacist
BPMH
performed by
pharmacist
BPMH
performed by
pharmacist
Both*
Paper*
EMR*
No*
BPMH
performed by
pharmacist
BPMH
performed by
pharmacist
Paper*
Limited*
No*
Paper*
EMR, CPOE*
No*
Lee, 20105
Pippins,
200822
Stone,
201023
Vira,
200624
Wong,
20084
Medication
reconciliation
by pharmacist
Medication
reconciliation
by pharmacist
or pharmacy
resident
No
Identification
of medically
complex
conditions
based on
published
guidelines
No
No
276
Yes (at one

site)
Not reported
(continued)
Study
Intervention
Schnipper,
200925
Medication
reconciliation
by physician
and confirmed
by pharmacist
or nurse
Medication
reconciliation
by physician
and reviewed
by pharmacist
Medication
reconciliation
by nurse
Medication
reconciliation
by nurse and
reviewed by
pharmacist
Dedhia,
200926
Jack,
200927
Koehler,
200928
Koehler,
200928
Medication
reconciliation
by pharmacist
with
supplementary
elderly care
bundle
Kramer,
200729
Medication
reconciliation
by pharmacist
and physician
Schnipper,
200630
Medication
reconciliation
by pharmacist
Medication
reconciliation
by physician
Showalter,
201131
Selection for
More
Complex
Patients
No
Age 65
None
Age 70, 5
medications,
3 chronic
comorbid
conditions,
requirement
for assistance
with 1 ADL
Age 70, 5
medications,
3 chronic
comorbid
conditions,
requirement
for assistance
with 1 ADL
One or more
of: 7
medications,
significant
comorbid
condition,
previous
admission for
ADR, 4 drug
allergies
None
None
Medication
History
Electronic- or
Paper-Based
Medication
Reconciliation
Electronic*
Institutional
Informatics
Functionality
Paper*
EMR, CPOE*
No*
Electronic*
EMR, CPOE*
No*
Not reported
Paper*
Limited*
No*
Not reported
Paper*
Limited*
No*
Structured
history
performed by
pharmacist
Electronic*
Limited
(pharmacist
electronic
order entry)*
Yes*
History
performed by
pharmacist
Not reported
Paper*
EMR, CPOE*
No*
Electronic*
EMR, CPOE,
electronic
discharge
program
(embedded as
force function)*
Not reported
BPMH
performed by
physician and
verified by
nurse and
pharmacist
History
performed by
physician and
reviewed by
pharmacist
Not reported
277
EMR, CPOE,
linkage to
Preadmission
Medication List
Builder*
Medication
Reconciliation
Became Order
Process
Partial (not at
time of study)*
(continued)
Study
Intervention
Selection for
Medication
Electronic- or
Institutional
Medication
More
History
Paper-Based
Informatics
Reconciliation
Complex
Medication
Functionality
Became Order
Patients
Reconciliation
Process
Walker,
Medication
One or more
History
Electronic*
EMR, CPOE,
No*
200932
reconciliation
of: 5
performed by
internal
by pharmacist
medications, pharmacist
electronic
1 targeted
pharmacy
medications**,
database*
medication
requiring
monitoring, 2
changes to
regimen,
dementia or
confusion, or
inability to
manage
medications
Abbreviations: BPMH, best possible medication history; CPOE, computerized physician order entry; EMR, electronic medical
record; ADL, activity of daily living; ADR, adverse drug reaction
*Information obtained by communication with authors.
**Targeted medications included digoxin, diuretics, anticoagulants, sedatives, opioids, asthma and/or chronic obstructive
pulmonary disease medications, angiotensin converting enzyme inhibitor and/or angiotensin receptor blocker.
Clinically Significant Unintended Medication Discrepancies

Studies varied in their definitions of clinical importance and categories of severity applied to
each medication discrepancy. However, all included studies reported a category that amounted to
trivial, minor, or unlikely to cause harm, with all other unintentional discrepancies deemed
to be clinically significant. This definition corresponds to the term, potential adverse drug
events (ADEs), though only three included studies explicitly used this term.21,25 We required that
assessments of clinical severity be performed by at least one clinician independent from the
medication reconciliation process. We also required an explicit statement that unintentional
discrepancies were distinguished from intentional medication changes, as well as a clear
description of the method for doing so.
As shown in Figure 2, rates for clinically significant discrepancies ranged from a low of 0.11
per patient to a high of 1.43. The only randomized controlled trial of medication reconciliation
vs. usual care yielded an estimate of 0.27 per patient, but this result included potential ADEs, not
just unintended discrepancies. This study is discussed in more detail below.
Across 13 medication reconciliation interventions, the median value for the number of
clinically significant unintentional discrepancies per patient was 0.35 (interquartile range [IQR]
0.25-0.88). Four interventions (2 from the same study21) reported notably higher values (Figure
2). No features of the intervention (Table 2), such as selection for high risk patients, inclusion of
additional interventions beyond medication reconciliation, or integration with clinical
informatics applications explained these outlying results.
278
Figure 2, Chapter 25. Median and interquartile range for the number of clinically significant
unintentional discrepancies per patient for the 13 included interventions
Only a minority of unintended discrepancies had clinical significance. The meta-analytic

mean for the proportion of unintended discrepancies that were clinically significant was 35.1%
(95% CI: 27.5%-43.6%). This result exhibited significant heterogeneity (I2 =92%) as the results
ranged from 15% to 54% (median: 34%, IQR 28%-49%). The meta-analytic average for the
proportion of patients with at least one clinically significant unintended discrepancy was 39.3%
(95% CI: 21.4%-60.5%). This result also exhibited significant heterogeneity (I2= 95%), due to a
wide range in values, from 15% to 60% (median 45%, IQR 31%-56%).
Only two randomized controlled trials25,30 evaluated the impact of medication reconciliation
in comparison with usual care using the established concept of adverse drug event (ADE). One
trial30 involved randomizing 178 patients being discharged from the medical service at a teaching
hospital in Boston to an intervention that included medication reconciliation and counseling by a
pharmacist, as well as a follow-up phone call within 5 days. For patients in the control arm,
nurses provided discharge counseling and pharmacists reviewed medication orders, but did not
perform a formal reconciliation process. Significantly fewer patients in the intervention arm
experienced preventable ADEs (1% vs. 11%; p=0.01), though total ADEs did not differ between
the two groups.
A subsequent, cluster randomized trial from the same research group involved 14 medical
teams at two teaching hospitals in Boston.25 The intervention included a web-based application
using the hospitals electronic medical record (which included ambulatory visits) to create a
preadmission medication list in order to facilitate the medication reconciliation process. Of note,
the interventions effect achieved statistical significance at one of the hospitals, with a relative
reduction of potential ADEs (equivalent to clinically significant unintended medication
discrepancies) of 0.72 (95% CI, 0.52-0.99), but not at the other (0.87, 95% CI, 0.57-1.32). The
authors attributed this difference to variation in the degree to which the two hospitals integrated
the medication reconciliation tool into the computerized order entry applications at discharge.
Emergency Department Visits and Readmission Within 30 Days

Across nine medication reconciliation interventions, the median proportion of patients with
emergency department visits or hospitalizations within 30 days of discharge was 28% (IQR,
279
20%-32%). The median rate of hospitalization or emergency department visits across seven
studies with control data was 30% (IQR, 22%-31%), a difference that was not statistically
significant.
Across three randomized controlled trials, readmissions and emergency department visits
were significantly reduced by 23% (95% CI, 5%-37%; I2 24%) (Figure 3). However, this pooled
result was driven by the statistically significant reduction achieved by an intensive intervention33
that included several interventions beyond medication reconciliation that were specifically aimed
at reducing readmissions.
Figure 3, Chapter 25. Emergency department visits and hospitalizations within 30 days of
discharge in three randomized controlled trials
Additional
interventions
No
Yes
No
Favors Intervention
Favors Control
With respect to the appropriate time period for observing an impact on post-discharge
utilization, it is particularly noteworthy that the two randomized controlled trials25,30 that
included no additional interventions beyond medication reconciliation did not reduce hospital
utilization within 30 days. However, one additional randomized controlled trial34 met all of our
inclusion criteria but was excluded because it measured hospital utilization at 12 months, rather
than 30 days, following discharge. This trial reported a statistically significant 16% reduction in
all visits to the hospital. The intervention consisted of intensive medication reconciliation in
which pharmacists identified drug related problems beyond unintended discrepancies, delivered
counseling to patients at admission and discharge, and telephoned patients 2 months after
discharge to ensure adequate home management of medications.
The lack of impact of medication reconciliation by itself on hospitalization utilization within
30 days of discharge may reflect the need to consider a longer window of observation to
demonstrate benefits from resolving unintended medication discrepancies. For instance,
inadvertent discontinuation of cholesterol lowering medications, antiplatelet or anticoagulant
agents, thyroid hormone replacement, anti-resorptive therapy for osteoporosis, and gastric acid
suppression agents all may produce adverse clinical effects requiring hospital utilization in the
long term, but not necessarily within 30 days of discharge.
Limitations of the Evidence

As emphasized at the outset, all but three of the 20 interventions that include any assessment
of the impacts of medication reconciliation involved clinical pharmacists as a key part of the
intervention. Thus, the literature provides evidence only for medication reconciliation supported
280
by pharmacists, which is not the intervention implemented in routine practice and required by
accreditation bodies in the U.S. and Canada.
In two RCTs that evaluated medication reconciliation using ADEs as the outcome, one30
reported a reduction in preventable ADEs, but the other21 (a comparably rigorous RCT from the
same research group) found only a reduction in potential ADEs at one of the two sites. The
remaining included studies evaluated the outcomes that have been judgments about the potential
clinical importance of detected medication discrepancies. These judgments are far from
straightforward. First, there is the usual problem with inter-rater reliability seen in studies of
adverse events and ADEs. Second, assessing the impact of unintended medication discrepancies
involves speculation about a number of factors, including not just the potential risk to a given
patient associated with the discrepancy, but also the likelihood that the discrepancy will persist
and for how long before it is eventually detected by the patient, an outpatient physician, or some
later health care encounter.
In the studies that reported particularly serious (e.g., potentially life-threatening)
discrepancies, few events were judged to be serious. Moreover, in the widely quoted study of
post-discharge adverse events,35 even though one of the examples of post-discharge adverse
events involved a medication discrepancy, the subsequent analysis of ADEs highlighted
problems with drug monitoring as the most common cause, not problems with medication
reconciliation.36

Mistakes in the medication reconciliation process have the potential to become hardwired
into the patient record. Once medication reconciliation has occurred, personnel caring for a given
patient may rely exclusively on the documented medication history and be less likely to confirm
the accuracy with the patient or other sources.
The larger issue with medication reconciliation concerns the opportunity costs. Clinical
pharmacists have proven roles in the prevention of adverse drug events,37-39 but they are in short
supply in most hospitals. Thus, involving pharmacists in medication reconciliation, the method
for which all the evidence of efficacy exists, risks taking these personnel away from other
important activities related to patient safety.
How Has the Patient Safety Practive Been Implemented, and in

What Context?
The number and intensity of medication reconciliation activities in the literature varies
substantially. Table 3 outlines a continuum of varying levels of medication reconciliation
intensity ranging from Bronze (simply a best possible medical history and admission
reconciliation) to Silver, Gold, Platinum and Diamond. The more advanced levels of
medication reconciliation involve progressions in interprofessional collaboration and patient
participation, integration of reconciled information into discharge summaries and prescriptions,
as well as the delivery of more comprehensive medication education and counseling to patients.
281
Table 3, Chapter 25. Medication reconciliation in varying levels of intensity as seen in published
studies
Level
Bronze
Key Components
BPMH with admission reconciliation
Silver
Bronze + reconciliation at discharge by

prescribing physician
Gold
Silver + discharge reconciliation is interprofessional

(e.g., prescribing physician and pharmacist)
+ Electronically generated discharge prescription
Gold + attention to broader medication issues, such as
appropriateness of medication choices (e.g., safe
prescribing in the elderly)
Platinum
Diamond
Platinum + additional elements, such as
pharmacist-led medication counseling prior to

discharge (including discussion of medication
changes)
communicating medication changes directly to

community pharmacy
post-discharge follow-up phone call to patient by

hospital clinician (e.g., nurse or pharmacist)
Published Examples
Cornish et al. 20052;
Kwan et al. 20073
Schnipper et al.
200925; Wong et al.
20084
Cesta et al. 200640;
Dedhia et al. 200926;
Schnipper et al. 200925
Dedhia et al. 200926;
Murphy et al. 200941;
Nazareth et al. 200142;
Al-Rashed et al. 200243
Gillespie et al. 200934;
Jack et al 200927;
Karapinar-arkit et al.
200944; Schnipper et
al. 200630; Walker et al.
2009)32
Are There Any Data About the Effect of Context On Effectiveness?

We did not identify any studies that explicitly assessed the differential effect of various
contexts on the effectiveness of medication reconciliation. However, we note that our review is
limited to interventions within hospitals, so that effectiveness in the outpatient setting is not
assessed, and further that the intervention needed to include a clinical pharmacist. Hence, to the
extent that context matters, these interventions have only been assessed in academicallyaffiliated hospitals using clinical pharmacists, and effectiveness in other contexts is not
established.

Some studies provided informal data on costs, loosely estimating the amount of time spent by
pharmacists performing medication reconciliation and equating that to a dollar value.19 One
model-based study45 considered the cost effectiveness of five pharmacist-led strategies for
reducing adverse drug events. In this analysis, pharmacist-led medication reconciliation carried a
reasonable probability of cost effectiveness (compared with no reconciliation) at 10,000
($16,272) per quality adjusted life year. The main limitation of this analysis is the uncertainty
surrounding assumptions about reductions in actual ADEs from reducing potential ADEs.

Medication reconciliation addresses the conceptually plausible and well-documented
problem of unintended medication discrepancies introduced at the time of transitions in care. The
frequency of non-trivial discrepancies varies across studiesthose studies that characterized
extremely severe discrepancies reported few such events. One well-designed randomized
controlled trial reported a significant reduction in potential ADEs at one of the two study
hospitals. The only study that reported a reduction on preventable ADEs (within 30 days of
discharge) found no difference in total ADEs. By itself, medication reconciliation probably does
282
not reduce hospital utilization within 30-days of discharge, but may do so when bundled with
other interventions aimed at improving transitions in care. It may by itself reduce hospital
utilization over timelines longer than 30 days.
Given limited resources, the paramount issue becomes how to target medication
reconciliation in order to direct resources most efficiently. This is especially important given that
all but three of the included interventions involved the use of pharmacists to conduct medication
reconciliation. Disappointingly, the studies that selected high-risk patients did not consistently
report higher rates of clinically significant unintentional discrepancies or show larger effects on
readmissions.
This null result could reflect the limited number of studies. But, the high risk criteria used
also have plausible limitations. For instance, elderly patients and patients with multiple chronic
conditions may take large numbers of medications. However, their medication regimens may
remain stable for some time and/or are well known to the patient or their caregivers. These risk
factors for unintended medication discrepancies do not account for such nuances. A more direct
risk factor is probably frequent or recent changes to medication regimens. Unfortunately, this
risk factor cannot be ascertained reliably without conducting a thorough medication history, not
unlike the BPMH required for medication reconciliation. A summary table is located below
(Table 4).
Scope of the
Strength of
the PSP
Effectiveness
Common/Low
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Moderate
Implementation Issues
How Much do We
Moderate/Moderate
References
1.
Tam VC, Knowles SR, Cornish PL, et al.

Frequency, type and clinical importance of
medication history errors at admission to
hospital: a systematic review. CMAJ. 2005
Aug 30;173(5):510-5. PMID 16129874.
5.
Lee JY, Leblanc K, Fernandes OA, et al.

Medication reconciliation during internal
hospital transfer and impact of computerized
prescriber order entry. Ann Pharmacother.
2010 Dec;44(12):1887-95. PMID 21098753.
2.
Cornish PL, Knowles SR, Marchesano R, et

al. Unintended medication discrepancies at
the time of hospital admission. Arch Intern
Med. 2005 Feb 28;165(4):424-9. PMID
15738372.
6.
3.
Kwan Y, Fernandes OA, Nagge JJ, et al.

Pharmacist medication assessments in a
surgical preadmission clinic. Arch Intern
Med. 2007 May 28;167(10):1034-40. PMID
17533206.
Bell CM, Brener SS, Gunraj N, et al.

Association of ICU or hospital admission
with unintentional discontinuation of
medications for chronic diseases. JAMA.
2011 Aug 24;306(8):840-7. PMID
21862745.
7.
Satterthwaite PS, Atkinson CJ. Using

reverse triage to create hospital surge
capacity: Royal Darwin Hospitals response
to the Ashmore Reef disaster. Emerg Med J.
2010 Oct 28PMID 21030549.
8.
Hospital: 2012 National Patient Safety

Goals. The Joint Comision.
www.jointcommission.org/hap_2012_npsgs/
Accessed Jan 8, 2012.
4.
Wong JD, Bajcar JM, Wong GG, et al.

Medication reconciliation at hospital
discharge: evaluating discrepancies. Ann
Pharmacother. 2008 Oct;42(10):1373-9.
PMID 18780806.
283
9.
Institute for Safe Medication Practices

Canada. Medication Reconciliation
(2011).Retrieved Jan 8, 2012. www.ismpcanada.org/medrec/ .
10.
Bayoumi I, Howard M, Holbrook AM, et al.

Interventions to improve medication
reconciliation in primary care. Ann
PMID 19737997.
11.
12.
13.
14.
15.
16.
17.
Weingart SN, Cleary A, Seger A, et al.

Medication reconciliation in ambulatory
oncology. Joint Commission journal on
quality and patient safety / Joint
Commission Resources. 2007
Dec;33(12):750-7. PMID 18200900.
Mueller SK, Sponsler KC, Kripalani S, et al.
Hospital-based medication reconciliation
practices: a systematic review. Arch Intern
Med. 2012 Jul 23;172(14):1057-69. PMID
22733210.
Midlov P, Bahrani L, Seyfali M, et al. The
effect of medication reconciliation in elderly
patients at hospital discharge. Int J Clin
Pharm. 2012 Feb;34(1):113-9. PMID
22207271.
Abdel-Qader DH, Harper L, Cantrill JA, et
al. Pharmacists interventions in prescribing
errors at hospital discharge: an observational
study in the context of an electronic
prescribing system in a UK teaching
hospital. Drug Saf. 2010 Nov
1;33(11):1027-44. PMID 20925440.
Steurbaut S, Leemans L, Leysen T, et al.
Medication history reconciliation by clinical
pharmacists in elderly inpatients admitted
from home or a nursing home. Ann
PMID 20736427.
Norris CM, Thomas V, Calvert PS. An audit
to evaluate the acceptability of a pharmacist
electronically prescribing discharge
medication and providing information to
GPs. Pharmaceutical Journal.
2001;276(Dec):857-9.
Climente-Marti M, Garcia-Manon ER,
Artero-Mora A, et al. Potential risk of
medication discrepancies and reconciliation
errors at admission and discharge from an
inpatient medical service. Ann
Pharmacother. 2010 Nov;44(11):1747-54.
PMID 20923946.
284
18.
Coffey M, Mack L, Streitenberger K, et al.

Prevalence and clinical significance of
medication discrepancies at pediatric
hospital admission. Acad Pediatr. 2009 SepOct;9(5):360-5 e1. PMID 19640822.
19.
Gleason KM, Groszek JM, Sullivan C, et al.

Reconciliation of discrepancies in
medication histories and admission orders of
newly hospitalized patients. Am J Health
Syst Pharm. 2004 Aug 15;61(16):1689-95.
PMID 15540481.
20.
Gleason KM, McDaniel MR, Feinglass J, et

al. Results of the Medications at Transitions
and Clinical Handoffs (MATCH) study: an
analysis of medication reconciliation errors
and risk factors at hospital admission. J Gen
Intern Med. 2010 May;25(5):441-7. PMID
20180158.
21.
Kripalani S, Roumie CL, Dalal AK, et al.

Effect of a Pharmacist Intervention on
Clinically Important Medication Errors
After Hospital Discharge: A Randomized
Trial. Ann Intern Med. 2012 Jul 3;157(1):110. PMID 22751755.
22.
Pippins JR, Gandhi TK, Hamann C, et al.

Classifying and predicting errors of inpatient
medication reconciliation. J Gen Intern Med.
2008 Sep;23(9):1414-22. PMID 18563493.
23.
Stone BL, Boehme S, Mundorff MB, et al.

Hospital admission medication
reconciliation in medically complex
children: an observational study. Arch Dis
Child. 2010 Apr;95(4):250-5. PMID
19948664.
24.
Vira T, Colquhoun M, Etchells E.

Reconcilable differences: correcting
medication errors at hospital admission and
discharge. Qual Saf Health Care. 2006
Apr;15(2):122-6. PMID 16585113.
25.
Schnipper JL, Hamann C, Ndumele CD, et

al. Effect of an electronic medication
reconciliation application and process
redesign on potential adverse drug events: a
cluster-randomized trial. Arch Intern Med.
2009 Apr 27;169(8):771-80. PMID
19398689.
26.
Dedhia P, Kravet S, Bulger J, et al. A

quality improvement intervention to
facilitate the transition of older adults from
three hospitals back to their homes. J Am
Geriatr Soc. 2009 Sep;57(9):1540-6. PMID
19694865.
27.
Jack BW, Chetty VK, Anthony D, et al. A

reengineered hospital discharge program to
decrease rehospitalization: a randomized
trial. Annals of internal medicine. 2009 Feb
3;150(3):178-87. PMID 19189907.
28.
Koehler BE, Richter KM, Youngblood L, et

al. Reduction of 30-day postdischarge
hospital readmission or emergency
department (ED) visit rates in high-risk
elderly medical patients through delivery of
a targeted care bundle. J Hosp Med. 2009
Apr;4(4):211-8. PMID 19388074.
29.
Kramer JS, Hopkins PJ, Rosendale JC, et al.

Implementation of an electronic system for
medication reconciliation. Am J Health Syst
Pharm. 2007 Feb 15;64(4):404-22. PMID
17299180.
36.
Forster AJ, Murff HJ, Peterson JF, et al.

Adverse drug events occurring following
hospital discharge. J Gen Intern Med. 2005
Apr;20(4):317-23. PMID 15857487.
37.
Kaboli PJ, Hoth AB, McClimon BJ, et al.

Clinical pharmacists and inpatient medical
care: a systematic review. Arch Intern Med.
2006 May 8;166(9):955-64. PMID
16682568.
38.
Kucukarslan SN, Peters M, Mlynarek M, et

al. Pharmacists on rounding teams reduce
preventable adverse drug events in hospital
general medicine units. Arch Intern Med.
2003 Sep 22;163(17):2014-8. PMID
14504113.
39.
Lee AJ, Boro MS, Knapp KK, et al. Clinical

and economic outcomes of pharmacist
recommendations in a Veterans Affairs
medical center. Am J Health Syst Pharm.
2002 Nov 1;59(21):2070-7. PMID
12434719.
40.
Cesta A, Bajcar JM, Ong SW, et al. The

EMITT study: development and evaluation
of a medication information transfer tool.
Ann Pharmacother. 2006 Jun;40(6):1074-81.
PMID 16735652.
41.
Murphy EM, Oxencis CJ, Klauck JA, et al.

Medication reconciliation at an academic
medical center: implementation of a
comprehensive program from admission to
discharge. Am J Health Syst Pharm. 2009
Dec 1;66(23):2126-31. PMID 19923314.
30.
Schnipper JL, Kirwin JL, Cotugno MC, et

al. Role of pharmacist counseling in
preventing adverse drug events after
hospitalization. Arch Intern Med. 2006 Mar
13;166(5):565-71. PMID 16534045.
31.
Showalter JW, Rafferty CM, Swallow NA,

et al. Effect of standardized electronic
discharge instructions on post-discharge
hospital utilization. J Gen Intern Med. 2011
Jul;26(7):718-23. PMID 21499825.
32.
Walker PC, Bernstein SJ, Jones JN, et al.

Impact of a pharmacist-facilitated hospital
discharge program: a quasi-experimental
study. Arch Intern Med. 2009 Nov
23;169(21):2003-10. PMID 19933963.
33.

trial. Ann Intern Med. 2009 Feb
3;150(3):178-87. PMID 19189907.
42.
Nazareth I, Burton A, Shulman S, et al. A

pharmacy discharge plan for hospitalized
elderly patients--a randomized controlled
trial. Age and ageing. 2001 Jan;30(1):33-40.
PMID 11322670.
34.
Gillespie U, Alassaad A, Henrohn D, et al.

A comprehensive pharmacist intervention to
reduce morbidity in patients 80 years or
older: a randomized controlled trial. Arch
Intern Med. 2009 May 11;169(9):894-900.
PMID 19433702.
43.
Al-Rashed SA, Wright DJ, Roebuck N, et al.

The value of inpatient pharmaceutical
counselling to elderly patients prior to
discharge. Br J Clin Pharmacol. 2002
Dec;54(6):657-64. PMID 12492615.
44.
35.
Forster AJ, Murff HJ, Peterson JF, et al. The

incidence and severity of adverse events
affecting patients after discharge from the
hospital. Annals of internal medicine. 2003
Feb 4;138(3):161-7. PMID 12558354.
Karapinar-Carkit F, Borgsteede SD, Zoer J,

et al. Effect of medication reconciliation
with and without patient counseling on the
number of pharmaceutical interventions
among patients discharged from the hospital.
Ann Pharmacother. 2009 Jun;43(6):1001-10.
PMID 19491320.
285
45.
Karnon J, Campbell F, Czoski-Murray C.

Model-based cost-effectiveness analysis of
interventions aimed at preventing
medication error at hospital admission
(medicines reconciliation). J Eval Clin Pract.

2009 Apr;15(2):299-306. PMID 19335488.
286
Chapter 26. Identifying Patients at Risk for Suicide: Brief

Review (NEW)
Steven C. Bagley, M.D.
Introduction
Patients are often hospitalized after suicide attempts or because of suicidal ideation.
However, hospitalization is not fully protective and the inpatient population remains at risk.
Many risk factors are associated with inpatient suicide, but as detailed below reported rates
vary widely, and the importance of this topic derives from the fatality of the outcome in close
proximity to care, not primarily from its frequency. Suicide has been frequently associated with
certain mental health diagnoses, especially depression and schizophrenia, but the risk of suicide
is not limited to patients psychiatrically hospitalized: medical and surgical patients have
profound risk factors, including severe pain, altered mental status, and progressive or terminal
diagnoses. For all patients, these risks persist, even if patients are placed on special observation
status with nursing personnel directly monitoring them.1
Assessing and reducing the suicide risk for inpatients has become a component of national
patient safety efforts. In 1998, The Joint Commission released a Sentinel Event Alert about
inpatient suicides based on a review of 65 cases, making brief recommendations about suicide
risk assessment, policy and procedures, staff training, and modification of the hospital to reduce
environmental risks.2 Although the 1998 Alert was not specific to behavioral health units, in
2010 the Joint Commission added a Sentinel Event Alert for inpatient suicide on
medical/surgical units and in emergency departments.3 The current Joint Commission (2011)
National Hospital Safety Goals include the goal of identifying patients at risk for suicide
(NPSG.15.01.0), with three elements of performance (perform risk assessment, identify
appropriate treatment environment and safety needs, and provide patient and their family with
suicide prevention information at discharge).4 National Quality Forums Serious Reportable
Events (2011) lists suicide, suicide attempts, and self-harm that results in serious injury.5
Medicare has placed inpatient suicide on the never events list. The Centers for Medicare and
Medicaid Services (CMS) announced in August 2007 that Medicare would no longer pay for
additional costs associated with many preventable errors, including those considered Never
Events. Since then, many states and private insurers have adopted similar policies.6
The Agency for Healthcare Research and Qualitys evidence report, Making Health Care
Safer: A Critical Analysis of Patient Safety Practices (2001)7, focuses on general safety
practices that would extend to psychiatry and other areas of medical practice, and on the relative
lack of evidence for behavioral health interventions within the patient safety remit.
Consequently, the authors did not specifically address inpatient suicide. The purpose of this
narrative literature review is to identify new developments and trends starting from the date of
the AHRQ report up to the present.
This review addresses three important questions related to the safety of medical, surgical, and
psychiatric inpatients at risk for suicide.
What is the evidence that clinical, organizational, or environmental programs work to
reduce attempts or completions for hospitalized patients?
What is the state of programs in use at this time?
287
What has been learned from their implementation?
To conduct the review, we searched PubMed in October 2011 using major heading search
terms Suicide, and Hospital or Inpatient or Safety Management, for English language articles
published starting in the year 2000. We expanded the search using the PubMed related
citations feature, and Google Scholar to search for citing articles of those retained for review;
we identified additional relevant articles by reference mining. Clinical trials, large observational
studies, reviews, and reports on implementations were given priority. Systematic reviews were
scored for methodologic quality using the 11-point AMSTAR scale;8 items rated Not Applicable
were not counted towards either the score or the total.
What Are the Practices for Reducing Inpatient Suicide?

Systematic reviews by Links9 (AMSTAR score 2/10) and Tishler10 (AMSTAR score 1/10),
and informal reviews and expert opinions11-14 have reached generally similar conclusions about
programs to reduce suicide risk for inpatients, including: (1) Suicide risk assessment at
admission, repeated especially during times of risk elevation such as personal crises, along with
careful and consistent chart documentation of these assessments. (2) Treating psychiatric
disorders that placed patients at risk, and addressing continuity and followup issues to maintain
the patient in treatment after discharge. (3) Removing risk factors in the physical environment.
(4) Staff training in risk assessment and communication. (5) Use of staff to observe high-risk
patients, and (6) Defining hospital policies in these areas, including those for collecting statistics
about suicide attempts and completions.

Identifying Patients at Risk
Bowers et al15 (AMSTAR score 5/11) conducted a systematic review of 98 articles published
in English, German, or Dutch since 1960 covering almost 15,000 inpatient suicides. Given the
breadth of articles surveyed, they found a great diversity in suicide rates, trends, risk factors, and
timing that reflected the national, cultural, social, and temporal variation. A personal history of
suicidal behavior was very consistently associated with suicide completions. Schizophrenia and
mood disorders (especially depression) were the leading psychiatric diagnoses. Mechanisms
varied with availability; hanging was consistently reported. The mechanisms and rates were
associated with location, because patients off-ward on a pass, or having eloped, are typically
considered to still have inpatient status, regardless of the actual site of their suicide. Similar
results were reported in articles by Kapur,16 Meehan,17 Hunt,18 Combs,19 (AMSTAR score 4/10).
Hunt20 reported an UK survey on suicides after absconding from the ward. Stewart21 reported on
a retrospective analysis of medical records from hospitals in London and surrounding areas,
finding that 10% of psychiatric inpatients made self-harm attempts, and 4% made suicide
attempts. Pompili22 (AMSTAR score 3/10) reported a literature review on suicide in patients
diagnosed with schizophrenia. Most of the reported deaths occurred while the patient was on
leave, or having eloped from the hospital. Specific risk factors for suicides on hospital wards
were not reported.
Ballard23 reviewed 12 case series comprising 335 general hospital suicides (including
patients off-ward on a pass), and found slightly different risk factors from those from inpatient
psychiatry. The most common medical diagnoses were cancer, cardiovascular and pulmonary
288
disease. The mental status of patients was infrequently and inconsistently reported. Jumping
from a building was the leading mechanism, unlike the pattern seen in psychiatric inpatients and
in the general population. Bostwick24 in an informal review of the same area based on a case
series of 50 psychiatric consultations from general medical/surgical wards concluded that
medical and surgical patients have different risk factors, and a different profile from psychiatric
patients, typically by lacking a strong personal history of suicide attempts, psychiatric diagnoses,
and substance abuse.
Risk factors, and the difficulties of risk prediction. Suicide is relatively rare, making it
difficult to predict even in populations with multiple risk factors and high relative risk. This
conclusion, long established for outpatients and the general population, holds true for inpatients
as well. Large25 (AMSTAR score 9/11) in a systematic review and meta-analysis of 29 studies
concluded that some specific risk factors are associated with inpatient suicide, but using the
presence of multiple risk factors to identify high-risk patients produces many false positives, and
misses some who will go on to commit suicide in the hospital. They concluded that reducing
environmental risks and improving systems of clinical care are likely to have greater effects on
suicide reduction than reliance on suicide prediction methods. The difficulties of accurate
prediction for inpatients are consistent with conclusions reached by others, including Busch,26
Cassells,27 Paterson,28 Bisconer,29 and the American Psychiatric Association Practice Guideline
for the Assessment and Treatment of Patients With Suicidal Behaviors.30
Environmental risk reduction factors. The removal of physical or structural risk factors from
the hospital environment has been frequently proposed. Lieberman31 and Cardell32 both report
expert opinions of this topic, and make specific suggestions for environmental modifications.
The modifications follow from the frequency with which hanging is used in inpatient suicide by
removing both materials that could form a noose and anchor points for the noose. Most of these
recommendations target inpatient psychiatric wards. Bostwick24 notes the difficulties of applying
these same recommendations to typically open general medical wards, which are more difficult
to secure; they recommended use of nursing observation for those areas.
Experiences of Specific Hospital Programs

A number of reports described implemented program or program components, mostly guided
by expert opinion or slight modifications of current practice. Few outcomes data were reported,
and the quality of the studies was poor in those that did.
Sullivan33 described a multi-component suicide reduction program implemented at Elmhurst
Hospital Center in Queens, NY, a teaching hospital affiliated with Mount Sinai School of
Medicine, with 117 inpatient psychiatric beds, including specialty units for Asians and Latinos.
The hospitals psychiatry service implemented a suicide reduction program that included a
formal assessment of suicide risk, encouraged accurate diagnosis (taking into consideration the
multicultural nature of the patients treated), replaced some use of one-to-one nursing observation
with close observation (visual observation at any distance, sometimes with a ratio of one nurse
for two patients), encouraged careful use of medications, used group sessions for inpatients (on
coping in the community, identifying triggers for suicidal thoughts, and listing information about
resources available in a crisis), added environmental rounds to remove safety hazards, along with
discharge planning and post-discharge followup. They reported a reduction in self-injurious
behaviors from 1.4 per 1000 before the intervention to 0.5 per 1000 afterwards. The reported
289
decrease was described as associated with the component involving the formal assessment of
suicide risk; unfortunately, the timing of the other components was not clearly described making
it difficult to assess their role in any reduction in suicides or attempts, and in the assignment of
causality to their intervention.
Other program experiences are described here more briefly. Temkin34 proposed a precaution
monitoring sheet to improve the consistency of documentation and communication within
treatment team, but did not report of evaluation of it. McAuliffe35 described the implementation
of a program at Trillium Health Centre, Ontario, Canada, reporting on their experiences with risk
assessment, staff surveys and focus groups, and training workshops; no outcomes data of
inpatient suicides were reported. Ellis36 reported on a program, called the Collaborative
Assessment and Management of Suicidality, underway at the Menninger Clinic in Houston. The
program began with the elaboration of suicide risk assessment into a comprehensive
collaborative framework for patient treatment and risk reduction. The framework does not appear
to be limited to inpatients. They noted the need for rigorous evaluation and planned to conduct a
randomized controlled trial of their program. Ballard37 proposed a framework for organizing the
response of a hospital to an inpatient suicide. No evaluation of this framework was reported.
Root Cause Analyses and Related Techniques

Root cause analysis (RCA) is a structured analysis technique originally developed for human
factors and systems engineering to retrospectively determine the interrelationship of component
elements in causing an observed malfunction or accident. It has been adapted for use in medical
and health care systems.
Dlugacz38 reported on the use of the results of RCAs of 17 suicides or suicide attempts at
North ShoreLong Island Jewish Health System, Great Neck, NY to design safety strategies.
They developed an inpatient suicide risk assessment and evaluation tool (apparently for use by
RNs), and an environmental suicide risk assessment tool used by a multidisciplinary hazard
surveillance team to identify environmental risks for all facilities with some specific additional
items for behavioral health units. They also developed an alcohol withdrawal protocol, as alcohol
problems had been relatively common in their RCA data. They reported no suicide attempts in
the acute care setting after implementing the alcohol withdrawal assessment protocol. Overall,
there had been 6 completed suicides and 11 attempts in the interval from April 1998 to
December 2001 represented in the RCAs; after making the implementations, there were no
suicides and one attempt from December 2001 to December 2002. No data were reported that
would allow assessment of the causal role of the other program components.
Mills39-41 reports on the Department of Veterans Affairs (VA) experience in using RCAs to
guide the development of policies and procedures. Their first study39 used information from
RCAs from completed suicides and parasuicidal behavior to identify the most common root
causes: communication issues (including documentation of risk), policies about suicide risk
assessment and treatment, patient stressors, and training or education for both staff and patients.
In the second study,40 they used VA RCA reports (presumably a superset of those in their
previous article) to identify the common locations (inpatient psychiatry) and means of suicide
(hanging). They also reported specific details on the anchor points and the material used as a
noose, by frequency. Outside of inpatient psychiatric units, drug overdoses were also common.
They made recommendations for reducing access to means through engineering interventions to
remove common anchor points, and for making regular environmental rounds using a
comprehensive checklist. Their environmental rounds checklist was described in detail in their
290
next report.41 No outcomes measures were reported. They also noted there was no evidence that
the checklist was being used correctly. The target location was inpatient psychiatric units; they
recommended using one-to-one observation for general medical units.
Janofsky42 reported on the use of Failure Mode Effects Analysis (FMEA), a structured,
systematic, prospective methodology from systems engineering, to identify possible system
failures, and used this analysis to redesign the communication flow related to observation of
psychiatric patients. No outcome results were reported.
Wu43 examined the use of RCAs in medicine generally, and noted a very wide range of skill
in performing RCAs accurately, a lack of best practices in reporting and followup, and the
absence of peer-reviewed evidence of the effectiveness of RCAs or their cost-benefits tradeoffs.
Observation of At-Risk Patients by Nursing Staff

One important area not frequently mentioned in some reviews is the use of nursing
observation. Nursing observation is regularly invoked for patients at risk of suicide (as well as
those with risks for violence, elopement, or falls). The practice varies considerably on multiple
dimensions. The intensity of the observation can range from intermittent through continuous, and
at specified distances from the at-risk patient. Observation also varies in who can initiate it,
whether by psychiatrists, psychologists, or nursing staff. There are also differences in the degree
of professional training needed to work as an observer, ranging from experienced psychiatric
nurses, thorough lower levels of nursing training, other staff, volunteers, or security personnel.
The terminology for the practice itself varies, being referred to constant observation, continuous
observation, enhanced observation, special observation, constant special observation, and suicide
precautions; all of these will be referred to here as observation status. Not considered are the
effects of nursing observation on staff morale, patients perceptions of caring, or the relationship
between staff and patients, although it would be expected that these could have second-order
effects on patient engagement in treatment and patient safety.
A 2006 Cochrane Systematic Review of non-pharmacological methods for the containment
of unsafe behavior found no evidence supported by any randomized controlled trials44
(AMSTAR score 7/7). A similar conclusion were reached by Manna45 (AMSTAR score 5/10).
Dodds46 reported an observational study with a before/after design at an inpatient psychiatric
ward in the UK, in which control-oriented formal observation of at-risk patients was replaced by
a care-oriented interventions on both an individual and group basis. They reported a two-thirds
decline in self-harm episodes in the following year, compared with the year before the
intervention. There was one inpatient suicide in the year before, and two the year after, both of
the latter while the patients were off the ward on leave. There were staffing changes and changes
in the size and demographics of the inpatients during the implementation of the program.
Bowers47 reported a survey of 128 psychiatric wards in the UK, finding no relationship
between the use of constant special observation and self-harm incidents, but an inverse
relationship for intermittent observation: greater use of intermittent observation was associated
with lower self-harm rates. This was an observational study, and causality cannot be inferred.
Stewart48 reported a longitudinal analysis of 16 wards at three London hospitals. Regression
modeling showed no statistical relation between the use of constant special observation (CSO),
when the staff person was either within reach of or in sight of the patient, and self-harm
incidents. No suicides were recorded. This was also an observational study and subject to the
same weaknesses in inference of causality. They noted a wide variation in the profiles of CSO
291
usage across time, wards, and hospitals, perhaps driven by idiosyncratic differences in staff
preferences for or against the use of CSO.
Bowers et al. in their literature review15 noted that suicide rates showed a mixed association
with the presence of nursing observation (at different levels of intensity) in force at the time of
the suicide. The cautions about inference from observational studies apply here.
Because the observer cannot be simultaneously engaged in other activities, use of nursing
observation can be expensive. More details about cost and the implementation of observation
programs at one Massachusetts hospital are reported by Harding.49
Most other articles note the lack of evidence that constant observation is efficacious. Issues
such as the quality or therapeutic effect of the observer-patient relationship have not been
addressed here, but common sense suggests they might vary widely, and have therapeutic or
counter-therapeutic effects, depending on the kind of interpersonal relationship between the
observer and the patient. Cutcliffe1 noted that suicides have occurred while the patient was on
observation status.
Alternatives to constant observation were explored by Cox,50 who proposed an alternative
nurse-team framework, with greater nurse autonomy and greater engagement with the patient,
along with the use of intermittent observation. These proposals have not been formally
empirically tested.
Jayaram51 reported an informal survey of the use of 15-minute checks (observation of the
patient at least once every 15 minutes), which showed considerable variation in the use of this
practice. No outcomes data were reported.
What Have We Learned About Practices for Reducing Inpatient

Suicide?
Patients at-risk for suicide are frequently hospitalized, but suicides can be completed by
inpatients on psychiatric, general medical, and surgical wards. Risk factors vary across these
groups, as do the available mechanisms, typically by hanging in behavioral health units, by
jumping or overdose in medical/surgical units. Risk factors are likely to be higher and involve
other means in patients for predicting risk suffer from unacceptably high error rates, falsely
predicting suicide in those who do not go on to commit it, and not predicting suicide in some
who do.
Most existing suicide reduction programs have not been formally or carefully evaluated.
Means reduction through careful periodic inspection and reengineering of the hospital wards
physical structure has been implemented, often based on results of root cause analyses of
suicides and suicide attempts. These programs have clear face validity, and are unlikely to
elevate risk. However, no controlled trials or high quality observational studies have been
performed so the magnitude of any risk-moderating effects is not known, limiting the ability to
make strong policy recommendations, or to develop cost-benefit analyses that could guide the
deployment of staff and capital resources.
Using staff to observe at-risk patients is a frequently used suicide prevention practice, but
there is no evidence from controlled trials showing the magnitude or even the direction of its
effect. Several observational studies have shown that the intensity of nursing observation is not
associated with reduction in self-harm episodes, but these did not control for the confounding
effect of the severity of the patients suicidality, which would be expected to both increase their
risk of suicide and increase the frequency with which nursing observation would be invoked for
their protection. Without controlled experiments, true causality cannot be inferred, and it remains
292
uncertain if nursing observation raises, lowers, or has no effect on the rates of suicide and selfharm for any given level of suicide risk. The psychological effects of nursing observation on
both staff and patients are not the focus of this review; however, these might be expected to have
second-order effects, including forging risk-lowering relationships between the at-risk patient
and a staff person or, conversely, raising risk by interfering with patient privacy and autonomy,
and increasing patient confinement and alienation.
What Methods Have Been Used To Improve Practices for Reducing

Inpatient Suicide?
Because there is little empirical evidence to support the suicide prevention practices in
current use, recommendations for improving practice have focused on the need for high quality
research44 including some specifics for making the results useful to both clinicians and
policymakers. Although data on completed suicides might seem to be the most valid outcome
measure, their use has been questioned because of problems in tracking and sampling, and the
statistical noise in the low rates, leading to instability in the measurements.52 Future work will
likely refer to structure or process measures of quality,53 in addition to, or in lieu of, hard
outcomes data. It is expected that continued efforts will necessitate periodic reassessment of this
topic area for consideration of review.

Current practice for the reduction of inpatient suicides is supported by tradition, expert
opinion, very limited observational studies of low quality, and the face validity of some of the
interventions.
The use of staff to observe at-risk patients is frequently employed, but there is no evidence
from controlled trials showing the magnitude or even the direction of its effect.
Recommendations for high quality research in this area, including some specifics for making
the results useful to both clinicians and policymakers, have been proposed.44 Although data on
completed suicides might seem to be the most valid outcome measure, their use has been
questioned because of problems in tracking and sampling, and the statistical noise in the low
rates, leading to instability in the measurements.52 Future work will likely refer to structure or
process measures of quality,53 in addition to or in lieu of hard outcomes data. It is expected that
continued efforts will necessitate a periodic reassessment of this topic area for consideration of
review. A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Rare/High
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Moderate
293
How Much do We
Little/Moderate
References
1.
2.
3.
Cutcliffe JR, Stevenson C. Never the twain?

Reconciling national suicide prevention
strategies with the practice, educational, and
policy needs of mental health nurses (part
two). Int J Ment Health Nurs.
2008;17(5):351-62.18789045.
The Joint Commission. Sentinel Event Alert
1998;(7). Sentinel Event Alert, Issue 7:
Inpatient Suicides: Recommendations for
Prevention, pp1-2;
ert_issue_7_inpatient_suicides_recommenda
tions_for_prevention/.
The Joint Commission. Sentinel Event Alert
2010;(46). Sentinel Event Alert, Issue 46:
A follow-up report on preventing suicide:
Focus on medical/surgical units and the
emergency department, pp1-5,
ert_issue_46_a_followup_report_on_preventing_suicide_focus_on
_medicalsurgical_units_and_the_emergency
_department/.
4.
The Joint Commission. National Patient

Safety Goals Effective July 1, 2011. 2011
[cited;
www.jointcommission.org/assets/1/6/NPSG
_EPs_Scoring_BHC_20110707.pdf.
5.
National Quality Forum. NQF Releases

Updated Serious Reportable Events. 2011
[cited;
www.qualityforum.org/News_And_Resourc
es/Press_Releases/2011/NQF_Releases_Upd
ated_Serious_Reportable_Events.aspx.
6.
7.
8.

Quality. AHRQ Patient Safety Network
(PSNet). [cited;
http://psnet.ahrq.gov/primer.aspx?primerID
=3.
analysis of patient safety practices. Evid Rep
Technol Assess (Summ). 2001;(43):i-x, 1668.PMID: 11510252.
Methodol. 2007;7:10. PMID: 17302989.
294
9.
Links PS, Hoffman B. Preventing suicidal

behaviour in a general hospital psychiatric
service: priorities for programming. Can J
Psychiatry. 2005;50(8):490-6. PMID:
16127967.
10.
Tishler CL, Reiss NS. Inpatient suicide:

preventing a common sentinel event. Gen
Hosp Psychiatry. 2009;31(2):103-9. PMID:
19269529.
11.
Baldwin D, Mayers A, Elgie R. Suicide in

psychiatric in-patients. CLINICAL RISK.
2003;9:229-32.
12.
Yeager KR, Saveanu R, Roberts RR, et al.

Measured Response to Identified Suicide
Risk and Violence: What You Need to
Know About Psychiatric Patient Safety
Brief Treatment and Crisis Intervention.
2005:121-41.
13.
Grant JE. Failing the 15-minute suicidewatch: Guidelines to monitor inpatients. .

Current Psychiatry. 2007;6(6):41-3.
14.
Lynch MA, Howard PB, El-Mallakh P, et al.

Assessment and management of hospitalized
suicidal patients. J Psychosoc Nurs Ment
Health Serv. 2008;46(7):45-52.PMID:
18686596.
15.
Bowers L, Banda T, Nijman H. Suicide

inside: a systematic review of inpatient
suicides. J Nerv Ment Dis. 2010;198(5):31528.PMID: 20458192
16.
Kapur N, Hunt IM, Webb R, et al. Suicide in

psychiatric in-patients in England, 1997 to
2003. Psychol Med. 2006;36(10):148592.PMID: 16863596.
17.
Meehan J, Kapur N, Hunt IM, et al. Suicide

in mental health in-patients and within 3
months of discharge. National clinical
survey. Br J Psychiatry. 2006;188:129-34.
PMID: 16449699.
18.
Hunt IM, Kapur N, Webb R, et al. Suicide in

current psychiatric in-patients: a casecontrol study The National Confidential
Inquiry into Suicide and Homicide. Psychol
Med. 2007;37(6):831-7. PM ID: 17306045.
19.
Combs H, Romm S. Psychiatric Inpatient

Suicide: A Literature Review Primary
Psychiatry. 2007;14(12):67-74.
20.
Hunt IM, Windfuhr K, Swinson N, et al.

Suicide amongst psychiatric in-patients who
abscond from the ward: a national clinical
survey. BMC Psychiatry. 2010;10:14.
PMID: 20128891.
21.
Stewart D, Ross J, Watson C, et al. Patient

characteristics and behaviours associated
with self-harm and attempted suicide in
acute psychiatric wards Journal of Clinical
Nursing. 2012 Apr;21(7-8):1004-13.
22.
23.
24.
25.
26.
27.
28.
29.
Pompili M, Mancinelli I, Ruberto A, et al.

Where schizophrenic patients commit
suicide: a review of suicide among
inpatients and former inpatients. Int J
Psychiatry Med. 2005;35(2):171-90. PMID:
16240974.
Ballard ED, Pao M, Henderson D, et al.
Suicide in the medical setting. Jt Comm J
Qual Patient Saf. 2008;34(8):474-81. PMID:
18714750.
Bostwick JM, Rackley SJ. Completed
suicide in medical/surgical patients: who is
at risk? Curr Psychiatry Rep. 2007;9(3):2426. PMID: 17521522.
Large M, Smith G, Sharma S, et al.
Systematic review and meta-analysis of the
clinical factors associated with the suicide of
psychiatric in-patients. Acta Psychiatr
Scand. 2011;124(1):18-9.
Busch KA, Fawcett J, Jacobs DG. Clinical
correlates of inpatient suicide. J Clin
Psychiatry. 2003;64(1):14-9. PMID:
12590618.
Cassells C, Paterson B, Dowding D, et al.
Long- and short-term risk factors in the
prediction of inpatient suicide: review of the
literature. Crisis. 2005;26(2):53-63. PMID:
16138741.
Paterson B, Dowding D, Harries C, et al.
Managing the risk of suicide in acute
psychiatric inpatients: A clinical judgement
analysis of staff predictions of imminent
suicide risk Journal of Mental Health.
2008;17(4):410-23.
Bisconer SW, Gross DM. Assessment of
Suicide Risk in a Psychiatric Hospital.
Professional Psychology: Research and
Practice. 2007;38(2):1439.
295
30.
Practice guideline for the assessment and

treatment of patients with suicidal behaviors.
Am J Psychiatry. 2003;160(11 Suppl):1-60.
PMID: 14649920.
31.
Lieberman DZ, Resnik HL, Holder-Perkins

V. Environmental risk factors in hospital
suicide. Suicide Life Threat Behav.
2004;34(4):448-53. PMID: 15585466.
32.
Cardell R, Bratcher KS, Quinnett P.

Revisiting suicide proofing an inpatient
unit through environmental safeguards: a
review. Perspect Psychiatr Care.
2009;45(1):36-44. PMID: 19154238.
33.
Sullivan AM, Barron CT, Bezmen J, et al.

The safe treatment of the suicidal patient in
an adult inpatient setting: a proactive
preventive approach. Psychiatr Q.
2005;76(1):67-83. PMID: 15757237.
34.
Temkin T, Crotty M. Suicide and Other Risk

Monitoring in Inpatient Psychiatry. J Am
Psychiatr Nurses Assoc. 2004;10(2):73-80.
35.
McAuliffe N, Perry L. Making it safer: a

health centres strategy for suicide
prevention. Psychiatr Q. 2007;78(4):295307. PMID: 17879162.
36.
Ellis TE, Allen JG, Woodson H, et al.

Implementing an evidence-based approach
to working with suicidal inpatients. Bull
Menninger Clin. 2009;73(4):339-54. PMID:
20025428.
37.
Ballard ED, Pao M, Horowitz L, et al.

Aftermath of suicide in the hospital:
institutional response. Psychosomatics.
2008;49(6):461-9. PMID: 19122122.
38.
Dlugacz YD, Restifo A, Scanlon KA, et al.

Safety strategies to prevent suicide in
multiple health care environments. Jt Comm
J Qual Saf. 2003;29(6):267-78. PMID:
14564745.
39.
Mills PD, Neily J, Luan D, et al. Actions

and implementation strategies to reduce
suicidal events in the Veterans Health
Administration. Jt Comm J Qual Patient Saf.
2006;32(3):130-41. PMID: 16617944.
40.
Mills PD, DeRosier JM, Ballot BA, et al.

Inpatient suicide and suicide attempts in
Veterans Affairs hospitals. Jt Comm J Qual
Patient Saf. 2008;34(8):482-8. PMID:
18714751.
41.
Mills PD, Watts BV, Miller S, et al. A

checklist to identify inpatient suicide
hazards in veterans affairs hospitals. Jt
Comm J Qual Patient Saf. 2010;36(2):87-93.
PMID: 20180441.
48.
Stewart D, Bowers L, Warburton F.

Constant special observation and self-harm
on acute psychiatric wards: a longitudinal
analysis. Gen Hosp Psychiatry.
2009;31(6):523-30. PMID: 19892210.
42.
Janofsky JS. Reducing inpatient suicide risk:

using human factors analysis to improve
observation practices. J Am Acad Psychiatry
Law. 2009;37(1):15-24. PMID: 19297628.
49.
Harding AD. Observation assistants: sitter

effectiveness and industry measures. Nurs
Econ. 2010;28(5):330-6. PMID: 21158254.
50.
43.
Wu AW, Lipshutz AK, Pronovost PJ.

Effectiveness and efficiency of root cause
analysis in medicine. JAMA.
2008;299(6):685-7. PMID: 18270357.
44.
Muralidharan S, Fenton M. Containment

strategies for people with serious mental
illness. Cochrane Database Syst Rev.
2008;3:CD002084. PMID: 16855984.
Cox A, Hayter M, Ruane J. Alternative

approaches to enhanced observations in
acute inpatient mental health care: a review
of the literature. J Psychiatr Ment Health
Nurs. 2010;17(2):162-71. 2 PMID:
0465762.
51.
Jayaram G, Sporney H, Perticone P. The

Utility and Effectiveness of 15-minute
Checks in Inpatient Settings. Psychiatry
(Edgemont). 2010;7(8):46-9.
52.
Desai RA, Dausey DJ, Rosenheck RA.

Mental health service delivery and suicide
risk: the role of individual patient and
facility factors. Am J Psychiatry.
2005;162(2):311-8. PMID: 15677596.
53.
Donabedian A. The Quality of Care: How

can it be assessed? JAMA.
1988;260(12):1743-48.
45.
46.
47.
Manna M. Effectiveness of formal

observation in inpatient psychiatry in
preventing adverse outcomes: the state of
the science. J Psychiatr Ment Health Nurs.
2010;17(3):268-73. PMID: 20465777.
Dodds P, Bowles N. Dismantling formal
observation and refocusing nursing activity
in acute inpatient psychiatry: a case study. J
Psychiatr Ment Health Nurs. 2001;8(2):1838. PMID: 11882126.
Bowers L, Whittington R, Nolan P, et al.
Relationship between service ecology,
special observation and self-harm during
acute in-patient care: City-128 study. Br J
Psychiatry. 2008;193(5):395-401. PMID:
18978321.
296
Chapter 27. Strategies To Prevent Stress-Related

Gastrointestinal Bleeding (Stress Ulcer Prophylaxis):
Brief Update Review
Stephanie Rennke, M.D.; Robert M. Wachter, M.D.; Sumant R. Ranji, M.D.
Introduction
Stress-related gastrointestinal ulceration is a known complication of critical illness.
Disruption of mucosal barriers and gastric acid hypersecretion lead to diffuse shallow mucosal
injury and discrete ulcerations in the proximal stomach and duodenum, which in turn can lead to
gastrointestinal (GI) bleeding and perforation.1-3 The prevalence of clinically significant bleeding
in patients with documented stress ulcers varies from 0.6-15%, and mortality associated with the
complication of GI bleeding can be nearly 50%.4-7
Independent risk factors for bleeding include respiratory failure requiring mechanical
ventilation for longer than 48 hours and coagulopathy.4,8 Other associated risk factors for
mechanically ventilated patients include shock of any cause, renal failure, and burns.8
Several pharmacologic therapies have been studied for the prevention of stress-induced
gastrointestinal bleeding, including proton pump inhibitors (PPIs), histamine-2 receptor (H-2)
antagonists, sulcrafate, and enteral nutrition. Despite decades of research, significant controversy
continues to surround standardization of prophylactic therapy, particularly because of evidence
that prophylaxis is associated with pneumonia, inappropriate use, and cost. Independent of
prophylactic therapy, rates of clinically significant bleeding have actually declined, likely related
to other patient safety practices around management of sepsis and enteral nutrition.9
Multicomponent or bundled interventions are becoming increasingly common as a method
of improving outcomes by preventing complications in ICU patients. Examples of these
approaches include the Surviving Sepsis Campaign,10 which includes stress ulcer prophylaxis
and deep venous thrombosis prophylaxis along with evidence-based clinical strategies to
improve sepsis outcomes, and the Institute for Healthcare Improvements Ventilator Bundle,
one of the key components of the 100,000 Lives campaign11 and the Keystone ICU project.12
Given the wide implementation of these bundles, the key issues around stress ulcer prophylaxis
involve not only standardization of therapy with the most efficacious agents but also
appropriateness of therapy based on risk assessment, and discontinuation of therapy when
appropriate.
The 2001 Making Health Care Safer report reviewed evidence on the epidemiology of stressrelated GI bleeding, and included an evaluation of two meta-analyses and one large randomized
controlled trial (RCT) on the effectiveness of pharmacologic therapies, including H2-antagonists
and sucralfate.13-15 Both H2-antagonists and sucralfate were found to be effective at preventing
clinically significant GI bleeding in ICU patients, but the overall magnitude of benefit was small.
The review found a relatively low incidence of clinically significant stress ulcer-related GI
bleeding and a higher cost-to-benefit ratio for low-risk patients. Concern was also raised
regarding a possible associated risk of hospital-acquired pneumonia with acid suppression.
Therefore, the review concluded that no evidence supported the institution of universal stress
ulcer prophylaxis in the ICU. The report recommended considering stress ulcer prophylaxis with
either an H2-antagonist or sucralfate for the prevention of GI bleeding in certain high risk ICU
297
patient populations, including patients with respiratory failure, coagulopathy, renal failure,
and/or burns, and considering enteral nutrition for other populations.
What Is Stress Ulcer Prophylaxis?

Pharmacologic acid suppressive therapy has been used to prevent stress-induced GI bleeding
in the critical care setting. Previous studies have reported decreased rates of bleeding with agents
such as H2-antagonists, PPIs, sucralfate, and prostaglandin inhibitors. The practice is to treat atrisk patients prophylactically with appropriate therapy to prevent stress-related gastrointestinal
ulceration and bleeding.
What Is the Context for the Use of Stress Ulcer Prophylaxis?

Guidelines from the American Society of Health Pharmacists recommend the use of stress
ulcer prophylaxis for high risk patients with any of the following conditions: mechanical
ventilation >48 hours, coagulopathy (platelet count <50,000 mm3, International Normalized
Ratio (INR) >1.5, or Prothrombin Time (PTT) >2 control value), or GI bleeding within the last
year; or 2 minor risk factors including >1 week ICU stay, sepsis, glucocorticoid therapy, or
occult GI bleeding 6 days.16
What Have We Learned About Stress Ulcer Prophylaxis?

In the past decade, several systematic reviews have been conducted on stress ulcer
prophylaxis. PPIs have increasingly replaced the use of H2-receptor antagonists and sucralfate,
despite a limited number of studies evaluating effectiveness in comparison to other agents. Thus,
the remainder of this chapter will present a recent review of the literature including specific
recommendations based on the evaluation of the evidence.
Recent Reviews and Systematic Evaluations

From 2010 to 2011, three systematic reviews compared the effectiveness of acid suppressive
therapies,13-15 including one systematic review that assessed studies on PPIs.17
Huang et al.17 conducted a meta-analysis of 10 RCTs, including 2092 patients, that directly
compared H2-antagonists and sucralfate in mechanically ventilated patients. The main outcome
measures were rates of clinically important gastrointestinal bleeding, ventilator-associated
pneumonia, gastric colonization, and ICU mortality. While there was a trend towards decreased
overt bleeding with H2-antagonists compared with sucralfate (OR = 0.87, 95% CI: 0.49 to 1.53),
sucralfate was associated with a decreased incidence of ventilator-associated pneumonia (OR =
1.32, 95% CI: 1.07 to 1.64). No difference between the agents was found for mortality (OR =
1.08, 95% CI: 0.86 to 1.34). The authors concluded that H2-antagonists were not more effective
in the prevention of overt GI bleeding than sucralfate, but were associated with higher rates of
ventilator-associated pneumonia.
Lin et al.18 evaluated 7 RCTs involving 936 patients that compared H2-antagonists with PPIs.
The meta-analysis reported on the incidence of stress-related upper gastrointestinal bleeding,
pneumonia, and ICU mortality. The review found no strong evidence that PPIs were significantly
different from H2-antagonists in the prevention of overt or clinically important upper GIl
bleeding (pooled risk difference -0.04, 95% CI: -0.09-0.01), pneumonia, or ICU mortality.
Marik et al.19 evaluated the effect of H2-antagonists compared with placebo, with specific
attention to the role of enteral nutrition as an effect modifier. The review found H2-antagonists
298
reduced the incidence of clinically significant GI bleeding, but only in patients not receiving
enteral nutrition. In patients receiving enteral nutrition, H2-antagonists did not affect the risk of
GI bleeding; however, this finding is based on only three trials enrolling a total of 262 patients.
The possibility that enteral nutrition may have a protective effect on patients baseline risk of
stress ulceration implies that routine acid suppressive therapy may not be necessary even in
patients with traditional risk factors. This finding, while exploratory, is certainly worthy of
further study.
These systematic reviews suggest that acid suppressive therapy, while effective in preventing
stress-related mucosal bleeding, is also associated with significant risks, including pneumonia.
PPIs, though widely used, do not appear to be superior to H2-antagonists in preventing clinically
significant GI bleeding.
No New Studies for Effectiveness of Acid Suppressive Therapy for

Stress Ulcer Prophylaxis
To date, no additional RCTs or large scale observational or cohort studies of adequate quality
have evaluated the effectiveness of pharmacologic acid-suppressive therapy for stress ulcer
prophylaxis, apart from those included in the recent systematic reviews discussed above.
PPI Use and Misuse Have the Potential for Harm

The only PPI that is FDA-approved for stress ulcer prophylaxis is omeprazole immediaterelease suspension. Overall, data demonstrate that PPIs are becoming the preferred agents of
choice for prophylaxis, despite no clear evidence that these agents are superior to H2-receptor
antagonists or placebo.20 Widespread use of PPIs, and inappropriate use, is common in
hospitalized patients and is associated with significant cost.21-25 A survey of trauma ICUs found
that the majority of patients continued stress ulcer prophylaxis after leaving the ICU.26 In a
retrospective chart review over a 3 month period, Wohlt found 357 patients received stress ulcer
prophylaxis in the ICU and 80% continued therapy following transfer out of the ICU. In 60% of
these cases, the authors judged that the therapy was continued inappropriately. Approximately
25% of patients were discharged from the hospital with inappropriate therapy, at a total cost of
$13,973.27
Several RCTs and systematic reviews have noted the association between acid suppressive
agents, specifically proton pump inhibitors and H2-receptor antagonists, and risk of nosocomial
pneumonia, community-acquired pneumonia and enteric infections, specifically Clostridium
difficile.13,15,17,18,28-31 The risk of hospital-acquired pneumonia extends to patients taking PPIs
outside of the ICU. A cohort study of 63,878 non-ICU patients demonstrated that PPI use was
associated with development of hospital-acquired pneumonia.29 Inappropriate continuation of
acid suppressive therapy, particularly PPIs, after discharge from the ICU therefore can have
adverse short-term effects for patients.
Costs and Implementation

Effective prevention of stress ulcer-related bleeding involves implementing methods to both
increase rates of appropriate prophylaxis and decrease inappropriate prophylaxis. Much of the
literature on increasing prophylaxis rates derives from studies of bundled approaches to ICU
preventive practices. The Keystone ICU Project, which ranks as one of the most successful
patient safety interventions of the past decade, used a ventilator bundle of five practices to
299
improve safety of mechanically ventilated patients, including stress ulcer prophylaxis.32 This
project was remarkably successful at preventing hospital-acquired infections and improving
other safety outcomes in the ICU, and also successfully increased stress ulcer prophylaxis rates.
Another successful approach to increasing prophylaxis was described by Krimsky et al, who
implemented a similar bundle approach incorporating several ICU prophylactic measures,
including stress ulcer prophylaxis. The implementation method emphasized team
communication, used prompts to providers to address the evidence-based measures on a daily
basis, and used a data wall to provide real-time feedback.33 This approach resulted in nearly
100% adherence to bundle use.
Evidence on efforts to control inappropriate prophylaxis use is limited. Coursol and Sanzari
described the implementation of an ICU algorithm with specific indications according to
guidelines on appropriate use, length of therapy, and cost.34 The algorithm was associated with a
reduction in inappropriate use of prophylaxis and costs.
Evidence on the cost of prophylaxis as it relates to implementation is also lacking. The cost
of acid suppressive therapy varies, with H2-receptor antagonists being less expensive than PPIs.
Decreasing inappropriate PPI use could likely be cost-saving for hospitals.

Acid suppressive therapy (H2-receptor antagonists and PPIs) and sucralfate are effective in
the prevention of bleeding from stress-related gastric ulceration in ICU patients. PPIs are widely
used, but are more expensive and no more effective than H2 receptor antagonists. Both types of
acid suppressive therapy appear to be used inappropriately, often being continued after patients
are discharged from the ICU. This practice raises safety concerns given the association between
acid suppressive therapy and pneumonia. While relatively strong evidence indicates that rates of
appropriate prophylaxis can be improved through the use of bundled approaches to ICU
prophylaxis, evidence on how to limit inappropriate prophylaxis is lacking. Further research in
this area is required in order to determine how to target prophylaxis most effectively to patients
who will receive the most benefit, while avoiding prophylaxis when it is not required. A
summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Rare/Moderate
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Moderate
(pneumonia)
Estimate of
Cost
Moderate
How Much do We
References
1.
2.
Cook DJ. Stress ulcer prophylaxis:

gastrointestinal bleeding and nosocomial
pneumonia. Best evidence synthesis. Scand
J Gastroenterol Suppl. 1995;210:48-52.
PMID: 8578207.
300
Fennerty MB. Pathophysiology of the upper

gastrointestinal tract in the critically ill
patient: rationale for the therapeutic benefits
of acid suppression. Crit Care Med.
2002;30(6 Suppl):S351-5. PMID: 12072660.
3.
Mutlu GM, Mutlu EA, Factor P. GI

complications in patients receiving
mechanical ventilation. Chest.
2001;119(4):1222-41. PMID: 11296191.
4.
Cook DJ, Fuller HD, Guyatt GH, et al. Risk

factors for gastrointestinal bleeding in
critically ill patients. Canadian Critical Care
Trials Group. N Engl J Med.
1994;330(6):377-81. PMID: 8284001.
5.
6.
Cook DJ, Griffith LE, Walter SD, et al. The

attributable mortality and length of intensive
care unit stay of clinically important
gastrointestinal bleeding in critically ill
patients. Crit Care. 2001;5(6):368-75.
PMID: 11737927.
Faisy C, Guerot E, Diehl JL, et al. Clinically
significant gastrointestinal bleeding in
critically ill patients with and without stressulcer prophylaxis. Intensive Care Med.
2003;29(8):1306-13. PMID: 12830375.
7.
Pimentel M, Roberts DE, Bernstein CN, et

al. Clinically significant gastrointestinal
bleeding in critically ill patients in an era of
prophylaxis. Am J Gastroenterol.
2000;95(10):2801-6. PMID: 11051351.
8.
Cook D, Heyland D, Griffith L, et al. Risk

factors for clinically important upper
gastrointestinal bleeding in patients
requiring mechanical ventilation. Canadian
Critical Care Trials Group. Crit Care Med.
1999;27(12):2812-7. PMID: 10628631.
9.
Daley RJ, Rebuck JA, Welage LS, et al.

Prevention of stress ulceration: current
trends in critical care. Crit Care Med.
2004;32(10):2008-13. PMID: 15483408.
10.
Dellinger RP, Levy MM, Carlet JM, et al.

Surviving Sepsis Campaign: international
guidelines for management of severe sepsis
and septic shock: 2008. Crit Care Med.
2008;36(1):296-327. PMID: 18158437.
11.
Berwick DM, Calkins DR, McCannon CJ, et

al. The 100,000 lives campaign: setting a
goal and a deadline for improving health
care quality. JAMA. 2006;295(3):324-7.
PMID: 16418469.
12.
Pronovost PJ, Berenholtz SM, Goeschel C,

et al. Improving patient safety in intensive
care units in Michigan. J Crit Care.
2008;23(2):207-21. PMID: 18538214.
301
13.
Cook D, Guyatt G, Marshall J, et al. A

comparison of sucralfate and ranitidine for
the prevention of upper gastrointestinal
bleeding in patients requiring mechanical
ventilation. Canadian Critical Care Trials
Group. N Engl J Med. 1998 Mar
19;338(12):791-7. PMID: 9504939.
14.
Cook DJ, Reeve BK, Guyatt GH, et al.

Stress ulcer prophylaxis in critically ill
patients. Resolving discordant metaanalyses. Jama. 1996 Jan 24-31;275(4):30814. PMID: 8544272.
15.
Messori A, Trippoli S, Vaiani M, et al.

Bleeding and pneumonia in intensive care
patients given ranitidine and sucralfate for
prevention of stress ulcer: meta-analysis of
randomised controlled trials. Bmj. 2000 Nov
4;321(7269):1103-6. PMID: 11061729.
16.
American Society of Health-System

Pharmacists. ASHP Therapeutic Guidelines
on Stress Ulcer Prophylaxis. ASHP
Commission on Therapeutics and approved
by the ASHP Board of Directors on
November 14, 1998. Am J Health Syst
Pharm. 1999 Feb 15;56(4):347-79. PMID:
10690219.
17.
Huang J, Cao Y, Liao C, et al. Effect of

histamine-2-receptor antagonists versus
sucralfate on stress ulcer prophylaxis in
mechanically ventilated patients: a metaanalysis of 10 randomized controlled trials.
Crit Care. 2010;14(5):R194. PMID:
21034484.
18.
Lin PC, Chang CH, Hsu PI, et al. The

efficacy and safety of proton pump
inhibitors vs histamine-2 receptor
antagonists for stress ulcer bleeding
prophylaxis among critical care patients: a
meta-analysis. Crit Care Med. 2010
Apr;38(4):1197-205. PMID: 20173630.
19.
Marik PE, Vasu T, Hirani A, et al. Stress

ulcer prophylaxis in the new millennium: a
systematic review and meta-analysis. Crit
Care Med. 2010 Nov;38(11):2222-8. PMID:
20711074.
20.
Heidelbaugh JJ, Goldberg KL, Inadomi JM.

Overutilization of proton pump inhibitors: a
review of cost-effectiveness and risk
[corrected]. Am J Gastroenterol. 2009
Mar;104 Suppl 2:S27-32. PMID: 19262544.
21.
Cornish P PJ, Saibil F. Audit of IV

pantoprazole: patterns of use and
compliance with guidelines. Can J Hosp
Pharm; 2002. p. 55:206.
22.
Enns R, Andrews CN, Fishman M, et al.

Description of prescribing practices in
patients with upper gastrointestinal bleeding
receiving intravenous proton pump
inhibitors: a multicentre evaluation. Can J
Gastroenterol. 2004 Sep;18(9):567-71.
PMID: 15457296.
23.
Farrell CP, Mercogliano G, Kuntz CL.

Overuse of stress ulcer prophylaxis in the
critical care setting and beyond. J Crit Care.
Jun;25(2):214-20. PMID: 19683892.
24.
Hwang KO, Kolarov S, Cheng L, et al.

Stress ulcer prophylaxis for non-critically ill
patients on a teaching service. J Eval Clin
Pract. 2007 Oct;13(5):716-21. PMID:
17824863.
25.
Perwaiz MK, Posner G, Hammoudeh F, et

al. Inappropriate Use of Intravenous PPI for
Stress Ulcer Prophylaxis in an Inner City
Community Hospital. J Clin Med Res. 2010
Oct 11;2(5):215-9. PMID: 21629543.
26.
27.
Barletta JF, Erstad BL, Fortune JB. Stress

ulcer prophylaxis in trauma patients. Crit
Care. 2002 Dec;6(6):526-30. PMID:
12493075.
Wohlt PD, Hansen LA, Fish JT.
Inappropriate continuation of stress ulcer
prophylactic therapy after discharge. Ann
PMID: 17848420.
302
28.
Eom CS, Jeon CY, Lim JW, et al. Use of

acid-suppressive drugs and risk of
pneumonia: a systematic review and metaanalysis. Cmaj. 2011 Feb 22;183(3):310-9.
PMID: 21173070.
29.
Herzig SJ, Howell MD, Ngo LH, et al. Acidsuppressive medication use and the risk for
hospital-acquired pneumonia. Jama. 2009
May 27;301(20):2120-8. PMID: 19470989.
30.
Laheij RJ, Sturkenboom MC, Hassing RJ, et

al. Risk of community-acquired pneumonia
and use of gastric acid-suppressive drugs.
Jama. 2004 Oct 27;292(16):1955-60. PMID:
15507580.
31.
Prodhom G, Leuenberger P, Koerfer J, et

al. Nosocomial pneumonia in mechanically
ventilated patients receiving antacid,
ranitidine, or sucralfate as prophylaxis for
stress ulcer. A randomized controlled trial.
Ann Intern Med. 1994 Apr 15;120(8):65362. PMID: 8135449.
32.
Bavishi C, Dupont HL. Systematic review:

the use of proton pump inhibitors and
increased susceptibility to enteric infection.
Aliment Pharmacol Ther. 2011 Dec;34(1112):1269-81. PMID: 21999643.
33.
Krimsky WS, Mroz IB, McIlwaine JK, et al.

A model for increasing patient safety in the
intensive care unit: increasing the
implementation rates of proven safety
measures. Qual Saf Health Care. 2009
Feb;18(1):74-80. PMID: 19204137.
34.
Coursol CJ, Sanzari SE. Impact of stress

ulcer prophylaxis algorithm study. Ann
Pharmacother. 2005 May;39(5):810-6.
PMID: 15811900.
Chapter 28. Prevention of Venous Thromboembolism: Brief

Update Review
Elliott R. Haut, M.D., FACS; Brandyn D. Lau, M.P.H.
Introduction
Deep venous thrombosis (DVT) refers to occlusion within the venous system, most
commonly of the lower extremities, which can lead to pulmonary embolism (PE), or embolism to
the pulmonary vasculature. Venous thromboembolism (VTE), comprising PE and DVT, is
estimated to account for 5 to 10 percent of all deaths among hospitalized patients,1,2 and also is
associated with significant morbidities. In 2008, the United States Surgeon General issued a Call
to Action to Prevent DVT and PE. The report brings to light the huge numbers of patients
afflicted by DVT (350,000-600,000) and killed by PE (>100,000) every year in the United
States.3 Even though high quality evidence exists for safe and effective strategies to reduce the
risk of VTE, studies continue to show that many hospitalized patients are not given riskappropriate VTE prophylaxis. One recent study across 32 countries found that only 59 percent of
at-risk surgical and 40 percent of at-risk medical patients received guideline-recommended VTE
prophylaxis4 and a United States registry study found that only 42 percent of patients diagnosed
with DVT during a hospitalization had received prophylaxis.5
The Agency for Healthcare Research and Quality (AHRQ) has indicated that delivery of
appropriate VTE prophylaxis is an essential patient safety practice and one that can prevent inhospital death.6 As of 2011, the National Quality Forum (NQF) has 10 VTE-related standards
and endorsed outcomes measures.7 Evidence-based best practice prophylaxis varies by primary
service (e.g. medicine, surgery, trauma, orthopedics) and patient risk factors. Risk of VTE
among hospitalized patients varies based on several risk factors including medical condition,
type of surgery, trauma, cancer, age, immobility, hypercoagulable state, and previous history of
VTE. Most hospitalized patients have one or more VTE risk factors, and well-developed
guidelines are available that specify which types of patients should receive prophylaxis
measures, and which specific measures are most appropriate.1
The original report, Making Health Care Safer, reviewed the effectiveness, safety, costeffectiveness, and indications for VTE prophylaxis. This review concluded that whereas VTE
prophylaxis shows clear benefits for a number of conditions and minimal concerns regarding
adverse events, the practice remains underused. A small number of interventions aimed at
improving use of prophylaxis were reviewed. The current review provides an update on the most
effective VTE prophylaxis regimens as well as on interventions aimed at improving adherence to
guidelines on the use of these preventive strategies. A MEDLINE search was conducted from
2001 to 2011 to identify studies that assessed the effectiveness and safety of VTE preventive
measures as well as those aimed at improving their use.
What Are the Practices for Preventing Venous Thromboembolism?

Both pharmacologic and mechanical prophylactic interventions have been demonstrated to be
effective in preventing many VTE events and have been evaluated for their appropriateness for
certain types of patients (medical vs. surgical) with certain risk factors.1,8 Pharmacologic
prophylaxis includes low dose unfractionated heparin; low-molecular weight heparins, including
303
enoxaparin, dalteparin, and fondaparinux; warfarin; and aspirin, along with newer classes of antithrombotic agents. Mechanical prophylaxis includes anti-embolic stockings and intermittent
pneumatic compression devices. Because the underlying approach of all prophylaxis medications
is to decrease clotting, they may increase the risk of bleeding. The balance between bleeding and
clotting must be considered in every patient, and the benefits and harms must be weighed before
administering these drugs. For this reason, patient risk stratification is paramount to ensure that
only at-risk patients are treated and that they receive the right prophylaxis. Ongoing clinical
research and evidence-based medicine reviews suggest that blanket approaches that give the
same medication to all patients without risk stratification may not be beneficial and may even
cause more harm than benefit.9-11
New Medications for VTE prophylaxis

New versions of low molecular weight heparins (LMWH) are being brought to market, with
additional newly approved indications by the U.S. Food and Drug Administration (FDA). In
addition, other medications with different pathways of action are being researched and approved.
Most recently in July 2011, rivaroxiban, an oral direct Factor Xa inhibitor, was approved by the
FDA for prophylaxis of DVT/PE in adults undergoing hip and knee replacement surgery.
Dabigitran, an oral direct thrombin inhibitor, is FDA approved for prevention of stroke in
patients with non-valvular atrial fibrillation. Although it is not currently approved for VTE
prophylaxis in the United States, it is being used in this capacity in some European countries and
Canada. A recent systematic review and meta-analysis of three novel oral agents, dabigatran,
apixaban and rivaroxaban, for VTE prophylaxis after total hip and total knee replacement
surgery found no difference in net clinical benefit. In fact, this review reported that success in
prevention of VTE was inversely associated with clinically relevant bleeding.12 These findings
are indicative of the diminishing returns associated existing medications developed to prevent
VTE and highlight the need to improve prescription of the best-practice medications currently
available.13
Inferior Vena Cava Filters

New technologic advances in devices to prevent DVT from becoming PE via mechanically
trapping the clot in the inferior vena cava before they can reach the heart and lungs may be
beneficial in some patient populations. Although originally designed for permanent use, multiple
approved devices can now be placed for temporary (also known as optional or retrievable)
prophylaxis and then removed at a later date. However, the evidence to support the use of this
technology is unclear.
For example, the placement of inferior vena cava filers (IVCFs) is rapidly increasing among
trauma patients14 for primary prophylaxis against PE even in patients without proven DVT.
Clinical uncertainty remains about whether prophylactic IVCFs should be used in trauma.
Current guidelines from the American College of Chest Physicians (ACCP)1 and the Eastern
Association for the Surgery of Trauma (EAST)15 have diametrically opposed opinions on the use
of IVCFs for primary PE prophylaxis. An ongoing AHRQ sponsored Evidence-based Practice
Center Systematic Review Protocol entitled Comparative Effectiveness of Pharmacologic and
Mechanical Prophylaxis of Venous Thromboembolism among Special Populations will assess
the role of IVCFs in the prevention of pulmonary embolism in trauma and other special
populations (including those patients undergoing bariatric surgery).
304
What Approaches Have Been Used To Improve Appropriate VTE

Prophylaxis?
Evolution of information technology is enabling development of more sophisticated clinical
decision support systems to improve compliance with guidelines. Several recent examples are
described below.
Lesselroth et al,16 developed a clinical decision support-enabled order menu in their
computerized patient record system (CPRS) to recommend appropriate VTE prophylaxis at the
time medication orders are written at the Portland Oregon VA Medical Center. After identifying
and addressing some key initial limitations (providers could unintentionally or intentionally
bypass the order menu and recommended guidelines), use of the order menu increased from 20
percent to 80 percent. This study underscores the need for interventions to integrate well into
provider workflow and ideally be mandatory without any possibility of ignoring or bypassing the
VTE algorithm. Alerts and systems are only effective if they consistently reach their intended
target.
In the study by Beeler et al,17 an electronic alert was displayed in the medical chart of every
hospitalized medical patient who did not have pharmacological or mechanical VTE prophylaxis
ordered within 6 hours after admission and had documented VTE risk. Rates of
thromboprophylaxis orders among medical patients significantly increased from preimplementation rates of 43.4 percent to 66.7 percent (p<0.0001) during the 4 months after
implementation. The following year, thromboprophylaxis orders increased further to 73.6
percent (p=0.011).
Kucher et al,18 proactively searched for hospitalized patients at risk for developing VTE who
were not prescribed prophylaxis (pharmacological or mechanical). Electronic alerts were sent to
providers of patients randomized to the intervention group that their patient was at risk for VTE.
Patients in the intervention group were significantly more likely to receive mechanical
prophylaxis (p<0.001) and significantly more likely to receive prophylactic doses of
unfractionated heparin (p<0.001). There were no significant changes to orders of enoxaparin
(p=0.18) or warfarin (p=0.11) between intervention and control groups. In addition, patients in
the intervention group were significantly more likely to be free from DVT or PE after 90 days
(p<0.001). This approach is reactive it identifies patients who were not initially ordered
prophylaxis and then attempts to correct the patient safety problem, rather than suggesting and
improving rates of prophylaxis at the appropriate time of initial treatment.
In 2008, a mandatory, computerized decision support-enabled VTE risk stratification order
set was implemented in the computerized provider order entry system at the Johns Hopkins
Hospital to recommend ACCP guideline-appropriate, service-specific (e.g. medicine, general
surgery, trauma, etc.) prophylaxis for an individual patients risk stratum.19,20 Within the first
year, adherence to guideline-appropriate VTE prophylaxis increased significantly hospital-wide
and rates of VTE have been on a decreasing trend. This system overcomes the downsides of the
Kucher approach since it requires proactive risk stratification during the completion of the
admission order set for all admitted patients and therefore is nearly 100 percent effective at
forcing providers to assign an appropriate risk stratum to all patients within 24 hours of hospital
arrival.21 However, this system remains fallible since the guideline-suggested VTE prophylaxis
is merely a recommendation; it is not mandatory and may be ignored.
305

What Are the Beneficial Effects of VTE Prophylaxis?
The original Making Health Care Safer report focused on the evidence for effectiveness of
specific clinical interventions (i.e. medications and mechanical prophylaxis) for specific clinical
situations, and concluded that there was extensive evidence supporting their effectiveness and
low cost, particularly after certain types of surgical procedures, trauma, and medical conditions
such as cerebrovascular accidents.22 Quality improvement-related interventions such as practice
guidelines, clinical decision support systems, and educational interventions to change provider
behavior were addressed in separate chapters in the original support. A few studies found
beneficial effects of clinical decision support systems and educational interventions, both
separately and combined.
The updated evidence for VTE prophylaxis in selected patients has been well-described in a
variety of recent evidence-based clinical guidelines and systematic reviews.1,10,15 The evidence
for clinical interventions for VTE prophylaxis remains strong in specified populations, and
prophylaxis is recommended by practice guidelines for those patients, although it should not be
applied universally. Since the availability of medications and condition-specific evidence is
rapidly evolving and these guidelines are regularly updated, this evidence is not summarized
here, and the remainder of this section focuses on interventions intended to improve compliance
with risk-appropriate VTE prophylaxis among different patient populations.
Interventions To Improve Prophylaxis Adherence

A systematic review of interventions to improve VTE prophylaxis use in hospitals, based on
literature searches from 1996-2003, found 30 eligible studies; only one was an RCT and only
three had concurrent controls. Strategies included passive dissemination, which had little effect
(50% compliance), single-strategy studies (12 studiesaudit and feedback, documentation aids,
and quality assurance activities all produced about 80% compliance), and clinical decision
support systems approached 100 percent compliance. Twelve studies incorporated two or more
strategies, usually including an educational component, and all demonstrated improvements in
use of VTE prophylaxis. In addition to the types of strategies used in the single-strategy studies,
these studies also included strategies such as advertising, appointment of specific implementation
staff, and recruitment of local change agents or opinion leaders. Most studies evaluated change in
provider behavior, not patient outcomes, and no study that evaluated outcomes demonstrated a
reduction in DVT or PE rates, often due to lack of adequate power.23
Interventions to improve adherence to prophylaxis include implementation of clinical
decision support tools, financial disincentives, and outcomes reporting. Clinical decision support
tools have the potential to improve adherence to guideline-appropriate prophylaxis ordering24,25
which may then have a sustained impact on clinical outcomes. While this method has classically
taken the form of paper-based order-sets, as computerized provider order entry systems are
adopted in hospitals across the country, an opportunity exists to build electronic clinical decision
support into these systems to evaluate, risk stratify patients based on individual patient risk
factors and recommend the appropriate VTE prophylaxis strategies.
Outcomes reporting is another approach to improve VTE prophylaxis, through feedback and
public reporting or the financial incentive of nonpayment for VTE events. The Centers of
Medicare and Medicaid Services (CMS) placed VTE after orthopedic hip/knee replacement on
their list of never events for which providers will not be reimbursed. However, even with best
306
practice, not all VTE events can be prevented;26,27 it has been estimated that best practice
prophylaxis may reduce incidence of DVT by up to 70 percent.1 Another potential limitation to
the use of DVT/PE rates alone to measure quality is the significant issue of surveillance bias
because many DVTs are clinically silent and therefore go undetected without routine screening.28
For example, in the field of trauma surgery, clinical ambiguity persists regarding the clinical and
cost effectiveness of the screening of high-risk asymptomatic trauma patients for DVT with
duplex ultrasound.28 As a result, certain providers and hospitals report higher DVT rates due
entirely to higher rates of diagnostic testing- a classic example of surveillance bias.29-31
Because of these issuesand variation in patient riskunadjusted VTE rates are likely not
appropriate for public reporting. A better definition of preventable harm may be obtained by
combining an outcome and process measure rather than relying on an outcome alone. For
example, it has been suggested that only VTE events occurring in patients who did not receive
adequate prophylaxis should be labeled a preventable VTE.28 This approach and specific
definition has been incorporated as one of the six Meaningful Use Quality Measures related to
VTE,22 although this measure has not yet been evaluated for its impact on VTE prophylaxis
compliance.

Strong evidence from numerous high-quality trials supports the effectiveness of VTE
prophylaxis for specific populations, although there are significant risks and risk stratification is
necessary to ensure that prophylaxis is targeted to appropriate patients. However, rates of VTE
prophylaxis are suboptimal, and VTE remains a difficult and elusive crisis in patient safety. Less
evidence exists on which interventions are effective for increasing rates of VTE prophylaxis in
appropriate populations. As with other patient safety interventions, educating providers on the
benefits of appropriate VTE prophylaxis alone is not an effective strategy to improve appropriate
use of VTE prophylaxis. Evidence, although mostly low-quality, non-randomized studies
without concurrent controls, supports that education combined with other quality improvement
strategies, and information technology approaches such as mandatory computerized clinical
decision support, appear to offer the most effective approaches to promote best practice
prophylaxis use and prevent patient harm resulting from VTE. A summary table is located below
(Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
High
Evidence or
Potential for
Harmful
Unintended
Consequences
Moderate
(bleeding)
Estimate of
Cost
Low
307
How Much do We
Little/Moderate
References
1.
2.
Geerts WH, Bergqvist D, Pineo GF, Heit

JA, Samama CM, Lassen MR, Colwell CW,
American College of Chest Physicians.
Prevention of venous thromboembolism:
American college of chest physicians
evidence-based clinical practice guidelines
(8th edition). Chest 2008 Jun;133(6
Suppl):381S-453S.
Anderson FA, Jr. Z,M., Heit JA, Fidan D,
Cohen AT. Estimated annual numbers of US
acute-care hospital patients at risk for
venous thromboembolism. Am J Hematol
2007 Sep;82(9):777-82.
3.
The Surgeon Generals Call to Action to

Prevent Deep Vein Thrombosis and
Pulmonary Embolism [Internet]; c2008
[cited 2011 August 30].
www.surgeongeneral.gov/topics/deepvein/.
4.
Cohen AT, Tapson VF, Bergmann J,

Goldhaber SZ, Kakkar AK, Deslandes B,
Huang W, Zayaruzny M, Emery L,
Anderson FA. Venous thromboembolism
risk and prophylaxis in the acute hospital
care setting (ENDORSE study): A
multinational cross-sectional study. The
Lancet 2008;371(9610):387-94.
5.
Goldhaber SZ, Tapson VF, DVT FREE

Steering Committee. A prospective registry
of 5,451 patients with ultrasound-confirmed
deep vein thrombosis. The American Journal
of Cardiology 2004;93(2):259-62.
6.
Maynard G, Stein J. Preventing hospitalacquired venous thromboembolism: A guide

for effective quality improvement. AHRQ
Publication No. 08-0075: Agency for
Healthcare Research and Quality, Rockville,
MD; 2008:Agency for Healthcare Research
and Quality.
7.
National Quality Forum. c2012. NQFEndorsed Standards.

www.qualityforum.org/Home.aspx.
8.
Cushman JG, Agarwal N, Fabian TC, Garcia

V, Nagy KK, Pasquale MD, Salotto AG,
EAST Practice Management Guidelines
Work Group. Practice management
guidelines for the management of mild
traumatic brain injury: The EAST practice
management guidelines work group. J
Trauma 2001 Nov;51(5):1016-26.
308
9.
Goldhaber SZ, Leizorovicz A, Kakkar AK,

Haas SK, Merli G, Knabb RM, Weitz JI,
ADOPT Trial Investigators. Apixaban
versus enoxaparin for thromboprophylaxis
in medically ill patients. N Engl J Med 2011
Dec 8;365(23):2167-77.
10.
Qaseem A, Chou R, Humphrey L, Starkey

M, Shekelle P. Venous thromboembolism
prophylaxis in hospitalized patients: A
clinical practice guideline from the american
college of physicians. Annals of Internal
Medicine 2011;155(9):625-32.
11.
Lederle FA, Zylla D, MacDonald R, Wilt

TJ. Venous thromboembolism prophylaxis
in hospitalized medical patients and those
with stroke: A background review for an
american college of physicians clinical
practice guideline. Ann Intern Med 2011
Nov 1;155(9):602-15.
12.
Gmez-Outes A, Terleira-Fernndez A,
Surez-Gea M, Vargas-Castrilln E.
Dabigatran, rivaroxaban or apixaban versus
enoxaparin for thromboprophylaxis after
total hip or knee replacement: Systematic
review, meta-analysis and indirect treatment
comparisons. BMJ 2012;344:e3675.
13.
Haut E, Lau B, Streiff M. New oral

anticoagulants for preventing venous
thromboembolism. BMJ 2012;344:e3820.
14.
Dossett LA, Adams RC, Cotton BA.

Unwarranted national variation in the use of
prophylactic inferior vena cava filters after
trauma: An analysis of the national trauma
databank. J Trauma 2011
May;70(5):1066,70; discussion 1070-1.
15.
Rogers FB, Cipolle MD, Velmahos G,

Rozycki G, Luchette FA. Practice
management guidelines for the prevention of
venous thromboembolism in trauma
patients: The EAST practice management
guidelines work group. J Trauma 2002
Jul;53(1):142-64.
16.
Lesselroth BJ, Yang J, McConnachie J,

Brenk T, Winterbottom L. Addressing the
sociotechnical drivers of quality
improvement: A case study of postoperative DVT prophylaxis computerised
decision support. BMJ Qual Saf
2011;20(5):381-9.
17.
Beeler PE, Kucher N, Blaser J. Sustained

impact of electronic alerts on rate of
prophylaxis against venous
thromboembolism. Thromb Haemost
2011;106(4):734-8.
25.
OConnor C, Adhikari NK, DeCaire K,

Friedrich JO. Medical admission order sets
to improve deep vein thrombosis
prophylaxis rates and other outcomes. J
Hosp Med 2009 Feb;4(2):81-9.
18.
Kucher N, Koo S, Quiroz R, Cooper JM,

Paterno MD, Soukonnikov B, Goldhaber
SZ. Electronic alerts to prevent venous
thromboembolism among hospitalized
patients. N Engl J Med 2005 Mar
10;352(10):969-77.
26.
19.
Streiff MB, Carolan H, Hobson DB, Kraus

PS, Holzmueller C, Demski R, Lau B,
Biscup-Horn P, Pronovost PJ, Haut ER.
Lessons from the john hopkins multidisciplinary venous thromboembolism
(VTE) prevention collaborative. BMJ
2012;344:e3935.
Haut E. Venous thromboembolism: Are

regulatory requirements reasonable? Critical
Connections [Internet]. [revised
2008:December 9, 2011.
www.sccm.org/Publications/Critical_Conne
ctions/Archives/April_2008/Pages/VenousT
hromboembolism.aspx.
27.
Streiff MB, Haut ER. The CMS ruling on

venous thromboembolism after total knee or
hip arthroplasty: Weighing risks and
benefits. JAMA 2009 Mar 11;301(10):10635.
28.
Haut ER, Pronovost PJ. Surveillance bias in

outcomes reporting. JAMA
2011;305(23):2462-3.
29.
Haut ER, Noll K, Efron DT, Berenholz SM,

Haider A, Cornwell EE3, Pronovost PJ. Can
increased incidence of deep vein thrombosis
(DVT) be used as a marker of quality of care
in the absence of standardized screening?
the potential effect of surveillance bias on
reported DVT rates after trauma. J Trauma
2007 Nov;63(5):1132,5; discussion 1135-7.
30.
Pierce CA, Haut ER, Kardooni S, Chang

DC, Efron DT, Haider A, Pronovost PJ,
Cornwell EE,3rd. Surveillance bias and deep
vein thrombosis in the national trauma data
bank: The more we look, the more we find. J
Trauma 2008 Apr;64(4):932,6; discussion
936-7.
31.
Haut ER, Chang DC, Pierce CA, Colantuoni

E, Efron DT, Haider AH, Cornwell EE,3rd,
Pronovost PJ. Predictors of posttraumatic
deep vein thrombosis (DVT): Hospital
practice versus patient factors-an analysis of
the national trauma data bank (NTDB). J
Trauma 2009 Apr;66(4):994,9; discussion
999-1001.
20.
21.
Haut ER, Streiff MB. Mandatory ComputerBased Decision Support System [computer
program]. North American Thrombosis
Forum (NATF); 2010.
Haut ER, Lau BD, Kraenzlin FS, Hobson
DB, Kraus PS, Carolan HT, Haider AH,
Holzmueller CG, Efron DT, Pronovost PJ, et
al. Improved prophylaxis and decreased
preventable harm with a mandatory
computerized clinical decision support tool
for venous thromboembolism (VTE)
prophylaxis in trauma patients. Archives of
Surgery in press.
22.

Wachter RM. Making health care safer: A
critical analysis of patient safety practices.
Research and Quality; 2001. Report nr 43.
23.
Tooher R, Middleton P, Pham C, Fitridge R,

Rowe S, Babidge W, Maddern G. A
systematic review of strategies to improve
prophylaxis for venous thromboembolism in
hospitals. Annals of Surgery 2005
Mar;241(3):397-415.
24.
Liu DS, Lee MM, Spelman T, Macisaac C,

Cade J, Harley N, Wolff A. Medication
chart intervention improves inpatient
thromboembolism prophylaxis. Chest
2012;141(3):632-41.
309
Chapter 29. Preventing Patient Death or Serious Injury

Associated With Radiation Exposure From Fluoroscopy and
Computed Tomography: Brief Review (NEW)
Introduction
Fluoroscopically- and computed tomography (CT)-guided diagnostic and interventional
procedures are being performed with increasing frequency worldwide. From 1980 to 2007,
annual performance of CT in the U.S. increased from 3 million1 to 80 million.2 With this rapid
increase in the use of imaging techniques, there has been a concurrent increase in patient
exposure to ionizing radiation.1
Effects associated with radiation can be categorized as either deterministic or stochastic.
Deterministic effects manifest themselves in a relatively short time after a high-intensity
exposure to radiation (e.g., 1 or more sieverts).3 In 1994, approximately 50 radiation-induced
burns were reported to the U.S. Food and Drug Administration (FDA). In 2000, a review of 73
reports of radiation-induced skin injuries4 identified fluoroscopically-guided procedures as the
cause of 38 severe skin injuries (e.g., chronic ulceration); 18 requiring skin grafts.5,6 Radiationinduced burns have also been reported after extended radiation exposure during CT brain
perfusion scans.7
Stochastic effects are increased risks of various conditions (e.g., cancer, heart disease) that
manifest themselves over a longer time period. Recent estimates indicate that CT scans
performed in the U.S. in 2007 will be related to approximately 29,000 future cancers; killing
nearly 15,000. Almost one half of the projected cancers will be due to scans of the abdomen and
pelvis.8 Experts indicate that more than 400 patients (across eight U.S. hospitals) who recently
received higher-than-expected radiation doses while undergoing CT brain perfusion scans may
now face long-term risks of cancer and brain damage.9
What Are the Practices for Reducing Ionizing Radiation Exposure?

The core principle governing the use of ionizing radiation is ALARA (As Low As
Reasonably Achievable). The goal of ALARA is to reduce both patient and technician exposure
to ionizing radiation without compromising diagnostic or therapeutic efficacy. Several measures
recommended by national organizations to reduce patients exposure to ionizing radiation are
discussed below.
Technical measures. The American College of Radiology (ACR), the Radiological Society of
North America (RSNA), the American Association of Physicists in Medicine (AAPM), and the
American Society of Radiologic Technologists (ASRT) are primary participating members in the
Image Wisely campaign.10 On its Web site (imagewisely.org), a list of technical mechanisms for
dose reduction during CT include x-ray beam filtration, x-ray beam collimation, tube current
modulation, peak kilovoltage optimization, improved detector efficiency, and noise-reduction
algorithms.11,12 In 2010, task force members of the U.S.-based Conference of Radiation Control
Program Directors (CRCPD)13 recommended technical methods during fluoroscopy:
310
Minimize x-ray beam time

Vary the site of the entrance port on the patient as clinically possible
Optimize collimation
Use the least amount of machine magnification possible
Position the x-ray source and image receptor optimally
Apply machine dose reduction features (e.g., last image hold feature, pulsed fluoroscopy)
Maintain equipment in good repair and calibration
Appropriate utilization. Steps to improve use of diagnostic imaging by referring physicians

include reexamining the need for more dose-intensive diagnostic imaging, which may affect the
number of self-referrals.11 As one of several U.S. physician groups participating in the Choosing
Wisely Campaign, the ACR recently identified imaging exams that, although commonly used,
might be unnecessary.14 To reduce unnecessary imaging, ACR recommended further physicianpatient discussion before scheduling five specific imaging exams. The list includes imaging for
uncomplicated headache absent specific risk factors for structural disease or injury and imaging
for suspected pulmonary embolism without moderate or high pre-test probability of pulmonary
embolism.15 The ACR recommendations were based on a review of professional guidelines and
published evidence.
The ACR also suggests that regularly posting individual physician ordering patterns, whether
appropriate or inappropriate, may positively influence physician ordering behavior through peer
pressure. This practice may be especially helpful for non-physicians (e.g., physician assistants,
nurse practitioners) who may be ordering imaging studies, and whose ordering patterns are likely
to reflect the behavior of their supervising physicians.16 The ACR also sponsors registries (e.g.,
Dose Index Registry), which provide participating facilities with feedback on their radiationexposure levels in comparison with nationwide levels and those from other institutions.17 Prior
and recent successes have been reported in providing physician feedback and the psychology
underlying it.18-20
Education and training. Referring physicians must be thoroughly educated on radiation safety
in order to routinely consider this factor when ordering imaging examinations. Technologists
should be trained to ensure that proper procedures and techniques are followed to prevent the
need for repeated imaging due to suboptimal image quality. Technologists can also notify a
radiologist when a duplicate questionable examination is ordered. Substituting less doseintensive modalities (e.g., MRI, ultrasound and radiography in lieu of CT) should also be
considered.16 According to the CRCPD, training of fluoroscopist and staff on the biological
effects of ionizing radiation is one of three components of a comprehensive radiation dose
management program. Two remaining components are monitoring and tracking of fluoroscopic
dose and patient follow-up.
Algorithms and protocols. CT-related strategies targeted to Imaging Physicians by the Image
Wisely campaign include use of adaptive iterative reconstruction and development of protocols
that maximize diagnostic yield while minimizing dose. A few preliminary studies have suggested
for example that more limited CT of the lower abdomen and pelvis (versus standard practice to
perform CT of the entire abdomen and pelvis) should be performed to evaluate conditions such
as suspected appendicitis.21-23 Adjustment of CT protocols to reduce radiation exposure
according to factors such as body mass is also a recommended strategy.24
311

Studies focusing on radiation exposure reduction measures during fluoroscopy and CT were
mostly conducted at single institutions at a university hospital setting. The largest study
examined efforts among 15 imaging centers involved in a Mid-west consortium.
Fluoroscopy. Lee et al. evaluated the effectiveness of a quality assurance (QA) protocol to
reduce radiation exposure during fluoro urodynamics.25 Prior to implementation, this institution
identified many unnecessary images that did not contribute to the diagnostic value of a patients
study. In this study, fluoroscopic imaging helped to visualize the anatomy of the lower urinary
tract in 97 patients diagnosed with urinary incontinence, urinary retention, and other conditions.
The QA protocol, limiting fluoroscopy to 4-5 static images, was distributed to all physicians,
nurses, and radiology technicians involved in the procedure. The importance of radiation safety
was emphasized to all staff involved in the procedure. This QA protocol was limited to
anteroposterior views in the sitting position so generalizability of this protocol may be limited.
Ngo et al.20 evaluated cases of unilateral ureteroscopy for stone disease. First steps to
implementation included working with operating room (OR) personnel to track fluoroscopy time
as an additional step in their post-procedural documentation. This process was not widely
publicized and required minimal changes to existing OR staff workflow.
The multicomponent QA protocol evaluated in Greene et al.26 started with a detailed review
of prior imaging, which was later placed in front of the scrubbed surgeon on a high-definition
monitor during the entire case. In addition, while previous radiation-reducing measures were
performed without regard for respiratory motion, the fluoroscopy-reducing protocol included
C-arm activation timed with the patients respiration. Key to implementation was participation of
a designated fluoroscopy technician acquainted with the protocol goals and completely
familiarized with the fluoroscopy machine usage and relevant urological anatomy.
Lakkireddy27 reported use of four high-dose lithium fluoride thermoluminescent dosimeters,
a direct method to measure patient exposure. As a relatively new technique, atrial fibrillation
(AF) catheter ablation involves a steep learning curve. Staff physicians, the primary operators
during the procedure, were described as having experience performing more than 400 AF
ablations. Three of the four studies described above stated adherence to the ALARA principle as
an external influencer.
Computed tomography. Implementation tools used in one study28 included the use of real
and distractor stickers to blind study radiologists to the location of the region of tenderness.
Staff participating in the study was also blinded to clinical information, including the patients
original radiology reports. Broder et al. indicated that targeted CT strategies that focus on scan
length optimization may be inappropriate under certain conditions such as the need to visualize
an entire structure (e.g., aorta) or when diffuse abdominal processes are strongly considered
(e.g., bowel obstruction). Changes in clinical practice in one study1 included the integration of
computed tomography angiography (CTA) as part of routine imaging in monitoring patients for
development of vasospasm.
The first step in implementing an imaging algorithm in another study was providing an
imaging protocol to the ED staff.29 A collaborative approach between imaging services
(including radiology and nuclear medicine) and the ED staff followed soon after. If ED staff
requested a CTPA [computed tomographic pulmonary angiography] for a patient with a normal
chest radiograph, an action that violated the protocol, a radiologist would followup with the ED
312
by phone or email to discuss the request. Implementing this patient safety practice has
encouraged the ED at this institution to implement additional radiation reduction measures for
other diagnoses (e.g., renal colic).
Use of prospective gating was a core element of radiation reduction measures in two
studies.30,31 Other measures used in one study30 included limiting scan length, minimizing tube
current or voltage according to body physique, use of small bowtie filters, and tube current
modulation during cardiac cycle. A collaborative effort amongst three sites was involved in
protocol development in one study.31 LaBounty states that two measures (prospective ECG
gating and 100-kV tube voltage imaging) were only used in 92% and 67% of patients, suggesting
that additional radiation reductions would have been possible if protocol compliance had been
higher. Lack of awareness, uncertainty regarding appropriate implementation, and concern about
the quality of studies that assessed reduction techniques were also described as barriers to
implementing multiple radiation reduction techniques in everyday practice. The generalizability
of implementing a similar initiative at less experienced sites may be limited because the patient
population involved in this study underwent cardiac computed tomography angiography (CCTA)
at three large-volume, experienced centers.
One large urban medical center benefitted from adding a decision support (DS) system to its
existing radiology order entry (ROE) system.32 Before DS integration, referring physicians
completed a ROE form to initiate a CT exam. After introduction of the DS component, a second
form was populated providing physician feedback on appropriateness of the exam (1-9
appropriateness score), alternate procedures to consider, and options to proceed or cancel the
request. Appropriateness scores, based on ACR Appropriateness Criteria scores and locally
developed indication and procedure pairs, are continuously reviewed and modified.
Locally derived evidence-based imaging guidelines were the basis for a DS tool at another
multispecialty integrated health care network.33 Rapid implementation of the DS tool was
attributed to pressure from local commercial payers and an institutional culture already vested in
evidence-based medicine (including evidence-based imaging protocols) and lean health care
management methodology.34 An audit of imaging requests to determine outcome for orders
initially denied by the DS system was described as a potential screening method to determine
whether providers had gamed (developed ways to order inappropriate studies) the system.
Lastly, Raff et al. described implementation efforts at 15 hospital imaging centers
participating in the Advanced Cardiovascular Imaging Consortium in Michigan. Hospital
imaging centers were located in both small community hospitals and large academic medical
centers (1,000+ beds). Best practice recommendations were developed based on data (including
radiation dose and image quality metrics) from CCTA scanning of 620 patients acquired during a
13-month control period. During an 8-month intervention period, recommendations created by a
team consisting of a physician program director, a consulting radiologic technician, and a
licensed medical physicist were distributed to participating sites at scheduled consortium
meetings, during on-site visits by coordinating center staff and through personal communication.
This Best-Practice Model for Scan Acquisition includes directives on topics such as medical
history, administration of beta blockers and nitroglycerin, and protocol parameters (e.g., field of
view, tube current modulation). Scanner manufacturers were involved in training on scannerspecific techniques. Responsibility for on-site implementation was designated to a physician and
radiology technologist. Raff et al. reported that the greatest reduction in dose occurred at lowvolume sites (30 scans per month).
313

We limited our research to studies implementing initiatives to reduce patients radiation
exposure from fluoroscopy and computed tomography in the United States from 2005 to the
present. Study designs of the 12 included studies were randomized controlled, non-randomized
comparison, prospective double-blind observational, retrospective cohort, pre-post observational,
and a time-series analysis.
Fluoroscopy. Two studies evaluated the effectiveness of single component initiatives to reduce
radiation exposure during diagnosis of urologic conditions. Several benefits were reported from
implementation of a QA protocol to limit fluoroscopy to 4-5 static images (unless clinically
warranted).25 Significant decreases at the 0.001 level were reported post-implementation for
mean fluoroscopy time (40.9 to 11.7 seconds per procedure), mean dose area product (energy
absorbed across the entire x-ray beam)(518.90 to 150.28 mGy), and mean air kerma (the energy
absorbed by ionizing radiation in a unit mass of air)(15.48 to 4.25 mGy). Increased physician and
staff awareness of radiation safety were also listed as benefits. Lee (2011) indicated that
significant reductions in outcomes did not change the treatment or diagnosis in 100% of the
fluoro urodynamics.
Ngo et al. reported a statistically significant reduction in mean fluoroscopy time (2.74-2.08,
p = 0.002) for unilateral ureteroscopy after physician feedback.20 Baseline data were collected
over a 9-month period. A continuous downward trend in mean fluoroscopy time was reported
over three consecutive years (263 cases) after surgeons received quarterly reports that showed
their mean fluoroscopy time and mean times of their peers. Multivariate analysis indicated that a
surgeons receiving feedback was an independent factor predicting decreased fluoroscopy time
(p = 0.0004).
Two studies evaluated the effectiveness of comprehensive radiation safety programs. In
2010, Greene et al. compared 30 ureteroscopy cases pre- and post-implementation of a QA
protocol. This multicomponent protocol consisted of use of a laser-guided C-arm, use of a
designated fluoroscopy technician, and substitution of visual for fluoroscopic cues during
ureteroscopy. Results included a significant reduction in mean fluoroscopy exposure from 86.1
seconds to 15.5 seconds (p<0.001).26 Greene et al. stated this represents an 82% reduction in
fluoroscopy time and consequently a proportional reduction in radiation exposure.
A comprehensive radiation-reducing program examined by Lakkireddy et al. included
(1) verbal reinforcement of previous fluoroscopy times; (2) effective collimation;
(3) minimizing source-intensifier distance; and (4) effective lead shield use.27 These techniques
were implemented during catheter ablation of atrial fibrillation, a procedure that requires
extensive fluoroscopy time with 15-20% of patients needing a second procedure. Patients were
randomized to either Group I (unexposed to program) or Group II (exposed to program).
Significant improvements were reported in Group II for lower dose area product (234120 vs.
548363 Gy cm2, p = 0.03) and mean patient peak skin dose (0.400.08 vs. 0.120.03 Gy,
p<0.001). Using five cancer deaths/mSv [millisievert] for assessing excess cancer risk, additional
lifetime cancer risk was reported as significantly lower in Group II patients (0.08 vs. 0.2%,
p<0.001).
Computed tomography. Broder et al. examined 93 emergency department (ED) patients who
had abdominal tenderness; 51 (55%) patients had abnormal CT results. Implementation of two
hypothetical z-axis restricted CT-reduced strategies, based on the region of tenderness, resulted
314
in reductions in mean radiation exposure by 70% (Strategy 1; 95% confidence interval [CI] 60%
to 78%) and 38% (Strategy 2: 95% CI 29% to 48%). The primary endpoint was the frequency of
complete inclusion of the acute pathologic region (detected on the complete CT scan) within the
scope of the two hypothetical z axis-restricted CT scans. Current standard practice indicates a CT
scan of the entire abdomen and pelvis. Abdominal pathology was completely included in limited
CTs in 17% to 36% of patients; completely or partially included in 84% to 92% of patients.
However, in 12 cases (eight from Strategy 1), the pathology detected at CT lay completely
outside the marked region of tenderness (see harms below).
Two studies examined use of algorithms to reduce radiation exposure from CT. Loftus et al.
examined use of an imaging algorithm to reduce radiation exposure in 60 patients with
aneurysmal subarachnoid hemorrhage (435 CT examinations).1 This imaging algorithm describes
the most appropriate time points at which to detect vasospasm with CTA and CT perfusion
imaging. Post-implementation results included a 12.1% decrease in cumulative radiation
exposure (p>0.05), a 25.6% reduction in mean number of CT examinations performed per
patient, and a 32.1% decrease in the number of CT perfusion examinations per patient. Stein et
al.29 implemented an imaging algorithm in which stable ED patients with a clinical suspicion of
pulmonary embolism underwent chest radiography followed by V/Q scanning (negative chest
radiograph) or CTPA (positive). Data indicates that when comparing CTPA to V/Q scanning, the
total effective dose from CTPA is almost five times greater; the dose to the female breast 20 to
40 times greater.35,36 After one year, results included a statistically significant 20% reduction in
mean effective dose (8.0 mSv to 6.4 mSv; p<0.0001); a 32% reduction in mean effective dose in
women younger than 40 years. From 2006 to 2007, no significant difference in the false-negative
rate (range, 0.8-1.2%) between CTPA and V/Q scanning occurred and CTPA usage in ED
patients with suspected PE declined from 64.6% to 39.4%.
Two studies evaluated the effectiveness of clinical DS systems to reduce unnecessary CT
imaging.32,33 Sistrom et al.32 reported results after integrating a new DS component to a
computerized ROE system at a large, integrated, multispecialty group practice. Significant
decreases were demonstrated in absolute growth (311 vs. 37; p<0.001) and growth rate (3% vs.
0.25%; p<0.001) of CT exams per quarter from 2004 to 2007. The authors reported that the
number of CT exams was essentially flat despite an increase in outpatient visits by almost
70,000 over the same period. One retrospective cohort study evaluated use of an evidence-based
clinical DS tool to reduce outpatient imaging use rates for several high-volume imaging
procedures.33 Two years after implementation, Blackmore et al. reported data from a single
commercial payer indicating a clinically and statistically significant decrease (-26%) in use of
sinus CT for suspected sinusitis (relative risk [RR], 0.73; 95% CI 0.65 to 0.82; p<0.001).
Secondary analysis indicated that use of the DS tool was also associated with a decrease in
overall volume of sinus CT studies, regardless of diagnosis.
The three remaining studies implemented several radiation reduction measures in cardiac
computed tomography angiography (CCTA);30,31,37 prospective gating was implemented in two
studies. Prospective gating was a core element in initiatives implemented in one study (n = 623)
by Choi et al.30 Results included a statistically significant difference in radiation dose between
the prospective (n = 384) and retrospective (n = 239) gating groups (2.0 vs. 9.6 mSv; p<0.0001).
In addition, median radiation doses per month decreased from 6.2 to 2.1 mSv over time due to
increased usage of prospective gating. One multisite study (n = 449) examined effectiveness of a
standardized BMI [body mass index]-based and heart rate-based protocol. Post-implementation,
LaBounty (2010) reported median radiation dose had decreased from 2.6 mSv (interquartile
315
range 2.0 to 4.2) to 1.3 mSv (interquartile range 0.8 to 1.9) due to use of the standardized
protocol (p<0.001). Statistically significant reductions (p<0.001 level) were also reported for
prospective (versus retrospective) electrocardiographic gating (-82%), reducing tube voltage
(-41% for 100 vs. 120 kV [kilovolts]), lowering tube current (-25% per -100 mA), and reducing
overall scan length (-6% per -1 cm). LaBounty also reported no differences between groups in
the frequency of interpretable studies on a per patient (96.4% vs. 95.5%; p = 0.66) or per artery
(99.1% vs. 98.5%; p = 0.26) basis.31
Lastly, Raff et al. (2009)38 reported improvements from dose reduction strategies from a
consortium of 15 imaging centers (n = 4,862). Radiation reduction measures involved
implementation of a best-practice model including techniques to minimize scan range, heart rate
reduction, electrocardiographic-gated tube current modulation, and reducing tube voltage in
suitable patients. Compared with the control period, patients estimated median radiation dose in
the follow-up period was reduced by 53.3% (dose-length product decreased from 1493 mGy x
cm [IQR 855-1823 mGy x cm] to 697 mGy x cm [IQR, 407-1163 mGy x cm]; p<0.001. A
statistically significant reduction in effective dose was also reported (21 mSv (IQR, 12-26 mSv)
to 10 mSv (IQR, 6-16 mSv) (P<0.001)). No significant changes were reported in median image
quality assessment (control vs. follow-up period) or proportion of diagnostic-quality scans.
Harms. Harms from a PSP were reported in one study when implementation of CT-reduced
strategies resulted in erroneous findings of no pathology in 12 patients.28 Three patients required
emergency treatment resulting in a laparoscopic appendectomy, stent placement, and admittance
for pyelonephritis.

A range of radiation-reduction measures have been successfully implemented by U.S.
institutions to lower risk of deterministic and stochastic injuries. Significant improvements were
reported for imaging time, number of images, and radiation dose (mostly measured by indirect
methods)measures that hypothetically correspond to reduction in patient exposure. Benefits
also included increased physician and staff awareness of radiation safety and no impact on
diagnostic interpretability.
Several studies provided moderately detailed descriptions of implementation but minimal
information on the influence of context on outcomes. Two studies included a discussion of
generalizability. One study described the expansion of radiation reduction measures for other
diagnoses. Two studies described reliance on national and local evidence-based guidelines to
assist in developing decision support systems.
Direct costs were not reported in these studies. However, initiatives were described as
inexpensive, easy to implement, and requiring minimal changes to current workflow. One study
described implementation of a comprehensive QA protocol with simple radiation-reducing
techniques as adding no technical difficulty. A summary table is following (Table 1).
316

Scope of the
Strength of
the PSP
Effectiveness
Rare/High
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Low
How Much do We
References
1.
Loftus ML, Minkowitz S, Tsiouris AJ, et al.

Utilization guidelines for reducing radiation
exposure in the evaluation of aneurysmal
subarachnoid hemorrhage: A practice
quality improvement project. AJR Am J
Roentgenol 2010 Jul;195(1):176-80. PMID:
20566813.
2.
Boone JM, Strauss KJ, Cody DD, et al. Sizespecific dose estimates (SSDE) in pediatric
and adult body CT examinations. College
Park (MD): American Association of
Physicists in Medicine (AAPM); 2011. 28 p.
(AAPM report; no. 204).
www.aapm.org/pubs/reports/RPT_204.pdf.
3.
4.
5.
6.
Committee on Radiation Source Use and

Replacement, National Research Council.
Radiation source use and replacement:
abbreviated version. Washington (DC):
National Academies Press; 2008. 200 p.
www.nap.edu/catalog.php?record_id=11976
.
Koenig TR, Mettler FA, Wagner LK. Skin
injuries from fluoroscopically guided
procedures: part 2, review of 73 cases and
recommendations for minimizing dose
delivered to patient. AJR Am J Roentgenol
2001 Jul;177(1):13-20.
www.ajronline.org/content/177/1/13.full.pdf
. PMID: 11418390.
Wagner LK, McNeese MD, Marx MV, et al.
Severe skin reactions from interventional
fluoroscopy: case report and review of the
literature. Radiology 1999 Dec;213(3):7736.
http://radiology.rsna.org/content/213/3/773.f
ull.pdf. PMID: 10580952.
Nahass GT, Cornelius L. Fluoroscopyinduced radiodermatitis after transjugular
intrahepatic portosystemic shunt. Am J
Gastroenterol 1998 Sep;93(9):1546-9.
PMID: 9732942.
317
7.
Domino D. Two-second CT scan turns into

65-minute ordeal for toddler. [internet]. San
Francisco (CA): Diagnostic Imaging; 2008
Nov 10 [accessed 2012 Feb 2]. [2 p].
www.diagnosticimaging.com/print/article/11
3619/1348813?printable=true.
8.
Berrington de Gonzalez A, Mahesh M, Kim

KP, et al. Projected cancer risks from
computed tomographic scans performed in
the United States in 2007. Arch Intern Med
2009 Dec 14;169(22):2071-7. PMID:
20008689.
9.
Bogdanich W. After stroke scans, patients

face serious health risks. [internet]. New
York (NY): New York Times; 2010 Jul 31
[accessed 2012 Feb 2]. [7 p].
www.nytimes.com/2010/08/01/health/01radi
ation.html?ref=radiation_boom&pagewante
d=print.
10.
Brink JA, Amis ES Jr. Image wisely: a

campaign to increase awareness about adult
radiation protection. Radiology 2010
Dec;257(3):601-2. PMID: 21084410.
11.
Imaging physicians for radiation safety in

adult medical imaging: image wisely.
[internet]. Reston (VA): American College
of Radiology; 2010 [accessed 2012 Feb 2].
[3 p]. www.imagewisely.org/ImagingProfessionals/ImagingPhysicians.aspx?CSRT=1089692503253052
5091.
12.
McCollough CH, Bruesewitz MR, Kofler

JM Jr. CT dose reduction and dose
management tools: overview of available
options. Radiographics 2006 MarApr;26(2):503-12.
http://radiographics.rsna.org/content/26/2/50
3.full.pdf. PMID: 16549613.
13.
14.
15.
H-31 Task Force. Technical white paper:

monitoring and tracking of fluoroscopic
dose. Frankfort (KY): Conference of
Radiation Control Directors, Inc. (CRCPD);
2010 Dec. 15 p. (CRCPD publication; no. E10-7). www.crcpd.org/Pubs/WhitePaperMonitoringAndTrackingFluoroDose-PubE10-7.pdf.
U.S. physician groups identify commonly
used tests or procedures they say are often
not necessary. [internet]. Washington (DC):
ABIM Foundation; 2012 Apr 4 [accessed
2012 Jul 10]. [3 p].
http://choosingwisely.org/wpcontent/uploads/2012/03/033012_ChoosingWisely-National-Press-Rls-FINAL.pdf.
ACR identifies list of commonly used - but
not always necessary - imaging exams as
part of Choosing Wisely campaign.
[internet]. Reston (VA): American College
of Radiology (ACR); 2012 Apr 4 [accessed
2012 Jul 10]. [3 p]. www.acr.org/AboutUs/Media-Center/Press-Releases/2012Press-Releases/20120404-ACR-IdentifiesList-of-Commonly-Used-Imaging-Examsas-Part-of-Choosing-Wisely-Campaign.
16.
Amis ES Jr, Butler PF, Applegate KE, et al.

American College of Radiology white paper
on radiation dose in medicine. J Am Coll
Radiol 2007 May;4(5):272-84. PMID:
17467608.
17.
DIR (dose index registry). [internet]. Reston

(VA): American College of Radiology;
[accessed 2012 Jan 17]. [2 p].
https://nrdr.acr.org/Portal/DIR/Main/page.as
px.
18.
19.
Francis JJ, Stockton C, Eccles MP, et al.

Evidence-based selection of theories for
designing behaviour change interventions:
using methods based on theoretical construct
domains to understand clinicians blood
transfusion behaviour. Br J Health Psychol
2009 Nov;14(Pt 4):625-46. PMID:
19159506.
Wilson K, MacDougall L, Fergusson D, et
al. The effectiveness of interventions to
reduce physicians levels of inappropriate
transfusion: what can be learned from a
systematic review of the literature.
Transfusion 2002 Sep;42(9):1224-9. PMID:
12430683.
318
20.
Ngo TC, Macleod LC, Rosenstein DI, et al.

Tracking intraoperative fluoroscopy
utilization reduces radiation exposure during
ureteroscopy. J Endourol 2011
May;25(5):763-7. PMID: 21388241.
21.
Fefferman NR, Roche KJ, Pinkney LP, et al.

Suspected appendicitis in children: focused
CT technique for evaluation. Radiology
2001 Sep;220(3):691-5.
22.
Gamanagatti S, Vashisht S, Kapoor A, et al.

Comparison of graded compression
ultrasonography and unenhanced spiral
computed tomography in the diagnosis of
acute appendicitis. Singapore Med J 2007
Jan;48(1):80-7.
www.sma.org.sg/smj/4801/4801a12.pdf.
PMID: 17245521.
23.
Rao PM, Rhea JT, Novelline RA, et al.

Helical CT technique for the diagnosis of
appendicitis: prospective evaluation of a
focused appendix CT examination.
Radiology 1997 Jan;202(1):139-44. PMID:
8988203.
24.
Gupta AK, Nelson RC, Johnson GA, et al.

Optimization of eight-element multidetector row helical CT technology for
evaluation of the abdomen. Radiology 2003
Jun;227(3):739-45.
25.
Lee CL, Wunderle K, Vasavada SP, et al.

Reduction of radiation during fluoroscopic
urodynamics: analysis of quality assurance
protocol limiting fluoroscopic images during
fluoroscopic urodynamic studies. Urology
2011 Sep;78(3):540-3. PMID: 21762969.
26.
Greene DJ, Tenggadjaja CF, Bowman RJ, et

al. Comparison of a reduced radiation
fluoroscopy protocol to conventional
fluoroscopy during uncomplicated
ureteroscopy. Urology 2011 Aug;78(2):28690. PMID: 21256553.
27.
Lakkireddy D, Nadzam G, Verma A, et al.

Impact of a comprehensive safety program
on radiation exposure during catheter
ablation of atrial fibrillation: a prospective
study. J Interv Card Electrophysiol 2009
Mar;24(2):105-12. PMID: 18987964.
28.
Broder JS, Hollingsworth CL, Miller CM, et

al. Prospective double-blinded study of
abdominal-pelvic computed tomography
guided by the region of tenderness:
estimation of detection of acute pathology
and radiation exposure reduction. Ann
Emerg Med 2010 Aug;56(2):126-34. PMID:
20074835.
29.
Stein EG, Haramati LB, Chamarthy M, et al.

Success of a safe and simple algorithm to
reduce use of CT pulmonary angiography in
the emergency department. AJR Am J
Roentgenol 2010 Feb;194(2):392-7. PMID:
20093601.
30.
Choi TY, Malpeso J, Li D, et al. Radiation

dose reduction with increasing utilization of
prospective gating in 64-multidetector
cardiac computed tomography angiography.
J Cardiovasc Comput Tomogr 2011 JulAug;5(4):264-70. PMID: 21723518.
31.
LaBounty TM, Earls JP, Leipsic J, et al.

Effect of a standardized qualityimprovement protocol on radiation dose in
coronary computed tomographic
angiography. Am J Cardiol 2010 Dec
1;106(11):1663-7. PMID: 21094371.
32.
Sistrom CL, Dang PA, Weilburg JB, et al.

Effect of computerized order entry with
integrated decision support on the growth of
outpatient procedure volumes: seven-year
time series analysis. Radiology 2009
Apr;251(1):147-55. PMID: 19221058.
33.
Blackmore CC, Mecklenburg RS, Kaplan

GS. Effectiveness of clinical decision
support in controlling inappropriate
imaging. J Am Coll Radiol 2011
Jan;8(1):19-25. PMID: 21211760.
34.
Bush RW. Reducing waste in US health care

systems. JAMA 2007 Feb 28;297(8):871-4.
PMID: 17327529.
35.
Einstein AJ, Henzlova MJ, Rajagopalan S.

Estimating risk of cancer associated with
radiation exposure from 64-slice computed
tomography coronary angiography. JAMA
2007 Jul 18;298(3):317-23. PMID:
17635892.
36.
Mettler FA Jr, Huda W, Yoshizumi TT, et

al. Effective doses in radiology and
diagnostic nuclear medicine: a catalog.
Radiology 2008 Jul;248(1):254-63.
319
37.
Raff GL, Chinnaiyan KM, Share DA, et al.

Radiation dose from cardiac computed
tomography before and after implementation
of radiation dose-reduction techniques.
JAMA 2009 Jun 10;301(22):2340-8. PMID:
19509381.
38.
ICS urodynamics course recognition

guidelines. Bristol (UK): International
Continence Society (ICS); 2003 Jun 1. 4 p.
Chapter 30. Ensuring Documentation of Patients

Preferences for Life-Sustaining Treatment: Brief Update
Review
Sydney Dy, M.D., M.Sc.
Introduction
Numerous studies have documented that the care patients receive at the end of life is often
not consistent with their preferences. In addition, communication about end-of-life issues is
suboptimal, and advance directive completion and documentation of health care proxies and
preferences for life-sustaining treatment often does not occur. For example, a study of advanced
cancer patients who died in one hospital and cancer center1 found that only 19 percent of patients
had documentation of an advance directive or a surrogate decisionmaker in the medical record.
Furthermore, only one of the 17 patients who received mechanical ventilation had documentation
of preferences regarding mechanical ventilation or documentation of why this information was
unavailable. Other research has found that a majority of physicians whose patients had advance
directives were not aware of them, that having an advance directive did not increase medical
record documentation of patient preferences, and that advance directives were often not used in
medical care.2
These gaps in quality can have several consequences. They can lead to patients receiving
care that is not consistent with their preferences: lack of appropriate documentation or other
miscommunication about the appropriate application of Do Not Resuscitate (DNR) orders or a
patients desire to not receive aggressive care at the end of life can lead to errors of providing
invasive treatments, such as intubation and resuscitation, that are inconsistent with patient
preferences and can lead to significant patient and family suffering. Poor documentation or
communication about these preferences can also lead to confusion among staff,
miscommunication with families, and errors in code situations. Significant harm may result if a
patients health care proxy is not documented or not followed, or if end-of-life decisions are
made with a surrogate who was not the patients surrogate of choice.
The original Making Health Care Safer report reviewed interventions aimed at increasing
individuals communication of their preferences for end-of-life care, through completion of
either advance directives or health care powers of attorney. This review found that although
policies to facilitate and increase rates of completion of such instruments completed are
widespread, evidence was lacking that these instruments actually improve compliance with
patients end-of-life care wishes. Based on a recent systematic review that we conducted on the
impact of quality improvement interventions on end-of-life care, this chapter updates the review
of interventions aimed at improving completion of advance directives, as well as interventions to
increase general communication about preferences and care at the end of life, and examines
evidence that communication or advance directives increase the likelihood that patients end-oflife care preferences are followed.3
320
What Is the Practice of Ensuring Documentation of Life-Sustaining

Treatment?
Ensuring documentation of preferences regarding life-sustaining treatment is included as one
element of the 2010 Update of the National Quality Forums 34 Safe Practices for Better
Healthcare. This patient safety practice (PSP) is described as follows4:
Ensure that written documentation of the patients preferences for
life-sustaining treatments is prominently displayed in his or her
chartOrganization policies, consistent with applicable law and
regulation, should be in place and address patient preferences for
life-sustaining treatment and withholding resuscitation. The
definition of life-sustaining treatment may include, but is not
limited to, mechanical ventilation, renal dialysis, chemotherapy,
antibiotics, and artificial nutrition and hydration.
As conceptualized by the National Quality Forum and Joint Commission, this PSP focuses on
encouraging advance directive completion and documentation in patients medical records.
Advance directives can document patients wishes about life-sustaining treatment (the living
will), choice of a surrogate decisionmaker (the durable power of attorney), or both. Advance
directives have many drawbacks, including frequent lack of applicability to decisions about lifesustaining treatment until patients are incapacitated and close to death, imprecise language, and
lack of translation into medical care, such as DNR orders. Therefore, this practice has become
more broadly conceptualized as advance care planning, defined as a process of communication
between a patient, family/health care proxy, and health care providers for the purpose of
identifying a surrogate decision-maker, clarifying treatment preferences, and developing
individualized goals of care about life-sustaining and other aggressive treatments. It can also
include ensuring appropriate communication about and completion of Do Not Resuscitate orders
when consistent with patients preferences; documentation practices to ensure that these orders
are followed when present; and newer forms of documentation to ensure that preferences are
honored across settings, such as POLST (Physician Orders for Life-Sustaining Treatment) and
similar programs (described below).

What Contexts?
A variety of interventions intended to improve end-of-life care of include improving care
planning as part of the intervention, such as palliative care services in hospitals and case
management. Descriptions of implementation and effectiveness of key policy initiatives are
described below.
In the United States, the Patient Self-Determination Act, passed by Congress in 1990,
requires that patients are given written notice upon admission to a variety of health care
institutions of their decision-making rights and policies regarding advance health care directives.
Although it is unclear if implementation of this legislation improved rates of advance directive
completion or elicitation of patient preferences, several expanding programs are improving
implementation of advance care planning at the community or state level. A program to promote
and implement advance care planning processes across an entire community, The Respecting
Choices program in Lacrosse, Wisconsin, (http://respectingchoices.org/),5 has achieved nearly
321
universal advance directive completion in the community, although no rigorous research to

document the impact of this initiative on end-of-life care in the community was identified in our
systematic review.3.
POLST (Physician Orders for Life-Sustaining Treatment) (www.ohsu.edu/polst/)6 is an
initiative to document patients preferences on an order form that is accepted across all settings
of care. Similar initiatives have been endorsed in at least 10 U.S. States. In our systematic
review, we did not find any evaluations of POLST or similar initiatives on the hospital or
community level that met our inclusion criteria of enrolling a comparison group.3
The Gold Standards Framework (www.goldstandardsframework.org.uk)7 is a systematic
approach for improving care at the end of life that has been widely implemented in primary care
practices, nursing homes, hospitals, and other settings throughout the United Kingdom. This
approach includes goals of increasing the percentage of patients with advance care planning
discussions and increasing concordance between patient preferences and care at the end of life.
In our systematic review,3 we identified one evaluation with patient-centered outcomes, a 2009
non-randomized trial8 addressing symptoms, needs, and coordination in 49 nursing homes, that
found statistically significant reductions in nursing home deaths and in crisis hospitalizations.
The Liverpool Care Pathway (www.liv.ac.uk/mcpcil/liverpool-care-pathway/)9 is a template
for structuring care at the end of life, including communication and documentation about which
treatments are appropriate. Programs based on this template have been implemented in a variety
of settings and countries, including the United Kingdom, Australia, and the Netherlands. In our
systematic review,3 we identified one evaluation, a 2010 non-randomized trial, which evaluated
the Liverpool Care Pathway in a variety of settings, including hospitals, nursing homes,
residential facilities, and home care in the Netherlands. The intervention did not statistically
significantly increase use of do-not resuscitate orders at the end of life, possibly because sample
sizes were small.10
What Have We Learned About Documenting Life-Sustaining

Treatments?
The issue of documentation of life-sustaining treatment was addressed in the 2001 Making
Health Care Safer report, as Advance Planning for End-of-Life Care. That review discussed early
studies on the POLST and Respecting Choices initiatives (described above) as well as several
studies of improving rates of advance directive completion.
More recent systematic reviews have found that interventions can significantly increase rates
of advance directive completion. However, these reviews have found few studies on the actual
impact of documentation of preferences for life-sustaining treatment such as advance directives
on patient outcomes, and have concluded that the few studies that have been conducted have not
generally found a significant impact.11,12
In a systematic review of the literature from 2000 through March 2011, we addressed quality
improvement interventions for end-of-life care, including the target of communication.3 We
included prospective studies with a comparison group, enrolling patients with serious or
advanced illness who were unlikely to recover or be cured, and assessing patient-centered
outcomes, including quality of care and health care utilization.
For the target of communication, we did not identify any studies addressing this practice
specifically (i.e., the impact of increasing or improving documentation of life-sustaining
treatment preferences on improved outcomes or decreased errors). However, we did identify a
number of studies focusing on increasing and improving end-of-life communication more
322
generally, primarily in the intensive care unit setting. These studies included quality
improvement interventions to increase the frequency and/or structure of family meetings to
address these issues, and the use of palliative care and ethics consultants. Fifteen studies (6
RCTs, 9 non-RCTs) evaluated health care utilization, such as intensive care unit length of stay,
as an outcome. We found moderate strength of evidence to support the impact of interventions
on this outcome: 73% of studies found a statistically significant improvement in the intervention
compared with the control group.
Observational studies have shown that patients with advance directives or related documents
are more likely to receive care consistent with their preferences. A retrospective evaluation of the
care provided for residents with POLST orders found that care provided was consistent with
residents preferences 98% of the time for resuscitation and 94% overall.13 Another retrospective
analysis of survey data found that patients who died with advance directives received care
strongly aligned with their preferences (97% of those who requested comfort care received it).
Subjects with living wills or who had a health care proxy were statistically significantly less
likely to receive aggressive care than those without advance directives.14 Patients who had
discussions about prognosis and end-of-life care with their providers also had care at the end of
life that was less invasive and more consistent with their preferences.

Achieving concordance between patients end-of-life care preferences and the care that they
receive is an accepted goal in health care, and improving communication about and
documentation of patients preferences is important. Errors such as resuscitation in a patient who
wanted comfort care because the correct documentation had not been completed can cause
significant harm and suffering for patients and families.
Recent systematic reviews have found moderate strength of evidence that interventions can
improve rates of advance directive completion and that interventions to improve end-of-life
communication can reduce health care utilization, which may be a marker for overly aggressive
care, at the end of life. However, insufficient evidence exists to support whether advance
directives or current policy initiatives to improve documentation of care preferences across
settings, such as POLST in the United States, improve the likelihood that patients receive care
consistent with their preferences at the end of life. Interventions to improve communication
about end-of-life care issues should be implemented in hospitals, particularly in the intensive
care unit settings. Emerging types of interventions, such as initiatives present in some hospitals
for requirements for code status or health care proxy documentation for all patients upon
admission or other initiatives to improve documentation of care preferences, deserve further
evaluation. A summary table is following (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low
323
How Much do We
Moderate/Moderate
References
1.
Dy SM, Lorenz KA, ONeill SM et al.

Cancer Quality-ASSIST supportive
oncology quality indicator set: feasibility,
reliability, and validity testing. Cancer 2010;
116(13):3267-75.
8.
Mazzocato C, Michel-Nemitz J, Anwar D,

Michel P. The last days of dying stroke
patients referred to a palliative care consult
team in an acute hospital. Eur J Neurol
2010; 17(1):73-7.
2.
Kass-Bartelmes BL, Hughes R. Advance

care planning: preferences for care at the
end of life. J Pain Palliat Care Pharmacother
2004; 18(1):87-109.
9.
3.
Dy, SM, R Aslakson, RF Wilson, et al.

Improving Health Care and Palliative Care
for Advanced and Serious Illness. Closing
the Quality Gap: Revisiting the State of the
Science. Evidence Report No. 208.
(Prepared by the Johns Hopkins University
Contract No. 290-2007-10061-I.) AHRQ
Publication No. 12(13)-E014-EF. Rockville,
Quality. October 2012.
The Marie Curie Palliative Care Institute.

Liverpool Care Pathway for the Dying
Patient (LCP).
www.liv.ac.uk/mcpcil/liverpool-carepathway/.
10.
van der Heide A, Veerbeek L, Swart S, van

der Rijt C, van der Maas PJ, van Zuylen L.
End-of-life decision making for cancer
patients in different clinical settings and the
impact of the LCP. J Pain Symptom Manage
2010; 39(1):33-43.
11.
Lorenz KA, Lynn J, Dy SM et al. Evidence

for improving palliative care at the end of
life: a systematic review. Ann Intern Med
2008; 148(2):147-59.
12.
Walling A, Lorenz KA, Dy SM et al.

Evidence-based recommendations for
information and care planning in cancer
care. J Clin Oncol 2008; 26(23):3896-902.
13.
Hickman SE, Nelson CA, Moss AH, Tolle

SW, Perrin NA, Hammes BJ. The
consistency between treatments provided to
nursing facility residents and orders on the
physician orders for life-sustaining treatment
form. J Am Geriatr Soc 2011; 59(11):20919.
14.
Silveira MJ, Kim SY, Langa KM. Advance

directives and outcomes of surrogate
decision making before death. N Engl J Med
2010; 362(13):1211-8.
4.

Better Healthcare 2010 Update: A
Consensus Report.
www.qualityforum.org/Projects/Safe_Practi
ces_2010.aspx.
5.
Gundersen Lutheran Medical Center.

Respecting Choices. 2011.
http://respectingchoices.org/.
6.
Oregon Health Sciences University.

Pysician Orders for Life-sustaining
Treatement Paradigms (POLST). 2008.
www.ohsu.edu/polst/.
7.
National Gold Standards Framework Centre.

www.goldstandardsframework.org.uk/.
324
Part 2b. Practices Designed To Improve Overall

System/Multiple Targets
Chapter 31. Human Factors and Ergonomics
Pascale Carayon, Ph.D.; Anping Xie, M.S.; Sarah Kianfar, M.S.

Many patient safety incidents are related to lack of attention to human factors and
ergonomics (HFE) in the design and implementation of technologies, processes, workflows, jobs,
teams and sociotechnical systems. For instance, a systems analysis of medication errors1
identified a range of proximal causes of medication errors, such as rule violations, memory slips
and lapses, poor communication with other services, and incorrect pump programming caused by
poor design of the pump interface. Lack of attention to HFE in areas such as technology design
can contribute to medication errors and preventable adverse drug events.
The Institute of Medicine (IOM) report on Medication Errors2 emphasizes the need for
addressing HFE issues, such as the design of medication labels and packages, and the design of
medication administration technologies (e.g., infusion pump). A study by Han et al.3 in a
pediatric hospital showed an increase in mortality rates after the implementation of
Computerized Provider Order Entry (CPOE); many factors that contributed to the increase were
related to HFE. For instance, the design of the CPOE interface required about 10 clicks per
order, thus significantly increasing time needed to enter orders. The poor usability of the CPOE
system and its lack of integration with clinician workflow contributed to delays in patient care
that were a major factor in the increased mortality rate after CPOE implementation.
The recent IOM report on Health IT and Patient Safety4 clearly indicates the need for HFE in
the design, implementation and use of health IT. The report proposes a sociotechnical approach
that emphasizes the need for health IT to support clinical workflows. Increased cognitive
workload associated with the implementation and use of health IT and lack of usability of health
IT are two HFE issues associated with patient safety incidents highlighted in the report.
A systematic review showed how environmental hazards can contribute to patient safety
incidents such as patient falls.5 For instance, the use of bedrails can contribute to patient falls by
contributing to entrapment injuries.
These studies provide evidence for the importance of HFE in patient safety; they highlight
the range of physical, cognitive and organizational HFE issues that can contribute to patient
safety incidents. There are many other examples of how lack of attention to HFE contributes to
patient safety incidents. For instance, a fatal medication error occurred on July 5, 2006 at St.
Marys hospital in Madison, WI: An epidural penicillin solution instead of an intravenous (IV)
penicillin retain was administered to a 16-year old pregnant patients IV line, causing her
immediate death. A root cause analysis identified several HFE issues that contributed to the fatal
error6. The IV and epidural bags had similar designs, and both medication bags could be
connected to IV and epidural tubing. A barcoded medication administration (BCMA) technology
had been recently introduced in the hospital, but the nurse did not use it. Because of the
technologys poor usability and lack of training (i.e., HFE issues), many nurses did not use the
technology and thus could not take advantage of its safety features.
Vincent et al.7 describe three groups of factors for explaining adverse surgical outcomes:
325
1. patient risk factors (e.g., increased body mass index, presence of comorbidity)
2. surgical skills (e.g., technical skills), and
3. operation profile (or system factors).
The operation profile includes a range of HFE-related system factors, such as operative
environment, team performance and communication, and decisionmaking processes. System
characteristics are factors that, in addition to patient characteristics and the skills of the surgery
team, can contribute to complications and adverse events7. A range of system factors can
influence the safety of surgery and can be addressed by using concepts, models, theories and
methods from HFE. Vincents approach can be extended to patient safety incidents in other care
settings besides surgery.8 Patient characteristics and clinician skills and knowledge are important
for patient safety; but poor system design can also contribute to patient safety incidents. HFE
helps to identify system design deficiencies and hazards that affect patient safety and provides
the concepts and methods to improve system design and, therefore, patient safety.

According to the International Ergonomics Association (IEA),9 Ergonomics (or human
factors) is the scientific discipline concerned with the understanding of the interactions among
humans and other elements of a system, and the profession that applies theoretical principles,
data and methods to design in order to optimize human well-being and overall system
performance.
Key Characteristics of Human Factors and Ergonomics

Human Factors and Ergonomics are synonymous names for the discipline; the discipline
is often referred to as Human Factors and Ergonomics or HFE. HFE covers a wide range of
physical, cognitive, and organizational issues involved in system design. Physical HFE issues
include physical dimensions of tools that do not fit physical characteristics of users (e.g., too
small font size on computer screen), inappropriately designed physical environments (e.g.,
lighting too bright and creating glare, noisy and distracting environment) and physical layout that
does not support clinician work (e.g., monitoring patients from the central nursing station).
Cognitive HFE issues include interactions between people and the rest of the system such as
perception, memory, attention, mental workload, and support for decisionmaking. At the
organizational level, HFE focuses on communication and coordination, teamwork, job design,
sociotechnical system, and system design, and change (e.g., participatory ergonomics). Other
examples of physical, cognitive and organizational (macroergonomic or sociotechnical) HFE
issues of relevance to patient safety can be found in textbooks and papers.10-12
Rather than attempting to fit the person to the system, HFE works to fit the system to the
person.13 Systems should be designed to accommodate the range of characteristics, needs, and
limitations of people. In this context, people means single individuals, teams, or larger
organizational units. According to the IEA definition, the objective of HFE-based system design
is to improve both human well-being and overall system performance. Patient safety can be
considered one aspect of overall system performance. From an HFE viewpoint, patient safety
activities should not only reduce medical errors and improve patient safety, but also improve
human well-being, such as job satisfaction, motivation and acceptance of technology. For
instance, patient safety programs that increase the workload of already busy clinicians would not
326
be considered well designed from the HFE perspective. See Table 1 for a summary of the key
characteristics of HFE.
Table 1, Chapter 31. Key characteristics of HFE and its application to patient safety
Definition of HFE by the
International Ergonomics
Association (www.iea.cc)
Name of the discipline
Range of HFE issues
Goal of HFE
Objectives of HFE
Ergonomics (or human factors) is the scientific discipline concerned with the
understanding of the interactions among humans and other elements of a system,
and the profession that applies theoretical principles, data and methods to design in
order to optimize human well-being and overall system performance.
Human Factors and Ergonomics are two different names for the same discipline.
Physical, cognitive and organizational (macroergonomic or sociotechnical) issues of
HFE are all relevant to patient safety.
The goal of HFE is to fit the system to the people instead of fitting people to the
system.
The objective of HFE-based system design is to improve both well-being and
system performance. Patient safety is one component of system performance.
Human Factors and Ergonomics Applications to Patient Safety

HFE contributes to five domains of patient safety: (1) usability of medical devices and health
information technology, (2) focus on human error and its role in patient safety, (3) role of health
care worker performance in patient safety, (4) system resilience and its role in patient safety, and
(5) HFE systems approaches to patient safety.
Usability of Medical Devices and Health IT

A significant focus of HFE in health care and patient safety has been the design of usable
medical devices and health IT.14,15 For instance, to improve medication management in medical
emergencies, HFE principles were used to redesign the code cart medication drawer16. User
testing was conducted to compare the medication retrieval time and number of wasteful actions
associated with the existing and prototype drawers. Compared with the existing drawer, the
prototype drawer resulted in shorter medication retrieval time and fewer wasteful actions. The
prototype drawer also received higher ratings for visibility, organization, and general usability. A
detailed example of the application of usability methods for improving safety of radiotherapy
treatment delivery is provided in the section on What are the beneficial effects of the Patient
Safety Practice?
Health IT can contribute to patient safety by eliminating hazards.17,18 However, it may also
create new hazards.19-21 Usability is one HFE design characteristic that can influence health ITs
patient safety benefits, or lack thereof.4 HFE methods have been used to improve the usability of
CPOE order sets,22 to design the user interface of a software application that was developed to
extract and present data relevant to the treatment of critically ill patients to providers,23,24 and to
improve the design of medication alerts.25,26 The second example in the section on What are the
beneficial effects of the Patient Safety Practice? provides information on the usability
evaluation of CPOE technology.
Human Error and Patient Safety

Another major focus of HFE in patient safety has been understanding the nature of human
error and identifying the mechanism of human error involved in patient safety.8,27 This probably
represents the largest contribution of HFE to patient safety.28 The Swiss Cheese model of
Reason29,30 describes the alignment of hazards (or holes) that can lead to an accident(e.g., a
patient safety event) and distinguishes between latent failures and active failures. Latent failures
327
result from decisions made by system designers and organizational decision-makers that lead to
unsafe conditions. Active failures are errors made by the operators of the system.
Vincent and colleagues8 have adapted Reasons Swiss Cheese model to patient safety. They
describe management decisions and latent failures that can influence error and create conditions
that produce safety violations. In turn, these conditions create problems for care delivery and
may lead to unsafe acts (i.e., errors and violations), which may then produce an incident if the
defenses and barriers are not appropriate. Reasons Swiss Cheese model and its patient safety
version by Vincent and colleagues include both errors and violations as active failures. Recent
HFE research has broadened the focus on human error and developed knowledge about the
contribution of violations to patient safety.31,32
Bogner33 has proposed another HFE model of human error and patient safety: the Artichoke
model defines layers of system factors that influence provider-patient interactions, such as legalregulatory-reimbursement, national culture, organization, physical environment, social
environment, and ambient environment. The frameworks proposed by Vincent and colleagues8
and Bogner33 can be used by health care organizations to investigate specific patient safety
incidents.
Health Care Worker Performance

Performance obstacles may endanger patients by making it difficult for clinicians to perform
tasks and procedures safely.34 A range of physical (e.g., lifting, injecting, charting), cognitive
(e.g., perceiving, attention, communicating, awareness) and social/behavioral (e.g., motivation,
decision-making) performance processes can influence patient safety.35 When obstacles are
present in the work environment, physical, cognitive and social/behavioral performance of
clinicians may be challenged and accidents may occur. Performance obstacles have been
identified for ICU nurses,36-38 staff in outpatient surgery centers,34 and hospital nurses.39
Information on performance obstacles can be used to improve working conditions of health care
professionals; these changes may produce patient safety benefits.40 When faced with
performance obstacles, clinicians have to improvise ways of getting their work done. HFE
research has characterized such work-arounds and their patient safety implications in nursing
medication administration,32 especially in the context of BCMA use.41,42 HFE has proposed a
range of approaches, including work teams and team training, to enhance health care worker
performance, and improve communication, coordination, and information flow.
System Resilience
Recently, HFE research in patient safety has focused on system resilience.46 Resilience has
been defined as the ability of systems to anticipate and adapt to the potential for surprise and
failure.47 Because not all errors may be prevented, HFE researchers have developed models to
understand how errors can be detected, corrected, mitigated, and dealt with by operators.48,49 The
WHO model of patient safety incorporates the concepts of error detection and mitigation.50,51
Strategies for error detection and recovery have been explored among nurses,52 in particular
critical care nurses,53,54 and among pharmacists.49,55 This line of HFE research can produce
information about mechanisms for achieving resilience, such as cross-checking.56,57 Resilience
engineering builds on and extends the work done by High Reliability Organization (HRO)
researchers. A key characteristic of HROs is mindfulness, i.e., the ability to prepare for the
unexpected and to be vigilant about hazards, and one aspect of mindfulness is organizational
commitment to resilience.58
328
Four factors contribute to resilience59:

1. knowing how to respond to disruptions and disturbances
2. monitoring events, in particular those likely to lead to an accident
3. anticipating developments, threats and opportunities
4. learning from patient safety incidents.
Further research is necessary to understand how these factors contribute to resilient
performance; this research should focus on understanding the role of distributed cognition (i.e.,
the distribution of knowledge across the social and physical environments as well as across time)
and situation awareness in demanding situations and the ways that clinicians react and deal with
surprising, demanding situations and other vulnerabilities or hazards.60
Human Factors and Ergonomics Systems Approaches to Patient Safety

The first four HFE approaches focus on specific aspects of HFE and patient safety--usability
of technology, human error, clinician performance, and resilience. A number of HFE approaches
have been proposed to describe more comprehensive systems of patient care. These systems
approaches address the following: (1) a broad range of system variables that can affect patient
safety, (2) interactions between system elements, and (3) interacting system levels.28,61 These
approaches include the systems approach proposed by Vincent and colleagues8,62 and the SEIPS
(Systems Engineering Initiative for Patient Safety) model of work system and patient safety
proposed by Carayon and colleagues.63
Vincent and colleagues8 defined seven types of system factors that can influence clinical
practice and lead to patient safety incidents:
1. patient factors
2. task and technology factors
3. individual (staff) factors
4. team factors
5. work environmental factors
6. organizational and management factors
7. institutional context factors.
This framework can be used to identify factors that contribute to patient safety incidents.
The SEIPS model of work system and patient safety63 (Figure 1) identifies a slightly different
set of system factors: (1) individual factors (which include characteristics of the staff and
patient), (2) tasks, (3) tools and technologies, (4) environment, and (5) organizational factors
(which include team factors). In addition to defining the system and emphasizing system
interactions,64 the SEIPS model describes how system design can influence care processes and
other connected processes (e.g., delivery of supplies, housekeeping, purchasing of medical
equipment). Based on the Structure-Process-Outcome framework of Donabedian,65 the SEIPS
model proposes that system design can contribute to deficiencies in care processes and thus to
patient safety incidents. Because the SEIPS model is anchored in HFE, employee and
organizational outcomes are addressed along with patient safety, reflecting the fact that patient
safety and worker safety and well-being are positively correlated and have common system
contributing factors.66
329
62
Figure 1, Chapter 31. SEIPS model of work system and patient safety
Reproduced from Work system design for patient safety: the SEIPS model. Carayon P, Schoofs Hundt A, Karsh BT, et al.
Quality & Safety in Health Care. 15(Supplement I):i50-i58. 2006 with permission from BMJ Publishing Group Ltd.
With recent emphasis on the role of health IT in patient safety, sociotechnical systems
approaches have been proposed, for instance, by the IOM report on Health IT and Patient
Safety.4 The work system of the SEIPS framework is a representation of a sociotechnical system
(the technology is part of a larger system and interacts with various system elements). The
sociotechnical system model proposed by the IOM4 includes all elements of the work system
model, except for the physical environment.
Human Factors and Ergonomics as a Patient Safety Practice

HFE as a patient safety practice can take three different forms:67
1. using HFE tools and methods,
2. increasing HFE knowledge, and
3. recruiting HFE engineers.
HFE tools and methods for patient safety include usability evaluation of technologies or
devices, work system assessment for performance obstacles and hazards, and risk assessment of
care processes. Other examples of HFE tools and methods are described in the section on the
beneficial impacts of HFE as a patient safety practice.
Increasing HFE knowledge may involve training and educating a range of health care
professionals and workers, including patient safety officers, quality improvement specialists, and
330
health IT staff, as well as leaders of health care organizations, policymakers, and vendors and
manufacturers of medical devices and health IT applications.
Health care organizations may hire HFE engineers in order to accelerate adoption and
dissemination of HFE. Integration of HFE engineers in health care organizations may enhance
the impact of HFE in a wide range of relevant departments and functions of health care
organizations,67,68 such as patient safety, risk management, quality improvement, employee
health, and health IT implementation and optimization.

Theories and models underlying the HFE approach to improving patient safety target the
design of work systems and care processes and aim to promote and facilitate performance of all
individuals involved. HFE focuses on how to design work systems and processes for supporting
safe behaviors and activities in both system design and operation. According to the SEIPS model
of work system and patient safety,63 HFE deficiencies in the design of work systems can
negatively influence the safe delivery of care processes, and therefore, lead to patient safety
incidents (see Figure 1).
HFE for patient safety is based on four mechanisms that connect system variables to patient
safety69 (see Table 2).
Table 2, Chapter 31. HFE mechanisms between system design and patient safety
1.
2.
3.
4.
HFE Mechanisms
A work system that is not designed according to
HFE design principles can create opportunities for
errors and hazards (see Table 3 for examples of
design principles).
Performance obstacles that exist in the work system
can hinder clinicians ability to perform their work
and deliver safe care.
A work system that does not support resilience can

produce circumstances where system operators may
not be able to detect, adapt to, and/or recover from
errors, hazards, disruptions and disturbances.
Because system components interact to influence
care processes and patient safety, HFE system
design cannot focus on one element of work in
63,69,77
isolation.
Objectives of System Design

The objective of HFE-informed system design is to
identify and remove system hazards from the design
through maintenance phases.
The objective of HFE system redesign is to identify and
remove performance obstacles.
If some obstacles cannot be removed, for instance,
because they are intrinsic to the job, then strategies
should be designed to mitigate the impact of
performance obstacles by enhancing other system
70-72
elements (i.e. Balance Theory of Job Design).
Work systems should be designed to enhance resilience
73
and support adaptability and flexibility in human work,
such as allowing problem or variance control at the
74-76
source.
Whenever there is a change in the work system, one
needs to consider how the change will affect the entire
work system, and the entire system needs to be
70-72
optimized or balanced.
Human Factors and Ergonomics Design Principles

A range of HFE design principles have been proposed for optimizing specific elements or
aspects of the work system. These principles can be used to design work systems to eliminate
hazards and performance obstacles. For instance, The Handbook of Human Factors in Medical
Device Design78 provides a comprehensive set of principles for medical device design. Usability
heuristics or rules of thumb for user interface design have been developed for health IT and
medical devices;79-81 these usability heuristics include consistency, a match between technology
and the users mental model, minimizing memory load, and users in control. The physical design
331
of the work system should minimize perception time, decision time, manipulation time, and the
need for excessive physical exertion, and optimize opportunities for physical movement.69,82,83
From an organizational HFE viewpoint, work systems should be designed so that tasks are
reasonably demanding physically and cognitively. Workers should have opportunities to learn,
adaptive levels of control over their work system, and access to social and instrumental support
(e.g., support from co-workers in case of emergency) within the work environment.84,85 Table 3
provides some examples of HFE design principles; additional information on HFE design for
specific work system elements can be found in the Handbook of Human Factors and
Ergonomics.86
Table 3, Chapter 31. Examples of HFE design principles
Focus of HFE
Physical HFE
Cognitive HFE
Organizational HFE
Examples of HFE Design Principles

Minimizing perception time, decision time, and manipulation time
Reducing or mitigating need for excessive physical exertion
Optimizing opportunities for physical movement
Consistency of interface design
Match between technology and the users mental model
Minimizing cognitive load
Allowing for error detection and recovery
Feedback to users
Worker opportunities to learn and develop new skills
Worker control over work system
Worker access to social support
Participation in system design
Given the systems focus of HFE, it is important not only that each component of the system
be designed appropriately, but also that system components be aligned75 and that system
interactions be optimized.64 For example, when a new BCMA system is introduced, it is
important to ensure that the technology is designed according to HFE principles (e.g., usability
heuristics). However, it is also important that the technology fits with the rest of the work
system. If there is not sufficient space in which to use the BCMA (interaction between the
technology and the physical environment) or if users are not provided with adequate training
(interaction between the technology and the organization), then BCMA may contribute to
diminished rather than improved clinician performance and patient safety.
The goal of HFE-informed design is work system that supports the work of individual and
teams.75,81 This is the essence of the user-centered design approach.87
HFE Implementation Principles

In addition to principles for designing work systems and processes, HFE has developed
principles for changing work systems. For instance, in the context of health IT, HFE
implementation principles, such as participation, communication and feedback, learning and
training, top management commitment, and project management are critical to realizing the
patient safety potential of health IT.88,89 These implementation principles are essential and
applicable to the implementation of all kinds of work system design. A key HFE system
implementation principle is user participation. Participatory ergonomics programs can be
implemented in health care and lead to substantial improvements in occupational health and
safety,90,91 and potentially in patient safety. However, it may be difficult to use participatory
ergonomics in a high-stress, high-pressure environment, such as an ICU, where patient needs are
critical and patients require immediate or continuous attention.90,91 Further research is needed to
332
refine and develop HFE implementation principles and methods for facilitating user participation
in designing work systems for patient safety.

The 2005 report by the U.S. Institute of Medicine and the National Academy of Engineering
stressed the importance of using HFE as a key systems engineering tool to design health care
systems and to improve quality of care and patient safety.92 Numerous studies use HFE tools and
methods to identify system factors that contribute to medical errors; based on these data,
researchers or other system designers devise recommendations for improving health care work
systems and processes.
These studies are useful for highlighting the importance of HFE to patient safety; however,
they do not provide empirical evidence for the value of HFE in improving patient safety.
Empirical studies of how HFE-based interventions affect patient safety are few; those that are
available have addressed usability of health care technologies, concomitant design of health care
technologies and work system, and design of health care processes. Further research is necessary
to document and demonstrate the value of HFE-based interventions and their impact on patient
safety. Evidence for the effectiveness of HFE-based interventions should include data on
changes in the work system, changes in the process and changes in outcomes (including both
patient safety and employee outcomes). In general, this evidence is provided through the use of
multiple quantitative and qualitative methods.
HFE-based interventions involve changes in work systems and processes and, like any
change, may produce unanticipated effects. However, a core principle of HFE is to ensure that
work systems and processes are designed to produce patient safety benefits. The purpose of an
HFE approach is (1) to anticipate potential negative patient safety consequences (e.g.,
conducting a work system or process analysis, or a proactive risk assessment), and (2) to learn
about potential negative effects on patient safety during the implementation process and fix
problems as quickly as possible (e.g, system resilience).
This review is not intended to be a systematic review of HFE-based interventions for patient
safety, especially given the broadly different clinical topics and the small number of studies in
each clinical topic. Rather, our objective is to highlight the variety of HFE applications and to
describe the details of a small number of HFE applications that produced patient safety
improvements. Thus we review only four studies to demonstrate various HFE applications.
These examples also show that HFE applications for patient safety do not have to wait for
accidents to occur; HFE is primarily a proactive system design approach.
Example 1: Human Factors and Ergonomics in the Design of

Radiotherapy Treatment Delivery System
In the first example, HFE methods were used in the design of a radiotherapy treatment
delivery system.93,94
Step 1: Human Factors and Ergonomics Analysis

The researchers first evaluated the existing radiotherapy treatment delivery process. Over a
3-month period, an HFE engineer conducted 30 hours of field observations of radiation therapists
performing their regular tasks. Workflows of radiation therapists, in particular their interactions
with the treatment-delivery system, were recorded. Based on these observations, the researchers
compiled a list of tasks regularly performed by radiation therapists during treatment delivery.
333
Step 2: Heuristic Usability Evaluation

One experienced therapist and two HFE engineers performed a heuristic evaluation of the
usability of treatment-delivery system. Since the two HFE experts were not authorized to operate
the system, the therapist performed the tasks and explained the workflow to the engineers. The
two HFE experts independently identified HFE issues based on 14 usability heuristics,79 and
evaluated the severity of each usability issue; they then compared their ratings and reached
consensus on a final list of usability issues and their severity. A total of 75 usability issues were
identified; of these, 18 were classified as having a high potential impact on patient safety (i.e.
high severity), 20 were classified as medium severity, and 37 were classified as low severity. For
instance, when the therapist entered notes into a patients file, the notes could be deleted without
warning if the therapist selected another patients file before saving the notes. This usability
issue violated the heuristics of feedback, error recovery, and ability to undo, and was rated with
high severity. The recommendation for technology redesign was to warn therapists that their
notes might be deleted if they have not saved them.
Step 3: System Redesign and Evaluation

The existing treatment delivery system was redesigned based on HFE design principles. Two
focus groups with experienced radiation therapists provided feedback on the redesigned
treatment delivery system, and the system was further refined. Finally, user testing with 16
radiation therapy students was conducted to compare the current and redesigned treatment
delivery systems. Using each of the two systems, students went through four scenarios related to
typical treatment-delivery tasks. Three of the four scenarios were designed with a high potential
for certain use errors to occur (overlooking an important note, shifting the treatment couch
incorrectly, and overlooking a change of approval dates). The error rates and overall time to
complete each scenario were measured. At the end of the testing, participants were asked to fill
out a questionnaire to compare various attributes of the two systems. Results showed that error
rates for overlooking an important note and for overlooking changes in approval dates decreased
significantly with the redesigned treatment-delivery system (from 73% to 33% and from 56% to
0% respectively). The redesigned treatment delivery system led to efficiency gains (the mean
task completion time was reduced by 5.5%) and improvement in user satisfaction.
Example 2: Human Factors and Ergonomics in the Design and

Implementation of Health IT
Various work system factors can affect the acceptance and effective use of health care
technologies.88,95 Inadequate planning for implementation and lack of integrating health care
technologies into existing work systems are associated with work-arounds and technologies
falling short of achieving their patient safety goal.42,96-98 HFE approaches, which emphasize
simultaneous design of the health care technology and the work system, are recommended for
achieving a balanced work system70-72 and fulfilling the full potential of health care technology
in improving patient safety.
Beuscart-Zphir and colleagues99 developed an HFE framework for health care technology
and work system design, along with a set of structured methods to optimize the work system.
The HFE framework includes 4 stages: (1) analysis of the sociotechnical system and the
demands of stakeholders, (2) cooperative design of the health care technology and the work
system with the institution, designers and developers, (3) iterative evaluation and redesign, and
(4) assessment of the new work system and its impact on patient safety and overall performance
334
of the sociotechnical system. The HFE framework was used to improve the design and
implementation of CPOE.100
Step 1: Analysis of Medication Use Process and Recommendations for System

Redesign
Researchers conducted a systematic qualitative analysis of the medication ordering
dispensingadministration process. Field observations and semi-structured interviews were
performed with nurses to identify nursing tasks in the medication administration process, to
characterize physiciannurse and nursenurse communication about medications, and to assess
nurses interactions with paper patient records. Then more than 7,000 paper medication order
sheets issued by physicians and the corresponding paper medication-administration records from
nurses were reviewed.
Step 2: Cooperative System Design

The results of observations, interviews, and document review were presented to nurses for
feedback; software engineering models (e.g., UML and Petri Nets) were created to model the
distribution of tasks observed. Factors contributing to the safety of medication process were
identified at three levels: individual (e.g., interactions between nurses and the technology when
administering medications), collective (e.g., verbal communications supporting cooperation
during the medication management process) and organizational (e.g., distribution of tasks across
different health care professionals). Recommendations for work system redesign were proposed,
such as the need to provide nurses with specific information at each step of the preparation and
administration of medications, and the need for regular physician-nurse communications about
patient treatment and changes to the plan of care (e.g., daily briefing either before or after
medical rounds).
Step 3: Usability Evaluation of CPOE Technology

The researchers also evaluated the usability of the proposed CPOE technology. Five
independent HFE experts evaluated the user interface of the software application, using a set of
HFE criteria.101 A total of 35 issues related to workload, compatibility, control, homogeneity,
guidance, and error prevention was identified and rated on a four-point scale for severity.
In laboratory user testing, 8 nurses used the think-aloud method in a simulation of the
preparation of medication dispensers and the validation and documentation of medication
administration. The laboratory test was designed to reproduce the nurses typical work
environment. Scenarios were created based on the results of the initial work system analysis.
Nurse participants identified a total of 28 usability issues during the test.
Step 4: Iterative Human Factors and Ergonomics Redesign

In the next phase of CPOE technology redesign, possible solutions for each of the identified
usability issues were proposed and evaluated with respect to costs and benefits. Mock-ups and
prototypes were developed for those solutions. Iterative usability evaluations and technology
redesigns were done until all critical usability issues were addressed. To evaluate the impact of
the HFE-based design of health care work system on patient safety, the researchers proposed to
link the system redesign to the actual identification of adverse events.
In a recent project, the researchers used statistical data mining methods to semi-automatically
identify adverse drug events and to link the identified adverse drug events to the analysis and
335
modeling of the work systems. The HFE framework of Beuscart-Zphir and colleagues is now
routinely integrated into the IT project management of the Centre Hospitalier Universitaire of
Lille, France.
Example 3: Human Factors and Ergonomics in the Physical Design of

Operating Rooms
In the third example, HFE is used to address infection-control problems in the operating
room (OR).102 To minimize infection risk, surgical devices were suggested to be positioned
within the clean airflow in the OR according to well-known design principles.103,104
Step 1: Benchmarking of System

A multidisciplinary team of hospital surgical staff learned from the experience of runway
operators at an international airport regarding marking, position of materials, traffic flows, safety
rules and regulations, and incident management. They applied this knowledge to OR traffic
flows, position of surgical tables and materials, safety management, and the process of incident
reporting.
Step 2: Human Factors and Ergonomics System Design

The multidisciplinary team designed and implemented floor marking to support consistently
correct positioning of surgical devices. The implementation was carried out in three steps:
1. temporary marking was implemented in 2 of 4 ORs in February 2009,
2. temporary marking was implemented in all four ORs by June 2009, and
3. permanent floor marking was implemented in all ORs in December 2009.
Step 3: Evaluation of System Redesign

Compliance with positioning of surgical devices within the clean airflow was evaluated by
observing a total of 182 surgeries before implementation of the floor marking. One month after
the implementation of the temporary floor marking in 2 ORs, compliance data were collected by
observing 195 surgeries in ORs with floor markings and 86 surgeries in ORs without floor
markings. Four months after implementation of the temporary floor markings in all four ORs,
167 surgeries were observed to collect compliance data. Finally, 199 surgeries were observed 1
month after the implementation of permanent floor markings. Floor marking resulted in
significantly increased compliance with recommended positionings of surgical devices in the
clean airflow. In addition, post-implementation interviews with 3 ophthalmic surgeons, 3
surgical and anesthesia nurses, and 2 managers showed enhanced safety awareness among
surgical staff. Although the researchers did not use the term HFE to describe their study, their
approach used a systematic work system analysis and led to a solution firmly rooted in the HFE
systems approach.105
Example 4: Human Factors and Ergonomics in the Design of Care

Processes
HFE can also help to improve the design of care processes.106,107 Proactive risk assessment
methods, such as failure mode and effects analysis (FMEA), are HFE methods that can be used
to evaluate high-risk processes in health care and provide input for health care process
design.108,109 The fourth study we review describes an FMEA of the IV medication
336
administration process conducted to assess the potential HFE and safety issues of a new IV
pump.110
Step 1: Formation and Training of FMEA Team

A multidisciplinary team consisting of representatives from anesthesiology, biomedical
engineering central supply, human factors engineering, internal medicine, nursing, pharmacy,
and quality improvement performed a health care failure mode and effects analysis (HFMEA)111
to evaluate the intravenous (IV) medication administration process using both current IV pump
and a Smart IV pump technology. The team members were trained for 1 to 2 hours in the VAs
HFMEA method.111
Step 2: FMEA Analysis Process

The FMEA process consisted of 46 hours of meetings over 4.5 months and unfolded in three
steps:
1. Process identification and mapping
2. Failure mode identification and scoring
3. Determination of interventions and outcome measures
Multiple data sources were used to develop the IV medication administration process map.
Two HFE experts conducted a total of 52 observations of nurses administering medications with
the current IV pump.112 Medication administration and IV pump events reported with the current
pump were retrieved from the hospitals event reporting system. The FMEA team mapped the
medication administration process with the current IV pump and then repeated the mapping
process with the Smart IV pump. In the process map with the current IV pump, the team
identified 10 steps for retrieving the medication and tubing, and 24 steps for pump programming
were identified. For the Smart IV pump, the team identified 14 unique pump programming steps
and new tubing setup and insertion steps.
Following process mapping, the team analyzed failure modes potentially associated with IV
pump use. About 200 failure modes were identified and scored with respect to severity and
probability of occurrence. A hazard score was calculated by using the product of the severity and
probability of occurrence ratings. Failure modes with low or lowmoderate hazard scores were
assessed for detectability, and only non-detectable failure modes were considered for further
action. All failure modes with moderate-to-high hazard scores were considered further.
Step 3: Recommendations for Process Redesign

Recommendations for prioritized failure modes were proposed and categorized into the five
elements of the work system63 (see Figure 1): (1) policies and procedures, (2) training or
education, (3) physical environment, (4) people, and (5) technology software or hardware
change. The evaluation of the impact of the FMEA on patient safety was based on: (1) audits of
programming of pumps for errors, (2) monitoring of end-user training for time to achieve
competency, and (3) monitoring and recording of IV medication administration event reports and
informal and formal complaints about pump functioning. Post-implementation results suggested
that the goal of mitigating risk to patients from potential or known failure modes was achieved.
337

What Contexts?
HFE can contribute to patient safety in a range of care settings. Table 4 describes selected
HFE issues in various care settings. The issues are categorized as physical, cognitive, or
organizational HFE issues, and are related to the various work system elements (see Figure 1).
These issues interact as part of the larger work system and produce the vulnerabilities that can
lead to patient safety incidents.
338
Table 4, Chapter 31. HFE issues in selected care settings

Care
Settings
Anesthesia
ED
Physical (P), Cognitive (C)

or Organizational (O) HFE
HFE Issues
Impact of fatigue and sleep deprivation on psychomotor performance and mood of
113
anesthesiology residents
114-116
Workload, production pressure and burnout of anesthesiologists
Poor display and control design of medical devices: auditory and visual alarms affect
117,118
vigilance and situation awareness of anesthesiologists
Working in a multidisciplinary team: anesthesiologists working with a new surgeon or
nurse may need extra effort to communicate effectively, in particular during stressful
113
conditions
Limited availability of information: patient history and other information are often not
117
easily accessible by ED clinicians who had no prior contact with patient
119,120
Design of ED physical environment: overcrowding, noise
Usability and workflow issues of ED status boards
Home Care
ICUs
x
x
113,119
339
x
x
Varied, dynamic, rapidly-changing condition of patients that require rapid clinician

113
responses
x
x
x
x
x
x
x
x
x
Design and implementation of guidelines and best practices, e.g., for infection
114
control
37,115,116
Workload, stress and burnout of ICU physicians and nurses
Information flow and decisionmaking in handoffs of patients: across units, across
113
services; patients discharged; shift changes
25,117
Design of alerts/alarms in medical devices and health IT
118
Design and implementation technology for remote monitoring of ICU patients
T/T
119-121
Impact of shift work on cognitive and work performance of ED clinicians in particular

119,121
during routine work
Limited opportunity for ED clinicians to maintain their skill level for risky and difficult,
119,121
but infrequent, procedures
Usability and acceptance of computer-based self-management tools for elderly
patients with disability and functional decline: usability of interface, functional and
122
physical accessibility
Design problems of telemedicine applications: poor usability, e.g., extensive amount
123
of text on screen
123
Informal care giving: fatigue, musculoskeletal injuries during personal care
Design of ICU patient rooms: open versus closed
Elements of the Work

System [I, T, T/T, E, O]*
x
x
x
x
x
x
x
x
x
x
Table 4, Chapter 31. HFE issues in selected care settings (continued)

Care
Settings
Long-Term
Care
Pediatrics
Primary
Care
Surgery
Physical (P), Cognitive (C)

or Organizational (O) HFE
HFE Issues
Poor working conditions and job stressors: understaffing, training, feeling unable to
meet resident needs, overtime, heavy workload, mostly standing and walking, risk of
124
back injuries due to moving patients
Physical environment: doors cannot accommodate a wheelchair, layout of facility
124
does not allow nursing station in a convenient place
Cognitive, communication, and speech limitations of children and their dependency
on adults result in communication challenges, and risk of delayed diagnosis or
125,126
misdiagnosis
Equipment not designed for children: CT scan with adjustable exposure for children,
125,126
cribs with adjustable height
Design problems of CPOE technology: weight-based dosing, age-dependent lab
125
normal values
Design problems of BCMA technology: barcodes with different sizes, packaging of
125
pediatric medications
Reliance on memory: missing diagnostic testing results, lack of tracking system;
127,128
physician needs to remember ordered tests
Multi-modal communication between patient and clinicians: retrieving and recording
127
information, information loss
Memory and information processing: patients with multiple problems, incomplete
127
patient charts
Workload and time pressure of clinicians: addressing several patient problems in
127
limited time
Operating room environment: clutter, noise, lighting, temperature, motion/vibration;
129
impact on surgical performance
Teamwork: miscommunication, lack of coordination, and lack of team familiarity and
129,130
stability contribute to errors during surgery
Poor design and implementation of technology affect acceptance and use: e.g.
integration of information across displays, unreliable audible alarms, shape of input
129,130
controls, and lack of proper training for surgeons
Impact of physical and mental workload on performance: task duration, strength
requirement, mental demands, and time pressure increase stress and fatigue, and
129
may affect cognitive processing
131
Design and implementation of surgery checklist
Elements of the Work

System [I, T, T/T, E, O]*
T/T
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Poor safety culture: lack of a culture to take responsibility for patient safety, report
errors, learn from mistakes, and adapt individual and organizational behavior based
x
x
x
129
on lessons learned from mistakes
* NOTE: Elements of the work system include the individual (I), his/her tasks (T), tools and technologies (T/T), the physical environment (E) and the organization (O) (see Figure
1).
340
Are There Any Data About Costs of the Patient Safety Practice?
The integration of HFE in health care and patient safety requires leadership and commitment
as well as resources (i.e., money, time, effort, knowledge, expertise, skills, methods and
structures).132-135 Health care organizations that invest in HFE typically engage in one or several
of the following activities: using HFE tools and methods, increasing HFE knowledge among
their staff, and recruiting human factors engineers.67 However, there is no information available
about the costs of these different HFE approaches.
It is important to recognize the key role of HFE in the early phase of system design.136 When
HFE is used early in the design process, system issues can be identified and solved more
efficiently and effectively, and with less risk that the fix to the system design will itself create
other hazards. This implies that designers, manufacturers, and vendors of health IT applications,
medical devices and other technologies must have in-house HFE expertise.
A case study of a medical device manufacturer demonstrates the challenges of implementing
HFE.137 Patient Controlled Analgesia (PCA) pumps that were introduced in 1988 were intended
to allow patients to administer small and frequent dosages of analgesia, and reduce nurse
workload. However, the poor HFE design of the device increased the likelihood of dosage
programming errors, which in some cases led to death.137 It took 6 years between the first
reported incident of patient death related to PCA pump programming error and the hiring of a
HFE engineer by the device manufacturer in 2001. Significant efforts may be required to speed
up the dissemination of HFE to improve patient safety across the health care industry.12,67
Fostering communication and collaboration between HFE and the health sciences and
professions is critical achieving significant improvements in patient safety. Clinicians and HFE
engineers need to learn to understand each others perspectives.132 Because the HFE knowledge
domain is broad and deep (see description above of the physical, cognitive and organizational
aspects of HFE), learning HFE can be a significant investment. It is not sufficient to have
physicians or nurses who have read a book or taken a seminar on HFE; this will not make them
HFE experts.132 On the other hand, HFE engineers need to understand health care before they
can have a significant impact on patient safety.67 The training of biculturals in both medicine
(or nursing or pharmacy or other health science) and HFE can accelerate the application of HFE
to improve patient safety.138 Because biculturals have deep knowledge of and training in both
HFE and a health science, they can help to translate and disseminate HFE knowledge and
tools.138

Despite the critical role of HFE in improving patient safety, the application of HFE to health
care may not be straightforward or spontaneous. More work is needed to understand the
challenges faced by health care organizations in adopting, implementing and institutionalizing
HFE in their operations. In the context of health care organizations, HFE can be conceptualized
as an innovation whose adoption, diffusion, and maintenance are associated with challenges.66
As described earlier, HFE patient safety practices include: using HFE tools and methods,
increasing HFE knowledge, and recruiting HFE engineers.66 A range of contextual factors can
affect the effectiveness of these HFE-based interventions or innovations, such as structural
characteristics of health care organizations (e.g., size, level of functional differentiation, and
level of centralization of decisionmaking), cultural characteristics of health care organizations
(e.g., leadership, strategic vision, approach to experimentation and risk, and learning style),
341
management and implementation tools (e.g., top management commitment, human resource
issues, funding, and communications), and wider environmental factors (e.g., legal and
regulatory requirements, efforts by national and international HFE organizations, and
collaboration between health care organizations).66,140 Cultural conflicts between the HFE
systems approach and health care can also impede the adoption and dissemination of HFE in
health care organizations (e.g., physician autonomy may hinder the team collaboration and
communication stressed by HFE).139,141
The case study described by Vicente138 shows that HFE is more likely to be integrated in the
organization of a medical device manufacturer if the manufacturer (1) has leaders who support
adoption of HFE, (2) experiences a profound performance crisis related to poor HFE
performances, and (3) operates in an environment in which advocacy for HFE can be found at all
levels of the complex sociotechnical system. Further research is needed to identify the key
contextual factors that can facilitate adoption and dissemination of HFE. Specifically, studies are
recommended for developing a theoretical framework to describe and evaluate contextual
elements and generating empirical evidence on how different contextual elements can influence
the success of HFE interventions.140

A study conducted by an HFE leader, Al Chapanis, and his colleague in the early 1960s
provided information on medication administration errors and the system factors that contributed
to these errors.143-145 Since then, awareness of the importance of HFE in medication safety and
other patient safety domains has significantly increased. Patient safety leaders call for increasing
involvement of HFE in helping not only to characterize system factors that contribute to patient
safety, but also to inform system design interventions.4,146,147 This chapter has described the
range of patient safety issues and care settings that HFE can be applied to. Further research is
needed to continue developing the evidence for the value of HFE-based interventions for patient
safety.
Numerous chapters in this report describe how patient safety practices can benefit from HFE.
For instance, chapter 6 reviews the evidence for the patient safety impact of Smart IV pump.
HFE problems in the design of the pump interface and alerts have limited the patient safety
impact of Smart IV pumps.148 HFE can provide the design principles and methods to improve
Smart IV pump technology (e.g., usability of pump interface design) and enhance its impact on
patient safety.110,149 Chapter 34 describes the strong empirical evidence for the impact of nursepatient ratio on patient safety. One potential mechanism for this impact is related to nursing
workload.115 HFE principles and methods can be used in the design of work systems to reduce or
mitigate nursing workload, and therefore, improve patient safety.37,115 Chapter 16 highlights
some of the HFE challenges that can be addressed with integrated information displays in the
OR, especially if these displays are designed to support team situation awareness and
coordination.
These examples show that many patient safety practices can benefit from HFE. Patient safety
practices target some aspect of the work system (see Figure 1) and should be designed and
implemented according to HFE principles to produce patient safety benefits. HFE is a core
element of patient safety improvement; therefore, every effort should be made to support HFE
applications in patient safety. A summary table is following (Table 5).
342

Scope of the
Strength of
the PSP
Effectiveness
Potentially applicable
to all patient safety
problems
Not assessed
systematically
but Moderateto-High
evidence for
some specific
applications
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Moderate
How Much do We
A lot/Moderate
References
1.
Leape LL, Bates DW, Cullen DJ, et al.

Systems analysis of adverse drug events.
Journal of the American Medical
Association. 1995;274(1):35-43.
2.
Institute of Medicine. Preventing

Medication Errors. Washington, DC: The
3.
Han YY, Carcillo JA, Venkataraman ST, et

al. Unexpected increased mortality after
implementation of a commercially sold
computerized physician order entry system.
Pediatrics. 2005;116:1506-1512.
Institute of Medicine. Health IT and Patient
Safety: Building Safer Systems for Better
Care. Washington, DC: The National
Academies Press; 2012.
4.
5.
6.
Hignett S, Masud T. A review of

environmental hazards associated with inpatient falls. Ergonomics. 2006;49(5-6):605616.
Smetzer J, Baker C, Byrne FD, Cohen MR.
Shaping systems for better behavioral
choices: Lessons learned from a fatal
medication error. Joint Commission Journal
on Quality and Patient Safety.
2010;36(4):152-163.
7.
Vincent C, Moorthy K, Sarker SK, Chang

A, Darzi AW. Systems approaches to
surgical quality and safety - From concept to
measurement. Annals of Surgery.
2004;239(4):475-482.
8.
Vincent C, Taylor-Adams S, Stanhope N.

Framework for analysing risk and safety in
clinical medicine. BMJ.
1998;316(7138):1154-1157.
343
9.
International Ergonomics Association (IEA).

The Discipline of Ergonomics. 2000;
www.iea.cc/ergonomics. Accessed August
22, 2004.
10.
Carayon P. Human factors of complex

sociotechnical systems. Applied
Ergonomics. 2006;37:525-535.
11.
Carayon P, ed Handbook of Human Factors

in Health Care and Patient Safety. Mahwah,
New Jersey: Lawrence Erlbaum Associates;
2007.
12.
Gurses AP, Ozok AA, Pronovost PJ. Time

to accelerate integration of human factors
and ergonomics in patient safety. BMJ
Quality & Safety. November 30 2011.
13.
Grandjean E. Fitting the Task to the Man:

An Ergonomic Approach. London: Taylor &
Francis; 1980.
14.
Gosbee J. Human factors engineering and

patient safety. Quality & Safety in Health
Care. 2002;11(4):352-354.
15.
Gosbee JW, Gosbee LL, eds. Using Human

Factors Engineering To Improve Patient
Safety. Oakbrook Terrrace, Illinois: Joint
Commission Resources; 2005.
16.
Rousek JB, Hallbeck MS. Improving

medication management through the
redesign of the hospital code cart medication
drawer. Human Factors. 2011;53(6):626636.
17.
Bates DW, Gawande AA. Improving safety

with information technology. The New
England Journal of Medicine.
2003;348(25):2526-2534.
18.
Kohn LT, Corrigan JM, Donaldson MS, eds.

To Err is Human: Building a Safer Health
System. Washington, D.C.: National
Academy Press.; 1999.
19.
Battles JB, Keyes MA. Technology and

patient safety: A two-edged sword.
Biomedical Instrumentation and
Technology. 2002;36(2):84-88.
29.
Reason J. Managing the Risks of

Organizational Accidents. Burlington,
Vermont: Ashgate; 1997.
30.
Reason J. The Human Contribution.

London: Ashgate Publishing Company;
2009.
31.
Alper SJ, Karsh B-T. A systematic review of

safety violations in industry. Accident
Analysis and Prevention. 2009;41(4):739754.
32.
Halbesleben JRB, Savage GT, Wakefield

DS, Wakefield BJ. Rework and
workarounds in nurse medication
administration process: Implications for
work processes and patient safety. Health
Care Management Review. 2010;35(2):124133.
33.
Pickering BW, Herasevich V, Gajic O.

Novel representation of clinical information
in the ICU. Applied Clinical Informatics.
2010;1(2):116-131.
Bogner MS. The artichoke systems approach

for identifying the why of error. In: Carayon
P, ed. Handbook of Human Factors in
Health Care and Patient Safety. Mahwah,
NJ: Lawrence Erlbaum; 2007:109-126.
34.
Ahmed A, Chandra S, Herasevich V, Gajic

O, Pickering BW. The effect of two
different electronic health record user
interfaces on intensive care provider task
load, errors of cognition, and performance.
Critical Care Medicine. 2011;39(7):16261634.
Carayon P, Gurses AP, Hundt AS, Ayoub P,

Alvarado CJ. Performance obstacles and
facilitators of healthcare providers. In:
Korunka C, Hoffmann P, eds. Change and
Quality in Human Service Work. Vol
Volume 4. Munchen, Germany: Hampp
Publishers; 2005:257-276.
35.
Karsh B-T, Holden RJ, Alper SJ, Or CKL. A

human factors engineering paradigm for
patient safety: Designing to support the
performance of the healthcare professional.
Quality & Safety in Health Care. 2006;15
(i6):i59-i65.
36.
Gurses AP, Carayon P. Performance

obstacles of intensive care nurses. Nursing
Research. 2007;56(3):185-194.
37.
Gurses A, Carayon P, Wall M. Impact of

performance obstacles on intensive care
nurses workload, perceived quality and
safety of care, and quality of working life.
Health Services Research. 2009;44(2):422443.
38.
Gurses A, Carayon P. A qualitative study of

performance obstacles and facilitators
among ICU nurses. Applied Ergonomics.
2009;40(3):509-518.
20.
Cook RI. Safety technology: Solutions or

experiments? Nursing Economic$.
2002;20(2):80-82.
21.
Reason J. Human Error. Cambridge:

Cambridge University Press; 1990.
22.
Chan J, Shojania KG, Easty AC, Etchells

EE. Does user-centred design affect the
efficiency, usability and safety of CPOE
order sets? Journal of the American Medical
Informatics Assocociation. 2011;18(3):276281.
23.
24.
25.
26.
Phansalkar S, Edworthy J, Hellier E, et al. A

review of human factors principles for the
design and implementation of medication
safety alerts in clinical information systems.
Journal of the American Medical
Informatics Association. 2010;17(5):493501.
Zachariah M, Phansalkar S, Seidling HM, et
al. Development and preliminary evidence
for the validity of an instrument assessing
implementation of human-factors principles
in medication-related decision-support
systems - I-MeDeSA. Journal of the
American Medical Informatics Association.
2011(September 21).
27.
Bogner MS, ed Human Error in Medicine.

Hillsdale, NJ: Lawrence Erlbaum
Associates; 1994.
28.
Waterson P. A critical review of the systems

approach within patient safety research.
Ergonomics. 2009;52(10):1185-1195.
344
39.
Tucker AL. The impact of operational

failures on hospital nurses and their patients.
Journal of Operations Management.
2004;22(2):151-169.
49.
Kanse L, Van der Schaaf TW, Vrijland ND,

Van Mierlo H. Error recovery in hospital
pharmacy. Ergonomics. 2006;49(5-6):503516.
40.
Sainfort F, Karsh B, Booske BC, Smith MJ.

Applying quality improvement principles to
achieve healthy work organizations. Journal
on Quality Improvement. 2001;27(9):469483.
50.
41.
Carayon P, Wetterneck TB, Hundt AS, et al.

Evaluation of nurse interaction with bar
code medication administration technology
in the work environment. Journal of Patient
Safety. 2007;3(1):34-42.
Runciman W, Hibbert P, Thomson R, Van

Der Schaaf T, Sherman H, Lewalle P.
Towards an international classification for
patient safety: Key concepts and terms.
International Journal for Quality in Health
Care. 2009;21(1):18-26.
51.
Runciman WB, Baker GR, Michel P, et al.

Tracing the foundations of a conceptual
framework for a patient safety ontology.
Quality & Safety in Health Care.
2010;19(6):1-5.
52.
Rogers AE, Dean GE, Hwang WT, Scott

LD. Role of registered nurses in error
prevention, discovery and correction.
Quality & Safety in Health Care.
2008;17(2):117-121.
53.
Henneman EA, Gawlinski A, Blank FS,

Henneman PL, Jordan D, McKenzie JB.
Strategies used by critical care nurses to
identify, interrupt, and correct medical
errors. American Journal of Critical Care.
2010;19(6):500-509.
54.
Faye H, Rivera-Rodriguez AJ, Karsh BT,

Hundt AS, Baker C, Carayon P. Involving
intensive care unit nurses in a proactive risk
assessment of the medication management
process. The Joint Commission Journal on
Quality and Patient Safety. 2010;36(8):376384.
55.
Rothschild JM, Churchill W, Erickson A, et

al. Medication errors recovered by
emergency department pharmacists. Annals
of emergency medicine. 2010;55(6):513521.
56.
Patterson ES, Roth EM, Woods DD, Chow

R, Gomes JO. Handoff strategies in settings
with high consequences for failure: Lessons
for health care operations. International
Journal for Quality in Health Care.
2004;16(2):125-132.
57.
Patterson ES, Woods DD, Cook RI, Render

ML. Collaborative cross-checking to
enhance resilience. In: The Human Factors
and Ergonomics Society, ed. Proceedings of
the Human Factors and Ergonomics Society
49th Annual Meeting. Santa Monica, CA:
The Human Factors and Ergonomics
Society; 2005:512-516.
42.
Koppel R, Wetterneck T, Telles JL, Karsh

B-T. Workarounds to barcode medication
administration systems: Their occurrences,
causes, and threats to patient safety. Journal
of the American Medical Informatics
Association. 2008;15:408-423.
43.
Helmreich RL, Foushee HC, Administration

USFA. Why Crew Resource Management?:
Conceptual Bases of Human Factors
Training in Aviation. Austin, TX: University
of Texas Press; 1991.
44.
Wiener EL, Kanki BG, Helmreich RL.

Cockpit Resource Management. San Diego,
CA: Academic Press; 1993.
45.
Weaver SJ, Salas E, King HB. Twelve best

practices for team training evaluation in
health care. Joint Commission Journal on
Quality and Patient Safety. 2011;37(8):341349.
46.
Jeffcott S, Ibrahim J, Cameron P. Resilience

in healthcare and clinical handover. Quality
and Safety in Health Care. 2009;18(4):256260.
47.
Woods DD, Hollnagel E. Prologue:

Resilience engineering concepts. In:
Hollnagel E, Woods DD, Leveson N, eds.
Resilience Engineering - Concepts and
Precepts. Hampshire, UK: Ashgate; 2006:16.
48.
Habraken MMP, van der Schaaf TW. If

only...: Failed, missed and absent error
recovery opportunities in medication errors.
Quality and Safety in Health Care.
2010;19(2):37-41.
345
58.
Weick KE, Sutcliffe KM. Managing the

Unexpected - Resilient Performance in an
Age of Uncertainty. Second edition ed. San
Francisco, CA: John Wiley & Sons; 2007.
69.
Carayon P, Alvarado C, Hundt AS. Work

design and patient safety. TIES-Theoretical
Issues in Ergonomics Science.
2007;8(5):395-428.
59.
Hollnagel E. Prologue: Prologue: The scope

of resilience engineering. In: Hollnagel EK,
Paries J, Woods D, Wreathall J, eds.
Resilience Engineering in Practice.
Burlington, VT: Ashgate; 2011:xxix-xxxvii.
70.
Carayon P. The Balance Theory and the

work system model... Twenty years later.
International Journal of Human-Computer
Interaction. 2009;25(5):313-327.
71.
60.
Cook RI, Nemeth C. Taking things in ones

stride: Cognitive features of two resilient
performances. In: Hollnagel E, Woods DD,
Leveson N, eds. Resilience Engineering Concepts and Precepts. Hampshire, UK:
Ahsgate; 2006:205-221.
Smith MJ, Carayon-Sainfort P. A balance

theory of job design for stress reduction.
International Journal of Industrial
Ergonomics. 1989;4(1):67-79.
72.
Smith MJ, Carayon P. Balance theory of job

design. In: Karwowski W, ed. International
Encyclopedia of Ergonomics and Human
Factors. London: Taylor & Francis;
2000:1181-1184.
73.
Hollnagel E, Woods DD, Leveson N, eds.

Resilience Engineering - Concepts and
Precepts. Burlington, VT: Ashgate; 2006.
74.
Cherns A. Principles of sociotechnical

design revisited. Human Relations.
1987;40(3):153-162.
75.
Clegg CW. Sociotechnical principles for

system design. Applied Ergonomics.
2000;31(5):463-477.
76.
Pasmore WA. Designing Effective

Organizations: The Sociotechnical Systems
Perspective. New York: John Wiley & Sons;
1988.
77.
Hollnagel E. Resilience - The challenge of

the unstable. In: Hollnagel E, Woods DD,
Leveson N, eds. Resilience Engineering Concepts and Precepts. Burlington, VT:
Ashgate; 2006:9-17.
78.
Weinger MB, Wiklund M, GardnerBonneau DJ, eds. Handbook of Human

Factors in Medical Device Design. Boca
Raton, FL: Taylor & Francis; 2011.
79.
Zhang J, Johnson TR, Patel VL, Paige DL,

Kubose T. Using usability heuristics to
evaluate patient safety of medical devices.
Journal of Biomedical Informatics.
2003;36:23-30.
80.
Nielsen J. Usability Engineering. Morgan

Kaufmann: Amsterdam, The Netherlands;
1993.
81.
Shneiderman B. Designing the User

Interface. Third ed. Reading, MA: AddisonWesley; 1998.
61.
Catchpole K, McCulloch P. Human factors

in critical care: Towards standardized
integrated human-centred systems of work.
Current Opinion in Critical Care.
2010;16(6):618-622.
62.
Vincent CA. Analysis of clinical incidents:

A window on the system not a search for
root causes. Quality & Safety in Health
Care. 2004;13:242-243.
63.
Carayon P, Hundt AS, Karsh B-T, et al.

Work system design for patient safety: The
SEIPS model Quality & Safety in Health
Care. 2006;15(Supplement I):i50-i58.
64.
Wilson JR. Fundamentals of ergonomics in

theory and practice. Applied Ergonomics.
2000;31(6):557-567.
65.
Donabedian A. The quality of medical care.

Science. 1978;200:856-864.
66.
Taylor JA, Dominici F, Agnew J, Gerwin D,

Morlock L, Miller MR. Do nurse and patient
injuries share common antecedents? An
analysis of associations with safety climate
and working conditions. BMJ Quality &
Safety. 2011.
67.
Carayon P. Human factors in patient safety

as an innovation. Applied Ergonomics.
2010;41(5):657-665.
68.
Carayon P. Top managements view on

human factors and patient safety: Do they
see it? In: Tartaglia R, Bagnara S, Bellandi
T, Albolino S, eds. Healthcare Systems
Ergonomics and Patient Safety. Florence,
Italy: Taylor & Francis; 2005:38-42.
346
82.
Helander MG, Willen BA. Design for

human assembly (DHA). In: Karwowski W,
Marras WS, eds. The Occupational
Ergonomics Handbook. Boca Raton, FL:
CRC Press; 1999:1849-1865.
83.
Kroemer KHE, Kroemer H, Kroemer-Elbert

K. Ergonomics - How to Design for Ease
and Efficiency. Second edition ed. Upper
Saddle River, NJ: Prentice Hall; 2001.
84.
Cherns A. The principles of sociotechnical

design. Human Relations. 1976;29(8):783792.
85.
House J. Work, Stress and Social Support.

Reading, MA: Addison-Wesley; 1981.
86.
Salvendy G, ed Handbook of Human

Factors and Ergonomics. Third ed. New
York, NY: John Wiley & Sons; 2006.
87.
Norman DA, Draper SW. User-Centered

System Design: New Perspectives on
Human-Computer Interaction. Hillsdale, NJ:
Lawrence Erlbaum Associates; 1986.
88.
Karsh B-T. Beyond usability: Designing

effective technology implementation
systems to promote patient safety. Quality
and Safety in Health Care. 2004;13:388-394.
89.
Carayon P, Alyousef B, Xie A. Human

factors and ergonomics in health care. In:
Salvendy G, ed. Handbook of Human
Factors and Ergonomics2011.
90.
Evanoff VA, Bohr PC, Wolf L. Effects of a

participatory ergonomics team among
hospital orderlies. American Journal of
Industrial Medicine. 1999;35:358-365.
91.
92.
93.
Bohr PC, Evanoff BA, Wolf L.

Implementing participatory ergonomics
teams among health care workers. American
Journal of Industrial Medicine.
1997;32(3):190-196.
Reid PR, Compton WD, Grossman JH,
Fanjiang G. Building a Better Delivery
System. A New Engineering/Health Care
Partnership. Washington, D.C.: The
Chan AJ, Islam MK, Rosewall T, Jaffray
DA, Easty AC, Cafazzo JA. The use of
human factors methods to identify and
mitigate safety issues in radiation therapy.
Radiotherapy & Oncology. 2010;97(3):596600.
347
94.
Chan AJ, Islam MK, Rosewall T, Jaffray

DA, Easty AC, Cafazzo JA. Applying
usability heuristics to radiotherapy systems.
Radiotherapy & Oncology. Jul 4 2011.
95.
Smith MJ, Carayon P. New technology,

automation, and work organization: Stress
problems and improved technology
implementation strategies. International
Journal of Human Factors in Manufacturing.
1995;5(1):99-116.
96.
Patterson ES, Cook RI, Render ML.

Improving patient safety by identifying side
effects from introducing bar coding in
medication administration. Journal of the
2002;9:540-553.
97.
Kaushal R, Barker KN, Bates DW. How can

information technology improve patient
safety and reduce medication errors in
childrens health care? Arch Pediatr Adolesc
Med. 2001;155(9):1002-1007.
98.
Karsh BT, Weinger MB, Abbott PA, Wears

RL. Health information technology:
Fallacies and sober realities. Journal of the
2010;17(6):617-623.
99.
Beuscart-Zephir MC, Elkin P, Pelayo S,

Beuscart R. The human factors engineering
approach to biomedical informatics projects:
State of the art, results, benefits and
challenges. Yearbook of Medical
Informatics. 2007:109-127.
100.
Beuscart-Zephir MC, Pelayo S, Bernonville

S. Example of a human factors engineering
approach to a medication administration
work system: Potential impact on patient
safety. International Journal of Medical
Informatics. 2010;79(4):e43-e57.
101.
Bastien C, Scapin DL. Ergonomic Criteria

for the Evaluation of Human-Computer
Interfaces: INRIA Rocquencourt;1993.
102.
de Korne DF, van Wijngaarden JDH, van

Rooij J, Wauben LSGL, Hiddema UF,
Klazinga NS. Safety by design: effects of
operating room floor marking on the
position of surgical devices to promote clean
air flow compliance and minimise infection
risks. BMJ Quality & Safety. August 18
2011.
103.
Mangram AJ, Horan TC, Pearson ML,

Silver LC, Jarvis WR. Guideline for
prevention of surgical site infection, 1999.
Hospital Infection Control Practices
Advisory Committee. Infection Control and
Hospital Epidemiology. 1999;20(4):250278; quiz 279-280.
104.
Pasquarella C, Sansebastiano GE, Ferretti S,

et al. A mobile laminar airflow unit to
reduce air bacterial contamination at
surgical area in a conventionally ventilated
operating theatre. Journal of Hospital
Infection. 2007;66(4):313-319.
105.
Norris BJ. Systems human factors: how far

have we come? BMJ Quality & Safety.
November 7 2011.
106.
Institute of Medicine Committee on Quality

of Health Care in America. Crossing the
Quality Chasm: A New Health System for
the 21st Century. Washington, DC: National
Academy Press; 2001.
107.
Leape LL. Smart pumps: A cautionary

tale of human factors engineering. Crit Care
Med. Mar 2005;33(3):679-680.
108.
Carayon P, Faye H, Hundt AS, Karsh B-T,

Wetterneck T. Patient safety and proactive
risk assessment. In: Yuehwern Y, ed.
Handbook of Healthcare Delivery Systems.
Boca Raton, FL: Taylor & Francis; 2011:1211.
109.
110.
111.
Bonnabry P, Despont-Gros C, Grauser D, et

al. A risk analysis method to evaluate the
impact of a computerized provider order
entry system on patient safety. Journal of the
2008;15(4):453-460.
Wetterneck TB, Skibinski KA, Roberts TL,
et al. Using failure mode and effects analysis
to plan implementation of Smart intravenous
pump technology. American Journal of
Health-System Pharmacy. 2006;63:15281538.
DeRosier J, Stalhandske E, Bagian JP,
Nudell T. Using health care failure mode
and effect analysis: The VA national center
for patient safetys prospective risk analysis
system. Joint Commission Journal on
Quality Improvement. 2002;28(5):248-267,
209.
348
112.
Carayon P, Wetterneck T, Hundt AS, et al.

Observing nurse interaction with infusion
pump technologies. In: Henriksen K, Battles
JB, Marks E, Lewin DI, eds. Advances in
Patient Safety: From Research to
Implementation. Vol 2. Rockville, MD:
Agency for Healthcare Resaerch and
Quality; 2005:349-364.
113.
Gurses AP, Winters B, Pennathur PR,

Carayon P, Pronovost PJ. Human factors
and ergonomics in intensive care units. In:
Carayon P, ed. Handbook of Human Factors
and Ergonomics in Health Care and Patient
Safety, Second Edition. 2nd ed. Boca Raton,
FL: Taylor & Francis; 2012.
114.
Gurses AP, Marsteller JA, Ozok AA, Xiao

Y, Owens S, Pronovost PJ. Using an
interdisciplinary approach to identify factors
that affect clinicians compliance with
evidence-based guidelines. Critical Care
Medicine. 2010;38(8 (supplement)):S282S291.
115.
Carayon P, Gurses A. Nursing workload and

patient safety in intensive care units: A
human factors engineering evaluation of the
literature. Intensive and Critical Care
Nursing. 2005;21:284-301.
116.
Embriaco N, Azoulay E, Barrau K, et al.

High level of burnout in intensivists:
Prevalence and associated factors. American
Journal of Respiratory and Critical Care
Medicine. 2007;175(7):686-692.
117.
Friesdorf W, Buss B, Gobel M. Monitoring

alarms - The key to patients safety in the
ICU? Intensive Care Medicine.
1999;25(12):1350-1352.
118.
Hoonakker PLT. Human factors in

telemedicine. In: Carayon P, ed. Handbook
of Human Factors and Ergonomics in Health
Care and Patient Safety. 2nd ed. Boca
Raton, FL: Taylor & Francis; 2012.
119.
Smith TJ, Schoenbeck K, Clayton S.

Macroergonomics of an open bay to a
private room NICU design transition. In:
Pikaar RN, Koningsveld EAP, Settels PJM,
eds. Proceedings of the IEA2006 Congress.
Amsterdam, The Netherlands: Elsevier;
2006.
120.
Wears RL, Perry SJ. Human factors and

ergonomics in the emergency department.
In: Carayon P, ed. Handbook of Human
Factors and Ergonomics in Health Care and
Patient Safety. Mahwah, NJ: Lawrence
Erlbaum Associates; 2007:851-863.
129.
Weinger MB. Human factors in

anesthesiology. In: Carayon P, ed.
Handbook of Human Factors and
Ergonomics in Health Care and Patient
Safety. 2nd ed. Boca Raton, FL: Taylor &
Francis; 2012.
121.
Wears RL, Woloshynowych M, Brown R,

Vincent CA. Reflective analysis of safety
research in the hospital accident &
emergency departments. Applied
Ergonomics. 2010;41(5):695-700.
130.
Weinger MB, Slagle J. Human factors

research in anesthesia patient safety. Journal
of the American Medical Informatics
Association. 2002;9(Suppl 6):S58-S63.
131.
122.
Bisantz AM, Pennathur PR, Guarrera TK, et

al. Emergency department status boards: A
case study in information systems transition.
Journal of Cognitive Engineering and
Decision Making. 2010;4(1):39-68.
Gaba DM, Howard SK, Jump B. Production

pressure in the work environment. California
anesthesiologists attitudes and experiences.
Anesthesiology. 1994;81(2):488-500.
132.
Wiegmann DA, Eggman AA, ElBardissi

AW, Parker SH, Sundt Iii TM. Improving
cardiac surgical care: A work systems
approach. Applied Ergonomics.
2010;41(5):701-712.
133.
Wiegmann DA, Duff S, Blocker R. In search

for surgical excellence: a work system
approach. In: Carayon P, ed. Handbook of
Human Factors and Ergonomics in Health
Care and Patient Safety. 2nd ed. Bocs Raton,
FL: Taylor & Francis; 2012.
134.
Gawande A. The Checklist Manifesto. New

York, NY: Metropolitan Books; 2009.
135.
Xiao Y, Fairbanks RJ. Speaking systems

engineering: Bilingualism in health care
delivery organizations. Mayo Clinic
Proceedings. 2011;86(8):719-720.
136.
Gopher D. Why is it not sufficient to study

errors and incidents: Human factors and
safety in medical systems. Biomedical
Instrumentation and Technology.
2004;38(5):387-409.
137.
Henriksen K. Human factors and patient

safety: Continuing challenges. In: Carayon
P, ed. Handbook of Human Factors and
Ergonomics in Health Care and Patient
Safety. Mahwah, NJ: Lawrence Erlbaum
Associates; 2007:21-37.
138.
Meister D, Enderwick TP. Human Factors in

System Design, Development, and Testing.
Mahwah, N.J.2001.
139.
Vicente KJ. What does it take? A case study

of radical change toward patient safety. Joint
Commission Journal on Quality and Safety.
2003;29(11):598-609.
123.
Scanlon M. Human factors and ergonomics

in pediatrics. In: Carayon P, ed. Handbook
Care and Patient Safety. Mahwah, NJ:
Lawrence Erlbaum Associates; 2007:865882.
124.
Scanlon MC, Karsh B, Densmore E. Human

factors and pediatric patient safety. Pediatric
Clinics of North America. 2006;53:11051119.
125.
Or CKL, Karsh B-T, Severtson DJ, Burke

LJ, Brown RL, Brennan PF. Factors
affecting home care patients acceptance of
a web-based interactive self-management
technology. Journal of the American
Medical Informatics Association.
2011;18(1):51-59.
126.
127.
128.
Zayas-Caban T, Brennan PF. Human factors

in home care. In: Carayon P, ed. Handbook
Care and Patient Safety. Mahwah, NJ:
Lawrence Erlbaum Associates; 2007:883897.
Zimmerman D, Antonova E. Human factors
and ergonomics in nursing home care. In:
Beasley JW, Escoto KH, Karsh BT. Human
factors and ergonomics in primary care. In:
349
140.
141.
142.
143.
Carayon P, Xie A. Decision making in

healthcare system design: When human
factors engineering meets health care. In:
Proctor RW, Nof SY, Yih Y, eds. Cultural
Factors in Systems Design: Decision
Making and Action. Boca Raton, FL: CRC
Press; 2011.
Research and Quality;2010.
Shortell SM, Singer SJ. Improving patient
safety by taking systems seriously. Journal
of the American Medical Association.
2008;299(4):445-447.
Safren MA, Chapanis A. A critical incident
study of hospital medication errors - Part 1.
Hospitals. 1960;34:32-34;57-66.
350
144.
Safren MA, Chapanis A. A critical incident

study of hospital medication errors - Part 2.
Hospitals. 1960;34(53):65-68.
145.
Chapanis A, Safrin MA. Of misses and

medicines. Journal of Chronic Diseases.
1960;12(4):403-408.
146.
Leape L. Human factors meets health care:

The ultimate challenge. Ergonomics In
Design. 2004;12(3):6-12.
147.
Pronovost PJ, Goeschel CA, Marsteller JA,

Sexton JB, Pham JC, Berenholtz SM.
Framework for patient safety research and
improvement. Circulation. 2009;119(2):330337.
148.
Carayon P, Hundt AS, Wetterneck TB.

Nurses acceptance of Smart IV pump
technology. International Journal of Medical
Informatics. 2010;79:401-411.
149.
Leape L. Smart pumps: A cautionary tale

of human factors engineering. Critical care
medicine. 2005;33(3):679-680.
Chapter 32. Promoting Engagement by Patients and Families

To Reduce Adverse Events (NEW)
Zack Berger, M.D., Ph.D.; Tabor Flickinger M.D.; Sydney Dy, M.D., M.Sc.

Patient-centeredness is now widely recognized as a central aspect of health care, including
hospital care. Each patient has unique needs that should be addressed by each hospital in order to
improve safety and quality.1 Through patient engagement in their own safety, they and their
families can help prevent adverse events.2-5 Such involvement is promoted by several
international organizations, and educational materials have been developed to facilitate patient
engagement in safety practices.6 In order to evaluate how patient engagement is being
implemented and the effectiveness of this safety practice, we performed a systematic review of
Medline, CINAHL, Embase, and Cochrane from 2000-2011, with a variety of synonyms for
patient engagement and patient safety, including physician-patient relations, patient participation,
and patient-centered care.

Compared with other patient safety practices, promoting patient and family engagement does
not lend itself to a precise definition as easily. Engagement can be seen as an umbrella term
incorporating various approaches rather than a specific process, team, or technology. In general,
definitions seem to center on patient and family participation in care, whether from the point of
view of humanism, consumer rights, or care coordination, being used to encourage the patient to
be active in reporting adverse events.7,8
For that reason, patient and family engagement can be understood as a patient safety practice
in various ways, not all of which are included in this chapter. First, patient engagement can be
approached as an overarching philosophy applicable to a number of patient safety practices
including reducing patient risk of suicide (Chapter 26) and improving care transitions at
discharge (Chapter 37).
Secondly, patient engagement can be understood as an implementation in its own right. Few
patient safety interventions are implemented with the sole primary goal of promoting patient and
family engagement. For example, it is relatively common for another PSP, such as Rapid
Response Systems or Rapid Response Teams (RRT), to be implemented with the primary goals
of improving care quality and safety. Promotion of patient engagement may be a secondary goal,
and data regarding the change in patient engagement after implementation of the RRT
intervention may not be reported. Also, in some patient safety interventions, patient engagement
may be treated as a contextual variable that may moderate the efficacy of the intervention.
Although engagement can be challenging to define, this review focused on the effectiveness
of interventions intended primarily to elicit patient or family involvement in reducing the
incidence of adverse patient safety events. In addition, patient/family engagement was examined
as part of the implementation of selected patient safety practices with other primary goals (eg.
RRT interventions).
351

Schwappach2 provided a conceptual framework for patient engagement based on the Theory
of Planned Behavior, which emphasizes the importance of beliefs and attitudes in creating
intentions and changing actual behavior. Since patients are the only members of the treatment
team who are always (theoretically) present at every treatment and visit, they provide important
information that may not be available from other sources, such as medical records. In addition,
many patients prefer to be involved in their care in general, which may also apply to the safety
and quality of care.2 Relating to this, patients have also been found to be highly motivated to
decrease the risk of harm and ensure good outcomes.9 Finally, since many safety problems occur
at the bedside and can be observed and potentially prevented by patients, they are both an
important source of information on potential problems and a potential mechanism for improving
safety.7

The systematic review of the literature resulted in 4,107 unique articles that were potentially
relevant to this topic. English-language studies from the U.S., UK, Canada, and Australia were
included in the present review, due to potentially significantly different cultural issues in patient
engagement in their health care outside of these countries, as well as potential differences in
tools for promoting engagement. We included studies that focused on hospital care settings only
(e.g., intensive care units), because we felt that patient engagement in safety in the home setting
would be difficult to differentiate from patient self-management of their medications and care,
when providers are not present. Finally, only systematic reviews focusing on effectiveness and
prospective, controlled studies were included.
A total of 4,061 of these articles were excluded during abstract screening, leaving 46 for full
article review. Of these 46 articles, 43 were excluded, leaving three articles that met inclusion
criteria for studies evaluating the effectiveness of interventions focusing on patient engagement
in patient safety in the hospital setting, focusing on the patient safety practices that are addressed
in other chapters in this report. We identified one systematic review of patient and family
engagement in safety.10 The authors found limited evidence, of poor quality, for the benefit of
patient involvement in patient safety, and found that the available studies were mostly concerned
with patient management of medications.7,11,12We identified three studies that evaluated the
impact of interventions for patient/family engagement in patient safety in the hospital setting
(Table 1).
One study13 was a randomized controlled trial (N=209) of an intervention to provide patients
with a personalized medication list. Both intervention and control groups were provided with
general education about drug safety. Measurements to determine incidence of adverse drug
events and close-calls included patient surveys and identification of incidents through interviews
of pharmacists, interviews of housestaff, and electronic medical record review. The study was
conducted in a teaching hospital without computerized physician order entry. This study found
no statistically significant benefits of the intervention compared with the control group: in 1,053
total patient-days at risk, the adverse drug event rate was 8.4 percent in the intervention group
and 2.9 percent in the control group (p=0.12). The close-call rate also showed no significant
change, 7.5 percent versus 9.8 percent (p=0.57).
McGuckin and colleagues14 conducted a pre-post study for hand hygiene among 35 patients
located on an inpatient rehabilitation unit in an acute-care hospital. Using a patient education
352
model, a 6-week study with a 3-month followup was conducted. During the intervention, patients
agreed (and were encouraged) to ask all health care workers with direct contact if they had
washed/sanitized their hands. Use of soap and sanitizer per resident-day was measured before,
during, and after the intervention. While the intervention itself was not multifaceted, multiple
methods were used to encourage patients to ask the handwashing question of their providers: a
visit by a premedical student with the patient to discuss hand hygiene (HH); an education
brochure; and multiple prompts for patients to ask their providers, including videos and visual
aids. Hand washing or sanitizing increased from five HH uses per resident-day during the
intervention to 9.7 HH per resident-day during the intervention (p<0.001), 6.7 HH per residentday post-intervention (at 6 weeks) (p<0.001) and 7.0 HH per resident day at 3 months (p<0.001).
Patients asked their physicians about hand hygiene 40 percent of the time, and their nurses 95
percent of the time.
Stone and colleagues15 carried out a pre-post study among 187 acute-care hospitals in the
National Health System of the United Kingdom that included patient engagement as part of a
multifaceted HH intervention. A HH campaign was introduced to these hospitals over a period of
7 months, including alcohol hand rub near the patients bedside, regularly changed wall posters
in the wards regarding HH, and materials telling patients to ask their providers to clean their
hands. Median use of alcohol hand rub increased statistically significantly in the participating
hospitals during the intervention period, from seven to 13 ml per bed-day of alcohol hand rub
over 6 months (p<0.001); the change in soap use was not statistically significant (p=0.06). Rate
of hospital-acquired infections did not change.
Because of the small number of studies meeting our criteria, and their heterogeneity, we
could not perform evidence grading.
Table 1, Chapter 32. Patient engagement in safety: effectiveness studies
Description of PSP
Outcomes: Benefits*
Multiple Interventions or
Multifaceted
Author, Year
Interventions
13
Weingart 2004
Proving patients with
personalized medication
list to help prevent
medication errors
14
McGuckin 2004
Asking all health care
workers who had direct
contact with them, Did you
wash/sanitize your hands?
15
Stone 2007
Instructing patients to ask
health care workers to
clean their hands.
*Statistical significance, Y (Yes) or N (No)
Study Design
Sample Size
RCT
Adverse drug event rate: N
209
Close-call rate: N
Pre-post
Hand hygiene per resident

day: Y
35
Pre-post
187
Total alcohol hand rub and

soap-use per patient day:
Y

None of the studies included in this review evaluated harms of interventions or surveys.
Interventions to increase patient engagement, such as reminding health care workers to wash
their hands, could theoretically adversely affect provider-patient relationships and patients trust
in their providers. Patients might fear adverse consequences, or health care providers could
become overly reliant on patient engagement and more lax in their own safety practices.6
353

What Contexts?
Patient engagement is part of several key organizations approach to patient safety. The
World Health Organization provides educational materials for patients, and The Joint
Commission National Patient Safety Goals include the Speak Up campaign to engage patients
in preventing wrong-site surgery. Bergal and colleagues assessed patients reliability in regard to
marking the site of planned surgery and found only partial compliance (68%). A review by
McGuckin and colleagues16 assessed the importance of patient role (which they term patient
empowerment) in HH interventions. Three of the cited studies (themselves authored by
McGuckin and colleagues) showed that, while 80 to 90 percent of patients reported willingness
to ask their health care workers to wash their hands, 60 to70 percent of patients actually did so.
Because of the paucity of literature, the authors were unable to conduct a systematic review.
Patient participation in safety practices may be influenced by societal norms and the health care
environment, including whether the organizational culture supports patients participation.7
Patient engagement interventions have been applied to a number of individual patient safety
practices addressed elsewhere in this report. These practices include hand hygiene, rapid
response teams, surgical site marking, and falls. Examples of implementation studies in patient
engagement are described below.
Patient Engagement in Implementation of Hand Hygiene Interventions

A review by McGuckin and colleagues17 addressed patient empowerment as an approach to
motivating strategies in hand hygiene (HH) interventions (Table 2). As summarized in Chapter 8
(Hand Hygiene Compliance) above, the authors estimated from the literature the proportion of
patients who stated their desire to be empowered, or engaged, in reminding health care workers
to wash their hands. However, the proportion of patients who endorsed readiness for engagement
(or empowerment) was not always congruent with the proportion of patients who asked their
health care workers about HH. The authors identify several barriers discussed in the literature to
patients activating their engagement, emphasizing the negative social reaction that patients
might feel when asking their HCWs about HH. Finally, McGuckin and colleagues emphasize
that the literature on patient empowerment in HH is lacking estimates of effectiveness.
Table 2, Chapter 32. Hand-hygiene intervention studies
Author, Year
17
McGuckin, 2011
Main Study Objective

Review of patient empowerment motivating
strategies in hand hygiene intervention
Tools: educational tools, motivation
and reminder tools, and role
modeling
Facilitators/barriers: Social barrier of
patient to confront health care
workers; Lack of evidence of
effectiveness
Patient Engagement in Implementation of Rapid Response Team

Interventions
Three studies considered patient engagement in the context of Rapid Response Teams (RRT)
(Table 3). Ray and colleagues18 implemented a pediatric RRT based on direct family activation,
an approach developed to empower family members to seek help when serious concerns arise.
The direct family activation itself was a direct telephone number to reach the RRT, which
354
families could reach from any room in the hospital. In addition, families were educated via
posters in patients rooms and flyers. In-person family awareness surveys assessed families
awareness of the family activation approach. Nurses were trained in how to explain the RRT
activation to families. They were also given reminders in the electronic medical record and given
feedback on levels of family awareness from the surveys. After implementation of family
activation, the number of RRT calls per 1,000 discharges increased from 16 to 24, though no
statistical tests were employed to assess significance.
Dean and colleagues19 described a similar early warning system that empowers patients and
families to serve as an additional line of patient-safety defense by integrating them into the RRT
system of a major childrens hospital. This study specified the conditions under which patients or
families were encouraged to call the RRT, including a noticeable medical change in a patient that
had been unaddressed; a breakdown in care or uncertainty regarding treatment; the
administration of a medication that causes an adverse effect or that the patient/family believed
had not been sufficiently explained; or the provision of treatment that the patient or family
believed was meant for another patient or contravened their doctors wishes. Apart from the
criteria under which the alert system was to be activated, this study did not detail how patients or
families were to be empowered or educated to overcome any barriers to using the system. From
September 2005 through August 2007, the early warning system responded to 42 calls from
patients and parents; the authors state that the root cause for all calls was miscommunication
between patient and provider.
Gerdik and colleagues20 studied the implementation of a family- and patient-activated rapid
response team in an acute care hospital. Pickers Eight Dimensions of Patient-Centered Care
provided the conceptual framework, including the involvement of family of friends, which
includes involving family in decision-making. However, no explicit attention to patient or
family engagement was otherwise given. Implementation of the family-/patient-activated RRT
involved written educational materials, informational signs, instructional labels for telephones,
and scripted education and training on the part of staff. After RRT activation, patient and family
satisfaction were assessed. Following implementation of the RRT, codes decreased significantly
outside the ICU, from 25/month to 17/month. Patients and families alike were found to be
satisfied with the RRT.
These studies of family and patient engagement in RRT evaluated implementation and did
not explicitly set out to evaluate effectiveness. However, in the study by Gerdik and colleagues20,
after implementation of the RRT, codes outside the ICU decreased, potentially representing a
measure of effectiveness.
355
Table 3, Chapter 32. Rapid response team intervention studies

Author, Year
18
Ray, 2009
Dean, 2008
19
Gerdik, 2010
20

To implement a pediatric RRT based on direct
family activation
To integrate patients and families into an RRS at a

childrens hospital
To implement a patient- and family-activated RRT

at an acute care hospital
Tools: direct telephone number to
reach the RRT which families could
reach from any room in the hospital,
posters, flyers
Staff/education: mock script to help
medical team discuss RRT activation
with patients/families
Facilitators/barriers: physicians
concerned that their role would be
undermined; providers
understanding of RRT as extension
of care they already provide
Tools: telephone number to activate
RRT available to patients/families 24
hours, 7 days a week
Staff/education: explanation by
admitting units nurse to patient and
family, reinforced by video and
brochure
Facilitators/barriers: leadership,
provider involvement
Tools: dedicated phone line
Staff/education: patient and family
education
Facilitators/barriers: concern that
resources would be overwhelmed;
endorsement of hospital
administration, physicians, and staff
Patient Engagement in Implementation of Fall Interventions

Chapter 19 addresses patient safety practices to reduce the incidence of in-facility falls.
While multi-modal interventions were found throughout the literature, including those
incorporating patient education, none of the studies reviewed specifically address patient
engagement.
Two studies were identified in the search for this review (Table 4). Krauss and colleagues21
implemented an educational intervention to reduce patient falls according to a quasiexperimental design among nursing staff, nursing secretaries, and patient care technicians in an
academic hospital. While patient or family activation or engagement were not mentioned
specifically as part of the implementation or its conceptual background, nurses were directed to
educate all patients in fall prevention. For patients with high risks of falling, nursing staff were
instructed to reinforce falls-prevention education with both patients and family. Staff received
feedback on fall rates on their unit during the implementation via meetings and flyers. The
nursing staffs knowledge and use of prevention strategies improved. The incidence of inhospital falls decreased for 5 months but the decrease for the full 9-month intervention period
was not significant.
van Gaal and colleagues22 implemented a multi-component intervention to reduce the risk of
pressure ulcers, falls, and urinary tract infections in ten wards from four hospitals and ten wards
from six nursing homes. Patient involvement was conceived as part of the intervention and was
included because it can enhance the implementation of innovations or improvements. Oral and
written information was given to patients at risk for specific adverse advents. Implementation on
every participating ward included educational meetings for nurses and informational brochures
356
for patients at risk for any one of the adverse events addressed. Fewer falls per patient week were
found the participating hospital wards and nursing homes, but the study was not powered to
determine the statistical significance of the decreased incidence of particular adverse events.
Table 4, Chapter 32. Falls prevention intervention studies
Author, Year
21
Krauss, 2008
van Gaal, 2011

To implement an educational intervention among
nursing staff, nursing secretaries, and patient care
technicians to reduce falls in an academic hospital
22
To implement a multi-component intervention,

including patient involvement, to reduce the risk of
pressure ulcers, falls, and urinary tract infections in
ten wards from four hospitals and ten wards from
six nursing homes
Tools: educating all patients and
families in fall prevention, patient
pamphlets
Staff/education: Nurses,
patient care technicians, and unit
secretaries all took part in education
modules
Facilitators/barriers: staff
turnover; high patient-to-nurse ratios;
high patient turnover
or high patient volume; competing
demands on nursing staff; lack of
buy-in from staff
Tools: education, patient involvement
and feedback on process and
outcome
Staff/education: Key nurses on each
unit implemented small-scale
educational program, two case
discussions on every ward, and
distributed a CD-ROM with
educational material
Facilitators/barriers: complexity of
intervention
Patient Engagement in Implementation of Surgical Checklist

Interventions
Most studies of inventions to prevent wrong-site surgery have focused on checklists for
surgeons or anesthesiologists to perform prior to surgery.23-25 Although patient interaction may
be part of the checklist, such as verbally verifying patient identity and surgical site,23 the
provider team is the target of the intervention. Only two studies have examined patient
engagement as a means to avoid wrong-site surgery (Table 5). One study, in the setting of a
private foot-and-ankle practice, gave patients written instructions to mark the limb not to be
operated on with the label NO and observed patient compliance of 59 percent on the day of
surgery.26 The other study, in the setting of a university-affiliated orthopedic practice, gave
patients both verbal and written instructions to mark the intended surgical site with the label
YES and provided a marking pen to do so.27 Patient compliance in this study was 68.2 percent,
with higher compliance in patients whose primary language was English and whose surgery
occurred sooner after instructions were given.
357
Table 5, Chapter 32. Surgical checklist intervention studies

Author, Year
27
Bergal, 2010
DiGiovanni, 2003

To investigate patient compliance in marking
surgical site
26
To investigate patient compliance in marking

surgical site
Tools: verbal and written instructions
to mark surgical site, marking pen
provided; assessment for compliance
on day of surgery
Barriers/facilitators: patients primary
language, cultural tendency to rely
on physicians, younger patient age,
time between enrollment and surgery
Tools: written instructions to mark
limb NOT to be operated on;
assessment for compliance on day of
surgery
Patient Engagement in Implementation of Care Transition

Interventions
Patient engagement is defined in a variety of ways. Depending on the definition, the practical
implications can be broader or more specific. While in the rest of this chapter we address
interventions centered on patient engagement as an independent element and on interventions
where patients assume a primary role in patient safety, an additional important and valuable
route to patient engagement is to encourage patient activation in an existing intervention, as one
part of a larger approach. Patient engagement in transitional care is an important example of this
broader approach.
Interventions to improve transitional care at the time of hospital discharge are examined in
Chapter 37. Patient engagement is one of many aspects of these patient safety practices. Predischarge interventions may include patient engagement in the form of patient and/or caregiver
education. Post-discharge interventions may include outreach to patients and/or caregivers by
means of follow-up phone calls or other methods. Bridging interventions may include a
combination of these and other components.
Only one intervention, the Care Transitions Intervention (CTI), had been implemented and
evaluated in multiple settings.28-32 The CTI is designed to provide patients and caregivers with
the tools and skills to take a more active role in their care. It is based on four pillars of
medication self-management: a patient-centered record, follow-up, and identification of red
flags with instructions on how to respond to them. Patients and care-givers received in-hospital
visits, telephone calls, home visits, encouragement to take an active role in care, and guidance
from a transition coach. Efficacy studies of CTI have shown reduced rates of readmission in
clinical trials set in a not-for-profit capitated delivery system28,29and a Medicare fee-for-service
system.33 Assessment of patient engagement as the mediator of reduced readmission have shown
that patients receiving CTI reported high levels of confidence in self-management, understanding
warning symptoms of worsening condition, ability to obtain needed information during followup visits, and understanding of how to take their medications.29
Implementation studies of CTI in real-world settings have also shown reduced
readmissions for patients who received coaching compared with those who did not.
Implementation issues included the training and time commitment of transition coaches and the
challenge of recruiting and retaining patients in the intervention.31 Studies that directly addressed
sustainability emphasized the importance of leadership, hospital-community partnerships,
tailoring to the needs of diverse communities and particular patient subgroups, and resource
allocation (staff and funding) as important factors in implementation. Mean patient activation
scores were moderately higher for sites with full sustainability plans than for sites with partial or
358
minor plans, suggesting that greater engagement in the program at the site level could affect
engagement by patients receiving the intervention.32 Qualitative data indicate that patient
perceptions of a caring relationship with transition coaches foster greater patient engagement in
the program, with implications for staff training.33

None of the reviewed studies directly evaluated the costs or cost-effectiveness of practices
designed to promote patient or family engagement with safety.

As noted above, McGukin14 recorded the frequency with which patients asked various
members of the care team about their HH practices; however, this outcome was not linked to any
model about how the outcome might be affected by the context in which the intervention was
implemented.

Patient and family engagement is an emerging area in patient safety research, with few
published effectiveness studies. However, it is an important part of key organizations patient
safety initiatives, and a number of recent studies have described implementation approaches and
challenges. Future work must address basic and applied concerns across the spectrum of
conceptual foundations and experimental design, including the research questions that need to be
answered: the definition and measurement of patient and family engagement; the safety
endpoints that should be addressed; and methodological issues around study design.
Also important to address in future work is the variety of approaches that have been taken to
promoting patient engagement, whether as an independent intervention or as part of an
intervention focused on an existing patient safety practice. Distinguishing the features of
instrumental patient engagement and independent patient engagement interventions will help
clarify the nature of patient engagement as a patient safety practice. A summary table is located
below (Table 6).
Scope of the
Strength of
the PSP
Effectiveness
Common
Emerging
practice (few
studies
available)
Evidence or
Potential for
Harmful
Unintended
Consequences
Uncertain
Estimate of
Cost
Low
How Much do We
Little/Moderate
References
1.
2.
The Joint Commission: Advancing Effective

Communication, Cultural Competence, and
Patient- and Family-Centered Care: A
Roadmap for Hospitals. Oakbrook Terrace,
IL: The Joint Commission, 2010.
359
Davis RE, Jacklin R, Sevdalis N, Vincent

CA. Patient involvement in patient safety:
what factors influence patient participation
and engagement? Health Expect 2007;
10(3):259-67.
3.
Entwistle VA. Differing perspectives on

patient involvement in patient safety. Qual
Saf Health Care 2007; 16(2):82-3.
4.
Koutantji M, Davis R, Vincent C, and

Coulter A. The patients role in patient
safety: engaging patients, their
representatives, and health professionals.
Clinical Risk 2005;11:99-104.
5.
Vincent CA, Coulter A. Qual Saf Health

Care: Patient safety: what about the patient?
2002; 11:76-80.
6.
Perneger T. The Council of Europe

recommendation Rec(2006)7 on
management of patient safety and
prevention of adverse events in health care.
Int J Qual Health Care 2008; 20(5):305-7.
7.
Schwappach DL. Med Care Res Rev:

Review: engaging patients as vigilant
partners in safety: a systematic review.
2010; 67:119-48.
8.
Longtin Y, Sax H, Leape LL, Sheridan SE,

Donaldson L, Pittet D. Patient participation:
current knowledge and applicability to
patient safety. Mayo Clin Proc. 2010
Jan;85(1):53-62.
9.
Lyons M. Qual Saf Health Care: Should

patients have a role in patient safety? A
safety engineering view. 2007; 16:140-2.
10.
Hall J, Peat M, Birks Y et al. Qual Saf

Health Care: Effectiveness of interventions
designed to promote patient involvement to
enhance safety: a systematic review. 2010;
19:e10.
11.
12.
13.
Peat M, Entwistle V, Hall J, Birks Y, Golder

S. J Health Serv Res Policy: Scoping review
and approach to appraisal of interventions
intended to involve patients in patient safety.
2010; 15 Suppl 1:17-25.
King A, Daniels J, Lim J, Cochrane DD,
Taylor A, Ansermino JM. Qual Saf Health
Care: Time to listen: a review of methods to
solicit patient reports of adverse events.
2010; 19:148-57.
Weingart SN, Toth M, Eneman J et al. Int J
Qual Health Care: Lessons from a patient
partnership intervention to prevent adverse
drug events. 2004; 16:499-507.
360
14.
McGuckin M, Taylor A, Martin V, Porten L,

Salcido R. Am J Infect Control: Evaluation
of a patient education model for increasing
hand hygiene compliance in an inpatient
rehabilitation unit. 2004; 32:235-8.
15.
Stone S, Slade R, Fuller C et al. Journal of

Hospital Infection: Early communication:
Does a national campaign to improve hand
hygiene in the NHS work? Initial English
and Welsh experience from the NOSEC
study (National Observational Study to
Evaluate the CleanYourHandsCampaign).
2007; 66:293-6.
16.
McGuckin M, Waterman R, Storr IJ et al.

Evaluation of a patient-empowering hand
hygiene programme in the UK. J Hosp
Infect 2001; 48(3):222-7.
17.
McGuckin M, Storr J, Longtin Y, Allegranzi

B, Pittet D. Patient empowerment and
multimodal hand hygiene promotion: a winwin strategy. Am J Med Qual 2011;
26(1):10-7.
18.
Ray EM, Smith R, Massie S et al. Family

alert: implementing direct family activation
of a pediatric rapid response team. Jt Comm
J Qual Patient Saf 2009; 35(11):575-80.
19.
Dean BS, Decker MJ, Hupp D, Urbach AH,

Lewis E, Benes-Stickle J. Condition HELP:
a pediatric rapid response team triggered by
patients and parents. J Healthc Qual 2008;
30(3):28-31.
20.

Wludyka PS, Panni MK. Successful
implementation of a family and patient
activated rapid response team in an adult
level 1 trauma center. Resuscitation 2010;
81(12):1676-81.
21.
Krauss MJ, Tutlam N, Costantinou E,

Johnson S, Jackson D, Fraser VJ.
Hosp Epidemiol 2008; 29(6):539-45.
22.
van Gaal BG, Schoonhoven L, Mintjes JA et

al. Int J Nurs Stud: Fewer adverse events as
a result of the SAFE or SORRY?
programme in hospitals and nursing homes.
Part I: Primary outcome of a cluster
randomised trial. 2011; 48:1040-8.
23.
Garnerin P, Ares M, Huchet A, Clergue F.

Health Care 2008; 17(6):454-8.
29.
Coleman EA, Parry C, Chalmers SA, Chugh

A, Mahoney E. The central role of
performance measurement in improving the
quality of transitional care. Home Health
Care Serv Q 2007; 26(4):93-104.
24.
Nilsson L, Lindberget O, Gupta A, Vegfors

M. Implementing a pre-operative checklist
to increase patient safety: a 1-year follow-up
of personnel attitudes. Acta Anaesthesiol
Scand 2010; 54(2):176-82.
30.
25.

surgery - Are we up to standard?
CentEurJMed 2010 Aug;5(4):407-10.
Parry C, Min SJ, Chugh A, Chalmers S,

Coleman EA. Further application of the care
transitions intervention: results of a
randomized controlled trial conducted in a
fee-for-service setting. Home Health Care
Serv Q 2009; 28(2-3):84-99.
31.
Voss R, Gardner R, Baier R, Butterfield K,

Lehrman S, Gravenstein S. The care
transitions intervention: translating from
efficacy to effectiveness. Arch Intern Med
2011; 171(14):1232-7.
32.
Parrish MM, OMalley K, Adams RI,

Adams SR, Coleman EA. Implementation of
the care transitions intervention:
sustainability and lessons learned. Prof Case
Manag 2009; 14(6):282-93; quiz 294-5.
33.
Parry C, Kramer HM, Coleman EA. A

qualitative exploration of a patient-centered
coaching intervention to improve care
transitions in chronically ill older adults.
Home Health Care Serv Q 2006; 25(3-4):3953.
26.
DiGiovanni CW, Kang L, Manuel J. Patient

compliance in avoiding wrong-site surgery.
J Bone Joint Surg Am 2003; 85-A(5):815-9.
27.
Bergal LM, Schwarzkopf R, Walsh M,

Tejwani NC. Patient participation in surgical
site marking: can this be an additional tool
to help avoid wrong-site surgery? J Patient
Saf 2010; 6(4):221-5.
28.
Coleman EA, Smith JD, Frank JC, Min SJ,

Parry C, Kramer AM. Preparing patients and
caregivers to participate in care delivered
across settings: the Care Transitions
Intervention. J Am Geriatr Soc 2004;
52(11):1817-25.
361
Chapter 33. Promoting a Culture of Safety

Sallie J. Weaver, Ph.D.; Sydney Dy, M.D., M.Sc.; Lisa H. Lubomski, Ph.D.; Renee Wilson, M.S.

A culture of safety has been suggested as a core mechanism underlying safe, effective, and
timely patient care. It has been implicated as a critical factor underlying continuous learning and
effective teamwork, as well as a key driver of safety behaviors such as error reporting, and safety
outcomes such as reduced adverse events.1,2
A number of studies have found associations between culture and safe care practices, such as
error reporting1,3-5 Other studies have found associations between patient safety culture and
patient outcomes, including reduced adverse event indicies6 and mortality.7,8 For example,
several studies have found relationships between safety culture and the AHRQ Patient Safety
Indicators.9,10 In one study that utilized a composite of 12 AHRQ patient safety indicators results
suggested that a 1 standard deviation increase in patient safety culture scores was associated with
a 10% decrease in the composite PSI risk.10 Other work has indicated that culture can account for
up to 6% of the variance in adverse events and 18% of the variance in patient willingness to
recommend a hospital to family and friends.11

Compared with other patient safety practices, interventions to promote patient safety culture
(PSC) are less easily defined. Effective safety cultures have been described as those in which
there is shared commitment to safety as the highest priority, where engaging in safety-promoting
behaviors is encouraged and reinforced by leaders and peers, and where glitches or near misses
are valued as opportunities for learning and improvement. Therefore, interventions to promote
safety culture include a broad range of interventions rooted in principles of promoting leadership,
creating effective teamwork, and behavior change rather than a single specific process, team, or
technology. For example, interdisciplinary rounding12,13 and executive walkrounds,14,15 as well as
interventions designed to enhance provider communication,16 to encourage error reporting,14,17
and team training18,19 have all been labeled as interventions to promote a culture of safety. In
the present review, for example, 11 studies described multi-faceted interventions (i.e.,
interventions or practices with multiple components or aspects).13,14,16,19-26
Precise definition is further complicated given that few patient safety interventions are
implemented with the sole primary goal of promoting safety culture. For example, it is relatively
common for another PSP, such as a Rapid Response System (RRS), to be implemented with the
primary goal of reducing code events or other negative patient outcomes. Promoting a culture of
safety may be a secondary goal of an RRS intervention, and data regarding the change in staff
perceptions of safety culture following implementation of the RRT intervention may be reported;
however, improving safety culture was not the primary stated goal of the intervention. In this
sense, some studies treat safety culture as a primary outcome variable, while others treat it as a
contextual variable that may moderate the efficacy of another PSP.
This review is specifically focused on studies in which the primary intervention goal was
explicitly to promote a culture of safety. We did not limit inclusion criteria based on a particular
362
type of intervention with the aim of identifying the full breadth of different practices being
described as interventions to promote safety culture.

Patient safety culture (PSC) is defined as a holistic snapshot of enacted norms, policies, and
procedures related to patient safety that guide the behaviors, attitudes, and cognitions of care
providers.27 In this sense, PSC is a social aspect of the work environment that shapes what
provides do, think, and feel during their day-to-day work activities. More concretely, PSC has
been described as a shared commitment to patient safety as the most important organizational
goal that provides cues to clinicians and staff about the relative priority of patient safety in
comparison to other unit or organizational goals. In this way, working in strong, positive safety
cultures motivates employees to behave in ways that support safety. In such work environments,
clinicians and employees feel a sense of obligation to speak up if they see a potential hazard, to
lend a hand to fellow team members, to ask for help if they need it, and believe that putting
patient safety first will be recognized and rewarded. Given its role as a motivational force that
helps to shape clinician behaviors and attitudes, and cognitions, PSC is important for
understanding issues of safety, care quality, error, and process improvement.
PSC is a facet-specific form of general organizational culture, meaning that is a specific form
of organizational culture that focuses on a narrowly defined aspect of performance, namely
patient safety. For example, general organizational culture refers to:
a pattern of shared basic assumptions learned by a group as it
solved problemswhich have worked well enough to be
considered valid and, therefore, to be taught to new members as
the correct way to perceive, think, and feel28
PSC therefore refers to the pattern of assumptions shared among members of a group (e.g., a
unit or organization) specifically related to patient safety and can be differentiated from general
organizational culture.29,30 More specifically, Sorra and Nieva31 cite the following the definition
of safety culture in their work dedicated to measuring patient safety culture:
the product of individual and group values, attitudes, perceptions,
competencies, and patterns of behavior that determine the
commitment to, and the style and proficiency of, an organizations
health and safety management. Organizations with a positive
safety culture are characterized by communications founded on
mutual trust, by shared perceptions of the importance of safety, and
by confidence in the efficacy of preventive measures.32
Definitions of safety culture also include a focus on employee safety27 and several authors
have started to develop theoretical models of patient safety culture.33-35 Overall though the
theoretical development of patient safety culture as a construct in the existing peer-reviewed
literature could be further developed and logic models underlying interventions to promote
culture are often not reported.
Safety Culture Versus Safety Climate?

Patient safety climate is a related term often used interchangeably with culture. Patient safety
climate refers specifically to provider perceptions of patient safety-related norms, policies, and
363
procedures that are shared among members of a group (care team, profession, unit, service,
department, organization, system). The difference between culture and climate is often reduced
to a difference in methodology, with studies involving surveys being categorized as measures of
climate and ethnographic studies that involve detailed, longitudinal observations being
categorized as studies of culture.
Practically, the differentiation between culture and climate is often viewed as a primarily
academic exercise. However, the dichotomy raises an important and very practical point. If you
are measuring safety culture using a survey, than you are measuring clinician and staff
perceptions of culture (i.e., safety climate) and therefore in order to change culture you have to
change perceptions of the relative priority of patient safety compared with other unit or
organizational goals, and it must be salient to providers that their actions and attitudes supporting
patient safety are actively reinforced by their peers and leaders. For example, it must be explicit
that patient safety comes first relative to other unit or organizational goals, such as efficiency,
and there must be visible recognition and positive outcomes related to engaging in safe
behaviors.
For the purposes of this review we included studies of both patient safety culture and patient
safety climate. In our discussion, we use the term patient safety culture to simplify the reporting
of results.
Measuring Patient Safety Culture

Patient safety culture has primarily been measured by patient safety climate surveys that
capture employee perceptions of social, technical, and environmental aspects of their workplace.
While the reviews of other patient safety practices focus on patient outcomes as the primary
dependent variable and culture as a contextual variable, the present review focuses on changes in
employee perceptions of patient safety culture as the primary dependent variable. We also
include concurrently reported patient outcomes.
At least five previous reviews have been dedicated to survey instruments used to measure
and assess patient safety culture.36-40 Their results indicate that the degree of psychometric
evidence for reliability and validity varies significantly among the surveys designed to measure
patient safety culture that have been published in the peer reviewed literature. Some of the
surveys with the greatest amount of psychometric evidence in the published literature to date
include the Hospital Survey on Patient Safety Culture,31 the Safety Attitudes Questionnaire,41
and the Patient Safety Climate in Healthcare Organizations survey.42 To ensure a foundational
level of psychometrically sound measurement, the inclusion criteria for this review required that
studies use a measure with evidence of reliability and validity available in the peer reviewed
literature.

The systematic review of the literature resulted in 2696 unique articles that were potentially
relevant to this topic. We excluded 2563 of these articles during abstract screening, leaving 133
for full article review. Of these 133 articles, we excluded 115, leaving 18 articles that addressed
the benefits of interventions to improve patient safety culture.
Our systematic review identified 18 primary studies dedicated to evaluating safety-oriented
interventions that were designed to promote a culture of patient safety, were conducted in an inpatient hospital, and measured patient safety culture/climate using a validated survey instrument.
364
We did not identify any prior systematic reviews specifically dedicated to interventions designed
to promote a culture of safety in health care.
Of the 18 studies reviewed, 12 were pre-post studies, 2 were cluster randomized control
trials, 2 were concurrent control studies, 1 was a pre-post with concurrent control, and 1 was a
quasi stepped-wedge design. Eleven studies measured patient safety culture/climate using the
Safety Attitudes Questionnaire (SAQ),41 4 studies used the AHRQ Hospital Survey on Patient
Safety (HSOPS),31 2 studies used the Patient Safety Climate in Healthcare Organizations survey
(PSCHO),42 and 1 study utilized the Safety Climate Scale (SCS).43 Though the primary method
of measuring patient safety culture/climate uses individual-level survey responses, culture is
considered a group-level phenomenon. Thus, the majority of studies operationalized culture at
the unit level of analysis. Three studies, however, operationalized culture at the hospital level of
analysis.15,24,44 Sample sizes returned ranged from 5461 individual responses nested within 144
units within a single hospital system to 28 individuals nested within a single unit. Response rates
ranged from 35% to 100% (see Summary Table and Evidence Tables).
The majority of interventions were multi-component interventions that combined several
improvement strategies under a single overarching initiative to promote a culture of safety. For
example, Belgen et al. 201019 utilized a three component approach that included team-training,
unit-based safety teams, and strategies for engaging patients in daily goal setting. Overall, 6
studies explicitly included teamwork and communication training and tools (e.g., structured
briefings or debriefings), 4 explicitly included some form of executive walk rounds, and 4
explicitly used a multi-component approach known as the Comprehensive Unit Based Safety
(CUSP) program (see Summary Table and Evidence Tables).
In terms of effectiveness, 9 of 18 (50%) reviewed studies reported a statistically significant
impact of the intervention on the overall culture score, the safety climate score, or on at least half
of reported survey items if analyzed at the item level. Several studies reported significant
improvements in teamwork climate, but did not find similar improvements in safety climate (see
Summary Table and Evidence Tables).20 None of the studies examining multi-component or
bundled interventions examined the relative effectiveness of individual intervention components.
Only one study directly compared the effectiveness of different interventions by comparing a
simulation-based team training intervention to didactic-only team-training intervention18. Results
indicated no change in safety culture survey scores for the didactic-only and control groups. An
increase in teamwork climate was reported for the simulation-training group; however, this
finding did not remain statistically significant after Bonferroni adjustment.
Seven studies also reported the impact of interventions on other outcomes, along with patient
safety culture. In terms of patient outcomes, one study that found significant improvements in
teamwork climate20 also found a significant decrease (0.56 vs. 0.15, p < .01) in the rate of
reported errors that resulted in patient harm after implementation of a multi-faceted suite of
interventions that included both cultural (e.g., feedback on errors in the form of posters and
emails, education and training) and system-focused changes (e.g., CPOE, medication
management protocols, changes to safety reports) implemented over 2.5 years. Another study
found that the number of rapid response system activations that led to code events decreased
from 29% to 22% following an intervention to promote safety culture in which paraprofessional
care providers learned how to utilize structured communication methods to communicate
changes in patient status. However, this difference was not tested statistically. Another study that
reported a marginal increase in teamwork climate18 also found that the experimental units
weighted adverse outcome score decreased by 37% after implementation of a team training
365
program designed to promote patient safety culture, compared with a 43% increase in a control
unit (p < .05). One study also provided descriptive information on reductions in nurse turnover
from 27% to 0% for two years following implementation of the CUSP program, but no statistical
analysis was reported (see Summary Table and Evidence Tables).26
The evidence to date suggests that several practices may help to promote a culture of safety;
however, methodologic issues related to variation in the practices studied and outcomes reported,
extremely small sample sizes, and lack of cluster randomized trials constrain the evidence for
intervention effectiveness available to date. Robust evaluations are needed that assess the impact
of practices to promote patient safety culture across multiple outcomes, based on theoretically
sound evaluation models. One previous review of interventions to promote safety culture also
noted methodological constraints in primary studies45 and one previous review dedicated to
strategies to improve culture concluded that there was no rigorous evidence available in the
current literature to demonstrate their effectiveness.46 While the criteria for the previous work
finding no studies that met inclusion criteria46 were more stringent than those utilized in the
present review, this conclusion supports our finding that the robustness with which interventions
to promote culture are studied, evaluated, and reported is in need of improvement.
In terms of grading, the strength of evidence for this topic was low. Risk of bias was
generally high due to study design issues - we identified only one true cluster RCT17 Major
issues affecting risk of bias for many studies included low response rates for surveys and
incomplete reporting (not reporting all units or hospitals where interventions were conducted,
and not reporting results for all parts of the culture survey but focusing only on those that were
statistically significant). Results were inconsistent: findings in half of the studies were not
statistically significant and significant findings were difficult to compare due to variations in
measurement methods. With respect to directness, the intervention was often not specifically
designed to improve patient safety culture, and actual patient safety outcomes were infrequently
reported. Finally, with respect to precision, a number of different survey instruments were used
and were often reported differently in the articles, so no conclusions could be drawn (see
Evidence Table on Risk of Bias).

None of the studies included in this review explicitly evaluated harms of culture
interventions or surveys.

What Contexts?
The effectiveness of the various methods for promoting a culture of safety likely varies based
upon the characteristics of the intervention and implementation processes. The interventions
reported in the 18 studies we reviewed differed in terms of the characteristics of the
organizations in which they were implemented, the level of leadership support and engagement
reported, and in the tools and strategies utilized to support implementation and to transfer the
intervention to daily care processes. Ten studies were conducted exclusively in academic
hospital settings, 3 studies were conducted in community based hospital settings, 4 studies
explicitly included a mix of academic and community hospitals, and several studies did not
explicitly address the hospital mix included in their sample. One study also reported that the gain
366
in safety culture scores was larger for faith-based hospitals; however statistical analyses of these
reported differences were not reported.23
This review also highlights the largely atheoretical nature of the research to date regarding
the development and implementation of interventions to promote patient safety culture. Only 3 of
the 18 reviewed studies (17%) noted any form of theoretical grounding for their improvement
and implementation strategy. Conceptual theories of patient safety culture and culture change are
foundational features of a clear logic model that guides intervention design and implementation.
Future studies should dedicate greater effort to developing and reporting the underlying
theoretical model driving intervention design and improvement efforts. Such models are a
critical element to furthering our understanding of the role context plays in moderating the
effects of various interventions to promote safety culture (Table 1).
Table 1, Chapter 33. Results of included studies on patient safety culture
Author, Year
20
Abstoss, 2011
Description of PSP
4 culture & 3 systemlevel interventions
Study Design
Pre-post
Adams24
Pizarro,2011
19
Blegen, 2010
Multi-component
intervention
Multi-component
intervention
Crisis resource
management training
Paraprofessionals
communication training
Pre-post
Outcomes: Benefits*
Culture survey: N
Teamwork: Y
Reported errors resulting in harm: Y
Overall reporting rate:
Culture survey: No statistical tests reported
Pre-post
Culture survey: Y
pre-post with
control hospitals
Pre-post
Culture survey: N
Multi-component
intervention
Executive walkrounds
Structured InterDisciplinary Rounds
Structured InterDisciplinary Rounds
Multiple interventions
Multi-component
Pre-post
Comprehensive UnitBased Safety Program
Quasi steppedwedge design
TeamSTEPPS training
Cluster RCT

Executive walkrounds
Pre-post
44
Cooper, 2008
Donahue, 2011
16
Edwards, 2008
47
15
Frankel, 2008
13
OLeary, 2010
OLeary, 2011
12
25
Paine, 2010
22
Pettker, 2009
21
Pettker, 2011
48
Pronovost, 2005
Riley, 2011
18
Sexton, 2011
23
Thomas, 2005
Tiessen, 2008
17
14
Multi-component
intervention
26
Timmel, 2010
*Overall results statistically significant Yes or no
Pre-post
Concurrent
control
Concurrent
control
Pre-post
Pre-post
Cluster RCT
Pre-post
Pre-post
367
Culture survey: No statistical tests reported

Use of structured communication: No
statistical tests reported
Rapid response events that led to code
events: No statistical tests reported
Culture survey: Y
Culture survey: Y
Culture survey: N
Teamwork: Y
Culture survey: Y
Teamwork: Y
Culture survey: Y
Culture survey: Y
Adverse outcomes: Y
Culture survey: Y
Nurse turnover: N
Length of stay: Y
Culture survey: N
Adverse outcomes: Y
Culture survey: Y
Culture survey: N overall; Y when analyzed
by exposure to intervention
Culture survey: N
Culture survey: Y
Nursing turnover: No statistical tests reported

None of the reviewed studies directly evaluated the cost-effectiveness of practices designed
to promote a culture of safety. However, one study attempted to examine how a multifaceted
intervention that included interdisciplinary rounding and regular interdisciplinary meetings
affected adjusted per-patient care costs. Compared with a control unit, adjusted costs of care
were reported as $24 less for intervention unit patients; however, the study was underpowered,
and this was not a statistically significant difference.13

None of the reviewed studies directly evaluated the effect of context on intervention
effectiveness; however, there is a clear need to determine the impact of contextual features in
promoting a culture of safety.

We found low strength of evidence that interventions to improve safety culture can improve
culture as an outcome, and insufficient evidence that interventions to improve culture can
improve patient safety outcomes, due to very few studies measuring these outcomes, and the
heterogeneity across interventions and types of safety outcomes reported. Although there is an
emerging evidence base dedicated to examining practices that promote a culture of safety, future
work must make large gains in robustness of experimental design and methodologies for
measuring culture in order to meaningfully advance our understanding of how to promote safety
culture effectively.
The evidence to date suggests that practices to promote patient safety culture may be
beneficial; however, evaluation designs and the rigor of available evidence do not support strong
causal conclusions. For example, the studies reviewed did not evaluate the mechanisms through
which these interventions impact culture. Additionally, most of the interventions reviewed had
multiple components, but none of the studies examined the incremental impact of each
component of the intervention. Future studies should strive to clearly evaluate the incremental
and differential impact of individual components of multifaceted intervention strategies.
Future research efforts should also aim to further our understanding of how providers
formulate their perceptions of safety culture. While there is a significant literature on the etiology
of safety culture (i.e., how individuals and groups formulate their perceptions of culture) in the
organizational sciences, little work has been done in health care to understand how providers
formulate their individual perceptions of safety culture and how these perceptions become shared
with others in their unit, department, or care team. Understanding how perceptions of patient
safety culture form and come to be shared among health care workers is a critical component of
understanding how to effectively promote and improve safety culture. There is a rich theoretical
and empirical literature from the organizational sciences that can be drawn upon to enhance the
strength of studies examining interventions to promote a culture of safety within health care.
Most important, this review underscores that the interventions designed to promote culture
need to be more rigorously tested and reported. Few studies to date have applied rigorous
evaluation designs or have clearly articulated critical aspects of study execution in peer reviewed
outlets. Findings from our review mirror those of other reviews45,46,49 that examined the
effectiveness of strategies to change organizational culture to improve health care performance.
Collectively results suggest some homogenous evidence for interventional strategies such as
368
team-training, executive engagement strategies, and unit-based improvement processes. They

also highlight the need for more rigorous evaluations of patient safety practices designed to
promote safety culture. A summary table is following (Table 2).
Scope of the
Strength of
the PSP
Effectiveness
Common/Low-to-high
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Uncertain
Estimate of
Cost
Low-tomoderate
(varies)
How Much do We
Moderate/Not difficult-toModerate (varies with

intervention)
References
1.
Braithwaite J, Westbrook MT, Travaglia JF,

Hughes C. Cultural and associated enablers
of, and barriers to, adverse incident
reporting. Qual Saf Health Care 2010;
19(3):229-33.
2.
Singer SJ, Gaba DM, Geppert JJ, Sinaiko

AD, Howard SK, Park KC. The culture of
safety: results of an organization-wide
survey in 15 California hospitals. Qual Saf
Health Care 2003; 12(2):112-8.
3.
Moody RF, Pesut DJ, Harrington CF.

Creating safety culture on nursing units:
Human performance and organizational
system factors that make a difference. J
Patient Saf 2006;2(4):198-206.
4.
5.
8. Sexton, J. B. (2002). A matter of life or death:

Social, psychological, and organizational
factors related to patient outcomes in the
intensive care unit. Dissertation. University
of Texas, Austin.
9.
Mardon RE, Khanna K, Sorra J, Dyer N,

Famolaro T. Exploring relationships
between hospital patient safety culture and
adverse events. J Patient Saf 2010; 6(4):22632.
10.
Singer S, Lin S, Falwell A, Gaba D, Baker

L. Relationship of safety climate and safety
performance in hospitals. Health Serv Res
2009; 44(2 Pt 1):399-421.
11.
Zohar D, Luria G. A multilevel model of

safety climate: cross-level relationships
between organization and group-level
climates. J Appl Psychol 2005; 90(4):61628.
Weaver, S.J. (2011). A configural approach

to patient safety climate: The relationship
between climate profile characteristics and
patient outcomes. Dissertation. University of
Central Florida, Orlando.
12.
Weingart, S. N., Farberstein, K., Davis, R.

B., & Phillips, R. S. (2004). Using a
multihospital survey to examine the safety
culture. Joint Commission Journal on
Quality and Patient Safety, 30(3), 125-132.
OLeary KJ, Haviley C, Slade ME, Shah

HM, Lee J, Williams MV. Improving
teamwork: impact of structured
interdisciplinary rounds on a hospitalist unit.
J Hosp Med 2011; 6(2):88-93.
13.
OLeary KJ, Wayne DB, Haviley C, Slade

ME, Lee J, Williams MV. Improving
teamwork: impact of structured
interdisciplinary rounds on a medical
teaching unit. J Gen Intern Med 2010;
25(8):826-32.
14.
Tiessen B. On the journey to a culture of

patient safety. Healthc Q 2008; 11(4):58-63.
6.
Naveh E, Katz-Navon T, Stern Z. Treatment

errors in healthcare: A safety climate
approach. Management Science.
2005:51(6);948-960.
7.
Estabrooks CA, Tourangeau AE, Humphrey

CK et al. Measuring the hospital practice
environment: a Canadian context. Res Nurs
Health 2002; 25(4):256-68.
369
15.
Frankel A, Grillo SP, Pittman M et al.

Revealing and resolving patient safety
defects: the impact of leadership
WalkRounds on frontline caregiver
assessments of patient safety. Health Serv
Res 2008; 43(6):2050-66.
16.
Donahue M, Miller M, Smith L, Dykes P,

Fitzpatrick JJ. A leadership initiative to
improve communication and enhance safety.
Am J Med Qual 2011; 26(3):206-11.
17.
Thomas EJ, Sexton JB, Neilands TB,

Frankel A, Helmreich RL. The effect of
executive walk rounds on nurse safety
climate attitudes: a randomized trial of
clinical units[ISRCTN85147255]. BMC
Health Serv Res 2005; 5(1):28.
18.
19.
20.
Riley W, Davis S, Miller K, Hansen H,

Sainfort F, Sweet R. Didactic and simulation
nontechnical skills team training to improve
perinatal patient outcomes in a community
hospital. Jt Comm J Qual Patient Saf 2011;
37(8):357-64.
Blegen MA, Sehgal NL, Alldredge BK,
Gearhart S, Auerbach AA, Wachter RM.
Improving safety culture on adult medical
units through multidisciplinary teamwork
and communication interventions: the TOPS
Project. Qual Saf Health Care 2010;
19(4):346-50.
Abstoss KM, Shaw BE, Owens TA, Juno
JL, Commiskey EL, Niedner MF. Increasing
medication error reporting rates while
reducing harm through simultaneous cultural
and system-level interventions in an
intensive care unit. BMJ Qual Saf 2011.
21.
Pettker CM, Thung SF, Raab CA et al. A

comprehensive obstetrics patient safety
program improves safety climate and
culture. Am J Obstet Gynecol 2011;
204(3):216.e1-6.
22.
Pettker CM, Thung SF, Norwitz ER et al.

Impact of a comprehensive patient safety
strategy on obstetric adverse events. Am J
Obstet Gynecol 2009; 200(5):492.e1-8.
23.
Sexton JB, Berenholtz SM, Goeschel CA et

al. Assessing and improving safety climate
in a large cohort of intensive care units. Crit
Care Med 2011; 39(5):934-9.
370
24.
Adams-Pizarro I, Walker ZA, Robinson J,

Kelly S, Toth M. Using the AHRQ Hospital
Survey on Patient Safety Culture as an
Intervention Tool for Regional Clinical
Improvement Collaboratives. 2008.
25.
Paine LA, Rosenstein BJ, Sexton JB, Kent

P, Holzmueller CG, Pronovost PJ. Assessing
and improving safety culture throughout an
academic medical centre: a prospective
cohort study. Qual Saf Health Care 2010;
19(6):547-54.
26.
Timmel J, Kent PS, Holzmueller CG, Paine

L, Schulick RD, Pronovost PJ. Impact of the
Comprehensive Unit-based Safety Program
(CUSP) on safety culture in a surgical
inpatient unit. Jt Comm J Qual Patient Saf
2010; 36(6):252-60.
27.
Zohar D. Thirty years of safety climate

research: Reflections and future directions.
Accident Analysis and Prevention.
2010;42(5):1517-1522.
28.
Schein EH. Organizational leadership and

culture, 4th edition. San Francisco: JosseyBass. 2010.
29.
Hartmann CW, Meterko M, Rosen AK et al.

Relationship of hospital organizational
culture to patient safety climate in the
Veterans Health Administration. Med Care
Res Rev 2009; 66(3):320-38.
30.
Singer SJ, Falwell A, Gaba DM et al.

Identifying organizational cultures that
promote patient safety. Health Care Manage
Rev 2009; 34(4):300-11.
31.
Sorra JS, Nieva VF. Hospital Survey on

Patient Safety Culture. AHRQ Publication
No. 04-0041, September 2004. Agency for
MD.
www.ahrq.gov/qual/patientsafetyculture/use
rgd.htm.
32.
Organising for Safety: Third Report of the

ACSNI (Advisory Committee on the Safety
of Nuclear Installations) Study Group on
Human Factors. Health and Safety
Commission (of Great Britain). Sudbury,
England: HSE Books, 1993.
33.
Reiman T, Pietikainen E, Oedewald P.

Multilayered approach to patient safety
culture. Qual Saf Health Care 2010;
19(5):e20.
34.
Kirk S, Parker D, Claridge T, Esmail A,

Marshall M. Patient safety culture in
primary care: developing a theoretical
framework for practical use. Qual Saf
Health Care 2007; 16(4):313-20.
43.
Sexton JB, Helmreich RL, Williams R, et al.

The Flight Management Attitudes Safety
Survey (FMASS). Research Project
Technical Report 01-01. Austin, TX: The
University of Texas; 2001.
35.
Parker D, Lawrie M, Carthey J, Coultous M.

Framework for Understanding the
Development of Organizational Safety
Culture. Clinical Risk. 2008;14(4):140-142.
44.
36.
Colla JB, Bracken AC, Kinney LM, Weeks

WB. Measuring patient safety climate: a
review of surveys. Qual Saf Health Care
2005; 14(5):364-6.
Cooper JB, Blum RH, Carroll JS et al.

Differences in safety climate among hospital
anesthesia departments and the effect of a
realistic simulation-based training program.
Anesth Analg 2008; 106(2):574-84, table of
contents.
45.
Morello RT, Lowthian JA, Barker AL,

McGinnes R, Dunt D, Brand C. Strategies
for improving patient safety culture in
hospitals: a systematic review. BMJ Qual
Saf. Jul 31. [Epub ahead of print].
46.
Parmelli E, Flodgren G, Beyer F, Baillie N,

Schaafsma ME, Eccles MP. The
effectiveness of strategies to change
organisational culture to improve healthcare
performance: a systematic review.
Implement Sci 2011; 6:33.
Notes: GENERAL NOTE: NLM: Original
Date Completed: 20110714.
37.
Flin R, Burns C, Mearns K, Yule S,

Robertson EM. Measuring safety climate in
health care. Qual Saf Health Care 2006;
15(2):109-15.
38.
Flin R. Measuring safety culture in

healthcare: A case for accurate diagnosis.
Safety Science. 2007:45; 653-667.
39.
Jackson J, Sarac C, Flin R. Hospital safety

climate surveys: measurement issues. Curr
Opin Crit Care 2010.
40.
Singla AK, Kitch BT, Weissman JS,

Campbell EG. Assessing patient safety
culture: A review and synthesis of the
measurement tools. J Patient Saf,
2006;2:105-115.
47.
Edwards P, Scott T, Richardson P et al.

Using staff perceptions on patient safety as a
tool for improving safety culture in a
pediatric hospital system. Journal of Patient
Safety 2008; 4(2):113-8.
41.
Sexton JB, Helmreich RL, Neilands TB et

al. The Safety Attitudes Questionnaire:
psychometric properties, benchmarking
data, and emerging research. BMC Health
Serv Res 2006; 6:44.
48.
Pronovost P, Weast B, Rosenstein B et al.

Implementing and validating a
comprehensive unit-based safety program.
Journal of Patient Safety 2005; 1(1):33-40.
49.
42.
Singer S, Meterko M, Baker L, Gaba D,

Falwell A, Rosen A. Workforce perceptions
of hospital safety culture: development and
validation of the patient safety climate in
healthcare organizations survey. Health Serv
Res 2007; 42(5):1999-2021.
Kapinos KA, Fitzgerald P, Greer N, Rutks I,

Wilt TJ. The Effect of Working Conditions
on Patient Care: A Systematic Review. VAESP Project #09-009; 2012.
371
Chapter 34. Effect of Nurse-to-Patient Staffing Ratios on

Patient Morbidity and Mortality
Paul G. Shekelle, M.D., Ph.D.

A small percentage of hospitalized patients die during or shortly after their hospitalization.
Evidence suggests that some proportion of these deaths could probably have been prevented with
more nursing care. For example, in one early study of 232,342 surgical discharges from several
Pennsylvania hospitals, 4,535 patients (2%) died within 30 days of the hospital admission; the
investigators estimated that the difference between 4:1 and 8:1 nurse-to-patient staffing ratios
might be approximately 1,000 deaths.1 Other studies have resulted in roughly similar estimates,
namely about 1 to 5 fewer deaths per 1000 inpatient days.

What the patient safety practice is remains unclear, because to date no intervention studies
have assessed the effect of a deliberate change in registered nurse (RN)topatient staffing
ratios. Most studies have been cross-sectional or longitudinal assessments of differences in
nursing staff variables (see below), with the most commonly assessed measure being the
proportion of RN time per a measure of inpatient load and the most commonly assessed outcome
being mortality. However, numerous other factors have been proposed as being causal with
respect to the relationship between nursing care and reductions in hospital mortality, potentially
in addition to or instead of a simple nursing staff- to- patient ratio: These factors include
measures of nursing burnout, job satisfaction, teamwork, nurse turnover, nursing leadership in
hospitals, and nurse practice environment.

Conceptual frameworks for why more effective nursing care may reduce inpatient mortality
have been proposed by Tourangeau and colleagues,2 Thornlow, Anderson and Oddone ,3 and
Despins, Scott-Cawiezell, and Rouder.4 Underlying all these conceptual frameworks is the belief
that surveillance is a critical factor that can be improved with more staff, better educated staff, or
a better working environment.5 As shown by Aiken and colleagues,6 nurse-patient ratios, along
with staffing skill mix, can lead to better surveillance, which along with a number of other
factors can influence the process of care and lead to better patient outcomes (see Figure 1).
372
Figure 1, Chapter 34. Hospital organization, nursing organization, and patient outcomes
Figure taken from Aiken et al., 20026

Aiken LH, Clarke SP, Sloane DM. Hospital staffing, organization, and quality of care: Cross-national findings. Nurs Outlook.
2002 Sep-Oct;50(5):187-94.with permission from Elsevier.
The model of Despins and colleagues (Figure 2) explicitly posits that better detection of
potential signals of patients at risk of poor outcomes is the mechanism by which more effective
nursing care exerts its beneficial effects; it further elaborates that organizational culture is an
important component of better signal detection (e.g., high reliability organizations instill in their
staff the value they place on safety). Internal factors such as nurse fatigue also play a role in
this model.
Figure 2, Chapter 34. Despins model on patient risk detection
Figure taken from Despins et al., 20104

Despins LA, Scott-Cawiezell J, Rouder JN. Detection of patient risk by nurses: a theoretical framework. Journal of Advanced
Nursing. 2010. Permission granted by John Wiley & Sons, Inc.
373
In the models proposed by Tourangeau (Figure 3) and by Thornlow (Figure 4), numerous
patient, system, nurse, nurse environment, and other factors are hypothesized to play an
important role in reducing inpatient mortality and other outcomes. The Tournageau model
explicitly posits that the use of care maps/protocols is associated with lowering the risk of
inpatient mortality.
Figure 3, Chapter 34. Tourangeaus model on determinants of 30-day mortality
Figure taken from Tourangeau et al., 20062

Tourangeau AE, Doran DM, Hall LM, et al. Impact of hospital nursing care on 30-day mortality for acute medical patients.
Journal of Advanced Nursing. 2006. Permission granted by John Wiley & Sons, Inc.
374
Figure 4, Chapter 34. Thornlows model on cascade iatrogenesis: postoperative respiratory failure
Notes: RN Registered Nurse; ASA American Society of Anesthesiologists

Figure taken from Thornlow et al., 20093
Reprinted from International Journal of Nursing Studies. 46(11), Thornlow DK, Anderson R, Oddone E. Cascade iatrogenesis:
Factors leading to the development of adverse events in hospitalized older adults. 1528-35, 2009 with permission from Elsevier.

Prior Studies and Reviews
Nurse staffing ratio is the most commonly assessed PSP in this category of practices and will
be the focus of this review. This portion of the review relied primarily on systematic reviews by
Kane and colleagues at the Minnesota Evidence-Based Practice Center7 (EPC) and by
Tourangeau;8 they scored 10 out of 10 relevant and 7 out of 9 relevant, respectively, on the
AMSTAR criteria. We supplemented these sources with an update search (described below). For
their review, the Minnesota EPC performed a thorough literature search through 2006 and
assessed the relationship between RN staffing ratios and the outcomes of inpatient mortality and
adverse patient events such as hospital-acquired pneumonia, failure to rescue, and surgical
wound infection. The review included 28 studies, of which 17 were cohort studies, 7 were crosssectional studies, and 4 were case-control studies (i.e., no experimental studies were identified).
Most were U.S. studies, and the average level of staffing was 3.0 patients per RN for the
intensive care unit (ICU) setting, 4.0 patients per RN in the surgical setting, and 4.4 patients per
RN for the medical setting. This review found a consistent association between higher RN
staffing and lower hospital-related mortality: An increase of one RN full-time equivalent (FTE)
per patient day was associated with a 9 percent reduction in the odds of death in the ICU, a 16
percent reduction in the odds of death in the surgical setting, and a 6 percent reduction in the
odds of death in the medical setting (see Table 1). The numbers of avoidable deaths per 1,000
patient days were, respectively, 5, 6, and 5. With respect to other outcomes, lower rates of
hospital-acquired pneumonia, pulmonary failure, unplanned extubation, failure to rescue, and
nosocomial bloodstream infections were associated with higher RN staffing in pooled analyses
of multiple studies. However, several other outcomes presumed to have strong sensitivity to
nurse staffing levels did not show consistent associations; these outcomes included falls, pressure
ulcers, and urinary tract infections.
375
Table 1, Chapter 34. Pooled odds ratios of patient outcomes corresponding to an increase of one
registered nurse full-time equivalent per patient day
No. Avoided Events/1000
Outcome
Studies
Odds Ratio (95% CI)
Hospitalized (95% CI)
All Patients
Mortality, intensive care units
5
0.91 (0.86; 0.96)
5 (2; 8)
Mortality, surgical patients
8
0.84 (0.8; 0.89)
6 (4; 8)
Mortality, medical patients
6
0.94 (0.94; 0.95)
5 (4; 5)
Hospital-acquired pneumonia
4
0.81 (0.67; 0.98)
1 (0; 2)
Pulmonary failure
5
0.94 (0.94; 0.94)
1 (1; 1)
Cardiopulmonary resuscitation
5
0.72 (0.62; 0.84)
2 (1; 2)
Intensive care units
Hospital-acquired pneumonia
3
0.7 (0.56; 0.88)
7 (3; 10)
Pulmonary failure
4
0.4 (0.27; 0.59)
7 (5; 9)
Unplanned extubation
5
0.49 (0.36; 0.67)
6 (4; 8)
Cardiopulmonary resuscitation
3
0.72 (0.62; 0.84)
2 (1; 2)
Surgical Patients
Failure to rescue
5
0.84 (0.79; 0.9)
26 (17; 35)
Nosocomial bloodstream infection
5
0.64 (0.46; 0.89)
4 (2; 5)
Table was adapted from Kane et al., 20077
Kane RL, Shamliyan TA, Mueller C, et al. The association of registered nurse staffing levels and patient outcomes - Systematic
review and meta-analysis. Medical Care. 2007 Dec;45(12):1195-204. Permission granted by Wolters Kluwer Health.
The EPC authors also conducted an indirect analysis of the potential for a dose-response
relationship. This analysis (Figure 5) assessed the effect across studies of additional RN-level
nurses per shift. In each case, comparisons of quartiles of nurse staffing levels showed the
expected relationship. In other words, if the association between nurse staffing and mortality is
causal, the difference in the risk for death should be greater between the 1st and the 3rd quartile
of nurse staffing than it is between the 1st and the 2nd quartile, because the difference in staffing
between the 1st and 3rd quartiles is greater than between the 1st and 2nd quartiles.
Figure 5, Chapter 34. Pooled odds ratio of quartiles of nurse staffing levels
Figure taken from Kane, 20077

Kane RL, Shamliyan TA, Mueller C, et al. The association of registered nurse staffing levels and patient outcomes - Systematic
review and meta-analysis. Medical Care. 2007 Dec;45(12):1195-204. Permission granted by Wolters Kluwer Health.
376
The EPC review concluded that a consistent relationship has been demonstrated but
identified numerous limitations in the literature with respect to establishing that this relationship
is causal. Ultimately, the authors concluded that the arguments for a causal relationship are
mixed, and they called for future research to address the role of nurse staffing and competence
on the effectiveness of patient care, taking greater cognizance of other relevant factors such as
patient and hospital characteristics and quality of medical care.
The Tourangeau search identified literature published through 2009 and was restricted to
studies that used hospital-related mortality as the outcome; the authors identified 17 studies (10
of which were not included in the Kane review, seven published since 2007). Although the
Tourangeau review was narrative (not a meta-analysis like the EPC review), the two had broadly
similar results: 14 of 17 studies found a statistically significant relationship between nurse
staffing variables and lower mortality rates (see Evidence Table in Appendix D). In addition,
Tourangeau and colleagues identified mixed findings for mortality among five studies assessing
the characteristics of the nurse work environment and work relationships, three studies assessing
nurses responses to work and the work environment (e.g., burnout), and seven studies assessing
nurses educational preparation and experience. Only one study assessed any nursing process-ofcare variables; it found a cross-sectional relationship between the use of care maps and lower
hospital-associated mortality, with an estimated effect size of 10 fewer deaths per 1000 acute
medicine discharged patients. Like the EPC review, the review by Tourangeau concluded that a
strong relationship exists but that more research is needed to understand the reasons that this
relationship between higher nurse staffing and lower hospital mortality might be causal; that is,
they called for a theoretical model that explains the relationship in ways that can be tested and
refined.
Thus, these two reviews came to broadly similar conclusions: Mostly cross-sectional studies
consistently report that higher RN staffing is associated with lower hospital-related mortality.
However, as Kane and colleagues ask, does this association reflect a causal relationship? If it
does not, then an intervention that simply hires more RN-level nurses may not achieve the
desired result. Indeed, mandates for fixed nurse-patient ratios have been critiqued as being an
inflexible solution which is unlikely to lead to optimal use of resources.9
Any number of factors might confound the observed relationship: In cross-sectional studies,
hospitals that are better in a variety of other ways might also be better staffed with RN-level
nurses. For example, one published study of electronic health record (EHR) implementation
showed that hospitals with EHRs have higher nurse staffing ratios and lower patient mortality.10
Longitudinal studies overcome these kinds of limitations in cross-sectional studies, but
imprecision in the measures of nurse staffing and of the severity of patient illness (which may
increase the risk of death via other, non-nursing-sensitive ways) constitute potential threats to the
validity of the association between nurse staffing and mortality.
Update Review
To supplement the two existing reviews, we used the Web of Science to conduct an update
search for articles published from 2009 onwards that cited any of four landmark articles in this
field. Our update search identified 546 titles, and 4 articles came from reference mining. From
550 titles, we identified 9 longitudinal studies and 1 new systematic review.11-20 The systematic
review included studies that assessed nurse staffing ratios and outcomes restricted to adult ICU
settings20 and reached conclusions similar to the previous reviews: a consistent relationship
between increased nurse staffing and better patient outcomes in observational studies, evidence
377
that falls short of causality. One longitudinal study narratively reported that increased nurse
staffing was related to significantly (P 0.01) decreased rates of decubiti, pneumonia, and
sepsis, but data were not presented.14
We discuss the 1 cross-sectional study because it addresses the effect of an intervention to
change nurse staffing ratios, implemented in response to a 2004 California law requiring
minimum nursepatient ratios in acute care hospitals.21 This legislation mandated patientnurse
staffing levels of 5:1, 4:1, and 2:1 for medical or surgical units, pediatric units, and ICUs,
respectively. The California legislative mandate does not require nurse staffing to be met with
RNs (that is, licensed vocational [practical] nurses can also meet the mandate).
Aiken and colleagues21 assessed the relationship between nurse staffing and mortality in
2006, 2 years after the California mandate, comparing data from California with those of two
states without mandates New Jersey and Pennsylvania. Data about workloads were drawn
from a survey of RNs in the three states22,336 nurses in totalwith a response rate of 35.4
percent. Hospital data came from the American Hospital Association, and patient and outcome
data came from State hospital discharge databases.
The authors reported that their survey data showed substantial compliance with the California
mandate, with 88 percent of medical/surgical, 85 percent of pediatric, and 85 percent of ICU
nurses reporting that on their last shift they were within the mandated staffing ratios. This level
of compliance is higher (sometimes considerably) than the values of 19 percent, 52 percent, and
63 percent for the same settings in New Jersey and 33 percent, 66 percent, and 71 percent in
Pennsylvania. In logistic regression analyses adjusted for a large number of patient
characteristics and three hospital characteristics (bed size, teaching status, and technology use),
Aiken and colleagues found statistically significant relationships between the estimation of the
average number of patients per nurse and two outcomes: 30-day mortality and failure to rescue
(Table 2).
Table 2, Chapter 34. Odds ratios indicating the effect of nurse staffing on 30-day inpatient
mortality and failure to rescue, in California, New Jersey, and Pennsylvania
State Hospital
Sample
California
Odds Ratios Estimating the Effect of Nurse Staffing on

30-day Inpatient Mortality
Failure to Rescue
1.13
1.15
(1.07-1.20)
(1.09-1.21)
New Jersey
1.10
1.10
(1.01-1.22)
(1.01-1.21)
Pennsylvania
1.06
1.06
(1.00-1.12)
(1.00-1.12)
Adjusted odds ratios are based on multivariate robust logistic regression models that controlled for 132 patient characteristics,
including age, sex, admission type, dummy variables for comorbidities and type of surgery, and interaction terms, and three
hospital characteristics, bed size, teaching status, and technology.
Table was adapted from Aiken et al., 2010.21

Aiken LH, Sloane DM, Cimiotti JP, et al. Implications of the California Nurse Staffing Mandate for Other States. Health Services
Research. 2010. Permission granted by John Wiley & Sons, Inc.
These associations were found for all three states. The authors also provided several
measures of nurse-assessed practice environment characteristics taken from their survey
responses, such as a reasonable workload and enough staff to get work done; all consistently
favored California over New Jersey or Pennsylvania. The authors concluded that, 2 years after
the California mandate, nurse patient care loads were significantly lower in California than in
either New Jersey or Pennsylvania; on average, these loads were one patient fewer, and in the
378
medical/surgical units they were closer to two patients fewer. California nurses were also more
likely to report favorable practice environment characteristics.
Although the study by Aiken and colleagues21 collected data after the implementation of
Californias staffing mandate, it did not test the effect of that mandate per se because it had no
comparison data from the period before the mandate went into effect. The possibility that the
relationship is causal is blunted by a longitudinal study that examined measures from before and
after the California mandate and showed the expected changes in nurse staffing and proportion of
licensed staff per patient but no improvement in two patient outcomes believed to be nursingsensitive: falls and pressure ulcers.11,13 In fact an unexpected statistically significant increase in
pressure ulcers was associated with a greater number of hours of care for the patient (which may
have been due to greater detection). This study did not assess mortality.
Five additional longitudinal studies add further information to this picture. The first is a
longitudinal assessment of nurse staffing and hospital mortality and failure to rescue in 283
California hospitals between 1996 and 2001, which had access to direct measures of nurse
staffing.15 In multivariable models that included numerous hospital market characteristics as well
as risk adjustment using the Medstat Disease Staging Methodology to produce a predicted
probability for complications or death, the authors found that an increase of one RN FTE per
1,000 inpatient days was associated with a statistically significant 4.3 percent decrease in
mortality.
The second longitudinal study assessed care at 39 Michigan hospitals between 2003 and
2006; it included adults admitted through the emergency department with acute myocardial
infarction, heart failure, stroke, pneumonia, hip fracture, or gastrointestinal bleeding.17 This
study simultaneously controlled for high hospital occupancy on admission, a weekend
admission, seasonal influenza, and nurse staffing levels. Each factor had a statistically significant
increased effect on in-hospital mortality. Each additional RN FTE per patient day was associated
with a 0.25 percent decrease in mortality.
The third longitudinal study assessed the effect of a mandate in three Western Australia
public hospitals to implement a new staffing method, the Nursing Hours Per Patient Day
(NHPPD).18 The study assessed three time periods: 20 months before implementation 7 months
of a transition period, and 2 months post implementation. The authors found that the total
nursing hours and RN nurse hours increased during the observation period. However, the
percentage of total nursing hours provided by RNs decreased (from 87% to 84%). Also, the
article stated that although the nursing hours increased for all three hospitals (in the postimplementation period), the changes were not statistically significant, Mortality rates improved
during this time period. Among a host of other outcomes, some improved, others did not, and
some changes were inconsistent across hospitals. Although the study was described as an
interrupted time series, it was analyzed as a before-and-after study.
The fourth longitudinal study assessed changes in nurse staffing over 9 years in 124 Florida
hospitals and related these to changes in Agency for Healthcare Research and Quality Patient
Safety Indicators.19 The study used both initial staffing ratios and changes in staffing ratios.
Results were mixed but generally favored better patient safety outcomes with higher RN staffing
levels.
The methodologically strongest longitudinal study is discussed here in more detail.16 In this
study, Needleman and colleagues used data over time from a single hospital to assess the
association between naturally occurring differences in levels of RN staffing within the same
hospital and inpatient mortality. This study is further characterized by a careful matching of
379
nurse staffing on a shift-by-shift basis with the actual patients cared for during that shift.
Knowing the actual patients cared for allowed for more sophisticated adjustments at the patient
level of risk-of-death. The study was carried out at a tertiary academic hospital between 2003
and 2006 and included 197,691 admissions and 176,696 nursing shifts, across 43 hospital units.
The patients themselves averaged 60 years of age, and about 50 percent were covered under
Medicare. The variable of interest was exposure of the patient to nursing care that was below the
target level (for that type of unit) for that shift, in other words the proportion of shifts below
target level staffing, on a patient basis. An additional exposure variable was a high turnover
shift (in other words, a shift with many admissions, discharges, or transfers). The authors found
that exposure to each shift of below-target staffing or high turnover was associated with a 2 to 7
percent increase in mortality, with higher levels of risk if the high-turnover or below-target shift
occurred in the first 5 days after admission (see Table 3). For patients who were not in an ICU,
this increased risk rose to 12 percent or 15 percent.
Table 3, Chapter 34. Risk of death associated with exposure to a shift with an actual RN staffing
level 8 hours or more below target, high patient turnover, and other variables
Variable
Total of 197,961 patients
Each shift with RN staffing level below target or high turnover
during first 30 days after admission
Shift with RN staffing level 8 hr or more below target
Shift with high patient turnover
Total of 171,041 patients with no shifts in an ICU
Hazard Ratio (95% CI)
1.02 (1.01-1.03)
1.04 (1.02-1.06)
1.03 (1.02-1.05)
1.07 (1.03-1.10)
1.04 (1.03-1.06)
1.07 (1.02-1.13)

1.15 (1.07-1.24)
Table adapted from Needleman et al., 201116
Needleman J, Buerhaus P, Pankratz VS, et al. Nurse Staffing and Inpatient Hospital Mortality. New England Journal of
Medicine. 2011 Mar;364(11):1037-45. Permission granted by Massachusetts Medical Society (MMS) publishers of the New
England Journal of Medicine.
The data from Needleman and colleagues contribute to the causality determination because
the study is longitudinal within one hospital, thus controlling for the hospital effect potentially
present in all cross-sectional studies, and has detailed measures of exposure and confounding
variables. These results, and the dose-response analysis from the EPC review, are the two
strongest pieces of evidence in support of causality.

One finding of the survey administered by the Aiken study, which collected data 2 years after
the California mandate for minimum nurse staffing ratios,21 was that some California nurses
perceived less support from the use of LVNs, unlicensed personnel, and non-nursing support
services (housekeeping, unit clerks) following implementation of the mandate. For example, 25
percent of RNs responded that they perceived decreased use of LVNs following the mandate,
whereas 10 percent perceived increased use and 56 percent reported that use remained the same.
380
The longitudinal assessments from California11 and Western Australia18 reported an increase in
pressure ulcers associated with increased nurse staffing, although this development may reflect
increased detection. Almost no other studies mentioned an explicit assessment of potential
unexpected adverse outcomes.

What Contexts?
Because we found no published studies of an assessment of an implementation per se, we
cannot answer this question directly. However, the cross- sectional and longitudinal studies that
have been published, and that have consistently shown an association between staffing levels and
patient outcomes, have included a broad array of hospitals, often all or almost all hospitals
(except for very small ones) in a state. Therefore, if the relationship between increased RN
staffing and inpatient mortality is a causal one, it very likely is applicable to most hospitals and
most contexts. This PSP is most likely to be carried out due to State or Federal policy.

Four simulation studies reported information about costs. The first used 2003 data from 28
Belgian cardiac surgery centers to assess the costs and outcomes of increasing nurse staffing.
Assuming a causal relationship between this staffing increase and an outcome of 5 fewer patient
deaths per 1000 elective hospitalizations, the authors concluded that the incremental costeffectiveness ratio was 26,372 Euros (approximately $35,000) per avoided death and 2,639
Euros (approximately $3500) per life-year gained.22
The second simulation study was conducted by the University of Minnesota Evidence-based
Practice Center, which produced the systematic review on nurse staffing.23 It used its own metaanalysis as the basis for estimating the potential monetary benefits of increased RN staffing.
Assuming that those relationships were causal and taking a societal perspective, the authors
concluded that increasing RN staffing by 1 FTE per patient day was related to positive savings
cost ratios across a broad range of clinical settings. For example, the net cost of adding 1 RN
FTE per 1000 hospitalized ICU patients was an estimated $590,000, whereas the net benefit (in
terms of life-years saved and productivity) was an estimated $1.5 million, for a benefitcost ratio
of 2.51. However, hospitals did not save money because the net cost of adding an extra nurse
FTE was not offset by the expected 24% decrease in length of stay.
A third simulation study used data from studies by Aiken and colleagues and Needleman and
colleagues to estimate benefits in mortality and length of stay, respectively, and estimated an
incremental cost-effectiveness ratio between $25,000 and $136,000 per life saved as patientRN
staffing ratios decreased from 8:1 to 4:1. The model was most sensitive to the estimate of effect
on mortality.24
Lastly, one additional study from Portugal estimated that increasing neonatal nurse staffing
to adequate would increase staff costs more than 30% of the current rate.25

As previously noted, the association between staffing and mortality that underpins this PSP
has been observed in a wide variety of hospitals and contexts. We believe that the effect, if it is
causal, is likely to be relatively insensitive to the usual effects of contexts considered in this
review. Of note, the recent study by Needleman and colleagues was conducted in a tertiary
381
medical center that has a lower-than-expected in-hospital mortality rate and a reputation for
excellence. Therefore, the association between increased RN staffing and lower mortality, if it is
causal, is potentially applicable even to high-performing hospitals.

Nurse staffing ratios have a consistent association with reductions in hospital-related
mortality. However, the strength of evidence for causality in this finding cannot be rated high,
given the lack of evaluations of a deliberate change in RN staffing from some initial value (for
example, 6 patients to 1 RN on general medical wards) to some higher RN staffing value (such
as 5-to-1 or 4-to-1). Such an evaluation should be possible, either as a time series analysis or as a
controlled before-and-after analysis. Studies evaluating a deliberate change in nurse staffing
ratios would greatly improve our understanding of the likelihood of causality. Developing a
testable conceptual framework for how increased staffing can influence outcomes would be an
important addition to these and other studies.
Therefore, given the consistent associations observed in multiple cross-sectional and a few
longitudinal studies, the indirect dose-response analysis by Kane and colleagues, and the
methodologically careful single-site study by Needleman and colleagues, we grade the strength
of evidence for increased RN staffing and lower hospital-related mortality as moderate. The
strength of evidence for other outcomes (hospital-acquired pneumonia, failure-to-rescue, falls,
pressure ulcers, etc.) remains low, owing to the sparseness of data, conflicting data, and/or lack
of evidence of a dose-response relationship.
If the relationship between nurse staffing and mortality outcomes is causal, then the wide
variety of hospital settings included in existing analyses suggests that the effect is likely to be
relatively insensitive to hospital contexts. However, some of the nurse work environment factors,
such as job satisfaction, burnout, teamwork, workload, and leadership, are potentially important
effect modifiers, and this area merits further study. Summary tables are located below (Tables 4
and 5).
Table 4, Chapter 34. Summary table for increasing nurse-to-patient staffing ratios to prevent death
Scope of the
Strength of
the PSP
Effectiveness
Common/High
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
High
How Much do We
A lot/Not difficult
Table 5, Chapter 34. Summary table for increasing nurse-to-patient staffing ratios to prevent falls,
pressure ulcers, and other nursing sensitive outcomes (other than mortality)
Scope of the
Strength of
the PSP
Effectiveness
Common/High
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
High
382
How Much do We
A lot/Not difficult
References
1.
Aiken LH, Clarke SP, Sloane DM, et al.

Hospital nurse staffing and patient mortality,
nurse burnout, and job dissatisfaction.
JAMA. 2002 Oct 23-30;288(16):1987-93.
PMID: 12387650.
2.
Tourangeau AE, Doran DM, Hall LM, et al.

Impact of hospital nursing care on 30-day
mortality for acute medical patients. Journal
of Advanced Nursing. 2006 Jan;57(1):32-44.
PMID: WOS:000243297400003.
3.
4.
5.
Thornlow DK, Anderson R, Oddone E.

Cascade iatrogenesis: Factors leading to the
development of adverse events in
hospitalized older adults. International
Journal of Nursing Studies. 2009
Nov;46(11):1528-35. PMID:
WOS:000270620000014.
Despins LA, Scott-Cawiezell J, Rouder JN.
Detection of patient risk by nurses: a
theoretical framework. Journal of Advanced
Nursing. 2010 Feb;66(2):465-74. PMID:
WOS:000273822000023.
Aiken LH, Sochalski J, Lake ET. Studying
outcomes of organizational change in health
services. Med Care. 1997 Nov;35(11
Suppl):NS6-18. PMID: 9366875.
6.
Aiken LH, Clarke SP, Sloane DM. Hospital

staffing, organization, and quality of care:
Cross-national findings. Nurs Outlook. 2002
Sep-Oct;50(5):187-94. PMID: 12386653.
7.
Kane RL, Shamliyan TA, Mueller C, et al.

The association of registered nurse staffing
levels and patient outcomes - Systematic
review and meta-analysis. Medical Care.
2007 Dec;45(12):1195-204. PMID:
WOS:000251538700011.
8.
Tourangeau AE. Mortality Rate as a NurseSensitive Outcome. Nursing Outcomes: The

State of the Science. Jones and Bartlett,
LLC; 2011.
9.
Griffiths P. RN plus RN = better care? What

do we know about the association between
the number of nurses and patient outcomes?
International Journal of Nursing Studies.
2009 Oct;46(10):1289-90. PMID:
WOS:000269275400001.
383
10.
Furukawa MF, Raghu TS, Shao BBM.

Electronic Medical Records, Nurse Staffing,
and Nurse-Sensitive Patient Outcomes:
Evidence from California Hospitals, 19982007. Health Services Research. 2010
Aug;45(4):941-62. PMID:
WOS:000279734400004.
11.
Burnes Bolton L, Aydin CE, Donaldson N,

et al. Mandated nurse staffing ratios in
California: a comparison of staffing and
nursing-sensitive outcomes pre- and
postregulation. Policy Polit Nurs Pract. 2007
Nov;8(4):238-50. PMID: 18337430.
12.
Cook A, Gaynor M, Stephens M, et al. The

effect of a hospital nurse staffing mandate
on patient health outcomes: Evidence from
Californias minimum staffing regulation.
Journal of Health Economics. 2012
Mar;31(2):340-8. PMID:
WOS:000303974900002.
13.
Donaldson N, Bolton LB, Aydin C, et al.

Impact of Californias licensed nurse-patient
ratios on unit-level nurse staffing and patient
outcomes. Policy Polit Nurs Pract. 2005
Aug;6(3):198-210. PMID: 16443975.
14.
Duffield C, Diers D, OBrien-Pallas L, et al.

Nursing staffing, nursing workload, the
work environment and patient outcomes.
Applied Nursing Research. 2011
Nov;24(4):244-55. PMID:
WOS:000297567500008.
15.
Harless DW, Mark BA. Nurse Staffing and

Quality of Care With Direct Measurement of
Inpatient Staffing. Medical Care. 2010
Jul;48(7):659-63. PMID:
WOS:000279428200014.
16.
Needleman J, Buerhaus P, Pankratz VS, et

al. Nurse Staffing and Inpatient Hospital
Mortality. New England Journal of
Medicine. 2011 Mar;364(11):1037-45.
PMID: WOS:000288439000009.
17.
Schilling PL, Campbell DA, Englesbe MJ,

et al. A Comparison of In-hospital Mortality
Risk Conferred by High Hospital
Occupancy, Differences in Nurse Staffing
Levels, Weekend Admission, and Seasonal
Influenza. Medical Care. 2010
Mar;48(3):224-32. PMID:
WOS:000275198200006.
18.
Twigg D, Duffield C, Bremner A, et al. The

impact of the nursing hours per patient day
(NHPPD) staffing method on patient
outcomes: A retrospective analysis of
patient and staffing data. International
May;48(5):540-8. PMID:
WOS:000291713900003.
19.
Unruh LY, Zhang NJ. Nurse Staffing and

Patient Safety in Hospitals New Variable
and Longitudinal Approaches. Nursing
Research. 2012 Jan-Feb;61(1):3-12. PMID:
WOS:000298158000002.
20.
McGahan M, Kucharski G, Coyer F. Nurse

staffing levels and the incidence of mortality
and morbidity in the adult intensive care
unit: A literature review. Australian Critical
Care. 2012 May;25(2):64-77. PMID:
WOS:000304456100002.
21.
Aiken LH, Sloane DM, Cimiotti JP, et al.

Implications of the California Nurse Staffing
Mandate for Other States. Health Services
Research. 2010 Aug;45(4):904-21. PMID:
WOS:000279734400002.
384
22.
Van den Heede K, Simoens S, Diya L, et al.

Increasing nurse staffing levels in Belgian
cardiac surgery centres: a cost-effective
patient safety intervention? Journal of
Advanced Nursing. 2010 Jun;66(6):1291-6.
PMID: WOS:000277204000012.
23.
Shamliyan TA, Kane RL, Mueller C, et al.

Cost Savings Associated with Increased RN
Staffing in Acute Care Hospitals: Simulation
Exercise. Nursing Economics. 2009 SepOct;27(5):302-+. PMID:
WOS:000270800900004.
24.
Rothberg MB, Abraham I, Lindenauer PK,

et al. Improving nurse-to-patient staffing
ratios as a cost-effective safety intervention.
Med Care. 2005 Aug;43(8):785-91. PMID:
16034292.
25.
Fugulin FMT, Lima AFC, Castilho V, et al.

Cost of nursing staffing adequacy in a
neonatal unit. Revista Da Escola De
Enfermagem Da Usp. 2011 Dec;45:1582-8.
PMID: WOS:000299523600007.
Chapter 35. Patient Safety Practices Targeted at Diagnostic

Errors (NEW)
Kathryn M. McDonald, M.M.; Despina Contopoulos-Ioannidis, M.D.; Julia Lonhart, B.S., B.A.;
Brian Matesic, B.S.; Eric Schmidt, B.A.; Noelle Pineda, B.A.; John P.A. Ioannidis, M.D.

The family of patient safety targets that includes diagnostic errors, diagnostic delays, and
other diagnostic misadventures is not fully defined with clear boundaries. However, one
operational definition adapted from the Australian Patient Safety Foundation by Mark Graber
and colleagues is that diagnosis is unintentionally delayed (sufficient information was available
earlier), wrong (another diagnosis was made before the correct one), or missed (no diagnosis
ever made), as judged from the eventual appreciation of more definitive information.1
Alternatively and similarly, Gordon Schiff and colleagues have defined diagnostic errors as any
mistake or failure in the diagnostic process leading to a misdiagnosis, a missed diagnosis, or a
delayed diagnosis.2
Depending on the definition and data source, the exact scope of the problem varies, although
its magnitude is consistently impressive. A systematic review of 53 different series of autopsies
reported a median error rate of 23.5 percent (range, 4.1% to 49.8%) for major errors (clinically
missed diagnoses involving a principal underlying disease or primary cause of death) and 9.0
percent (range, 0% to 20.7%)1 for class I errors (the most serious subset of major errors being
those likely to have affected patient outcomes).3 These data translate to approximately 35,000
patients who might have survived to discharge from United States hospitals annually had
misdiagnosis not happened.(3) A Harris poll found that three in five Americans (63%) are very or
extremely concerned that a diagnostic error can take place.4
Numerous disease-specific studies show that 2 percent to 61 percent of patients experienced
missed or delayed diagnoses.5 Examining potential causes of delay in diagnosis for colorectal
cancer (CRC), 161 of 513 patients (31.4%) with newly diagnosed CRC had at least one
previously missed opportunity for their physician to initiate diagnostic workup. These patients
averaged 4.2 missed imaging initiation opportunities despite a mean of 5.3 clinical indications
for diagnostic workup for CRC.6 In a study of 587 patients diagnosed with lung cancer, 37.8
percent experienced missed clinical opportunities due to failure in recognizing predefined
clinical indications for follow-up or failure to complete requested follow-ups. Patients with
missed opportunities experienced a significantly longer median time to diagnosis than patients
without missed opportunities (132 vs. 19 days, respectively; p < .001). Patient non-adherence to
physician recommendations was present in 44 percent of patients with missed opportunities.7 In a
survey administered to academic, community, and trainee pediatricians, 54 percent reported
making a diagnostic error at least once per month and 45 percent noted making diagnostic errors
that harmed patients at least once per year. Survey respondents reported that lack of pertinent
historical or clinical information and team processes such as care coordination were contributors
to errors.8 Furthermore, research on variation in patient outcomes related to diagnosis timing
suggests room for improvement for some high stakes conditions. For example, early
identification of sepsis (along with protocols for treatment pathways) has been associated with
decreased mortality in surgical intensive care.9 Improving diagnostic speed, accuracy and triage
385
to treatment of such high risk, rapidly developing conditions is another important frontier for
those seeking to improve consequential diagnostic delays.
Problems in care related to diagnosis are particularly prevalent among precipitating causes
for lawsuits, with studies reporting 25 percent to 59 percent of malpractice claims attributable to
diagnostic errors.5,10,11 A recent study of 91,082 diagnosis-related malpractice claims from 1986
to 2005 estimated payments summing to 34.5 billion dollars (inflation-adjusted to 2010 dollars),
well over one billion dollars per year. The mean per-claim payout was $378,858 (interquartile
range: $72,250 to $472,000).12 Diagnosis-related claims made up 29.1 percent of total claims and
accounted for the highest proportion of total payments (35.6%). In terms of severity, lethal
injuries accounted for 40 percent of total payments. Another study of 10,739 malpractice claims
from the 2005-2009 National Practitioner Data Bank found that diagnosis-related reasons
accounted for 45.9 percent of paid claims from outpatient settings (95% confidence interval [CI],
44.4 to 47.4), the most frequently cited reason from that setting. Diagnostic reasons were the
second-most frequently cited for paid claims in the inpatient setting (21.1%; 95% CI, 20.0 to
22.3) and when both settings were involved (26.7%; 95% CI, 23.9 to 29.5).13
Some have asserted that diagnostic errors are more likely to be preventable and more likely
to result in patient harms than other types of errors (e.g., treatment-related errors, such as wrongsite surgery or incorrect medication dose), making the problem particularly important as well as
useful to address.14 Given this potential, the purpose of this review is to assess the multitude of
interventions to prevent diagnostic errors and better understand their effectiveness.

Many types of patient safety practices (PSPs) have been devised to address diagnostic errors,
and a number haven even been tailored to specific types of diagnostic error, root causes for the
error, technologies available, and other factors. Studies of the epidemiology and etiology of
diagnostic errors offer the foundation for an even richer and more robust set of potential PSPs in
this area. In an analysis of physician-reported errors, Schiff and colleagues found that the most
common missed or delayed diagnoses that physicians recalled were pulmonary embolism, drug
reactions or overdose, various cancers, acute coronary syndrome, and stroke.2 Incidence rates
could not be calculated from the convenience sample: The study focused on understanding the
potential root causes of the errors. They determined that errors occurred throughout the
diagnostic process and classified the reported cases using the Diagnostic Error Evaluation and
Research (DEER) project tool. From analysis of the subgroup of major diagnostic errors, over
43 percent were related to clinician assessment (including failure/delay in considering the
diagnosis, placing too much weight on competing/coexisting diagnosis) and 42 percent to
laboratory and radiology testing (including failure to order needed tests, technical errors in
processing specimens/tests, erroneous reading of a test). Some PSPs are designed to target these
failure areasfor example, the design and application of algorithms, checklists, and related tools
to help identify and weight potential diagnoses.
Viewing diagnostic errors from specific departments or specialties is another approach to
understanding contributing factors and designing interventions to mitigate these in specific
settings. As an example, Crosby developed a human- and system-oriented framework based on a
decade of reviewing emergency department (ED) cases from an urban, public, teaching
hospital.15 This framework examined ten areas, each one tied to points of leverage for
development and testing of PSPs, and together demonstrating the broad scope of possible
interventions to reduce diagnostic errors:
386
Patient factors: systems may be designed around areas that are more prone to risk (e.g.,
improved staffing with translators).
Human/clinician factors: interventions may aim at errors of planning separately from
errors of execution, and may also be designed to address cognitive error, skill-set error,
task-based error, and/or personal impairment.
Outside care systems, ED access, and triage: consideration of these three framework
areas aims to understand patterns of failure and errors that affect patients before their
arrival in the ED or initiation of care.
Teamwork: interventions in this area focus on communication, coordination, conflict
resolution, personnel assignment practices (e.g., considerations of capability, workload),
and training.
Local ED environment, hospital environment, hospital administration and third parties,
and community level: systems and resources at each of these four additional levels of the
framework have potential for effective interventions to reduce diagnostic errors within
the ED and after the patient leaves.
Within the above framework, human and clinician factors have received significant attention
from researchers interested in diagnosis. Cognitive factors may affect diagnostic accuracy
through rote over-learned actions or through purposive reasoning and decisionmaking processes.
The cluster of automatic or quasi-automatic decisionmaking processes may be classified as
heuristics, or rule-based decisionmaking processes. Heuristics aid in making decisions quickly
and are important for keeping cognitive capacity high for other, more demanding, cognitive
tasks. However, the very thing that makes heuristics helpful, decisions based on logical
assumptions gained from experience, can also lead to systematic bias and incorrect
decisionmaking when assumptions are wrong.16 Other cognitive processes affecting diagnosis
involve working memory in conjunction with learned knowledge, or more plainly, information
that is purposefully stored, recalled and used for completing a current goal. An example of these
cognitive processes can be seen in physicians listening to their patients describe symptoms. The
physician cognitively stores symptomatic information in the short term until she or he can
classify the symptoms into a more general descriptive category of a diagnosis. This process is
also subject to error when attention is pulled away from the task at hand or cognitive capacity is
altered for others reasons (e.g., lack of sleep). The process of metacognition involves continued
focusing and re-focusing attention on these cognitive processes so as to reflect on ones own
potential for biases, incorrect assumptions, and reduced cognitive capacity.17 Ultimately, both
human factors and the systems within which they operate have long been recognized as unique
contributors to human error.18
PSPs relevant to diagnostic error are also being actively developed by those bringing more
attention to this important patient safety target, and drawing on previous work in the research
domains of medical problem solving, decision analytic/normative decisionmaking, and clinical
diagnostic decision support.19 As health information technologies become more pervasive,
electronically-supported workflow and system redesign might target preventing or mitigating
diagnostic errors. PSPs in this area would be akin to computerized physician order entry with
clinical decision support, though more aptly named something like computerized diagnosis
management.
387

Many types of interventions, spanning a range of specialties and settings, are potentially
applicable to reducing diagnostic errors. Thus, it is impossible to answer the question of why
these interventions should work with one general statement. In addition to some of the
frameworks described above as the bases for logic models, recent commentaries and focus group
reports offer examples of why specific approaches could work (e.g., electronic clinical
documentation, checklists, interventions to decrease the frequency of missed test results).20-22 For
electronic documentation, for example, researchers have suggested goals and features of
redesigned systems for improved diagnosis (e.g., aid cognition through aggregation, trending,
contextual relevance, and minimizing of superfluous data) tied to specific roles for that
particular approach (e.g., providing access to information).20

A recently published systematic review on system-related interventions addressing
organizational vulnerabilities to diagnostic errors23 based on a search from 2000 to 2009
included 43 studies. A companion piece focused on cognitively-related interventions.24 To build
on the previous work, we conducted a separate systematic review, encompassing a longer time
period, and with broader inclusion criteria to provide a high-level summary of categories of
interventions studied. We searched MEDLINE, PSNet, bibliographies of background articles and
previous systematic reviews to identify literature about effects of practices with implications for
errors and delays in diagnosis. For further detail, see Appendix C.
Although numerous articles proposed or described interventions, few reported evaluations of
these interventions. Singh and colleagues summarized 37 studies with no evaluations, classifying
them along five process dimensions: provider-patient encounter, diagnostic test performance and
interpretation, follow-up and tracking, referral-related issues, and patient-related issues.23 Their
review also identified six evaluations of interventions, of which only three reported diagnostic
outcomes (incidence of delayed diagnosis of injury, incidence of missed injuries, misdiagnosis
rates), and none provided information on patients downstream clinical course.23
Graber and colleagues summarized 141 articles on improving congition and human factors
affecting diagnosis, 42 of which reported evaluation of interventions.24 These investigators
classified the literature along three dimensions. In the first dimension, interventions to increase
knowledge and expertise, the authors identified seven evaluation studies, only one of which
provided information on diagnostic outcomes and clinical course for actual patients. The second
dimension included interventions to improve intuitive and deliberate considerations. Among the
five studies evaluating interventions for this dimension, none reported resultant effects on
documented diagnoses with actual patients during clinical course of care. In the largest group of
studies, interventions assigned to the third dimension of getting help from colleagues, consultants
and tools, 16 of 28 studies evaluated diagnostic outcomes in actual patients. Graber and
colleagues note the current scarcity of evidence for any single intervention targeting cognitive
and human factors in reducing diagnostic error. The authors highlighted potential for
interventions that target content-focused training, feedback on performance, simulation-based
training, metacognitive training, second opinion or group decision-making, and the use of
decision support tools and computer-aided technologies.
Our review identified 94 studies of PSPs targeted at patient diagnosis. These studies reported
missed diagnosis, misdiagnosis, delayed diagnosis, or some other diagnostic discrepency with
388
potential for clinical consequence. The Supplementary Evidence Table (see Appendix D, Table
2) provides basic descriptions of targeted diagnostic errors, intervention descriptions, patient
outcome, study design and results with respect to the effectiveness of the proposed interventions.
Drawing from frameworks proposed by others, we classified interventions into one or more
of the following six types (Figure 1):
Technique (introduction of novel technologies for testing, adaptations of testing
equipment, or changes in medical interventions potentially affecting diagnostic
performance)
Additional Review Methods (introduction of additional steps from the interpretation
through reporting of test results)
Personnel Changes (introduction of additional health care members and/or replacing
certain professionals with others)
Educational Interventions (implementation of educational strategies)
Structured Process Changes (implementation of feedback systems or additional stages in
the diagnostic pathway)
Technology-based Systems Interventions (implementation of technology-based tools at
the system levelcomputer assistive diagnostic aids, decision support algorithms, text
message alerting, pager alerts, etc.)
Figure 1, Chapter 35. Interventions by type
Interventions by Type
(% of Total)
Technique
13%
Educational
9%
Technology-based
Systems
22%
Structured Process
22%
Personnel
5%
Additional Review
M ethods
29%
This pie chart illustrates the percentage of studies as categorized to the six intervention types: Technique, Educational,
Technology-Based Systems, Personnel Changes, Additional Review Methods, and Structured Process Changes.
All six of the evaluative studies identified by Singh and colleagues,23 many of the evaluative
studies identified by Graber and colleagues,24 and most of the studies included in our systematic
review, reported beneficial effects along the diagnostic pathway for a broad array of intervention
types. Because the evidence is predominantly from uncontrolled before-after study designs or
other uncontrolled study types (Table 1) with markedly different outcomes, the strength of the
evidence about interventions to reduce diagnostic errors is insufficient to draw any strong
conclusions. Furthermore, the magnitude of difference attributable to interventions varied by
study and clinical process. For example, some researchers demonstrated what would be
moderate-to-large effects on diagnosis if the assumption of causality were made (e.g., Perno and
colleagues, 2005),25 although methodologies were not designed to test causality, whereas other
389
studies were designed to demonstrate the absence of change in diagnostic outcomes despite
intervention (e.g., Thomas and colleagues, 2003).26
Table 1, Chapter 35. Study design distribution
Study Design
Description
Number of
Studies*
14
Randomized,
controlled trial
(RCT)
A standard randomized controlled trial involving two groups; a control and the
intervention group.
Experimental
Design
Study with a concurrent usual care control group, or another method for
controlling and comparing between experimental and usual care nonexperimental groups (but not including the pre/post method)
12
Pre/Post
A before and after study comparing pre-intervention to post-intervention

results.
Evaluative studies not matching the aforementioned designs.
16
Other
58
*Number of studies adds to more than 95 because several had multiple designs.
As a result of the state of the science in this area, no meta-analyses have been conducted.
Pooled analysis may be feasible in the near future as the evaluative literature is growing rapidly
in some intervention categories. Figure 2 shows particular increases for several classes of
interventions: Additional Review Methods, Technology-based Systems Interventions, and
Structured Process Changes. The other intervention types have not been studied much over the
entire period.
Figure 2, Chapter 35. Intervention studies by year
The graph illustrates a timeline of the included studies broken down by the six intervention types.
Few studies (5 randomized, controlled trials and 8 other designs) have evaluated patient-level
clinical outcomes to reduce diagnostic errors.9,27-38 Diagnostic errors have a complex relationship
with direct patient outcomes because they can play a role at many different time points in a
patients care; that is, many opportunities exist to catch diagnostic errors. If a diagnostic error is
caught at any of these opportunities, then negative effects on clinical outcomes could potentially
be avoided. Thus, examining the direct relationship between diagnostic errors and clinical
390
outcomes is complex and explains why many of the articles do not report on hard patient
outcomes. The remainder of this section summarizes the findings of the review.
Results of Randomized, Controlled Trials

Primary and secondary comparative quantitative outcomes data were available in 13
randomized trials, and are summarized in Appendix Table 1 (See Appendix D). Seven trials (9
comparisons) addressed diagnostic accuracy outcomes, and 3 trials (5 comparisons) addressed
outcomes related to further diagnostic test use. Six trials (8 comparisons) addressed outcomes
related to further therapeutic management. Five trials (7 comparisons) addressed direct patientrelated outcomes. Three trials addressed composite outcomes (diagnostic accuracy and
therapeutic management, and therapeutic management and patient outcome). One trial addressed
time to correct therapeutic management, and another trial addressed time to diagnosis.
Trials evaluated various interventions. The control group used most often was usual care. No
trials had high risk of bias, whereas 9 and 5 trials had moderate and low risk of bias,
respectively.
Statistically significant improvements were seen for at least 1 outcome in all but 3 trials. Of
the 3 trials with nonstatistically significant improvements, one was a noninferiority trial that
showed no more diagnostic errors occurred during work-up of abdominal pain among patients
given morphine and those not given morphine26. Two trials that involved patients with mental
conditions38,39 reported no beneficial diagnostic error effects from computerized decision-support
systems. Only 1 trial34 reported improvements in direct patient outcomes; whether improvements
were related to the comparison against the randomized concurrent control group or a
preintervention period was unclear.
Use of Additional Review Methods

The most common intervention type evaluated was the review of test interpretation
(n=36).9,29-31,40-71 Most studies showed a positive impact on diagnostic performance of an
additional review step (usually by a separate reader, sometimes from the same specialty and
other times from another specialty). However, in some cases, the detection of errors came at a
high cost in terms of additional false positives. Not all studies reported the tradeoffs between
sensitivity and specificity. Some of the studies targeted higher risk patients for enriched review.
However, the systems to support such targeting were neither described nor evaluated.
Diagnostic Techniques
The studies of interventions related to medical techniques (n=14)26,31,72-83 demonstrated that
technologies as well as diagnostic test selection might either enhance diagnosis (e.g., visual
enhancements via ultrasound-guided biopsy, changes to number of biopsy cores, cap-fitted
colonoscopy) or impede it (e.g., medical interventions for pain relief in patients with abdominal
pain). In the latter cases, the interventions hypothesized to impede diagnosis did not have that
effect, and interventions expected to enhance diagnostic accuracy did not always do so.
Personnel Changes
Six studies36,37,67,69,84,85 compared the impact on diagnosis of substituting one type of
professional for another, or adding another professional to the care team. The three studies67,69,85
that added a specialist to examine the interpretation of a test result reported an increase in case
detection, although the studies were quite small and targeted narrow patient populations.
391
Educational Interventions
Ten studies employed educational interventions35,61-64,86-90 for various targets: consumers,
community doctors, and intensive care unit doctors and nurses. Strategies targeted at
professionals produced improvements. Only two studies targeted consumers (parents, candidates
for screening) and both intervened on a behavior that occurs much earlier than actual diagnosis
(e.g., awareness of symptom seriousness with the intent of reducing office visits in ways that
would not adversely affect diagnosis)86
Structured Process Changes

Twenty six studies25,35,36,38,39,63,65,69,70,79-82,89,91-102 examined interventions that added structure
to the diagnostic process; this structure included, among other things, triage protocols, feedback
steps, and quality improvement processes (Q-Track, Toyota Production Process). Most
interventions included the addition of a tool, often a checklist or a form (i.e., to guide and
standardize physical examination of a patient). Some of the studies centered on laboratory
processes, whereas others occurred during clinical management. Results were mixed for these
types of interventions, with positive results (e.g., improved diagnosis) only among studies that
were not randomized, controlled trials (RCTs). Two of the three RCTs tested interventions in
mental health diagnosis.
Technology-Based Systems Interventions

Twenty nine studies9,27,28,32-34,103-117 included computerized decision support systems and
alerting systems (e.g., for abnormal lab results), most associated with improvements to processes
on the diagnostic pathway (e.g., critical laboratory value relayed to clinician in a more timely
manner).
Some interventions related to specific symptoms (e.g., computer aided diagnostic tool for
abdominal pain interpretation), while others intervened at the level of a particular test (e.g.,
electronic medical record alert for positive fecal occult blood (FOBT) cancer screening test
results).
Studies With Interventions that Corresponded to Multiple Categories

Twenty-four studies9,31,35,36,38,61-63,65-70,79-83,85,89,90,102,118 combined approaches in a variety of
ways and also covered a broad range of clinical areas, with mixed results. These studies are
included in the categories above. Twenty of the 23 studies combined two categories of
intervention in almost every permutation possible (11 of 15 combinations). All but three studies
included at least one of the two predominant categories in this set of multiple category
interventions: Additional Review Methods (11/23) and Structured Process Changes (13/23).
With combined approaches comes an inherent complexity in the intervention. However, the
results from studies of combined intervention strategies largely parallel those reported above.
With only one to four studies for any combination set, it is not possible to draw any conclusions
about whether benefits are enhanced with more complex interventions. In addition, these more
complex approaches may be more costly, but this information was not reported.
Notifying Patients of Test Results

Another potential grouping of PSPs focuses on the interface between the system and the
patient. Indeed interim care processes such as patient notification of test results has gained
attention at the national level.119 However, no studies evaluated this intervention with
392
comparative designs. The review by Singh and colleagues identified seven studies of patient
preferences or satisfaction with different options for receipt of test results.23 However, they also
found no studies that tested ways to reduce error using an intervention that affected test
notification. One of the articles identified in the Singh review by Casalino and colleagues found
a 7.1 percent rate of apparent failures to inform patients of an abnormal test result, and identified
an association between use of simple processes by physician practices for managing results and
lower failure rates.120 A systematic review of failures to follow-up test results with ambulatory
care patients reported that failed follow-ups ranged from 1 percent to 62 percent depending on
type of test result, and these failures were associated with missed cancer diagnoses. Electronic
record systems appeared to exert a mild protective effect against failed follow-ups, although the
authors note the pool of literature was small in this analysis.121

In general evaluations of PSPs have not assessed unintended adverse effects. However, some
of the screening test literature is applicable to maintaining a balanced perspective on diagnostic
error reduction. For example, an excluded study by Molins and colleagues122 reported on the
negative effects of multiple mammogram screening (patient anxiety, higher costs, poorer
subsequent screening attendance). Although this study did not involve an intervention to reduce
diagnostic error per se, it was similar to some of the included interventions with added testing.
Although none of the studies in our review evaluated direct patient harm, some reported false
positive rates.
How Has the Patient Safety Practice Been Implemented, and In

What Contexts?
The context in which a PSP is implemented depends on the specific type of diagnostic error and
PSP being examined. The studies identified in our literature search covered a range of
subspecialties, settings, and patient populations, with varying contexts. Most of the interventions
studied have not been tested in more than one site, with some even more appropriately
categorized as proof of concept. For diagnostic practice, another important context is the
sequence of events and the role of time itself. Sometimes these factors are embedded in the
patient safety target analyzed, as is the case for delayed diagnosis, which was an outcome in 26
studies included in the Appendix Supplementary Evidence Table.

The main source of information about costs related to diagnostic error is derived from
malpractice claims, as noted in an earlier section. In terms of costs of implementing some of the
PSPs reviewed, no information was reported, but would likely range from low to high depending
upon the PSP. For example, a PSP that involves an additional reviewer of imaging tests might
double the cost of that step in the diagnostic process for all patients, meaning a relatively large
investment per diagnostic error averted. For PSPs that compared the results of one technology to
another, the cost might be more or less, though often, technologies that perform with greater
accuracy cost more because they deliver a clinical benefit. For PSPs that revise a workflow to
follow a structured process, the start-up cost would depend on whether a structured process is
already available and can be adapted inexpensively or if workgroups have to spend significant
time to reengineer a local process. In either case, the cost may still be relatively low compared
393
with interventions that have ongoing incremental costs. Finally, information technology PSPs to
reduce diagnostic errors may be relatively expensive, though these costs could vary as well.

The evidence base for this topic does not yet include an examination of the influence of
contextual factors during implementation.

The original Making Health Care Safer report did not consider diagnostic errors because
just a decade ago, few studies had quantified the prevalence and clinical consequence of this
patient safety target. As a result, much of the literature over this period has focused on
quantifying the scope of the problem, and elucidating potential causal pathways that result in
failures in diagnosis. Very few intervention studies have tested strategies to reduce diagnostic
errors. However, frameworks for filling in the evidence gaps are beginning to emerge.
This review identified over 90 evaluations of interventions to reduce diagnostic errors, many
of which had a reported positive effect on at least one end point, including statistically significant
improvements in at least one end point in 10 of the 13 randomized trials. Mortality and morbidity
end points were seldom reported.
We also identified two previous systematic reviews of cognitive and systems-oriented
approaches to improve diagnostic accuracy that mostly found proof-of-concept strategies not yet
tested in practice. Our review built on the previous systematic reviews by grouping PSPs
targeting diagnostic errors from an organizational perspective into changes that an organization
might consider more generically (techniques investment; personnel configurations; additional
review steps for higher reliability; structured processes; education of professionals, patients,
families; and information and communications technologybased enhancements), as opposed to
individual clinicians looking for ways to improve their own cognitive processing in specific
diagnostic contexts. Although many of the PSPs tested thus far target diagnostic pathways for
specific symptoms or conditions, grouping interventions into common leverage points will
support future development in this field by the various stakeholders who seek to reduce
diagnostic problems. Involvement of patients and families has received minimal attention, with
only two studies addressing education of consumers.
Data synthesis is difficult because few studies have used randomized designs, comparable
outcomes, or similar interventions packages. The existing literature may be susceptible to
reporting biases favoring positive results for different interventions. It is expected that with
heightened awareness of the problem, the number of studies in this field will increase further in
the future, including more randomized trials and studies testing different approaches: for
example, policy-level efforts. However, the range of outcomes assessed in the studies that we
reviewed highlights the known lack of tools to routinely measure the effect of interventions to
decrease diagnostic errors. Additional work is needed on appropriate measurements of diagnostic
errors and consequential delays in diagnosis. A final limitation, especially for synthesis, is the
diversity of interventions that are reverse-engineered on the basis of the many diagnostic targets;
the diverse tailored needs for each clinical situation (for example, protocols designed for specific
work-up pathways); and the variety of specialized personnel, and even patients, receiving
educational or cognitive-support approaches.
Evidence is also lacking on the costs of interventions and implementation, particularly how
to reduce diagnostic errors without producing other diagnostic problems, such as overuse of
394
tests. Eventually reaching the correct diagnosis with inefficient testing strategies (for example,
some sequences of multiple test ordering) is not the appropriate pathway to improved diagnostic
safety. Our review found a paucity of studies that assessed both sensitivity and specificity of
interventions addressing diagnostic performance in the context of mitigating diagnostic errors.
Thus, although we found several promising interventions, evaluations need to be strengthened
before any specific PSPs are scaled up in this domain.
Alongside the literature scoping the problem and generating ideas for potential solutions,
some are also working on policy level efforts. Singh and Vij describe potential institutional-level
policies for communicating test results within the clinical team and to the patient.123 These types
of policies respond to national attention (e.g., the Joint Commission Patient Safety Goals),
spotlighting this part of the diagnostic pathway as ripe for intervention. They note that the area of
notifying patients about their test results is an emerging area for intervention testing.
In conclusion, our review demonstrates that the nascent field of diagnostic error research is
growing, with new interventions being tested that involve technical, cognitive, and systemsoriented strategies. The framework of intervention types developed in the review provides a basis
for categorizing and designing new studies, especially randomized, controlled trials, in these
areas. A summary table is located below (Table 2).
Scope of the
Strength of
the PSP
Effectiveness
Common/High
Emerging
practice (few
studies
available)
Evidence or
Potential for
Harmful
Unintended
Consequences
Uncertain
Estimate of
Cost
Varies
How Much do We
Varies
References
1.
Graber ML. Next steps: envisioning a

research agenda. Advances in health
sciences education : theory and practice.
2009;14 Suppl 1:107-12. Epub 2009/08/12.
doi: 10.1007/s10459-009-9183-1. PubMed
PMID: 19669917.
2.
Schiff GD, Hasan O, Kim S, Abrams R,

Cosby K, Lambert BL, et al. Diagnostic
error in medicine: analysis of 583 physicianreported errors. Archives of internal
medicine. 2009;169(20):1881-7. Epub
2009/11/11. doi:
10.1001/archinternmed.2009.333. PubMed
PMID: 19901140.
395
3.
Shojania KG, Burton EC, McDonald KM,

Goldman L. Changes in rates of autopsydetected diagnostic errors over time: a
systematic review. JAMA.
2003;289(21):2849-56. Epub 2003/06/05.
doi: 10.1001/jama.289.21.2849 [doi]
289/21/2849 [pii]. PubMed PMID:
12783916.
4.
Newswire P. Americans Are Concerned

About Hospital-Based Medical and Surgical
Errors. Wall Street Journal Onlines Health
Industry Edition. 2004.
5.
Schiff GD, Kim S, Abrams R, Cosby K,

Lambert B, Elstein AS, et al. Diagnosing
Diagnosis Errors: Lessons from a Multiinstitutional. In: Henriksen K, Battles JB,
Marks ES, Lewin DI, editors. Advances in
Implementation (Volume 2:. Advances in
Patient Safety. Rockville (MD)2005.
6.
7.
8.
9.
10.
11.
Singh H, Daci K, Petersen LA, Collins C,

Petersen NJ, Shethia A, et al. Missed
opportunities to initiate endoscopic
evaluation for colorectal cancer diagnosis.
Am J Gastroenterol. 2009;104(10):2543-54.
Epub 2009/06/25. doi: ajg2009324 [pii]
10.1038/ajg.2009.324. PubMed PMID:
19550418; PubMed Central PMCID:
PMC2758321.
Singh H, Hirani K, Kadiyala H, Rudomiotov
O, Davis T, Khan MM, et al. Characteristics
and predictors of missed opportunities in
lung cancer diagnosis: an electronic health
record-based study. J Clin Oncol.
2010;28(20):3307-15. Epub 2010/06/10.
doi: JCO.2009.25.6636 [pii]
10.1200/JCO.2009.25.6636. PubMed PMID:
PMC2903328.
Singh H, Thomas EJ, Wilson L, Kelly PA,
Pietz K, Elkeeb D, et al. Errors of diagnosis
in pediatric practice: a multisite survey.
Pediatrics. 2010;126(1):70-9. Epub
2010/06/23. doi: peds.2009-3218 [pii]
10.1542/peds.2009-3218. PubMed PMID:
PMC2921702.
Moore LJ, Jones SL, Kreiner LA, McKinley
B, Sucher JF, Todd SR, et al. Validation of a
screening tool for the early identification of
sepsis. The Journal of trauma.
2009;66(6):1539-46; discussion 46-7. Epub
2009/06/11. doi:
10.1097/TA.0b013e3181a3ac4b 00005373200906000-00005 [pii]. PubMed PMID:
19509612.
Phillips RL, Jr., Bartholomew LA, Dovey
SM, Fryer GE, Jr., Miyoshi TJ, Green LA.
Learning from malpractice claims about
negligent, adverse events in primary care in
the United States. Quality & safety in health
care. 2004;13(2):121-6. Epub 2004/04/08.
PubMed PMID: 15069219; PubMed Central
PMCID: PMC1743812.
Selbst SM. Pediatric emergency medicine:
legal briefs. Pediatric emergency care.
2005;21(3):214-8. Epub 2005/03/04.
PubMed PMID: 15744204.
396
12.
Tehrani AS, Lee H, Mathews S, Shore A,

Frick KD, Makary M, et al., editors. 20-year
Summary of U.S. Malpractice Claims for
Diagnostic Errors from 1985-2005. 33rd
Annual Meeting of the Society for Medical
Decision Making; 2011; Chicago, IL.
13.
Bishop TF, Ryan AM, Casalino LP. Paid

malpractice claims for adverse events in
inpatient and outpatient settings. JAMA.
2011;305(23):2427-31. Epub 2011/06/16.
doi: 305/23/2427 [pii]
10.1001/jama.2011.813. PubMed PMID:
21673294.
14.
Ely JW, Graber ML, Croskerry P. Checklists

to reduce diagnostic errors. Acad Med.
2011;86(3):307-13. Epub 2011/01/21. doi:
10.1097/ACM.0b013e31820824cd [doi].
15.
Cosby KS. A framework for classifying

factors that contribute to error in the
emergency department. Annals of
emergency medicine. 2003;42(6):815-23.
Epub 2003/11/25. doi:
10.1016/S0196064403006358. PubMed
PMID: 14634609.
16.
Tversky A, Kahneman D. Judgment under

Uncertainty: Heuristics and Biases. Science.
1974;185(4157):1124-31. Epub 1974/09/27.
doi: 185/4157/1124 [pii]
10.1126/science.185.4157.1124. PubMed
PMID: 17835457.
17.
Metcalfe J, Shimamura AP. Metacognition :

knowing about knowing. Cambridge, Mass.
London: MIT Press; 1994.
18.
Reason J. Human error: models and

management. BMJ. 2000;320(7237):768-70.
Epub 2000/03/17. PubMed PMID:
PMC1117770.
19.
Berner ES. Diagnostic error in medicine:

introduction. Advances in health sciences
education : theory and practice. 2009;14
Suppl 1:1-5. Epub 2009/08/12. doi:
10.1007/s10459-009-9187-x. PubMed
PMID: 19669914.
20.
Schiff GD, Bates DW. Can electronic

clinical documentation help prevent
diagnostic errors? N Engl J Med.
2010;362(12):1066-9. Epub 2010/03/26.
doi: 362/12/1066 [pii]
10.1056/NEJMp0911734 [doi]. PubMed
PMID: 20335582.
21.
Winters BD, Aswani MS, Pronovost PJ.

Commentary: Reducing diagnostic errors:
another role for checklists? Acad Med.
2011;86(3):279-81. Epub 2011/02/25. doi:
10.1097/ACM.0b013e3182082692 [doi]
00001888-201103000-00011 [pii]. PubMed
PMID: 21346432.
22.
Wahls TL, Cram P. Proposed interventions

to decrease the frequency of missed test
results. Advances in health sciences
Suppl 1:51-6. Epub 2009/08/12. doi:
10.1007/s10459-009-9180-4. PubMed
PMID: 19669920.
23.
24.
25.
26.
27.
Singh H, Graber M, Kissam SM, Sorensen

AV, Lenfestey NF, Tant EM, et al. Systemrelated interventions to reduce diagnostic
errors: a narrative review. BMJ Qual Saf.
2011. doi: 10.1136/bmjqs-2011-000150.
Graber ML, Kissam S, Payne VL, Meyer
AN, Sorensen A, Lenfestey N, et al.
Cognitive interventions to reduce diagnostic
error: a narrative review. BMJ Qual Saf.
2012;21(7):535-57. Epub 2012/05/01. doi:
bmjqs-2011-000149 [pii] 10.1136/bmjqs2011-000149. PubMed PMID: 22543420.
Perno JF, Schunk JE, Hansen KW, Furnival
RA. Significant reduction in delayed
diagnosis of injury with implementation of a
pediatric trauma service. Pediatric
emergency care. 2005;21(6):367-71. Epub
2005/06/09. doi: 00006565-20050600000003 [pii]. PubMed PMID: 15942513.
Thomas SH, Silen W, Cheema F, Reisner A,
Aman S, Goldstein JN, et al. Effects of
morphine analgesia on diagnostic accuracy
in Emergency Department patients with
abdominal pain: a prospective, randomized
trial. J Am Coll Surg. 2003;196(1):18-31.
Epub 2003/01/09. doi: S10727515(02)01480-1 [pii]. PubMed PMID:
12517545.
Bogusevicius A, Maleckas A, Pundzius J,
Skaudickas D. Prospective randomised trial
of computer-aided diagnosis and contrast
radiography in acute small bowel
obstruction. Eur J Surg. 2002;168(2):78-83.
Epub 2002/07/13. doi:
10.1080/11024150252884287. PubMed
PMID: 12113275.
397
28.
Kuperman GJ, Teich JM, Tanasijevic MJ,

MaLuf N, Rittenberg E, Jha A, et al.
Improving response to critical laboratory
results with automation: results of a
randomized controlled trial. J Am Med
Inform Assoc. 1999;6(6):512-22. Epub
1999/12/01. PubMed PMID: 10579608;
PubMed Central PMCID: PMC61393.
29.
Dudley M, Channer KS. Assessment of the

value of technician reporting of
electrocardiographs in an accident and
emergency department. J Accid Emerg Med.
1997;14(5):307-10. Epub 1997/10/08.
PMCID: PMC1343097.
30.
Nam YS, Pikarsky AJ, Wexner SD, Singh

JJ, Weiss EG, Nogueras JJ, et al.
Reproducibility of colonic transit study in
patients with chronic constipation. Dis
Colon Rectum. 2001;44(1):86-92. Epub
2002/01/24. PubMed PMID: 11805568.
31.
Beigi B, Uddin JM, McMullan TF, Linardos

E. Inaccuracy of diagnosis in a cohort of
patients on the waiting list for
dacryocystorhinostomy when the diagnosis
was made by only syringing the lacrimal
system. Eur J Ophthalmol. 2007;17(4):4859. Epub 2007/08/03. PubMed PMID:
17671919.
32.
Major K, Shabot MM, Cunneen S. Wireless

clinical alerts and patient outcomes in the
surgical intensive care unit. Am Surg.
2002;68(12):1057-60. Epub 2003/01/09.
33.
Etchells E, Adhikari NK, Wu R, Cheung M,

Quan S, Mraz R, et al. Real-time automated
paging and decision support for critical
laboratory abnormalities. BMJ Qual Saf.
2011;20(11):924-30. Epub 2011/07/05. doi:
bmjqs.2010.051110 [pii]
10.1136/bmjqs.2010.051110. PubMed
PMID: 21725046.
34.
Fitzgerald M, Cameron P, Mackenzie C,

Farrow N, Scicluna P, Gocentas R, et al.
Trauma resuscitation errors and computerassisted decision support. Arch Surg.
2011;146(2):218-25. Epub 2011/02/23. doi:
146/2/218 [pii] 10.1001/archsurg.2010.333.
35.
Chern CH, How CK, Wang LM, Lee CH,

Graff L. Decreasing clinically significant
adverse events using feedback to emergency
physicians of telephone follow-up outcomes.
Annals of emergency medicine.
2005;45(1):15-23. Epub 2005/01/07. doi:
S0196064404012880 [pii]
10.1016/j.annemergmed.2004.08.012 [doi].
36.
Vernon DD, Furnival RA, Hansen KW,

Diller EM, Bolte RG, Johnson DG, et al.
Effect of a pediatric trauma response team
on emergency department treatment time
and mortality of pediatric trauma victims.
1999/01/26. PubMed PMID: 9917434.
37.
38.
39.
40.
41.
Duijm LE, Groenewoud JH, Fracheboud J,

de Koning HJ. Additional double reading of
screening mammograms by radiologic
technologists: impact on screening
performance parameters. J Natl Cancer Inst.
2007;99(15):1162-70. Epub 2007/07/27.
doi: djm050 [pii] 10.1093/jnci/djm050 [doi].
42.
Howard J, Sundararajan R, Thomas SG,

Walsh M, Sundararajan M. Reducing missed
injuries at a level II trauma center. J Trauma
Nurs. 2006;13(3):89-95. Epub 2006/10/21.
43.
Manion E, Cohen MB, Weydert J.

Mandatory second opinion in surgical
pathology referral material: clinical
consequences of major disagreements. The
American journal of surgical pathology.
2008;32(5):732-7. Epub 2008/03/25. doi:
10.1097/PAS.0b013e31815a04f5. PubMed
PMID: 18360282.
44.
Nordrum I, Johansen M, Amin A, Isaksen

V, Ludvigsen JA. Diagnostic accuracy of
second-opinion diagnoses based on still
images. Hum Pathol. 2004;35(1):129-35.
Epub 2004/01/28. doi: S0046817703004982
[pii]. PubMed PMID: 14745735.
45.
Schriger DL, Gibbons PS, Langone CA, Lee

S, Altshuler LL. Enabling the diagnosis of
occult psychiatric illness in the emergency
department: A randomized, controlled trial
of the computerized, self-administered
PRIME-MD diagnostic system. Annals of
PubMed PMID: ISI:000166803800002.
Raab SS, Stone CH, Jensen CS, Zarbo RJ,

Meier FA, Grzybicki DM, et al. Double
slide viewing as a cytology quality
improvement initiative. Am J Clin Pathol.
2006;125(4):526-33. Epub 2006/04/22. doi:
TKNM9GRGPL2JT95Y [pii]
10.1309/TKNM-9GRG-PL2J-T95Y [doi].
46.
Ciatto S, Del Turco MR, Morrone D, Catarzi

S, Ambrogetti D, Cariddi A, et al.
Independent double reading of screening
mammograms. J Med Screen. 1995;2(2):99101. Epub 1995/01/01. PubMed PMID:
7497164.
Singh P, Warnakulasuriya S. The two-week

wait cancer initiative on oral cancer; the
predictive value of urgent referrals to an oral
medicine unit. British dental journal.
2006;201(11):717-20; discussion 4. Epub
2006/12/13. doi: 10.1038/sj.bdj.4814304.
47.
Bruner JM, Inouye L, Fuller GN, Langford

LA. Diagnostic discrepancies and their
clinical impact in a neuropathology referral
practice. Cancer. 1997;79(4):796-803. Epub
1997/02/15. doi: 10.1002/(SICI)10970142(19970215)79:4<796::AIDCNCR17>3.0.CO;2-V [pii]. PubMed PMID:
9024718.
Sakr M, Angus J, Perrin J, Nixon C, Nicholl

J, Wardrope J. Care of minor injuries by
emergency nurse practitioners or junior
doctors: a randomised controlled trial.
Lancet. 1999;354(9187):1321-6. Epub
1999/10/26. doi: S0140673699024472 [pii].
Rollman BL, Hanusa BH, Lowe HJ, Gilbert
T, Kapoor WN, Schulberg HC. A
randomized trial using computerized
decision support to improve treatment of
major depression in primary care. J Gen
Intern Med. 2002;17(7):493-503. Epub
2002/07/23. doi: jgi10421 [pii]. PubMed
PMID: 12133139; PubMed Central PMCID:
PMC1495078.
398
48.
Canon CL, Smith JK, Morgan DE, Jones

BC, Fell SC, Kenney PJ, et al. Double
reading of barium enemas: is it necessary?
AJR American journal of roentgenology.
2003;181(6):1607-10. Epub 2003/11/25.
49.
Carew-McColl M. Radiological
interpretation in an accident and emergency
department. Br J Clin Pract. 1983;37(1112):375-7. Epub 1983/11/01. PubMed
PMID: 6671078.
50.
51.
56.
Parameswaran L, Prihoda TJ, Sharkey FE.

Diagnostic efficacy of additional stepsections in colorectal biopsies originally
diagnosed as normal. Hum Pathol.
2008;39(4):579-83. Epub 2008/02/22. doi:
S0046-8177(07)00476-5 [pii]
10.1016/j.humpath.2007.08.015 [doi].
57.
Espinosa JA, Nolan TW. Reducing errors

made by emergency physicians in
interpreting radiographs: longitudinal study.
BMJ. 2000;320(7237):737-40. Epub
2000/03/17. PubMed PMID: 10720354;
Raab SS, Grzybicki DM, Mahood LK,

Parwani AV, Kuan SF, Rao UN.
Effectiveness of random and focused review
in detecting surgical pathology error. Am J
Clin Pathol. 2008;130(6):905-12. Epub
2008/11/21. doi: R8245P6P760X8402 [pii]
10.1309/AJCPPIA5D7MYKDWF [doi].
58.
Galasko CS, Monahan PR. Value of reexamining x-ray films of outpatients

attending accident services. Br Med J.
1971;1(5750):643-4. Epub 1971/03/20.
PMCID: PMC1795440.
Robson N, van Benthem PP, Gan R, Dixon

AK. Casualty X-ray reporting: a student
survey. Clin Radiol. 1985;36(5):479-81.
4075715.
59.
Thiesse P, Ollivier L, Di Stefano-Louineau

D, Negrier S, Savary J, Pignard K, et al.
Response rate accuracy in oncology trials:
reasons for interobserver variability. Groupe
Francais dImmunotherapie of the
Federation Nationale des Centres de Lutte
Contre le Cancer. J Clin Oncol.
1997;15(12):3507-14. Epub 1997/12/13.
60.
Westra WH, Kronz JD, Eisele DW. The

impact of second opinion surgical pathology
on the practice of head and neck surgery: a
decade experience at a large referral
hospital. Head & neck. 2002;24(7):684-93.
Epub 2002/07/12. doi: 10.1002/hed.10105.
61.
Seltzer SE, Hessel SJ, Herman PG,

Swensson RG, Sheriff CR. Resident film
interpretations and staff review. AJR
American journal of roentgenology.
1981;137(1):129-33. Epub 1981/07/01.
62.
McPhee SJ, Bird JA, Jenkins CN, Fordham

D. Promoting cancer screening. A
randomized, controlled trial of three
interventions. Archives of internal medicine.
1989;149(8):1866-72. Epub 1989/08/01.
52.
Kwek BH, Lau TN, Ng FC, Gao F. Nonconsensual double reading in the Singapore
Breast Screening Project: benefits and
limitations. Annals of the Academy of
Medicine, Singapore. 2003;32(4):438-41.
12968545.
53.
Lind AC, Bewtra C, Healy JC, Sims KL.

Prospective peer review in surgical
pathology. Am J Clin Pathol.
1995;104(5):560-6. Epub 1995/11/01.
54.
Lufkin KC, Smith SW, Matticks CA,

Brunette DD. Radiologists review of
radiographs interpreted confidently by
emergency physicians infrequently leads to
changes in patient management. Annals of
Epub 1998/02/24. PubMed PMID: 9472181.
55.
Murphy R, Slater A, Uberoi R, Bungay H,

Ferrett C. Reduction of perception error by
double reporting of minimal preparation CT
colon. Br J Radiol. 2010;83(988):331-5.
Epub 2009/08/05. doi: 65634575 [pii]
10.1259/bjr/65634575 [doi]. PubMed
PMID: 19651707.
399
63.
Kundel HL, Nodine CF, Krupinski EA.

Computer-displayed eye position as a visual
aid to pulmonary nodule interpretation.
Invest Radiol. 1990;25(8):890-6. Epub
1990/08/01. PubMed PMID: 2394571.
64.
Thomas HG, Mason AC, Smith RM,

Fergusson CM. Value of radiograph audit in
an accident service department. Injury.
1992;23(1):47-50. Epub 1992/01/01. doi:
0020-1383(92)90126-D [pii]. PubMed
PMID: 1541500.
65.
Ross PD, Huang C, Karpf D, Lydick E, Coel

M, Hirsch L, et al. Blinded reading of
radiographs increases the frequency of
errors in vertebral fracture detection. J Bone
Miner Res. 1996;11(11):1793-800. Epub
1996/11/01. doi: 10.1002/jbmr.5650111124
[doi]. PubMed PMID: 8915788.
66.
Jiang Y, Nishikawa RM, Schmidt RA,

Toledano AY, Doi K. Potential of computeraided diagnosis to reduce variability in
radiologists interpretations of
mammograms depicting microcalcifications.
Radiology. 2001;220(3):787-94. Epub
2001/08/30. PubMed PMID: 11526283.
67.
Trotter MJ, Bruecks AK. Interpretation of

skin biopsies by general pathologists:
diagnostic discrepancy rate measured by
blinded review. Arch Pathol Lab Med.
2003;127(11):1489-92. Epub 2003/10/22.
doi: OA3050 [pii] 10.1043/15432165(2003)127<1489:IOSBBG>2.0.CO;2
68.
Peldschus K, Herzog P, Wood SA, Cheema

JI, Costello P, Schoepf UJ. Computer-aided
diagnosis as a second reader: spectrum of
findings in CT studies of the chest
interpreted as normal. Chest.
2005;128(3):1517-23. Epub 2005/09/16.
doi: 10.1378/chest.128.3.1517. PubMed
PMID: 16162752.
69.
Tsai JJ, Yeun JY, Kumar VA, Don BR.

Comparison and interpretation of urinalysis
performed by a nephrologist versus a
hospital-based clinical laboratory. Am J
Kidney Dis. 2005;46(5):820-9. Epub
2005/10/29. doi: S0272-6386(05)01050-4
[pii] 10.1053/j.ajkd.2005.07.039 [doi].
400
70.
Goodyear N, Ulness BK, Prentice JL,

Cookson BT, Limaye AP. Systematic
assessment of culture review as a tool to
assess errors in the clinical microbiology
laboratory. Arch Pathol Lab Med.
2008;132(11):1792-5. Epub 2008/11/04.
doi: 2008-0037-OA [pii] 10.1043/15432165-132.11.1792. PubMed PMID:
18976017.
71.
Hamady ZZ, Mather N, Lansdown MR,

Davidson L, Maclennan KA. Surgical
pathological second opinion in thyroid
malignancy: impact on patients
management and prognosis. Eur J Surg
Oncol. 2005;31(1):74-7. Epub 2005/01/12.
doi: S0748-7983(04)00244-6 [pii]
10.1016/j.ejso.2004.08.010. PubMed PMID:
15642429.
72.
Attard AR, Corlett MJ, Kidner NJ, Leslie

AP, Fraser IA. Safety of early pain relief for
acute abdominal pain. BMJ.
1992;305(6853):554-6. Epub 1992/09/05.
PMCID: PMC1883284.
73.
Resnick NM, Brandeis GH, Baumann MM,

DuBeau CE, Yalla SV. Misdiagnosis of
urinary incontinence in nursing home
women: prevalence and a proposed solution.
Neurourol Urodyn. 1996;15(6):599-613;
discussion -8. Epub 1996/01/01. doi:
10.1002/(SICI)15206777(1996)15:6<599::AIDNAU2>3.0.CO;2-A [pii]
10.1002/(SICI)15206777(1996)15:6<599::AIDNAU2>3.0.CO;2-A [doi]. PubMed PMID:
8916113.
74.
Borgstein PJ, Gordijn RV, Eijsbouts QA,

Cuesta MA. Acute appendicitis--a clear-cut
case in men, a guessing game in young
women. A prospective study on the role of
laparoscopy. Surg Endosc. 1997;11(9):9237. Epub 1997/09/19. PubMed PMID:
9294274.
75.
Vermeulen B, Morabia A, Unger PF,

Goehring C, Grangier C, Skljarov I, et al.
Acute appendicitis: influence of early pain
relief on the accuracy of clinical and US
findings in the decision to operate--a
randomized trial. Radiology.
1999;210(3):639-43. Epub 1999/04/20.
76.
77.
78.
79.
80.
81.
Prieto VG, Argenyi ZB, Barnhill RL, Duray

PH, Elenitsas R, From L, et al. Are en face
frozen sections accurate for diagnosing
margin status in melanocytic lesions? Am J
Clin Pathol. 2003;120(2):203-8. Epub
2003/08/23. doi: 10.1309/J1Q0-V35EUTMV-R193 [doi]. PubMed PMID:
12931550.
82.
Ravery V, Goldblatt L, Royer B, Blanc E,

Toublanc M, Boccon-Gibod L. Extensive
biopsy protocol improves the detection rate
of prostate cancer. The Journal of urology.
2000;164(2):393-6. Epub 2000/07/14.
83.
Weatherburn G, Bryan S, Nicholas A, Cocks

R. The effect of a picture archiving and
communications system (PACS) on
diagnostic performance in the accident and
2000;17(3):180-4. Epub 2000/05/20.
PMCID: PMC1725383.
84.
Hewett DG, Rex DK. Cap-fitted

colonoscopy: a randomized, tandem
colonoscopy study of adenoma miss rates.
Gastrointest Endosc. 2010;72(4):775-81.
Epub 2010/06/29. doi: S00165107(10)01558-0 [pii]
10.1016/j.gie.2010.04.030 [doi]. PubMed
PMID: 20579648.
de Lacey G, Barker A, Harper J, Wignall B.

An assessment of the clinical effects of
reporting accident and emergency
radiographs. Br J Radiol. 1980;53(628):3049. Epub 1980/04/01. PubMed PMID:
7378697.
85.
Brossner C, Madersbacher S, Bayer G,

Pycha A, Klingler HC, Maier U.
Comparative study of two different TRUSguided sextant biopsy techniques in
detecting prostate cancer in one biopsy
session. European urology. 2000;37(1):6571. Epub 2000/02/15. PubMed PMID:
10671788.
Jacobs MJ, Edmondson MJ, Lowry JC.

Accuracy of diagnosis of fractures by
maxillofacial and accident and emergency
doctors using plain radiography compared
with a telemedicine system: a prospective
study. Br J Oral Maxillofac Surg.
2002;40(2):156-62. Epub 2002/08/16.
86.
McCarthy PL, Sznajderman SD, LustmanFindling K, Baron MA, Fink HD,

Czarkowski N, et al. Mothers clinical
judgment: a randomized trial of the Acute
Illness Observation Scales. J Pediatr.
1990;116(2):200-6. Epub 1990/02/01.
87.
Fridriksson S, Hillman J, Landtblom AM,

Boive J. Education of referring doctors
about sudden onset headache in
subarachnoid hemorrhage. A prospective
study. Acta Neurol Scand. 2001;103(4):23842. Epub 2001/05/01. doi: ane065 [pii].
88.
Thaler T, Tempelmann V, Maggiorini M,

Rudiger A. The frequency of
electrocardiographic errors due to electrode
cable switches: a before and after study. J
Electrocardiol. 2010;43(6):676-81. Epub
2010/07/02. doi: S0022-0736(10)00203-7
[pii] 10.1016/j.jelectrocard.2010.05.007
Kokki H, Lintula H, Vanamo K, Heiskanen

M, Eskelinen M. Oxycodone vs placebo in
children with undifferentiated abdominal
pain: a randomized, double-blind clinical
trial of the effect of analgesia on diagnostic
accuracy. Arch Pediatr Adolesc Med.
2005;159(4):320-5. Epub 2005/04/06. doi:
159/4/320 [pii] 10.1001/archpedi.159.4.320.
Naughton CK, Miller DC, Mager DE,

Ornstein DK, Catalona WJ. A prospective
randomized trial comparing 6 versus 12
prostate biopsy cores: impact on cancer
detection. The Journal of urology.
2000;164(2):388-92. Epub 2000/07/14.
Presti JC, Jr., Chang JJ, Bhargava V,
Shinohara K. The optimal systematic
prostate biopsy scheme should include 8
rather than 6 biopsies: results of a
prospective clinical trial. The Journal of
urology. 2000;163(1):163-6; discussion 6-7.
10604337.
401
89.
Gleadhill DN, Thomson JY, Simms P. Can

more efficient use be made of x ray
examinations in the accident and emergency
department? Br Med J (Clin Res Ed).
1987;294(6577):943-7. Epub 1987/04/11.
PMCID: PMC1246007.
90.
Linver MN, Paster SB, Rosenberg RD, Key

CR, Stidley CA, King WV. Improvement in
mammography interpretation skills in a
community radiology practice after
dedicated teaching courses: 2-year medical
audit of 38,633 cases. Radiology.
1992;184(1):39-43. Epub 1992/07/01.
91.
92.
93.
Enderson BL, Reath DB, Meadors J, Dallas

W, DeBoo JM, Maull KI. The tertiary
trauma survey: a prospective study of
missed injury. The Journal of trauma.
1990;30(6):666-9; discussion 9-70. Epub
1990/06/01. PubMed PMID: 2352294.
Klassen TP, Ropp LJ, Sutcliffe T, Blouin R,
Dulberg C, Raman S, et al. A randomized,
controlled trial of radiograph ordering for
extremity trauma in a pediatric emergency
department. Annals of emergency medicine.
1993;22(10):1524-9. Epub 1993/10/01. doi:
S0196-0644(05)81252-1 [pii]. PubMed
PMID: 8214829.
Biffl WL, Harrington DT, Cioffi WG.
Implementation of a tertiary trauma survey
decreases missed injuries. The Journal of
trauma. 2003;54(1):38-43; discussion -4.
Epub 2003/01/25. doi:
10.1097/01.TA.0000046379.29204.CE
94.
Soundappan SV, Holland AJ, Cass DT. Role

of an extended tertiary survey in detecting
missed injuries in children. The Journal of
trauma. 2004;57(1):114-8; discussion 8.
15284560.
95.
Ursprung R, Gray JE, Edwards WH, Horbar

JD, Nickerson J, Plsek P, et al. Real time
patient safety audits: improving safety every
day. Quality & safety in health care.
2005;14(4):284-9. Epub 2005/08/04. doi:
14/4/284 [pii] 10.1136/qshc.2004.012542.
PMCID: PMC1744058.
402
96.
Raab SS, Andrew-Jaja C, Condel JL, Dabbs

DJ. Improving Papanicolaou test quality and
reducing medical errors by using Toyota
production system methods. Am J Obstet
Gynecol. 2006;194(1):57-64. Epub
2006/01/04. doi: S0002-9378(05)00972-5
[pii] 10.1016/j.ajog.2005.06.069. PubMed
PMID: 16389010.
97.
Raab SS, Grzybicki DM, Sudilovsky D,

Balassanian R, Janosky JE, Vrbin CM.
Effectiveness of Toyota process redesign in
reducing thyroid gland fine-needle
aspiration error. Am J Clin Pathol.
2006;126(4):585-92. Epub 2006/08/30. doi:
NJQ1L7KA10UKV93Q [pii]
10.1309/NJQ1L7KA10UKV93Q [doi].
98.
Raab SS, Tworek JA, Souers R, Zarbo RJ.

The value of monitoring frozen sectionpermanent section correlation data over
time. Arch Pathol Lab Med.
2006;130(3):337-42. Epub 2006/03/08. doi:
CP5722 [pii] 10.1043/15432165(2006)130[337:TVOMFS]2.0.CO;2
99.
Raab SS, Jones BA, Souers R, Tworek JA.

The effect of continuous monitoring of
cytologic-histologic correlation data on
cervical cancer screening performance. Arch
Pathol Lab Med. 2008;132(1):16-22. Epub
2008/01/10. doi: 2007-0197-CPR1 [pii]
10.1043/15432165(2008)132[16:TEOCMO]2.0.CO;2
100.
Mueller CA, Klaassen-Mielke R, Penner E,

Junius-Walker U, Hummers-Pradier E,
Theile G. Disclosure of new health problems
and intervention planning using a geriatric
assessment in a primary care setting. Croat
Med J. 2010;51(6):493-500. Epub
2010/12/17. PubMed PMID: 21162161;
101.

AM, Smorenburg SM, Gouma DJ,
Boermeester MA. Prevention of surgical
safety checklist. Ann Surg.
2011;253(3):624-8. Epub 2011/01/07. doi:
10.1097/SLA.0b013e3182068880. PubMed
PMID: 21209590.
102.
Lewis G, Sharp D, Bartholomew J, Pelosi

AJ. Computerized assessment of common
mental disorders in primary care: effect on
clinical outcome. Fam Pract.
1996;13(2):120-6. Epub 1996/04/01.
103.
Wexler JR, Swender PT, Tunnessen WW,

Jr., Oski FA. Impact of a system of
computer-assisted diagnosis. Initial
evaluation of the hospitalized patient. Am J
Dis Child. 1975;129(2):203-5. Epub
1975/02/11. PubMed PMID: 1091140.
104.
105.
106.
107.
108.
Wellwood J, Johannessen S, Spiegelhalter

DJ. How does computer-aided diagnosis
improve the management of acute
abdominal pain? Ann R Coll Surg Engl.
1992;74(1):40-6. Epub 1992/01/01. PubMed
PMC2497469.
Selker HP, Beshansky JR, Griffith JL,
Aufderheide TP, Ballin DS, Bernard SA, et
al. Use of the acute cardiac ischemia timeinsensitive predictive instrument (ACI-TIPI)
to assist with triage of patients with chest
pain or other symptoms suggestive of acute
cardiac ischemia. A multicenter, controlled
clinical trial. Ann Intern Med.
1998;129(11):845-55. Epub 1998/12/29.
Poon EG, Kuperman GJ, Fiskio J, Bates
DW. Real-time notification of laboratory
data requested by users through
alphanumeric pagers. J Am Med Inform
Assoc. 2002;9(3):217-22. Epub 2002/04/25.
PMCID: PMC344581.
Gur D, Sumkin JH, Rockette HE, Ganott M,
Hakim C, Hardesty L, et al. Changes in
breast cancer detection and mammography
recall rates after the introduction of a
computer-aided detection system. J Natl
Cancer Inst. 2004;96(3):185-90. Epub
2004/02/05. PubMed PMID: 14759985.
Kakeda S, Moriya J, Sato H, Aoki T,
Watanabe H, Nakata H, et al. Improved
detection of lung nodules on chest
radiographs using a commercial computeraided diagnosis system. AJR American
journal of roentgenology. 2004;182(2):50510. Epub 2004/01/23. PubMed PMID:
14736690.
403
109.
Cupples TE, Cunningham JE, Reynolds JC.

Impact of computer-aided detection in a
regional screening mammography program.
2005;185(4):944-50. Epub 2005/09/24. doi:
10.2214/AJR.04.1300. PubMed PMID:
16177413.
110.
Ramnarayan P, Winrow A, Coren M,

Nanduri V, Buchdahl R, Jacobs B, et al.
Diagnostic omission errors in acute
paediatric practice: impact of a reminder
system on decision-making. BMC medical
informatics and decision making. 2006;6:37.
Epub 2006/11/08. doi: 10.1186/1472-69476-37. PubMed PMID: 17087835; PubMed
Central PMCID: PMC1654143.
111.
Fenton JJ, Taplin SH, Carney PA, Abraham

L, Sickles EA, DOrsi C, et al. Influence of
computer-aided detection on performance of
screening mammography. N Engl J Med.
2007;356(14):1399-409. Epub 2007/04/06.
doi: 10.1056/NEJMoa066099. PubMed
PMC3182841.
112.
Park HI, Min WK, Lee W, Park H, Park CJ,

Chi HS, et al. Evaluating the short message
service alerting system for critical value
notification via PDA telephones. Ann Clin
Lab Sci. 2008;38(2):149-56. Epub
2008/05/13. doi: 38/2/149 [pii]. PubMed
PMID: 18469361.
113.
Piva E, Sciacovelli L, Zaninotto M,

Laposata M, Plebani M. Evaluation of
effectiveness of a computerized notification
system for reporting critical values. Am J
Clin Pathol. 2009;131(3):432-41. Epub
2009/02/21. doi: 131/3/432 [pii]
10.1309/AJCPYS80BUCBXTUH. PubMed
PMID: 19228648.
114.
Singh H, Wilson L, Petersen LA, Sawhney

MK, Reis B, Espadas D, et al. Improving
follow-up of abnormal cancer screens using
electronic health records: trust but verify test
result communication. BMC medical
Epub 2009/12/17. doi: 1472-6947-9-49 [pii]
10.1186/1472-6947-9-49. PubMed PMID:
PMC2797509.
115.
David CV, Chira S, Eells SJ, Ladrigan M,

Papier A, Miller LG, et al. Diagnostic
accuracy in patients admitted to hospitals
with cellulitis. Dermatol Online J.
2011;17(3):1. Epub 2011/03/24. PubMed
PMID: 21426867.
116.
Olsson SE, Ohlsson M, Ohlin H, Dzaferagic

S, Nilsson ML, Sandkull P, et al. Decision
support for the initial triage of patients with
acute coronary syndromes. Clin Physiol
Funct Imaging. 2006;26(3):151-6. Epub
2006/04/28. doi: CPF669 [pii]
10.1111/j.1475-097X.2006.00669.x.
117.
Pozen MW, DAgostino RB, Selker HP,

Sytkowski PA, Hood WB, Jr. A predictive
instrument to improve coronary-care-unit
admission practices in acute ischemic heart
disease. A prospective multicenter clinical
trial. N Engl J Med. 1984;310(20):1273-8.
Epub 1984/05/17. doi:
10.1056/NEJM198405173102001. PubMed
PMID: 6371525.
118.
Thomas DC, Spitzer WO, MacFarlane JK.

Inter-observer error among surgeons and
nurses in presymptomatic detection of breast
disease. J Chronic Dis. 1981;34(12):617-26.
Epub 1981/01/01. PubMed PMID: 7309826.
119.
Davis Giardina T SH. Should patients get

direct access to their laboratory test results?:
An answer with many questions. JAMA:
The Journal of the American Medical
Association. 2011;306(22):2502-3. doi:
10.1001/jama.2011.1797.
120.
Casalino LP, Dunham D, Chin MH, Bielang

R, Kistner EO, Karrison TG, et al.
Frequency of Failure to Inform Patients of
Clinically Significant Outpatient Test
Results. Arch Intern Med.
2009;169(12):1123-9.
121.
Callen JL, Westbrook JI, Georgiou A, Li J.

Failure to Follow-Up Test Results for
Ambulatory Patients: A Systematic Review.
J Gen Intern Med. 2011. Epub 2011/12/21.
doi: 10.1007/s11606-011-1949-5. PubMed
PMID: 22183961.
404
122.
Molins E, Comas M, Roman R, RodriguezBlanco T, Sala M, Macia F, et al. Effect of

participation on the cumulative risk of falsepositive recall in a breast cancer screening
programme. Public Health.
2009;123(9):635-7. Epub 2009/09/08. doi:
S0033-3506(09)00198-X [pii]
10.1016/j.puhe.2009.07.007 [doi]. PubMed
PMID: 19733372.
123.
Singh H, Vij M. Eight recommendations for

policies for communicating abnormal test
results. Jt Comm J Qual Patient Saf.
2010;36:226-32.
Chapter 36. Monitoring Patient Safety Problems (NEW)


Adverse events (AEs) associated with medical treatments are a major source of morbidity
and mortality.1-4 Studies showed that the incidence of AEs varied from 3 percent to 17 percent of
hospitalized patients,5 and about 50 percent of the AEs were judged to be preventable. Most AEs
resulted in minor or temporary disability, but a proportion of the AEs, 4 percent to 21 percent,
contributed to death.5
In 1999, the Institute of Medicine (IOM) published a landmark report on medical errors titled
To Err Is Human: Building a Safer Health Care System.6 Since IOM released the report,
several studies have examined progress in patient safety and have found little evidence of
systematic improvements in the health care system.1,7-10 According to a 2008 Healthcare Cost
and Utilization Project statistical brief, drug-related adverse outcomes were noted in nearly
1.9 million inpatient hospital stays (4.7% of all stays) and 838,000 treat-and-release emergency
department visits (0.8% of all visits).10 The Institute for Healthcare Improvement estimated that
nearly 15 million instances of medical harm occur in the United States (U.S.) each year.11 Over
the five years from 2004 to 2008, drug-related adverse outcomes in the inpatient setting
increased by 52 percent.10 This increase in AEs could be the result of the intensified effort in
incident reporting; still, keeping patients from being harmed by preventable medical errors will
continue to be a challenging goal for the medical community.
As used in this review, an AE is defined as an event that results in unintended harm to the
patient by an act of commission or omission rather than by the underlying disease or condition of
the patient.12 A medical error is the failure of a planned action to be completed as intended or the
use of a wrong plan to achieve an aim.12 AEs include medical errors as well as more general
substandard care that can result in harm, such as harm caused by incorrect diagnoses or lack of
patient monitoring during treatment.13 Therefore, AEs do not always involve errors, negligence,
or poor quality of care and may not always be preventable.
Near miss is another term often used in patient safety monitoring. It is defined as an event
or a situation that did not produce patient harm, but only because of intervening factors, such as
patient health or timely intervention.12

Health care organizations use a wide array of methods to uncover and monitor AEs and
errors in medical care. These methods include incident reporting, direct observation of patient
care, chart review, analysis of malpractice claims, patient complaints and reports to risk
management, executive walk rounds, trigger-tool use, patient interviews, morbidity and mortality
conferences, autopsy, and clinical surveillance.14-19 These methods vary in the timing of finding
AEs (retrospective or real-time), and each has advantages and limitations.14-18
Historically, medical errors were revealed retrospectively through morbidity and mortality
committees, autopsy, and malpractice claims data.6,14-16,18 While these methods provide valuable
information on medical errors, they are not appropriate for measuring the incidence or
prevalence of the errors or events. They might also be limited by hindsight bias (e.g., a tendency
405
to rate care in the context of a bad outcome as substandard), if the evaluators are not blinded to
outcome.15
Chart review was often used as the benchmark for estimating the extent of medical harms in
hospitals or as the gold standard in patient safety studies to quantify AE rates.15,16 However,
chart review is generally resource-intensive.15 Incomplete documentation in the medical record
can affect the ability to detect the potential causes of AEs.15 Near misses that produce no injury
are rarely detected by this method.15,16 The reliability (precision) of the judgments about the
presence of AEs by chart reviewers could also be low.15
Incident reporting systems are a popular mechanism that the majority of hospitals rely on to
uncover internal threats to patient safety.1-4,20,21 Since IOM endorsed using incident reporting
systems in its landmark report on patient safety, 27 states and the District of Columbia have
established hospital AE reporting systems.13,22 Reporting systems include surveys of providers
and structured interviews and can provide rich information about medical errors that lead to AEs.
Incident reporting systems can identify latent errors (system problems) not uncovered by some
other methods; thus, they can be used to improve patient safety.13,15,16 In comparison with
comprehensive chart reviews, incident reporting is also relatively inexpensive.15
However, like other methods for detecting safety problems, incident reporting has its own
limitations. Incident reporting systems alone cannot reliably measure incidence and prevalence
rates of errors and AEs.20,23 Providers may not report errors because of busy schedules, concerns
about potential lawsuits, fear their reputations could be tarnished, or misperceptions about what
constitutes patient harm.20 As a result, reported incidents may represent only a portion of serious
incidents and may misguide detailed investigation efforts to less important targets.16,18,20,23,24
Additionally, the rates of incidents reported over time may not reflect real changes in safety in an
institution, because an increased rate may simply indicate an improved commitment by the
institution to identify medical errors rather than a true rise in medical hazards.23
In recent years, with the adoption of electronic medical records, computerized surveillance,
including using electronic triggers, has become an increasingly popular method for identifying
certain types of medical errors or AEs, particularly those related to use of medications.25-28 By
integrating multiple data sources (e.g., electronic medical records, laboratory, pharmacy, billing),
computerized surveillance may efficiently detect medical errors and AEs and could provide realtime information for preventing harm to patients from errors in medical treatment.25-28 Use of
electronic medical record-based surveillance of diagnostic errors was also reported.29 However,
the accuracy and reliability of the tools for computerized surveillance need further study.30 The
initial cost of the systems remains another barrier to implementation.25
Similarly, other methods for detecting and monitoring patient safety problems (e.g., chart
audits assisted with trigger tools, direct observation of patient care, executive walk rounds,
administrative data analysis, data warehouses) also have their strengths and weaknesses. We
have identified several documents that provide an overview of these methods based on
systematic or targeted literature review.14-16 Table 1 summarizes the purposes of different patient
safety problem detecting methods. Tables 2 and 3 summarize the strengths and weaknesses of
these methods. Because the original documents used different taxonomies for the methods, we
compiled the adapted tables together in this chapter to provide a more comprehensive overview.
As a result, some contents in these tables may overlap.
406
Table 1, Chapter 36. Overview of the purposes of different methods for detecting patient safety
problems
Method
Adverse Event Counting

(Frequency Assessment)
Adverse Event Understanding

(Root Cause Analysis)
Latent Causes/
Contributory Factors
Harm
Active Errors
Review of medical records
Studies based on interviews with healthcare providers
Direct observation
Use of incident reporting systems
External audit and confidential inquiries
NR
NR
Studies of claims and complaints
NR
NR
Use of information technology and

electronic medical records
NR
NR
Analysis of administrative data
NR
NR
Analysis of autopsy reports
NR
NR
Analysis of mortality and morbidity data
NR
14
Source: Michel P. 2003. Used with permission.

X
Method is relevant for the purpose
?
Relevance of the method for the purpose is to be confirmed
NR
not relevant
407
Table 2, Chapter 36. Advantages and disadvantages of different methods used to measure errors
and adverse events in health care (from the Thomas and Petersen study)
Method
Advantages
Disadvantages
Morbidity and mortality

conferences and autopsy
Can suggest latent errors

Familiar to health care providers and
required by accrediting groups
Malpractice claims analysis
Provides multiple perspectives

(patients, providers, lawyers)
Can detect latent errors
Error reporting systems
Can detect latent errors

Provide multiple perspectives over
time
Can be a part of routine operations
Utilizes readily available data
Inexpensive
Administrative data
analysis
Chart review
Utilizes readily available data

Commonly used
Electronic medical record
Inexpensive after initial investment

Monitors in real time
Integrates multiple data sources
Observation of patient care
Potentially accurate and precise

Provides data otherwise unavailable
Detects more active errors than other
methods
Clinical surveillance
Potentially accurate and precise for

adverse events
Source: Thomas EJ and Petersen LA. 2003.15 Used with permission.
408
Hindsight bias
Reporting bias
Focused on diagnostic errors
Infrequently and nonrandomly utilized
Hindsight bias
Reporting bias
Nonstandardized source of data
Reporting bias
Hindsight bias
May rely upon incomplete and inaccurate

data
The data are divorced from clinical
context
Judgments about adverse events not
reliable
Expensive
Medical records are incomplete
Hindsight bias
Susceptible to programming and/or data
entry errors
Expensive to implement
Not good for detecting latent errors
Expensive
Difficult to train reliable observers
Potential Hawthorne effect
Potential concerns about confidentiality
Possible to be overwhelmed with
information
Potential hindsight bias
Expensive
for internal patient safety problems (from the Shojania study)
Method
Traditional incident
reporting
Advantages
Process already ubiquitous
Can identify latent errors
(system problems)
Stimulated/facilitated
incident reporting
Builds on an existing process

Improves frequency of events and
broadens range of events
Can contribute to improvements in
culture
Patient complaints
Process or data already exist

Highlights important problems about
patient experience often not captured
elsewhere
Malpractice claims
Process or data already exist

Details about causes of the event and
its impact on the patient usually
collected as part of medicolegal
process
Complements incident reporting for
capturing rare but serious events
(e.g., wrong-site surgery)
Probably similar to malpractice
claims, but not clear
Engages frontline staff without
requiring much work for them
Provides a human face to problems
management usually learns about
through impersonal pie charts and
time trends
Alerts management to problems
faced daily by frontline staff
Risk management reports

Executive walk rounds
409
Disadvantages
Underreporting of serious incidents
Frequent reporting of events
not suited to individual analysis
(e.g., falls)
Can be demoralizing when staff
do not perceive meaningful
improvements resulting from
incidents they have reported
Cannot assess changes in safety
(over time)
More labor intensive than traditional
incident reporting
Greater engagement of staff
increases importance of making
meaningful improvements (i.e., even
more demoralizing than usual
if improvements not made)
May be dismissed by clinicians as
service problems
May require more up-front work
(compared with incident reporting) to
identify incidents worth analyzing in
detail for potential safety
improvements
Heavily biased toward detecting
diagnostic issues and procedural
complication (though these are
usually not detected by other
systems)
Probably similar to malpractice
claims, but not clear
Demoralizing to frontline staff if
management focuses only on
improved public relations
(management cares) and does not
seriously address the problems
identified
Tempting for management to focus
on easy fixes (e.g., related to
equipment) not deeper problems or
those requiring substantial
investments of resources
(e.g., staffing, skill mix, or work-load
problems)
for internal patient safety problems (from the Shojania study) (continued)
Method
Chart audits (commonly
operationalized with
trigger tools)
Electronic triggers
(e.g., drug-lab
combinations, use of
antidotes suggestive of
medication errors)
Performance indicators
derived from
administrative data
Data warehouses
Modifying traditional
morbidity and mortality
rounds with modern
patient safety framework
Advantages
Types of events captured may be
more likely to engage frontline
clinicians (especially physicians)
Produce rates that can be monitored
over time, not just counts or
frequencies susceptible to changes in
reporting biases
Very efficient
Potentially high sensitivity capture for
the events captured
Data easily available
In principle, event rates can be
tracked over time, but in practice
probably applies only for frequent
event types in large health care
systems
Richness of detail (e.g., from

medications data, laboratory results,
time stamps) addresses many of the
limitations of administrative data
Can generate data that will engage
both managers and clinicians
Event rates can be followed over time
Builds on format familiar to clinicians
Types of events captured and
richness of detail more likely to
engage physicians
Discrepancies between
clinical and autopsy
diagnoses
Builds on a traditional process of

improvement
Detects problems likely to engage
clinicians
Monitoring pathologic
discrepancies
(e.g., between cytology
and histology or
antemortem biopsies and
autopsies)
Corrected laboratory
results/reports
Relatively efficient
Can identify patterns of problems
amenable to substantial improvement
projects
Relatively efficient
Event rates can probably be followed
over time
410
Disadvantages
Requires willing clinicians to
participate
Many important events not
documented in charts and
contributing factors for documented
events typically unclear
Many triggers have low specificity
Captures only certain types of events
(small subset of events involving
medications or laboratory tests)
Trade-offs between sensitivity and
specificity
Low signal-to-noise ratio
Various methodologic problems
leading to misleading
characterizations of performance
Managers and clinicians tend to
distrust these data (often with good
reason)
Requires intensive effort to
investigate if poor performance is real
and further effort to determine causes
Requires substantial up-front
investments and appropriate clinical
and methodological expertise
Requires organizational culture and
management structures conducive to
driving change on the basis of these
novel data
Care required to avoid traditional
focus on individual errors and
blaming other departments
New processes required to follow-up
systematically on issues identified
(traditional rounds heavy on
discussion, with follow-up occurring
only haphazardly)
Addressing problems identified often
requires host department to engage
and collaborate with other
departmentsdepartures from
traditional norm
Likely to succeed only in select
hospitals because of low autopsy
rates and decreased interest in
autopsies among clinicians and
pathologists at most hospitals
Requires interest on the part of
pathologists to undertake this
nontraditional form of quality
assurance and willingness of clinical
departments to collaborate in
improvement projects
Fairly narrow focus
Requires interested laboratory
medicine personnel
for internal patient safety problems (from the Shojania study) (continued)
Method
Natural language
screening of electronic
portions of medical
records
Advantages
Relatively efficient once implemented
Reasonable sensitivity and specificity
for certain types of events
Direct observation
(e.g., audits of hand
hygiene compliance,
medication administration,
operating room
procedures daily rounds)
Richer, more accurate data than by

many other methods
Identifies problems particularly
difficult to detect by other means
Active surveillance
(combination of chartbased trigger tool applied
in quasi-real time,
stimulated reporting, and
other interactions with
frontline staff)
Telephone calls to patients
(can be automated)
Rich data that are more likely to

include information about causal
factors than record review alone
Process can engage frontline staff
and stimulate them to participate in
subsequent improvement efforts
Identifies problems typically not
captured by other methods
(e.g., post-discharge adverse events
and problems occurring between
ambulatory visits)
Disadvantages
Requires appropriate technical
expertise and initial investment of
time to develop and refine
combinations of search terms with
acceptable sensitivity and specificity
for safety problems
Somewhat labor intensive (but short
periods of measurement may provide
ample data)
Requires appropriately trained
observers
Care must be taken not to create
mistrust among frontline staff
Somewhat labor intensive (but short
periods of measurement may provide
ample data)
Requires appropriately trained
observers
Care must be taken not to create
mistrust among frontline staff
Requires appropriate technology and,
even with automation, still requires
investment of personnel time (e.g., at
least one nurse case manager and a
physician) to respond in real-time to
clinical problems
Source: Shojania KG. 2010.16 Used with permission.
As Tables 2 and 3 have demonstrated, health care organizations have been using a wide array
of methods to detect AEs and medical errors.14-19 Many of these methods (e.g., trigger tools) can
be further categorized by the targeted problems (e.g., medication-related medical errors or
iatrogenic infections), tools, algorithms, and data source used. Given the limited timeframe for
this review, we focus this chapter on general approaches to detecting patient safety problems that
involve using multiple methods (e.g., incident reporting, executive walk rounds, clinical
surveillance, chart review, and trigger tools) to collect data.
We primarily reviewed studies that compared the utilities of different methods. However,
comparison studies that used any method as a gold standard to validate another method were not
included for this chapter, because, in essence, these studies still focused on one individual
method (i.e., the method being validated). We believe that understanding the strengths and
weaknesses of various methods is crucial for decisionmakers who need to form an effective
strategy for monitoring patient safety problems that is appropriate for their organizations.
Readers who are seeking information on individual methods can refer to studies and reviews
specifically focusing on those methods. As we reviewed the literature for this chapter, we
identified a large number of publications focusing on individual methods, particularly in the
areas of incident reporting, chart review, and trigger tools. Some systematic or targeted reviews
provided insightful summaries about commonly used methods.13,18,22,31,32

Detection of AEs is a primary step to achieving a safe health care system. In the report, Safe
Practices for Better Healthcare2010 Update, the National Quality Forum stated that health
411
care organizations must systematically identify and mitigate patient safety risks and hazards with
an integrated approach to continuously drive down rates of preventable patient harm.33 As
several landmark studies have suggested, medical errors are often a system failure where care
practices are inconsistent among health care professionals.6,34,35 By systematically uncovering
these errors and analyzing their causes, health care institutions can identify defects in processes
of care and design system changes to prevent the errors.18,19,23
In the 1999 report, To Err is Human: Building a Safer Health Care System, the IOM also
acknowledged the need to learn from medical errors and recommended establishing mandatory
incident reporting systems as part of an approach to improving safety.6 The report noted that one
of the causes of medical errors is lack of reliable data on the number of medical errors, which
limits the ability to identify the problems origins and develop initiatives to resolve the problem.
A subsequent IOM report, Crossing the Quality Chasm: A New Health System for the 21st
Century, reinforced the need for reliable data and noted a need for evidence-based policies and
practices.36 By performing root-cause analyses (an in-depth examination of the data to identify
factors in the care process that contribute to the errors) and implementing corrective action plans,
health care organizations may be able to address system and process failures to ensure that
potential errors are prevented in the future.5,7,22,23

Measuring the beneficial effects of a safety-problem detection approach is not always
straightforward. Few studies have measured changes in health outcomes that are brought about
by implementing a safety-problem detection method. Designing rigorous studies to establish a
direct connection between the method and any patient safety outcomes is challenging. The
effectiveness, if any, of a safety-problem detection method may not always translate into better
patient outcomes. These outcomes rely not only on how promptly and accurately the problems
are identified but also on how the safety data are used in root-cause analyses and whether the
corrective action plans are implemented effectively.5,6,22,23,36 If the safety data were
misinterpreted or the action plans were not executed successfully, no improvement in safety
outcomes would be observed regardless of the effectiveness of the detection method per se.6,22
Additionally, accurately estimating the true prevalence of safety problems is almost
impossible, particularly with medical errors that did not cause any harm.23 While chart review
has been used as the gold standard in some patient safety studies to quantify AE rates, it rarely
detects medical errors that produce no harm and may also miss other safety problems because of
incomplete documentation in the medical record.15,16 When an increased number of medical
errors is identified, determining whether the finding reflects a deteriorating performance in risk
management or is the result of improved efforts in uncovering these errors is difficult. Likewise,
a decreased number of detected safety problems could be the result of effective risk management
or simply reflect inadequate efforts to find the problems.
Because of these reasons, empirically measuring the impacts of safety-problem detection
methods on patient outcomes is almost unlikely. The beneficial effects of these detection
methods have often been judged partly on data and partly on assumptions. If data suggest that a
method helped detect medical errors that had not been found via other means or detected more
errors in a more timely fashion than other mechanisms, the method would be assumed beneficial
to patient safety. While these assumptions appear reasonable, the data do not provide direct
evidence that the detection method will lead to improved patient safety outcomes.
412
As discussed, we primarily reviewed studies that compared the utility of different methods.
Our search identified one systematic review published by the World Health Organization (WHO)
in December 2003.14 This study by Michel reviewed methods for assessing the nature and scale
of harm caused by health systems. The objective of the study was to identify the strengths and
weaknesses of available methods according to a defined set of criteria. These criteria included
the following:14
Effectiveness in capturing the extent of harm (in different environments).
Availability of reliable data (judged by interobserver reliability).
Suitability for large-scale or small, repeated studies. (Large-scale studies refer to national
and regional studies. Small, repeated studies are carried out for a limited period at the
hospital or local level.)
Costs (financial, human resources, time, and burden on system).
Effectiveness in influencing policy (focused on national, regional, or local policy or
strategic programs).
Effectiveness in influencing hospital and local safety procedures and outcomes.
Synergy with other domains of quality of care.
This set of criteria was defined by the WHO Working Group in Patient Safety: Rapid
Assessment Methods for Assessing Hazards in December 2002. The first four criteria focused
on the intrinsic characteristics of the methods, their validity, reliability, and cost. The last three
criteria were more related to the ability of the methods to trigger improvements in safety cultures
and the quality of safety programs. The study reviewed 262 relevant studies. With the exception
of comparative studies available for the assessment of effectiveness in capturing the extent of
harm, the literature consisted mostly of descriptive studies. For the review, Michel considered
the data reported in the included studies as well as the opinions of the authors of the studies.
The study rated each method on all seven criteria to produce a summary of its key strengths
and limitations.14 When valid information was available, the author rated the criteria from 1
(least favorable) to 4 (most favorable). A study was defined as valid when an appropriate
description of the method (sampling strategy, data collection, and data analysis) in line with
current standards was available. The lowest level (1) indicates low effectiveness, suitability, or
availability, or it means very high cost. Where the amount of evidence-based data was small, the
author noted to be confirmed. The evidence-based ratings for each method in the seven areas
are provided in Table 4. In the absence of valid data, the author used a subjective rating scale
from 1 (least favorable) to 4 (most favorable), based on the opinions of the studies being
reviewed. These opinion-based ratings are provided in Table 5. Both Table 4 and Table 5 were
based on literature from developed countries. The author also reviewed literature from
developing countries, but that information is not discussed in this chapter.
The WHO study revealed that the methods for assessing the nature and scale of harm caused
by health systems have different purposes (Table 1), strengths, and limitations (Table 4 and
Table 5).14 The main conclusion of the study was that these methods do not compete with each
other. Instead, they complement each other by providing different levels of qualitative and
quantitative information. The list of methods and the illustrative ratings (Table 4 and Table 5)
provided by the study may serve as a starting point for choosing appropriate methods for
detecting harms caused by health organizations. The author suggested that identification of
appropriate methods must take into account the distinct environmental factors faced by each
health care organization or region.
413
This WHO study also had limitations.14 First, the studies included for review varied in
quality and quantity. For some methods of interest, such as interviews with health care providers,
analysis of administrative data, or confidential inquiries, few studies were available for review.
Some other methods that the author thought might be useful for detecting safety problems, such
as single case analysis and focus group discussions, were not covered by the study. Second, the
rating systems and criteria used in the study for judging the strengths or weaknesses of the
methods were not adequately validated. The assessment of the methods was generally subjective
rather than objective. Third, because of the wide variety of studies reviewed, the author was not
able to use explicit criteria for quality assessment of the studies.
In addition to the WHO review, we also identified several primary studies that compared the
utility of various methods for monitoring AEs or medical errors. In 2007, Olsen and colleagues
compared the use of incident reporting, pharmacist surveillance, and local real-time record
review for the recognition of clinical risks associated with hospital inpatient care.37 Using the
three methods, they prospectively collected data on AEs on 288 patients discharged from an
850-bed general hospital in the National Health System in the UK. The study found little overlap
in the nature of events detected by the three methods. Record review detected 26 AEs and
40 potential AEs (PAEs) occurring during the index admission. Incident reporting detected
11 PAEs and no AEs. Pharmacy surveillance found 10 medication errors, all of which were
PAEs. The study concluded that incident reporting does not provide an adequate assessment of
clinical AEs and that a variety of methods need to be used to provide a full picture of the safety
condition in a health care organization.
In 2008, Wetzels and colleagues compared the validity and usefulness of five methods for
identifying AEs in general practice.38 The five methods included physician reported AEs,
pharmacist reported AEs, patients experiences of AEs, assessment of a random sample of
medical records, and assessment of all patients who died. In this prospective observational study,
a total of 68 events were identified using these methods. The patient survey identified the highest
number of events and the pharmacist reports identified the fewest. No overlap among the
methods was detected. The authors concluded that a mix of methods is needed to identify AEs in
general practice.
A study by Ferranti and colleagues compared results from two adverse drug event (ADE)
detection methodsvoluntary reporting and computerized surveillanceat a large academic
medical center.39 This 2008 study analyzed the medications most likely to cause harm and
evaluated the strengths and weaknesses of each detection system. During a 7-month period,
computerized surveillance detected 710 ADEs (6.93/1,000 patient days), whereas voluntary
reporting identified 205 ADEs (1.96/1,000 patient days). For each major drug category
(anticoagulants, hypoglycemia, narcotics and benzodiazepines, and miscellaneous), the two
methods detected significantly different event rates.39 Most surveillance-identified events were
hypoglycemia-related, whereas most voluntarily-reported events were in the miscellaneous
category. Of all unique ADEs (875), only 40 were common between the systems. The studys
findings underscored the synergistic nature of the two ADE detection approaches. Although
surveillance provides quantitative data to estimate the actual rate of ADEs, voluntary reporting
contributes qualitative evidence to prompt future surveillance rule development and identify
areas of emerging risk. The authors concluded that the two detection methods should be used
together to provide a full picture of ADE-related patient safety problems.
In 2010, Levtzion-Korach and colleagues published a study that examined and compared five
AE detection methods in one hospital.40 The methods included a Web-based voluntary incident
414
reporting system, medical malpractice claims, patient complaints, the hospital risk management
database, and executive walk rounds. These methods varied in the timing of the reporting
(retrospective or prospective), severity of the events, and profession of the reporters. The five
disparate data sources at the hospital captured about 15,000 problems. The authors systematically
classified the detected problems into 23 categories using a taxonomy that they developed. The
study found that each method identified important safety problems that were generally not
captured by any of the other methods.40 The following are the common categories of safety
problems detected using the five methods compared in the study:
Spontaneous reporting: patient identification issues, falls, and medication problems
Malpractice claims: issues with clinical judgment related to diagnosis and treatment,
communication, and technical skills and problems with medical records (incomplete,
illegible, or missing)
Patient complaints: issues with communication, ancillary services (e.g., patient transport,
kitchen, housekeeping), and administration (admission and discharge processes,
scheduling)
Risk management: issues with technical skills, patient and family behavior (compliance
issues, unusual behavior by a patient or family members), administration, and clinical
judgment
Executive walk rounds: problems with equipment, electronic medical records and other
such technologies, and infrastructure (work environment, security)
Communication problems were common among patient complaints and malpractice claims.
Clinical judgment problems were the leading category for malpractice claims. Walk rounds
identified issues with equipment and supplies. AE reporting systems highlighted identification
issues, especially mislabeled specimens. The authors concluded that, to obtain a comprehensive
picture of their patient safety problems and to develop priorities for improving safety, hospitals
should use a broad portfolio of approaches and then synthesize the messages from all individual
approaches into a collated and cohesive whole.
In another 2010 study, the Office of Inspector General of the Department of Health and
Human Services compared the usefulness of five safety event screening methods: nurse reviews,
analysis of present-on-admission (POA) indicators, Medicare beneficiary interviews, hospital
incident reports, and analysis of patient safety indicators.41 The study used a sample of 278
Medicare beneficiary hospitalizations selected from all Medicare discharges from acute care
hospitals in two selected counties during a 1-week period in August 2008. The investigators
compared events flagged by each screening method to the 120 events identified and/or confirmed
through physician reviews. The study found that nurse reviews and POA analysis identified the
greatest number of safety events. Nurse reviews identified 93 of the 120 confirmed safety events
and POA analysis identified 61 events. Beneficiary interviews identified 22 events, and the
remaining two screening methods identified 8 events each. Of the 120 events, 55 (46%) were
identified by only one screening method. Nurse reviews identified 35 events (29% of the 120
events) not flagged by any other screening method. POA analysis alone flagged 14 events (12%
of the 120 events).
We also reviewed a study by Tinoco and colleagues that compared a computerized
surveillance system (CSS) with manual chart review (MCR) for detecting inpatient ADEs and
hospital-associated infections (HAIs).25 The authors retrospectively analyzed the events detected
using the two methods by type of events. From a sample of 2,137 patient admissions between
415
October 2000 and December 2001, the authors identified AEs that were detected only by MCR,
only by CSS, or by both methods. The study found that CSS detected more HAIs than MCR
(92% vs. 34%); however, a similar number of ADEs was detected by both systems (52% vs.
51%). The agreement between systems was 26 percent and 3 percent for HAIs and ADEs
respectively. The study also found that MCR detected events missed by CSS using information
in physician narratives and that some events found by MCR were missed by CSS. The authors
concluded that integrating information from physician narratives with CSS using natural
language processing would improve the detection of ADEs more than HAIs.
A compelling theme emerged among the findings of the studies reviewed for this section.
That is, different methods for detecting patient safety problems overlap very little in the safety
problems they detect. These methods complement each other and should be used in combination
to provide a comprehensive safety picture of the health care organization. Detailed information
on the studies reviewed in this section (except for the WHO report14) is provided in Appendix D.
Because the body of evidence consists of studies of different designs, the overall strength of
evidence is not assessed.
416
Table 4, Chapter 36. Evidence-based rating of the main methods used in developed countries for estimating hazards in health care systems
Criteria
Effectiveness in
capturing the
extent of harm
Availability of
reliable data
Ad Hoc Studies Based on Epidemiological

Designs and Systematic Data Collection
Review of Studies Based
Direct
Medical
on Interviews
Observation
Records
With HealthCare Providers
Methods Based on
Reporting
Incident
External
Reporting
Audit and
Systems
Confidential
Inquiries
No evidence
available
4 for harm
assessment, to
be confirmed*
3, to be
confirmed*
No evidence
available
Analysis of Routinely Collected and Existing Data

Studies of
Claims and
Complaints
Electronic
Medical
Records
Administrative
Data
Autopsy
Reports
Mortality
and
Morbidity
Conferences
No evidence
available
2, to be
confirmed*
No evidence
available
No evidence
available
3, to be
confirme
d*
1
No evidence
available
No evidence
available
No
evidence
available
2
No evidence
available
Not
applicable
4, to be
confirmed*
No
evidence
available
No
evidence
available
No
evidence
available
No
evidence
available
No
evidence
available
Suitability for
large-scale
studies
Suitability for
small, repeated
studies
Costs
4, to be
confirmed*
1 for prospective;
2 for crosssectional
No evidence
available
No evidence
available
Effectiveness in
influencing policy
No evidence
available
No evidence
available
3, to be
confirmed*
3, to be
confirmed*
No
evidence
available
4
No evidence
available
Effectiveness in
No evidence
3, to be
No evidence
3, to be
3, to be
3
No evidence
influencing
available
confirmed*
available
confirmed*
confirmed*
available
hospital and local
safety
procedures and
outcomes
Synergy with
4
4
4
2
No evidence
3
4
4
other domains of
available
quality of care
Adapted from Michel P. Strengths and weaknesses of available methods for assessing the nature and scale of harm caused by the health system.14
Rating scale from 1 to 4, the most favorable level being 4
* to be confirmed where the amount of evidence-based data is small
417
3, to be
confirmed*
Table 5, Chapter 36. Subjective rating, where there was no evidence-based data, of the main methods used in developed countries for
estimating hazards in health care systems
Ad Hoc Studies Based on
Epidemiological Designs and
Systematic Data Collection
Review
Studies
Direct
of
Based on
Observation
Medical
Interviews
Records
With HealthCare
Providers
Incident
Reporting
Systems
External
Audit and
Confidential
Inquiries
Studies of
Claims and
Complaints
1-2 for cause

analysis
4 for harm
assessment,
2-3 for cause
analysis
2-3
2-3
Suitability for largescale studies
1 for global
assessment,
3 for limited
focus
Suitability for small,

repeated studies
2-3
(videotaping)
Costs
Criteria
Effectiveness in
capturing the extent of
harm
Availability of reliable
data
Methods Based on
Reporting
Analysis of Routinely Collected and Existing Data
Electronic
Medical
Records
Administrative
Data
Autopsy
Reports
Mortality and
Morbidity
Conferences
3-4
Not
applicable
1-2
3-4
Effectiveness in
*
3-4
3-4
*
*
*
3-4
influencing policy
Effectiveness in
*
2-3
3-4
*
1-2
*
*
influencing hospital
and local safety
procedures and
outcomes
Synergy with other
*
*
*
*
2-4
*
*
domains of quality of
care
Adapted from Michel P. Strengths and weaknesses of available methods for assessing the nature and scale of harm caused by the health system.14
Rating scale from 1 to 4, the most favorable level being 4
* Evidence-based rating is available (see Table 4)
418
3-4

None of the studies that we reviewed reported any harm directly caused by the
implementation of a method for monitoring patient safety problems. However, in theory, a
method that often fails to capture important AEs or medical errors may mislead the health care
organization about its true safety status and cause a delay in addressing safety problems, likely
leading to patient harms. Additionally, various detection methods may compete with each other
for the limited resources available for risk management in an organization. Adopting a relatively
ineffective method might shift resources from more effective alternatives and, thus, decrease the
organizations overall performance in uncovering safety problems. This loss of detection
capability, in turn, could lead to an increase in harms to the patients treated in the organization.
However, designing rigorous studies to empirically test these hypotheses is difficult.

What Contexts?
As previously described, a wide variety of methods exist for detecting patient safety
problems, and this chapter focuses on evidence only from studies that compared these methods.
The methods being compared were implemented differently due to their differences in the
primary problems targeted, tools used, resources required, staff involved, and the timing
(retrospective or real-time) of the detection (see Appendix D, the Description of PSP
column).14-19
Some of these methods (e.g., incident reporting and trigger tools) can be further categorized
(e.g., mandatory or voluntary incident reporting systems), and each method in the subcategories
can also be implemented differently. For example, at least 27 states and some Government
agencies (e.g., the U.S. Food and Drug Administration) in the U.S. have established some form
of incident reporting systems. These various incident reporting systems or programs could be
implemented differently in terms of data collected, tools used, reporting process, and how data
are shared or used.1,2,13,18,21,22
Similarly, many different types of trigger tools and automated systems exist.31,32 These tools
or systems target different problems (e.g., general AEs, ADEs, nosocomial infection, decubitus
ulcers, surgical complications) and may involve different data sources, equipment, software, or
algorithms. It is not feasible for this chapter to cover the implementation issues for all these
methods. Therefore, we describe only the relevant information reported in the comparison
studies reviewed for this chapter. This information is provided in Appendix D (refer to the
Description of PSP and the Context columns).

Accurately estimating the cost associated with implementing strategies for detecting patient
safety problems is difficult. The direct cost for this activity may include expenditures for
equipment and materials (e.g., computers, software, photocopy machines, paper), facilities and
space, and labor for collecting and analyzing data. Indirect, overhead expenses also may need to
be counted. These direct and indirect costs vary across health care organizations and regions and
constantly change over time.
Our search identified sporadic data about costs for implementing safety problem detection
methods. The most recent and relevant data came from the study by Levtzion-Korach.40 This
study estimated the direct cost of the five methods used in one hospital (Table 6). It showed that
419
the hospitals expenditures on these systems were estimated to be a one-time cost of $120,000
and an annual cost of almost $1 million. Additionally, we identified some general discussions
about which detection methods are generally more expensive or labor-intensive (Table 2 and
Table 3). Our search did not identify any full economic evaluation (e.g., cost-effectiveness
analysis from the publics perspective) of the burden related to the implementation of various
methods for detecting AEs or medical errors.
Table 6, Chapter 36. Estimated costs of systems for detecting patient safety problems in one
hospital
Incident
Reporting
Patient
Complaints
Risk
Management
Malpractice
Claims
Executive Walk
Round
One-time
expense
$72,400
$42,580
$0
Not directly
supported by the
institution
$0
Annual support
$9,000
$3,395
$0
Not directly
supported by the
institution
$0
0.5 FTE PS
manager:
$43,340
12 FTE PS
analyst:
$540,000
3.5 FTE risk

management
analyst:
$318,500
Not directly
supported by the
institution
0.2 FTE PS
manager:
$17,380
Software
Manpower
Annual support
0.2 FTE RM
analyst: $18,000
0.3 FTE PS
analyst: $12,780
0.1 FTE PS
analyst: $4,500
A weekly hour of
CEO, CMO,
CNO, and COO:
$10,500
Sum
One-time
expense
$72,400
$42,580
$0
Not directly
supported by the
institution
$0
Annual support
$74,840
$543,395
$318,500
Not directly
supported by the
institution
$40,660
Source: Levtzion-Korach O. et al. 2010.40 Used with permission.

Abbreviations: FTE PS=full-time equivalent patient safety; FTE RM=full-time equivalent risk management; CEO=chief
executive officer; CMO=chief medical officer; CNO=chief nursing officer; COO=chief operating officer.

For this chapter, we focus on the evidence only from studies that compared different methods
for detecting patient safety problems. It is not feasible for the chapter to review the effect of
context on effectiveness for each individual method. We collected data only on the context for
the methods being compared in the included studies. These data fall into five categories: the
external context, organizational characteristics, teamwork, leadership, and culture (see Appendix
D, the Context column). However, based on the data collected, no conclusion can be drawn
regarding the effect of context on the effectiveness of the detection methods, mainly because
these studies were not designed to assess such links.
420
Nevertheless, the importance of strong leadership, teamwork, and organization-wide safety

culture to successful implementation of patient safety practices as a whole has been well
documented in literature that is beyond the scope of this chapter.3,6,22,23 It is reasonable to expect
that leadership, teamwork, and safety culture have the same impact on the implementation of
patient safety monitoring strategies. Additionally, the external factors (e.g., how governments or
the Joint Commissions use the safety data reported by hospitals) should also have a significant
impact on the effectiveness of the strategies.13,18,21

The studies reviewed for this chapter consistently suggested that each method for detecting
AEs or medical errors has advantages and disadvantages. These various methods do not compete
with each other. They identify fairly distinct problems and complement each other by providing
different levels of qualitative and quantitative information about patient safety.
Health care organizations are generally faced with a variety of safety problems, such as
misdiagnoses, misidentified patients, falls, procedural complications, and medication-related
errors. All these problems need to be identified adequately so that hospitals can effectively
prioritize the problems on the basis of the burden of harm and costs associated with the
problems, the availability of effective prevention strategies, and the likelihood of local success in
implementing such strategies.3,6,22,23 Therefore, health care organizations should use a broad
portfolio of methods to uncover safety problems and then synthesize the data collected into a
comprehensive picture.40
For administrators and risk management professionals, a primary challenge is how to make a
rational choice among a large number of methods to build a portfolio appropriate for their
organizations.40 While no simple formula exists to guide the decisionmaking process, the
composition of the portfolio generally depends on the safety problems most relevant to the
organization and the resources available for the risk management effort.16 The bottom line is that
the choice of a specific method by a health care organization might not be as important as the
decision to use more than one method.16 The information that we compiled in this chapter is
intended to serve as a starting point for health care organizations to reconsider their general
approach to monitoring patient safety problems. Future research needs to assess the effectiveness
of different portfolios of methods and provide practical guidance on how to combine the
information collected using different methods into one safety picture. A summary table is located
below (Table 7).
Scope of the
Strength of
the PSP
Effectiveness
Common/Low-to-High
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
High
421
How Much do We
Moderate/Difficult
References
1.
2.
3.
4.
5.
6.
7.
8.
Indiana medical error reporting system:

report for 2009. Indianapolis (IN): Indiana
State Department of Health; 2010 Aug 30.
56 p.
www.in.gov/isdh/files/2009_MERS_Report.
pdf.
Utah Department of Health, HealthInsight,
UHA, Utah Hospitals & Health Systems
Association. 2009 Utah Sentinel events data
report: identifying opportunities for
improvement. Utah: Utah Department of
Health; 2010 Mar. 9 p.
http://health.utah.gov/psi/pubs/sentinel_even
ts09.pdf.
Adverse health care events reporting system:
what have we learned? St. Paul (MN):
Minnesota Department of Health; 2009 Jan.
32 p.
Thompson WC Jr. The high costs of weak
compliance with the New York State
hospital adverse event reporting and
tracking system: policy report. New York
(NY): New York State Department of
Health; 2009 Mar 9. 54 p.
www.comptroller.nyc.gov.
Zegers M, de Bruijne MC, de Keizer B, et
al. The incidence, root-causes, and outcomes
of adverse events in surgical units:
implication for potential prevention
strategies. Patient Saf Surg 2011;5:13.
27749/pdf/1754-9493-5-13.pdf. PMID:
21599915
Institute of Medicine Committee on Quality
of Health Care in America. Kohn LT,
Corrigan JM, Donaldson MS, editors. To err
is human: building a safer health system.
Washington (DC): National Academy Press;
1999 Nov 1. 223 p.
http://stills.nap.edu/books/0309068371/html.
Wachter RM. Patient safety at ten:
unmistakable progress, troubling gaps.
Health Aff (Millwood) 2010 JanFeb;29(1):165-73.
http://content.healthaffairs.org/content/29/1/
165.full.pdf. PMID: 19952010.
Leape LL, Berwick DM. Five years after To
Err Is Human: what have we learned. JAMA
2005 May 18;293(19):2384-90. PMID:
15900009.
422
9.
Altman DE, Clancy C, Blendon RJ.

Improving patient safety--five years after the
IOM report. N Engl J Med 2004 Dec
11;351(20):2041-3. PMID: 15537902.
10.
Lucado J, Paez K, Elixhauser A.

Medication-related adverse outcomes in
U.S. hospitals and emergency departments,
2008: statistical brief #10. [internet].
Research and Quality (US); 2011 Apr
[accessed 2011 Sep 13]. PMID: 21595139.
11.
Institute for Healthcare Improvement.

Protecting 5 million lives from harm.
[internet]. Cambridge (MA): Institute for
Healthcare Improvement; [accessed 2011
Nov 10].
12.
National Quality Forum (NQF). NQF patient

safety terms and definitions. Washington
(DC): National Quality Forum (NQF); 2009
Dec. 6 p.
www.qualityforum.org/Topics/Safety_Defin
itions.aspx.
13.
Office of Inspector General. Adverse events

in hospitals: state reporting systems.
Washington (DC): Department of Health
and Human Services; 2008 Dec. 37 p.
http://oig.hhs.gov/oei/reports/oei-06-0700471.pdf.
14.
Michel P. Strengths and weaknesses of

available methods for assessing the nature
and scale of harm caused by the health
system: literature review. Geneva: World
Health Organization (WHO); 2003. 59 p.
www.who.int/patientsafety/research/P_Mich
el_Report_Final_version.pdf.
15.
Thomas EJ, Petersen LA. Measuring errors

and adverse events in health care. J Gen
Intern Med 2003 Jan;18(1):61-7. PMID:
12534766.
16.
Shojania KG. The elephant of patient safety:

what you see depends on how you look. Jt
Sep;36(9):399-401. PMID: 20873672.
17.
Michel P, Quenon JL, de Sarasqueta AM, et

al. Comparison of three methods for
estimating rates of adverse events and rates
of preventable adverse events in acute care
hospitals. BMJ 2004 Jan 24;328(7433):199.
PMID: 14739187.
18.

draft guidelines for adverse event reporting
and learning systems. Geneva (Switzerland):
World Health Organization; 2005. 80 p.
www.who.int/patientsafety/events/05/Report
ing_Guidelines.pdf.
27.
Ferranti J, Horvath MM, Cozart H, et al.

Reevaluating the safety profile of pediatrics:
a comparison of computerized adverse drug
event surveillance and voluntary reporting in
the pediatric environment. Pediatrics 2008
May;121(5):e1201-7. PMID: 18450863.
19.
Leape LL. Reporting of adverse events. N

Engl J Med 2002 Nov 14;347(20):1633-8.
PMID: 12432059.
28.
20.
Levinson DR. Hospital incident reporting

systems do not capture most patient harm
[OEI-06-09-00091]. Washington (DC):
Department of Health and Human Services,
Office of Inspector General; 2012 Jan. 42 p.
Rommers MK, Teepe-Twiss IM, Guchelaar

HJ. A computerized adverse drug event
alerting system using clinical rules: a
retrospective and prospective comparison
with conventional medication surveillance in
the Netherlands. Drug Saf 2011 Mar
1;34(3):233-42. PMID: 21332247.
29.
Singh H, Giardina TD, Forjuoh SN, et al.

Electronic health record-based surveillance
of diagnostic errors in primary care. BMJ
Qual Saf 2012 Feb;21(2):93-100.
http://qualitysafety.bmj.com/content/21/2/93
.full.pdf. PMID: 21997348.
30.
Murff HJ, FitzHenry F, Matheny ME, et al.

Automated identification of postoperative
complications within an electronic medical
record using natural language processing.
JAMA 2011 Aug 24;306(8):848-55. PMID:
21862746.
31.
Triggers and targeted injury detection

systems (TIDS) expert panel meeting:
conference summary [AHRQ Publication
No. 090003]. In: Agency for Healthcare
Research and Quality. 2008 Jun 30-Jul 1;
Rockville (MD). Agency for Healthcare
Research and Quality, U. S. Department of
Health and Human Services; 2009 Feb. p. 157. www.ahrq.gov/qual/triggers/.
32.
Govindan M, Van Citters AD, Nelson EC, et

al. Automated detection of harm in
healthcare with information technology: a
systematic review. Qual Saf Health Care
2010 Oct;19(5):e11. PMID: 20671081.
33.
National Quality Forum. Safe practices for

better healthcare 2010 update, a consensus
report. [internet]. Washington (DC):
National Quality Forum; 2010 Apr
[accessed 2011 Nov 10]. [54 p].
/Safe_Practices_for_Better_Healthcare_%e2
%80%93_2010_Update.aspx.
34.
Reason J. Understanding adverse events:

human factors. Qual Health Care 1995
Jun;4(2):80-9. PMID: 10151618.
21.
22.
United States Government Accountability

Office. Health-care-associated infections in
hospitals: an overview of state reporting
programs and individual hospital initiatives
to reduce certain infections. Washington
(DC): United States Government
Accountability Office; 2008 Sep. 55 p.
www.gao.gov/products/GAO-08-808.
The power of safety: state reporting provides
lessons in reducing harm, improving care.
[internet]. Rockville (MD): Agency for
Healthcare Research and Quality (AHRQ);
2010 [accessed 2011 Jul 27]. [2 p].
http://psnet.ahrq.gov/resource.aspx?resource
ID=18516.
23.

analysis of patient safety practices. AHRQ
2001;(43):i-x, 1-668.
24.
Brunicardi FC, Heggeness MH, Andersen

DK, et al. Orthopedic surgery. In:
Schwartzs principles of surgery [database
online]. 9th. New York (NY): McGraw-Hill
Companies, Inc.; 2010 [accessed 2011 May
26]. [2 p].
www.accessmedicine.com/content.aspx?aID
=5028211.
25.
Tinoco A, Evans RS, Staes CJ, et al.

Comparison of computerized surveillance
and manual chart review for adverse events.
J Am Med Inform Assoc 2011 JulAug;18(4):491-7. PMID: 21672911.
26.
Kilbridge PM, Noirot LA, Reichley RM, et

al. Computerized surveillance for adverse
drug events in a pediatric hospital. J Am
Med Inform Assoc 2009 Sep-Oct;16(5):60712. PMID: 19567791.
423
35.
Vincent C. Understanding and responding to

adverse events. N Engl J Med 2003 Mar
13;348(11):1051-6. PMID: 12637617.
36.
Institute of Medicine, Committee on Quality

of Health Care in America. Corrigan JM, et
al, editors. Crossing the quality chasm: a
new health system for the 21st century.
Washington (DC): National Academy Press;
2001. 337 p.
http://search.nap.edu/books/0309072808/ht
ml/.
37.
Olsen S, Neale G, Schwab K, et al. Hospital

staff should use more than one method to
detect adverse events and potential adverse
events: incident reporting, pharmacist
surveillance and local real-time record
review may all have a place. Qual Saf
Health Care 2007 Feb;16(1):40-4.
64933/pdf/40.pdf. PMID: 17301203.
424
38.
Wetzels R, Wolters R, van Weel C, et al.

Mix of methods is needed to identify
adverse events in general practice: a
prospective observational study. BMC Fam
Pract 2008;9:35. PMID: 18554418.
39.
Ferranti J, Horvath MM, Cozart H, et al. A

multifaceted approach to safety: the
synergistic detection of adverse drug events
in adult inpatients. J Patient Saf 2008;4:18490.
http://analytics.dhts.duke.edu/wysiwyg/dow
nloads/Ferranti_JPS_adults.pdf.
40.
Van Der Linden W, Warg A, Nordin P.

National register study of operating time and
outcome in hernia repair. Arch Surg 2011
Oct;146(10):1198-203. PMID: 22006880.
41.
Levinson DR. Adverse events in hospitals:

methods for indentifying events [OEI-06-0800221]. Washington (DC): Department of
Health and Human Services, Office of
Inspector General; 2010 Mar. 60 p.
Chapter 37. Interventions To Improve Care Transitions at

Hospital Discharge (NEW)
Stephanie Rennke, M.D.; Marwa H. Shoeb, M.D., M.S.; Oanh K. Nguyen, M.D.; Yimdriuska
Magan, B.S.; Robert M. Wachter, M.D.; Sumant R. Ranji, M.D.

The term transitions of care refers to any instance in which a patient moves from one
health care setting to another.1,2 More often than not, the care of a patient with chronic illness
involves multiple settings (e.g., inpatient, outpatient, or long term care) and several different
health professionals. These transitions are inevitable, but because they can increase the risk of
adverse events and poor clinical outcomes, they warrant particular attention.
Hospital discharge represents a particularly risky care transition, especially for older adults.
Multiple studies document that adverse events occur in approximately one in five adult medical
patients within 3 weeks of discharge.3,4 Nearly 20 percent of older Medicare patients discharged
from a hospital will be readmitted within 30 days.5 A broad spectrum of adverse events can
occur after discharge, including both diagnostic and therapeutic errors, but adverse drug events
(ADEs) are particularly common and harmful. Recent studies indicate that nearly 100,000
elderly patients are hospitalized every year due to ADEs.6 Additionally, 1 in 67 emergency
hospitalizations are the result of an ADE. Particularly in the face of an aging population,
ensuring safe care transitions for patients with complex, chronic illnesses will remain an
important patient safety issue.
The Patient Protection and Affordable Care Act (PPACA) contains provisions specifically
focused on decreasing preventable readmissions and improving care transitions. Under this
legislation, hospitals will be financially penalized for high readmission rates. The Centers for
Medicare & Medicaid Services already publicly reports hospitals risk-adjusted 30-day
readmission rates for specific diagnoses on its Web site, Hospital Compare.7 Therefore, hospitals
and health care organizations have considerable incentives to improve transitional care at
hospital discharge.

A wide range of interventions have been proposed and studied in order to ensure smooth
transitions of care at hospital discharge. Therefore, this patient safety practice (PSP) comprises
multiple interventions. Broadly speaking, we defined a transitional care strategy as an
intervention or a series of interventions that occurs among health care practitioners and across
settings in order to ensure the safe and effective transfer of patients from one level of care to
another or from one type of setting to another. This definition is based on two widely used
definitions of the broader concept of transitional care,2,8 which both refer to the movement
patients make between health care practitioners and settings as their condition and care needs
change during the course of a chronic or acute illness.
Within this broader definition, we sought to more specifically define PSPs targeting the
particular problems of adverse events, readmissions, and emergency department visits after
hospital discharge. Adverse events (AEs) have been previously defined as an adverse outcome or
injury resulting from medical management, and can range in severity from laboratory
425
abnormalities to symptoms to permanent disability and death.3,4AEs can be further categorized as

preventable, ameliorable and not preventable. Prior studies have found that the most common
preventable AEs after hospital discharge include procedural complications, hospital-acquired
infections, and adverse drug events (ADE).3,4 An ADE is defined as harm associated with the
appropriate or inappropriate use of a drug.9
For this report, the PSP refers to any intervention to improve transitions from acute care
hospitals to the outpatient setting, with the goals of (1) bridging gaps in continuity of care and
coordination of care across the health care continuum and (2) preventing adverse events,
emergency department (ED) visits, and rehospitalizations after hospital discharge. This definition
explicitly excludes formal care programs that do not primarily target discharge from the acute
hospital setting. Examples of such excluded interventions include disease management programs,
emergency services-based programs, Hospital at Home programs, day hospital care programs
(including psychiatric day hospitals), palliative care and hospice programs, and interventions
targeting discharge from the hospital to other acute or subacute settings.
In order to analyze a disparate body of literature, we developed a taxonomy of transitional
care interventions based on analysis of existing systematic reviews in the field8,10-14 and expert
consensus. We grouped individual interventions into three broad categories: pre-discharge, postdischarge, and bridging interventions (Box 1).
Box 1. Taxonomy of interventions to improve transitional care at hospital discharge
Pre-discharge
interventions
Assessment of risk for adverse events or readmissions

Patient engagement (for example, patient or caregiver education)
Creation of an individualized patient record (customized document in lay language
containing clinical and educational information for patients use after discharge)
Facilitation of communication with outpatient providers
Multidisciplinary discharge planning team
Dedicated discharge advocate or coach
Postdischarge
interventions
Outreach to patients (including follow-up phone calls, patient-activated hotlines,

and home visits)
Facilitation of clinical follow-up (including facilitated ambulatory provider follow-up)
Medication reconciliation after discharge
Bridging
interventions
Inclusion of at least one pre-discharge component and at least one post-discharge

component
Bridging interventions included both pre- and post-discharge components, and often
emphasized longitudinal relationships in the pre- and the post-discharge periods, as well as the
role of the patient or caregiver in maintaining safe transitions.15
The purpose of this systematic review is to analyze published literature to determine the
effectiveness of the kinds of interventions described in Table 1 to reduce adverse events, ED or
unscheduled acute care visits, and readmissions after hospital discharge of adult patients, and to
assess the feasibility of implementing successful interventions on a larger scale. We included
randomized controlled trials (RCTs) and non-randomized clinical controlled trials (CCTs) that
evaluated one or more of the above interventions in adult general medical patient populations,
utilized at least one intervention prior to discharge, and reported rates of ED visits, readmissions,
426
or adverse events (AEs) after discharge. We included studies that reported costs if they also
reported one of the other targeted outcomes.

AEs after discharge and readmissions have been attributed to many factors, including poor
communication and transfer of information between inpatient and outpatient providers;16
medication changes during hospitalization;17 inadequate patient comprehension of diagnoses,
medications, and follow-up needs;18 and failure to complete planned outpatient diagnostic or
treatment plans.19 Interventions to improve care transitions after hospital discharge generally
target one or more of these documented deficiencies in care. In addition to these specific factors,
more general patient-related factors and health care system-related factors may influence an
individual patients risk for AEs or readmissions after discharge. Figure 1 in Appendix C depicts
the theoretical construct underpinning the role of interventions in reducing AEs and readmissions
after discharge. In brief, patients risk factors for an AE depend on patient and health care system
factors. Specific interventions target known deficiencies in care transitions, aiming to improve
continuity of care and decrease AEs after discharge. A reduction in AEs after discharge should,
in theory, result in fewer readmissions and ED visits.
This framework has two main limitations: (1) many readmissions may not be preventable,
and (2) transitional care interventions for general medical patients are comparatively less welldefined than are those for disease-specific populations, where the link between interventions and
improved outcomes is clearer.20
No clear consensus exists on the proportion of readmissions of adult patients that are
preventable. A recent study21 has suggested that as few as one in five 30-day readmissions may
be truly preventable, and that the proportion of preventable readmissions may vary widely
among individual hospitals. Another recent study22 found that no method reliably predicts an
individual patients readmission risk. Given these limitations, hospitals face difficulties
determining which patients should be targeted for transitional care interventions. Indeed, Hansen
and colleagues15 recently published a systematic review of interventions to reduce 30-day
rehospitalization, and found that no single intervention was consistently associated with reduced
risk.
Prior research in this field has identified some interventions that have reduced readmission
risk, but these successes have largely been achieved in disease-specific populations, such as
patients with congestive heart failure. Multiple systematic reviews23,24 have found that
multidisciplinary transitional care programs are associated with reduced readmission risk and
improved mortality in elderly CHF patients. Naylor and colleagues8 summarized 21 randomized
clinical trials of transitional care interventions targeting chronically ill adults, including both
disease-specific studies and studies conducted in general medical populations. They identified
nine interventions that demonstrated positive effects on measures related to hospital
readmissions. Many of the successful interventions shared similar features, such as assigning a
nurse as the clinical manager or leader of care and including in-person home visits to discharged
patients. However, the majority of these successful interventions were conducted in diseasespecific patient populations. A 2010 Cochrane review conducted by Shepperd and colleagues
examined RCTs that compared an individualized discharge plan with routine discharge care in
both general and disease-specific populations.14 They found that a structured and individualized
discharge plan led to small reductions in hospital length of stay and readmission rates for older
people admitted with a medical condition; but again, most of the successful studies focused on a
427
specific disease process. Unlike the Naylor review, this review did not consider interventions
that occurred after the patient was discharged.
While some aspects of transitional care interventions for disease-specific populations may
apply broadly to general medical populations, others may not be generalizable or may be less
effective. In CHF patients, for example, a clear link exists between dietary and medication
adherence and readmission risk; therefore, many successful interventions incorporate extensive
patient and caregiver counseling around diet, medication adherence, and weighing daily at home.
However, an elderly patient who is debilitated after a lengthy hospitalization for pneumonia may
not derive the same level of benefit from medication and dietary counseling, as would a younger
CHF patient, but might benefit from an intervention emphasizing restoring functional status and
close clinical follow-up.
As several recently published systematic reviews evaluated the role of transitional care
interventions in disease-specific populations and because the outcomes of such interventions
appear to be different in disease-specific and more undifferentiated patient populations, we chose
to focus our review on studies that evaluated only interventions conducted in adult general
medical populations. In contrast to another recent review15 that evaluated only studies of
interventions to reduce readmissions, we also included studies that sought to reduce adverse
events or ED utilization after discharge.

We conducted a systematic literature search of Medline, CINAHL, EMBASE, and the
Cochrane Database of Controlled Trials using a search strategy developed with the assistance of
a medical librarian. We identified 15,905 citations, of which 454 underwent full-text review
(Appendix C, Figure 2, Chapter 37). Forty-three studies met all inclusion criteria, including 25
RCTs25-50 and 18 CCTs51-67 (Appendix D, Table 1, Chapter 37). Studies used an average of 4
separate interventions (range 1-8) based on our taxonomy. Thirty-one studies used a bridging
intervention, of which 21 were RCTs,25-30,32,34-38,40-45,47,48,50,52,53,55,56,59,64-66,68 and 12 studies (3
RCTs)31,33,46,49,51,57,58,60-63,67 included only hospital-based interventions. We used the Cochrane
Effective Practice and Organization of Care (EPOC) criteria to evaluate the methodologic quality
of included studies (Appendix D, Table 2, Chapter 37). Included studies generally had fair
methodologic quality.
The interventions assessed in the studies included a variety of components (Appendix D,
Table 3, Chapter 37). Five studies included risk assessment as part of the intervention.33,39,48,63,66
Thirteen of the 43 included studies used an individualized health record that included a list of
diagnoses, warning signs or symptoms, medication list with side effects, and contact
information.25-28,32,34,35,39,41,43,53,68 Most studies (34) included patient engagement, with varying
levels of interaction that ranged from patient education to counseling to symptom
management.25-30,32-35,37,39-48,50-54,56,58,60,62-66 Twenty-two studies included direct communication
between inpatient and outpatient providers,25,26,32-35,37,39,41,43-46,48,50,53,55,60,62,64,66 and 16 included
facilitated clinical follow-up either through directly scheduled appointments or telephone
availability following hospitalization.25,26,28,30,36,39,43,44,47,48,50,53,55,56,64,66 Only 11 studies included
medication reconciliation prior to discharge,26,33-35,37,41,45,46,60,62,66 and 10 studies included postdischarge medication reconciliation,26,27,32,34,37,41,43,45,65,68 done either by telephone or in the home
visit. Of the 31 studies that included a bridging intervention, 24 included an identified health
provider who took a primary role in the transitional period, with contact in the hospital and in the
outpatient setting.26-28,30,32,34-36,39,40,43,45,47-51,53,55,56,64,67,68 Seventeen studies included a
428
multidisciplinary team, including at least two providers, as part of the

intervention.28,34,35,38,41,42,47-50,55,57,59,61,63,64,67,69 Twenty-eight studies included post-hospitalization
outreach, either by telephone (22),5,25-30,32,35-37,39,40,43,45,47,50,55,59,65,66,68,69 home visit (18)26-29,3739,41-43,47,48,52,53,55,59,65
or both telephone contact and at least one home visit (10).26-29,37,43,55,59,65,68
None of the studies specifically addressed end of life issues, palliative care, or counseling as part
of the interventions. However, studies evaluating the Care Transitions Intervention (CTI) did
include advanced directives in the patient-centered health record.27,43,65
Interventions To Reduce 30-Day Readmissions

All but one study reported readmission rates, including 18 studies (10 RCTs) that reported
ED visit or hospital readmission rates 30 days or less after discharge25,27-30,34,35,37,43,45,52,56,57,63,65-68
(Appendix D, Table 4, Chapter 37). Sixteen of these studies (10 RCTs) reported these outcomes
at 30 days after discharge,25,27-30,34,35,37,43,45,52,56,57,63,65-68 and two studies reported 14-day
readmission rates52,63. We focused our analysis on the studies reporting 30-day ED visit and/or
readmission rates, given the policy importance of this outcome (i.e., Medicares decision to use
this time horizon for public reporting and readmission penalties).
We identified six studies (four RCTs, two CCTs) that reported significant reductions in 30day ED visit or readmission rates.27,34,35,43,65,68 Overall, these studies were of similar fair
methodologic quality compared with the other studies. All of these studies used a bridging
strategy with five or more separate interventions. Coleman 2004 (CCT) and Coleman 2006
(RCT) evaluated the CTI in hospitalized geriatric patients in large managed care and capitated
delivery systems respectively.27,68 This transitional care program focuses on engaging patients
and caregivers to be active participants in self care in four areas (pillars): medication selfmanagement, a flexible and dynamic patient-centered record, outpatient provider followup, and
identification and management of red flags including signs or symptoms of a worsening
condition. The intervention includes hospital and home visits and several telephone contacts, all
of which emphasize the importance of self care of chronic illness through education, role
modeling, and counseling during the transitions period. Two subsequent studies implemented the
CTI in Medicare fee-for-service populations in Colorado and Rhode Island.43,65 Both of these
studies also found reductions in 30-day readmission rates, reaching statistical significance in the
Rhode Island study.
Jack and colleagues evaluated the ReEngineered Discharge Program (Project RED) in a
single site RCT at a large urban safety net hospital.34 The intervention focuses on an in-hospital
component, where a nurse discharge advocate develops a comprehensive patient-centered afterhospital care plan, including medication and contact information, pending tests and
appointments, and a post-hospitalization pharmacist telephone call that includes communication
with primary providers. The study reported significant reductions in ED utilization after
discharge; readmission rate was reduced as well, but this outcome did not achieve statistical
significance.
A 2009 report by Koehler and colleagues evaluated a supplemental geriatric care bundle as
part of a multidisciplinary team-based program with care coordinators and pharmacists around
patient education on medications and self-management (including use of a personal health
record), as well as post-discharge telephone follow-up calls.35 A 2009 report by Courtney and
colleagues evaluated a nursing and physiotherapy program for hospitalized elders that included
individualized exercise instruction, nurse-led discharge planning with a focus on activities of
429
daily living, medical treatment, social support, and followup with a home visit and telephone
contact in the post-hospitalization period.28
These six studies share several similarities. Five studies were done in geriatric
populations.27,28,35,43,65,68 All had bridging interventions that included five or more separate
interventions, including a dedicated transitional provider across the continuum of care,
individualized personal health records, and post-hospitalization outreach to patients. All six
studies also involved patient contact at multiple points during and after hospitalization. These
interventions likely require a considerable amount of time, resources, and additional staff
(dedicated transitional provider) to facilitate the coordination of care from hospital to home.
Although the relative intensity of the interventions could not be measured directly, the
multifaceted nature of these interventions means they likely were more intensive than those
described in studies that did not find reduced readmission rates. The CTI is the only program
shown to reduce readmissions in multiple studies in different health care settings.27,43,65,68
Interventions To Prevent Adverse Events After Discharge

A total of nine studies reported adverse events (AEs) following discharge25,30-33,40,44,45,58
(Appendix D, Table 5, Chapter 37). Of these, five specifically reported rates of adverse drug
events (ADEs)32,33,44,45,58 and/or reactions--i.e., events that could be attributed to the use of a
drug. Five studies reported more generally on rates of other types of AEs,25,30,31,33,40 including
falls, post-discharge infection rates, failure to complete recommended outpatient follow-up, and
composite rates of all AEs. All studies except for one were RCTs.58
Only three studies demonstrated a significant decrease in event rates (specifically, ADEs)
following implementation of a transitional care intervention.32,45,58 Gillespie and colleagues
reported that a comprehensive pharmacist intervention in elderly patients 80 years of age and
over resulted in fewer medication-related (re)admissions. Hellstrom and colleagues reported that
a comprehensive pharmacist-led inpatient intervention, including systematic medication
reconciliation on admission and discharge, resulted in a reduction in the composite rate of drugrelated admissions and emergency department visits. Schnipper and colleagues reported that an
intervention consisting of pharmacist medication reconciliation at discharge, patient counseling,
and telephone follow-up resulted in a lower rate of preventable ADEs 30 days after hospital
discharge.
Each of these successful interventions was pharmacist led, while among unsuccessful
interventions, only one was pharmacist led. In addition, all successful interventions had
substantial and multi-faceted inpatient components, including some form of medication
reconciliation and patient education focused on enabling patient self-management. Two of the
three interventions also had bridging components, including a follow-up phone call by a
pharmacist after patient discharge.32,45 One intervention also included the creation of an
individualized patient record of medications, which was faxed to the outpatient provider at
discharge.32 In contrast, the majority of unsuccessful interventions had only inpatient
components that were focused on intervening at a single step of the discharge process.
Regarding intervention context, all three studies were performed at teaching hospitals. Two of
the three studies32,58 took place in Sweden; only one of the three was based in the U.S.45
Authors used varying strategies to classify events as ADEs. Gillespie and colleagues used the
electronic medical record to ascertain if admissions were medication-related physicians caring
for patients were blinded to study assignments and were required to record if an admission was
thought to be medication-related. Hellstrom and colleagues had a multidisciplinary team who
430
were blinded to group allocationreview electronic medical records for unscheduled hospital
readmissions and ED visits to determine if they were drug related. Schnipper and colleagues
used a combination of structured screening via patient report by the Bates method70 and chart
review by blinded physician reviewers using the Naranjo algorithm to assess causality.71

None of the studies reported any harms associated with transitional care interventions. One
study reported a significantly increased rate of readmission in the intervention group,50 which
was considered a result of heightened vigilance on the part of providers and patients to identify
issues arising after hospitalization.

What Contexts?
Heterogeneity of Target Populations and the Exclusion of High-Risk
Groups
To maximize the generalizability of our findings, we limited our analysis to studies
examining the effectiveness of transitional care interventions in general medical inpatients only
(31 of 43 studies) 25,26,28-32,34-38,40-42,45-47,49,51-53,55-59,61,62,66,67or mixed patient populations (12 of 43)
27,33,39,43,44,48,50,60,63-65,68
. Despite attempting to capture studies aimed at a general medical
population, we found that the majority of studies targeted a specific demographic among medical
patients. Twenty six studies (60%) 26-29,31,32,35-43,46,47,49,52,53,58,59,63-65,68,72 were interventions
targeted specifically at elderly populations, although definitions of elderly varied widely ( >5580 years of age). 26 Seven (16%) of studies targeted patients with a specific payor,25,37,43,50,51,65,68
including members of a specific health plan (three studies 25,51,68); Medicare or Medicare fee-forservice (three studies 37,43,65); or individuals receiving care through the Veterans Administration
Health Care System (one study50). Eleven studies (26%) targeted individuals who were thought
to be at high risk for readmissions or adverse events,27,28,33,35,37,39,48,53,57,63,66 although definitions
of high risk were inconsistent across studies. Eight studies targeted individuals based on
medication-related indications, 26,35,37,41,44,46,62,66 including polypharmacy, or being on a highrisk medication; again, definitions of polypharmacy and high risk were inconsistent across
studies. The heterogeneity of target populations for interventions may limit the generalizability
of study findings to a general medical inpatient population at a single given institution.
Additionally, individuals with characteristics that may place them at higher than average risk
for readmission and adverse events were often excluded from study populations. The most
common clinically relevant exclusion criteria were as follows: presence of cognitive impairment
or dementia (14 studies); 27,28,31,34-37,40,42,43,50,52,53,65 non-English speaking, or not fluent in
dominant language of country in which intervention took place (15 studies); 27,3335,37,40,41,43,45,47,50,52,65,66,72
no telephone (ten studies); 27,33-35,37,40,43,45,50,72 terminal illness or too ill
31,33,41,42,47-50,53
(nine studies);
homeless (four studies); 27,29,37,40,69 presence of mental illness (four
studies); 26,27,36,53 and inadequate caregiver support (one study).37 The exclusion of these
individuals may limit the generalizability of study findings to specific groups generally
considered to be at lower risk for readmission and adverse events.
431
Limited Generalizability Due To Wide Variation in Health Care System

Factors
Most studies were conducted at teaching hospitals (25 studies or 57%;28,30; of these, five were multi-site studies39,50,60,62,65). Six studies
took place in a community hospital setting;37,41,43,46,49,63 of these, three were multi-site
studies.41,46,63 Four studies took place in safety net systems.25,34,48,56
Only about one-third of studies (14 studies) reported the health system context in which the
intervention was implemented. Three studies took place within the context of an integrated
delivery system;25,26,68 two studies took place within an HMO or capitated system;27,51 four
studies were in a safety net system;25,34,48,56 and six studies took place in a variety of other
settings, ranging from open non-integrated systems43,65,68 in countries with national health
systems36,46 to the Veterans Administration health system.50 Virtually no studies reported on
aspects of local quality improvement structures or safety culture that could influence intervention
success.
Only about half of analyzed studies (22 studies) were conducted within the
U.S.5,25,27,29,33,35,37,39,40,43-45,47,49-51,55,56,65-68 Of the remaining 21 studies, 4 took place in the United
Kingdom,26,38,41,52 3 took place in Canada,30,59,60and 14 took place in other countries, including
Australia,28,36,48,53,61 Sweden,32,58,64 Ireland,31,46,62 Germany,42 New Zealand,57 and Belgium63.
Given the heterogeneity of hospital sites, health care system contexts, and countries in which
the interventions took place, data are insufficient to allow broad generalization of various study
findings across different types of health care settings. Additionally, the large number of studies
taking place within academic settings may limit the generalizability of study findings to care
settings without an infrastructure and resources similar to those found within academic settings.
36,42,44,45,47,50,51,53,55,56,58,60-62,65-67,72
Limited Information on Resources Needed To Initiate and Sustain

Transitional Care Interventions
Fewer than one-third of studies (11 studies 25,33-35,38,41,44,46,60,63,65) described training protocols
or resources needed to implement a transitional care intervention. Most studies included at least a
general outline of the intervention (30 studies25-27,30-35,37,39-47,49-52,55,56,59,61,65,66,68) and a majority
(25 studies) reported a detailed timeline 26,27,29,30,34,35,37-43,45-48,50,52,55,58,62,63,65,68with explicit
descriptions of the components of the intervention. No studies reported a plan for sustainability
or plans for long-term incorporation of the intervention into current clinical practice. Thus,
information on the types of resources and/or training needed to conduct an intervention was
limited, and data on sustainability of interventions over time were markedly absent. However,
our results suggest that the most effective interventions also tended to be the most resource
intensive. Both the paucity of data on what resources are necessary for implementation and
sustainability, and the fact that the level of needed resources for a successful intervention is
likely to be quite high may represent significant barriers to implementation of transitional care
interventions in most settings.
Lack of Demonstrated Replicability of Interventions, Except for the

Care Transitions Intervention
We found that only one intervention, the CTI,68 had been implemented and evaluated in
multiple settings. The five studies of the CTI 27,43,65,68,73 have been conducted in a range of
hospitals, including tertiary care academic medical centers and community hospitals with and
without teaching programs, and in both integrated and non-integrated health care systems. All
432
other studies that demonstrated reductions in 30-day readmissions or ED visits were singlecenter studies that have not been replicated in other settings or patient populations.
One study73 evaluated the implementation of the CTI in ten California hospitals, using a
qualitative approach to identify key factors associated with successful implementation.
Leadership support and early engagement of hospital and community stakeholders were
identified as important steps in ensuring early implementation success; maintaining a cadre of
funded transition coaches was thought to be essential for ensuring CTI sustainability.

Cost outcomes were reported in 14 studies, although no studies actually reported the costs
associated with intervention implementation itself. The studies that did report costs generally
compared the health care utilization and associated costs for patients in the intervention group
with those of patients receiving usual care. These costs were measured over varying intervals
after discharge, and used cost estimates from different sources. As a result, it is difficult to draw
any firm conclusions on the effect of transitional care interventions on overall health care costs.
Prior systematic reviews of interventions conducted in disease-specific and general medical
populations also did not reach any definitive conclusions regarding cost savings from transitional
care interventions.8,14,15
The lack of information on the cost of intervention implementation is particularly
problematic for health care organizations that are planning strategic approaches to reducing
readmissions. We found that only relatively intensive bridging interventionswhich generally
required additional personnel and other resourcessuccessfully reduced readmissions. This
finding suggests that hospitals may have to make considerable up-front investments in order to
implement such programs. Doing so will likely require a strong business case that the investment
will eventually be at least cost-neutral, if not cost-saving (perhaps driven by upcoming CMS
penalties on excessive readmissions). However, the data required to make this business case are
currently lacking.

As the CTI is the only method evaluated in different patient populations and health care
systems, we are not able to draw conclusions regarding the effect of context on effectiveness. As
discussed above, only a minority of studies reported important contextual details such as the
structure of the health care system in which the study was conducted or relevant measures of
culture or teamwork, and at the patient level, studies generally excluded patient populations that
might be at a higher risk of readmission. Transitional care is inherently complex, with myriad
patient- and system-level factors that may influence the success of an intervention. It is therefore
quite likely that contextual factors do influence the effectiveness of transitional care strategies;
however, this issue is not well explored in the existing literature.

Hospitals and health care organizations are under increasing pressure to improve transitional
care, particularly at hospital discharge, due to a growing body of literature documenting
unacceptably high rates of AEs after discharge and short-term ED visits and readmissions. We
systematically reviewed the literature to identify Patient Safety Practices that were effective at
433
reducing AEs, ED visits, and readmissions after discharge, and determine what is known about
the influence of contextual and implementation factors on the success of these interventions.
Only a limited number of relatively high-intensity bridging interventions appear to reduce
readmissions and ED visits, and only one of these (the Care Transitions Intervention) has been
implemented in multiple contexts. Pharmacist-led interventions do appear successful at reducing
ADEs after discharge, but the overall literature base of interventions specifically targeting
common AEs after discharge is small. The studies we identified unfortunately provided little
information about implementation factors, contextual factors affecting the success of the
intervention, or costs of implementation. Such information will be needed to allow health care
system leaders and policymakers to plan strategically as they consider implementing programs to
prevent readmissions and other harms associated with transitions of care. A summary table is
located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Moderate-tohigh
How Much do We
Little/Difficult
References
1.
Naylor M, Keating SA. Transitional care.

Am J Nurs. 2008;108(9 Suppl):58-63; quiz
18797231.
2.
Coleman EA, Boult C. Improving the

quality of transitional care for persons with
complex care needs. J Am Geriatr Soc.
2003;51(4):556-7.12657079.
3.
Forster AJ, Murff HJ, Peterson JF, et al. The

incidence and severity of adverse events
affecting patients after discharge from the
hospital. Ann Intern Med. 2003;138(3):1617.12558354.
4.
Forster AJ, Clark HD, Menard A, et al.

Adverse events among medical patients after
discharge from hospital. Cmaj.
2004;170(3):345-9.14757670.
5.
Jencks SF, Williams MV, Coleman EA.

Rehospitalizations among patients in the
Medicare fee-for-service program. N Engl J
Med. 2009;360(14):1418-28.19339721.
6.
Budnitz DS, Lovegrove MC, Shehab N, et

al. Emergency hospitalizations for adverse
drug events in older Americans. N Engl J
Med. 2011;365(21):2002-12.22111719.
434
7.
Hospital Compare. U.S. Department of

Heatlh & Human Services Web site.
www.hospitalcompare.hhs.gov. Updated
April 19, 2012. Accessed April 26, 2012.
8.
Naylor MD, Aiken LH, Kurtzman ET, et al.

The Importance Of Transitional Care In
Achieving Health Reform. Health Affairs.
2011;30(4):746-54.2011061422.
9.
Nebeker JR, Barach P, Samore MH.

Clarifying adverse drug events: a clinicians
guide to terminology, documentation, and
reporting. Ann Intern Med.
2004;140(10):795-801.15148066.
10.
Parker SG, Peet SM, McPherson A, et al. A

systematic review of discharge arrangements
for older people. Health Technol Assess.
2002;6(4):1-183.12065067.
11.
Richards S, Coast J. Interventions to

improve access to health and social care
after discharge from hospital: a systematic
review. J Health Serv Res Policy.
2003;8(3):171-9.12869344.
12.
Mistiaen P, Francke AL, Poot E.

Interventions aimed at reducing problems in
adult patients discharged from hospital to
home: a systematic meta-review. BMC
Health Serv Res. 2007;7:47.17408472.
13.
Chiu WK, Newcomer R. A systematic

review of nurse-assisted case management
to improve hospital discharge transition
outcomes for the elderly. Prof Case Manag.
2007;12(6):330-6; quiz 7-8.18030153.
14.
Shepperd S, McClaran J, Phillips CO, et al.

Discharge planning from hospital to home.
2010;(1):CD000313.20091507.
15.
16.
17.
Hansen LO, Young RS, Hinami K, et al.

Interventions to reduce 30-day
rehospitalization: a systematic review. Ann
Intern Med. 2011;155(8):520-8.22007045.
Kripalani S, LeFevre F, Phillips CO, et al.
Deficits in communication and information
transfer between hospital-based and primary
care physicians: implications for patient
safety and continuity of care. JAMA.
2007;297(8):831-41.17327525.
Coleman EA, Smith JD, Raha D, et al.
Posthospital medication discrepancies:
prevalence and contributing factors. Arch
Intern Med. 2005;165(16):18427.16157827.
18.
Olson DP, Windish DM. Communication

discrepancies between physicians and
hospitalized patients. Arch Intern Med.
2010;170(15):1302-7.20696951.
19.
Moore C, McGinn T, Halm E. Tying up

loose ends: discharging patients with
unresolved medical issues. Arch Intern Med.
2007;167(12):1305-11.17592105.
20.
Phillips CO, Wright SM, Kern DE, et al.

Comprehensive discharge planning with
postdischarge support for older patients with
congestive heart failure: a meta-analysis.
JAMA. 2004;291(11):1358-67.15026403.
21.
van Walraven C, Jennings A, Taljaard M, et

al. Incidence of potentially avoidable urgent
readmissions and their relation to all-cause
urgent readmissions. Cmaj.
2011;183(14):E1067-72.21859870.
22.
Kansagara D, Englander H, Salanitro A, et

al. Risk prediction models for hospital
readmission: a systematic review. JAMA.
2011;306(15):1688-98.22009101.
23.
Holland R, Battersby J, Harvey I, et al.

Systematic review of multidisciplinary
interventions in heart failure. Heart.
2005;91(7):899-906.15958358.
435
24.
McAlister FA, Stewart S, Ferrua S, et al.

Multidisciplinary strategies for the
management of heart failure patients at high
risk for admission: a systematic review of
randomized trials. J Am Coll Cardiol.
2004;44(4):810-9.15312864.
25.
Balaban RB, Weissman JS, Samuel PA, et

al. Redefining and redesigning hospital
discharge to enhance patient care: a
randomized controlled study. J Gen Intern
Med. 2008;23(8):1228-33.18452048.
26.
Bolas H, Brookes K, Scott M, et al.

Evaluation of a hospital-based community
liaison pharmacy service in Northern
Ireland. Pharm World Sci. 2004;26(2):11420.15085948.
27.
Coleman EA, Parry C, Chalmers S, et al.

The care transitions intervention: results of a
randomized controlled trial. Arch Intern
Med. 2006;166(17):1822-8.17000937.
28.
Courtney M, Edwards H, Chang A, et al.

Fewer emergency readmissions and better
quality of life for older adults at risk of
hospital readmission: a randomized
controlled trial to determine the
effectiveness of a 24-week exercise and
telephone follow-up program. J Am Geriatr
Soc. 2009;57(3):395-402.19245413.
29.
Dellasega CA, Zerbe TM. A multimethod

study of advanced practice nurse
postdischarge care. Clin Excell Nurse Pract.
2000;4(5):286-93.11858450.
30.

Effect of a nurse team coordinator on
outcomes for hospitalized medicine patients.
Am J Med. 2005;118(10):114853.16194647.
31.
Gallagher PF, OConnor MN, OMahony D.

Prevention of potentially inappropriate
prescribing for elderly patients: a
randomized controlled trial using
STOPP/START criteria. Clin Pharmacol
Ther. 2011;89(6):845-54.21508941.
32.

Intern Med. 2009;169(9):894900.19433702.
33.
Graumlich JF, Novotny NL, Stephen Nace

G, et al. Patient readmissions, emergency
visits, and adverse events after softwareassisted discharge from hospital: cluster
randomized trial. J Hosp Med.
2009;4(7):E11-9.19479782.
43.
Parry C, Min SJ, Chugh A, et al. Further

application of the care transitions
intervention: results of a randomized
controlled trial conducted in a fee-forservice setting. Home Health Care Serv Q.
2009;28(2-3):84-99.20182958.
34.

trial. Ann Intern Med. 2009;150(3):17887.19189907.
44.
Schillig J, Kaatz S, Hudson M, et al. Clinical

and safety impact of an inpatient
Pharmacist-Directed anticoagulation service.
J Hosp Med. 2011;6(6):322-8.21834113.
45.
35.

a targeted care bundle. J Hosp Med.
2009;4(4):211-8.19388074.

hospitalization. Arch Intern Med.
2006;166(5):565-71.16534045.
46.
Scullin C, Scott MG, Hogg A, et al. An

innovative approach to integrated medicines
management. J Eval Clin Pract.
2007;13(5):781-8.17824872.
47.
Siu AL, Kravitz RL, Keeler E, et al.

Postdischarge geriatric assessment of
hospitalized frail elderly patients. Arch
Intern Med. 1996;156(1):76-81.8526700.
48.
Stewart S, Pearson S, Luke CG, et al.

Effects of home-based intervention on
unplanned readmissions and out-of-hospital
deaths. J Am Geriatr Soc. 1998;46(2):17480.9475445.
49.
Thomas DR, Brahan R, Haywood BP.

Inpatient community-based geriatric
assessment reduces subsequent mortality. J
Am Geriatr Soc. 1993;41(2):101-4.8426028.
50.
Weinberger M, Oddone EZ, Henderson WG.

Does increased access to primary care
reduce hospital readmissions? Veterans
Affairs Cooperative Study Group on
Primary Care and Hospital Readmission. N
Engl J Med. 1996;334(22):1441-7.8618584.
51.
Adler A, Lipkin C, Cooper L, et al. Effect of

social work intervention on hospital
discharge transition planning in a special
needs population. Manag Care.
2009;18(11):50-3.19999254.
52.

discharge. Br J Clin Pharmacol.
2002;54(6):657-64.12492615.
36.
Lim WK, Lambert SF, Gray LC.

Effectiveness of case management and postacute services in older people after hospital
discharge. Med J Aust. 2003;178(6):2626.12633482.
37.
Lipton HL, Bird JA. The impact of clinical

pharmacists consultations on geriatric
patients compliance and medical care use: a
randomized controlled trial. Gerontologist.
1994;34(3):307-15.8076871.
38.
Martin F, Oyewole A, Moloney A. A

randomized controlled trial of a high support
hospital discharge team for elderly people.
Age Ageing. 1994;23(3):228-34.8085509.
39.
Naylor MD, Brooten D, Campbell R, et al.

Comprehensive discharge planning and
home follow-up of hospitalized elders: a
randomized clinical trial. JAMA.
1999;281(7):613-20.10029122.
40.
Naylor MD. Comprehensive discharge

planning for hospitalized elderly: a pilot
study. Nurs Res. 1990;39(3):15661.2188217.
41.
42.

trial. Age Ageing. 2001;30(1):3340.11322670.
Nikolaus T, Specht-Leible N, Bach M, et al.
A randomized trial of comprehensive
geriatric assessment and home intervention
in the care of hospitalized patients. Age
Ageing. 1999;28(6):543-50.10604506.
436
53.
Brand CA, Jones CT, Lowe AJ, et al. A

transitional care service for elderly chronic
disease patients at risk of readmission. Aust
Health Rev. 2004;28(3):275-84.15595909.
54.
Coleman EA, Min SJ, Chomiak A, et al.

Posthospital care transitions: patterns,
complications, and risk identification.
Health Serv Res. 2004;39(5):144965.15333117.
55.
Cowan MJ, Shapiro M, Hays RD, et al. The

effect of a multidisciplinary
hospitalist/physician and advanced practice
nurse collaboration on hospital costs. J Nurs
Adm. 2006;36(2):79-85.16528149.
56.
Einstadter D, Cebul RD, Franta PR. Effect

of a nurse case manager on postdischarge
follow-up. J Gen Intern Med.
1996;11(11):684-8.9120655.
57.
Gow P, Berg S, Smith D, et al. Care coordination improves quality-of-care at South

Auckland Health. J Qual Clin Pract.
1999;19(2):107-10.10408752.
58.
Hellstrom LM, Bondesson A, Hoglund P, et

al. Impact of the Lund Integrated Medicines
Management (LIMM) model on medication
appropriateness and drug-related hospital
revisits. Eur J Clin Pharmacol.
2011;67(7):741-52.21318595.
59.
60.
61.
62.
Hogan DB, Fox RA. A prospective

controlled trial of a geriatric consultation
team in an acute-care hospital. Age Ageing.
1990;19(2):107-13.2337005.
Makowsky MJ, Koshman SL, Midodzi WK,
et al. Capturing outcomes of clinical
activities performed by a rounding
pharmacist practicing in a team
environment: the COLLABORATE study
[NCT00351676]. Med Care.
2009;47(6):642-50.19433997.
Mudge A, Laracy S, Richter K, et al.
Controlled trial of multidisciplinary care
teams for acutely ill medical inpatients:
enhanced multidisciplinary care. Intern Med
J. 2006;36(9):558-63.16911547.
Scullin C, Hogg A, Luo R, et al. Integrated
medicines management - can routine
implementation improve quality? J Eval
Clin Pract. 2011.21504517
437
63.
Steeman E, Moons P, Milisen K, et al.

Implementation of discharge management
for geriatric patients at risk of readmission
or institutionalization. Int J Qual Health
Care. 2006;18(5):352-8.16861721.
64.
Styrborn K. Early discharge planning for

elderly patients in acute hospitals--an
intervention study. Scand J Soc Med.
1995;23(4):273-85.8919370.
65.
Voss R, Gardner R, Baier R, et al. The care

efficacy to effectiveness. Arch Intern Med.
2011;171(14):1232-7.21788540.
66.

study. Arch Intern Med. 2009;169(21):200310.19933963.
67.
Palmer HC, Jr., Halperin A, Elnicki M, et al.

Effect of a patient care partnership project
on cost and quality of care at an academic
teaching hospital. South Med J.
2002;95(11):1318-25.12540000.
68.
Coleman EA, Smith JD, Frank JC, et al.

Preparing patients and caregivers to
participate in care delivered across settings:
the care transitions intervention. Journal of
the American Geriatrics Society.
2004;52(11):1817-25.2005044404.
69.
Crossing the Quality Chasm: A New Health

Care System for the 21st Century. 2001.
70.
Bates DW, Cullen DJ, Laird N, et al.

Incidence of adverse drug events and
potential adverse drug events. Implications
for prevention. ADE Prevention Study
Group. JAMA. 1995;274(1):29-34.7791255.
71.
Leape LL, Brennan TA, Laird N, et al. The

nature of adverse events in hospitalized
patients. Results of the Harvard Medical
Practice Study II. N Engl J Med.
1991;324(6):377-84.1824793.
72.
Naylor MD, McCauley KM. The effects of a

discharge planning and home follow-up
intervention on elders hospitalized with
common medical and surgical cardiac
conditions. J Cardiovasc Nurs.
1999;14(1):44-54.10533691.
73.
Parrish MM, OMalley K, Adams RI, et al.

Implementation of the care transitions
intervention: sustainability and lessons
learned. Prof Case Manag. 2009;14(6):28293; quiz 94-5.19935345.
438
Chapter 38. Use of Simulation Exercises in Patient Safety

Efforts
Eric M. Schmidt, B.A.; Sara N. Goldhaber-Fiebert, M.D.; Lawrence A. Ho, M.D.; Kathryn M.
McDonald, M.M.

Every July, patients are treated by new residents and fellows who have started their training
programs only days before. Given that error rates drop with experience,1-4 an important challenge
is how to train physicians while minimizing the potential for patient harm. Many medical
educators now regard the traditional medical training model See one, Do one, Teach one, as
unstructured and inadequate.5,6 For example, before performing a procedure like airway
intubation on a patient for the first time, a trainee likely would have read about the theory with
some visuals, but may or may not have touched the equipment, discussed the detailed steps and
common pitfalls, or necessarily observed an experienced physician performing the procedure.
Before being a team leader, for example before coordinating a cardiac arrest resuscitation, a
senior resident would likely have participated in prior resuscitations but may never have formally
debriefed the experience to learn from it. However, simulation exercises or drills allow residents
and trainees to practice the necessary steps in a safe situation. Then, a debrief is often structured
with dedicated time to reflect upon what worked well and what did not during the exercise to
identify ways to improve the next resuscitation. Without such simulations, the team leader would
by definition never have practiced being the leader before going live in this role for the first
time. While all physicians still must perform procedures on and manage critical events for an
actual patient for the first time, simulation is likely to make this first time safer and more
efficient.
However, the use of patient safety efforts that involve simulation target not just inexperience.
Clinical expertise and mastery within a specialty does not increase simply as a function of
experience,7,8 and, likewise, patient safety issues are not likely to decrease simply as a function
of more practice hours: experienced surgeons are involved in a notable proportion of malpractice
claims.9 Simulation with debriefing, or other forms of deliberate practice and reflection are
needed to continuously improve care. With experience, comes increased pressure in supervising
roles and increased clinical demand to treat the most complex, difficult, and rare conditions. The
likelihood of desirable outcomes can be increased, even among experienced clinicians, as
simulation allows adjustment of the complexity of the procedures and regular practice with
treatment for rare conditions. At times, the most senior clinicians will be called on to perform
tasks outside of their specialty due to medical necessity or availability of resources (e.g., in rural
areas). Simulation may provide a mechanism for clinicians to practice responding to these highstakes situations without harm to patients. In addition to developing these advanced
techniques, experienced clinicians must maintain proficiency in a wide array of skills, some of
which are known to deteriorate over time without practice.10,11 Simulation can serve to maintain
clinical skills and may be part of maintenance of board certification, as is the case for The
American Board of Anesthesiology.12
Simulation can also address team and organizational issues that challenge patient safety.
Technical errors may occur in a team care environment as a result of nontechnical factors (e.g.,
439
communication), and specific simulation-based training protocols have been developed for
enhancing team performance. These protocols include Anesthesiology Crisis Resource
Management.13 Likewise, simulation may provide a way to assess the efficiency and safety of
system-level practices that may be difficult to control in real time or unsafe to test empirically.
Simulation has received increased recognition since the release of Making Health Care
Safer and seminal Institute of Medicine reports on preventable medical errors and mortality.14,15
The Agency for Healthcare Research and Quality (AHRQ) continues to fund a number of
projects dedicated to enhancing patient safety through simulation.16,17 Accrediting bodies from
the American Association of College of Nurses, American College of Surgeons, American
Association of Anesthesiologists, and Society for Simulation in Healthcare have all supported
simulation training centers.18-20
Studies evaluating the relationship between the benefits of simulation exercises and patient
safety outcomes, including potential harms, have not been thoroughly evaluated. Therefore, the
purpose of this chapter is to systematically review evidence on the benefits and harms of using
simulation to improve patient safety in medicine.

The use of simulation has a long history in health services education.21-23 However, recent
advances in simulation have been inspired by aviation and other high-reliability, high-stakes
industries that emphasized the inherent value of allowing initial practice of all crucial skills
during simulation, as a lower-stakes context, but with sufficient realism for the skills to transfer
effectively. In the 1990s, Gaba and Howard developed simulation courses for Anesthesia Crisis
Resource Management built upon Crew Resource Management (CRM) principles from aviation
(see the chapter Team-Training in Healthcare: Brief Review for more information on CRM in
health care).13 In addition to enhancing team performance, use of simulations to improve
understanding of technological advances and effective practices among individuals and care
systems has increased substantially in the past decade.
Simulation is considered a technique rather than any one patient safety practice. Importantly,
this technique is versatile as it may be applied across specialties and levels of intervention.24
Simulation to enhance patient safety has four general purposes25 :
1. Education
2. Assessment
3. Research
4. Health System Integration
These purposes are not mutually exclusive, and each may span a range of complexity.
Simulation serves an educational role in transitioning trainees from content knowledge to
experiential practice, in continuing education, and in moving toward advanced practices. A
classic low-fidelity example is training intramuscular medication administration through
simulated practice inserting a needle into an orange. More complicated partial task training
may now include high-fidelity simulations that utilize anatomically accurate mannequins with
realistic surgical equipment monitored by computer. Patient safety may also be enhanced
through full scenario management, a fully-simulated care environment such as entire operating
rooms and care teams. In addition to educational aims, simulators can provide structure for
critical assessment in quality control or quality improvement through systematic research into
clinician behaviors, care team processes, and integrating health system-level processes.
440
Simulation is particularly useful for research into processes that cannot be varied in a study
without undue harm to patients. Specific considerations in implementing simulation are detailed
in a later section of this chapter.
On a basic level, simulations improve patient safety by allowing physicians to become better
trained without putting patients at risk and, importantly, by providing a protected time for
reflection and debriefing. The challenge is matching the best simulation method and training
details for the desired learning objectives, while recognizing the costs of each method.24 Because
simulation is a broad technique, rather than a specific technology, faculty training and time are
often considered a more important investment than are specific expensive simulation equipment.
Given that most clinicians are not trained in simulation and debriefing, specific practitioners with
interest in the area must be appropriately trained to effectively use simulation techniques, as well
as any specific desired technologies, in order to accomplish the relevant training or systems
probing goals. The significant investment of time requires that faculty are provided time to
develop and teach meaningful simulations. Besides some procedural skills that may require
mainly repetitive practice, most simulations require an extensive debriefing component, during
which much of the learning takes place. Learning to effectively facilitate debriefings is often the
most time-intensive component of faculty training.26,27
The simulation itself needs to feel real enough for participants to be able to suspend
disbelief, acting and thinking much as they would in a similar but real scenario.25 If the learning
objective is mainly to practice cognitive skills for diagnosis or treatment, a verbal simulation
such as What would you do if may be sufficient. In contrast, if time pressure and/or team
communication are the focus, a more accurate replication of the actions and team presence
become important for the simulation experience.

Research demonstrating the benefits of simulation comes from studies about simulation as
well as studies using simulation.28 Research about simulation directly examines the effect of a
simulation technique as an intervention on behaviors and actions at the health professional or
team level that could directly improve patient safety if that training were widely implemented.
Studies using simulation harness these techniques as a laboratory to investigate new technologies
and human performance for insights into potential causal pathways to improve safety. Strategies
that employ simulation techniques may do so as part of a multicomponent intervention or may
focus on the simulation alone. The versatility of simulation techniques affords those working to
improve patient safety a number of benefits.
First, simulation is initiated on command and may be practiced repeatedly. Additionally, the
content of simulated exercises may be structured to meet particular goals. Thus simulation has
been utilized to enhance the reliability of clinician behaviors in order to reduce medical error
associated with inexperience or undesirable levels of competency.1-4 A systematic review of
simulation with deliberate practice added to traditional training models reported in favor of
simulation for enhancing clinicians technical skills performance (pooled effect size = 0.71, 95%
CI, 0.65 to 0.76, p < .001).29 Simulation has been associated with improved knowledge
acquisition and clinical reasoning beyond traditional training in other studies as well,30 and one
meta-analysis reported that training with computerized virtual patients enhanced performance in
actual patient care (pooled effect size = 0.50; 95% CI, 0.34 to 1.19).31 The nature of simulation
also allows practitioners experience with rare and unpredictable care scenarios (e.g., mass
441
casualty incidents),32 with rare clinical events, and with advancing to mastery levels in a
specialty.7,8
Second, simulation may be particularly valuable in duplicating complex high-stakes
scenarios that involve care teams and complex factors that affect performance.33,34 During an
Anesthesiology Crisis Resource Management simulation (including realistic emergency
scenarios, team communication, and complex decisionmaking), Blum and colleagues planted
probes of clinically pertinent information known only to a single member of the care team.35
These researchers found that participants shared only 27 percent of the probes with the team,
electing instead to share redundant information already available to others. Authority and power
differentials between senior and junior clinicians may also hinder effective communication, and
reluctance to challenge superiors may result in medical errors in high-stakes scenarios. The field
of aviation uses a method called the two-challenge rule to train all team members in a technique
known as advocacy-inquiry, giving junior team members the language to speak up while
seeking clarification. To study use of the two-challenge concept in debriefing anesthesiology
trainees, Pian-Smith and colleagues structured emergency simulations with contraindicated
decisions made by attending clinicians.36 The researchers discovered that trainees were initially
reluctant to challenge their superiors, but after debriefing, they were significantly more likely to
make an effective and clear challenge to contraindicated decisions by attending clinicians.
Third, errors are allowed in simulation and utilized as learning experiences through reflection
and debriefing. Training with real patients requires that supervising clinicians intervene if certain
errors occur, disallowing trainees to carry out the remainder of the procedure or to protect time
for debriefing. Additionally, as Fanning and colleagues note, reflective practices are considered a
cornerstone of life-long education in medicine.26 Simulation may facilitate highly accurate
measures of specific care-related behaviors and processes (technical or non-technical) for the
purpose of debriefing and reflection.26,27 Debriefing has exerted an increased performance effect
over self-study in high-fidelity simulated assessments.37
Fourth, simulation may be used to test new technologies, especially complex ones that
involve new learning curves in even the most experienced of clinicians.38 Likewise, new team
processes or innovative practices may be simulated prior to implementing in real time.39
Simulation may also help resolve disputes between best care practices. For example,
cardiopulmonary arrest of a woman pregnant for greater than 20 weeks requires cesarean
delivery within 5 minutes of onset, to protect both mother and fetus. However, debate has existed
as to whether the procedure should be performed in the labor room or in an operating room after
transporting the patient. Clearly, one would not consider placing actual patients in harms way
just for the purpose of settling this debate in research. However, utilizing simulation in a
randomized design, Lipman and colleagues determined that the average time to incision was
three and half minutes longer in teams instructed to transport patients to the operating room.40
Lastly, teams can simulate patient care flow in situ, for critical events41-43 as well as new
facility usual care, to check for the proper equipment (e.g., severe hemorrhage drills on various
hospital units or new operating rooms, respectively).44,45 Researchers have utilized simulationbased communication training to enhance inter-agency processes such as telephone referrals,46
but this use remains a relatively under-studied area.

Questions regarding the details of a simulation intervention and its measurable impact on
patient outcomes are difficult to answer for many reasons, so there have been relatively few
442
studies in this arena. The measurable outcome may vary substantially based on the specific
behavior or process under investigation. For example, small gains may be expected on patient
outcomes from additional simulation exercises when assessing clinicians who must first
demonstrate a minimum acceptable level of competency before practicing on actual patients.
Likewise, comparing experienced to expert clinicians requires a highly sensitive and specific
methodology that is still being developed currently among simulation experts.7,8 A plethora of
simulation research has been conducted on what translational science refers to as T1, or
laboratory-only, outcomes.12,47 As noted above, T1 studies, or research about simulation,
support the logic for why simulation matters for patient safety. This section on beneficial effects
of simulation focuses on studies that reported T2 or T3 outcomes and that were published since
Making Health Care Safer, or those studies that translate interventions directly to patient-level
and systems-level outcomes, respectively.12,47 Studies are grouped into simulation exercises that
assessed patient outcomes related to practitioner technical performance, team-level, and systemlevel outcomes. Following these sections is an in-depth review of literature on patient outcomes
related to central venous catheterization.
Practitioner Performance
This section focuses on the technical aspects of physician performance during procedures.
Although technical skills are a crucial component to effective and safe health care, there is
potential for simulation to improve cognitive and other decision-making processes that impact
the delivery of services.
Diagnostic procedures. In a randomized trial with first-year gastroenterology fellows,
simulation-trained fellows out-performed traditionally-trained fellows during standardized
assessment of performance on the first 80 colonoscopy procedures (p = 0.03). The difference
between groups did not persist beyond 80 procedures, and the researchers determined that both
groups required the entire 200-colonoscopy training experience to achieve a desirable level of
procedure mastery.48 Another randomized study of first-year gastroenterology fellows reported
similar results, with simulation-trained clinicians successfully reaching the cecum in 38 percent
of their first 15 colonoscopies compared with 20 percent in the control group (p = 0.027).
Differences between groups were also observable through 30 procedures, but differences did not
persist beyond 30 procedures.49 Other studies have reported earlier acquisition and higher
performance in first colonoscopies performed among trainees who received simulation
training,50-52 and one reported that simulation-trained clinicians reached the cecum 4.5 times
more often on average than did clinicians in the control group during their first 10 procedures
(95% CI, 1.89 to 11.60, p = 0.001).50 Similar effects of simulation on increased early
performance enhancement, and, subsequently, absence of differences between study groups,
were reported for upper gastrointestinal endoscopy.53 In a randomized study, bronchoscopy
simulation exerted no effect on procedure time.54 Safety outcomes reported in these studies
focused primarily on patient discomfort (e.g., insufflation), and simulation training was
associated with lower discomfort in one study,50 no difference in another,48 and higher patient
discomfort in a third.49
In a before and after design, thoracenteses performed after implementing simulation in a
training curriculum involved fewer pneumothoraces (8.7% before vs. 1.1% after, p = 0.003) and
procedures advancing to thoracostomy (6% before vs. 0% after, p = 0.003) than those performed
prior to simulation-based training.55 In a similar type of research design examining
443
cordocentesis, procedure time was lower (6.4 min vs. 13.2 min, p < 0.001) and success rate was
higher (98.8% vs. 94.8%, p < 0.001) after implementing simulation training. There was no
difference between the before- and after-simulation study periods in procedure-related fetal loss
or overall fetal loss.56
Surgical procedures. A meta-analysis of laparoscopic training with virtual reality simulators
reported that procedure time was no faster, but was more accurate (standardized mean difference
[SMD] 0.68, 95% CI 0.05 1.31) in simulation-trained clinicians compared with traditional
video-trained clinicians. These authors reported no difference between simulation-trained and
other-trained clinicians in conversion rate to open surgery.57 Surgical residents who were
randomized to simulation training on laparoscopic cholecystectomy exhibited fewer errors in
exposing (control mean = 53.4, simulator mean = 15.0, p < 0.04), clipping (control mean = 7.1,
simulator mean = 1.9, p < 0.008), and dissection (control mean = 29.5, simulator mean = 11.5, p
< 0.03) during their first ten cholecystectomies. Three-fold fewer total errors and an eight-fold
decreased variation in error making totals were found among simulation-trained clinicians.58
Another study reported that warming up with laparoscopic simulation led to increased
observed structured assessment of technical skills (OSATS) global rating on actual
cholecystectomies in trainee- and experienced clinicians (warm-up mean = 28.50, control mean
= 19.25, p = 0.042), but no significant differences were reported on any one technical skill.59
Compared with no simulation training among surgery residents, simulation training for
extraperitoneal hernia repair was associated with larger increases in OSATS scores over a
baseline procedure for knowledge of procedure (p < 0.05), knowledge of instruments (p = 0.05),
and use of assistants (p < 0.05), but not global score.60 Researchers have utilized simulation for
other surgical modalities as well. In one study, urology residents trained with simulators
performed prostate resection faster (p = 0.025), and with higher performance scores on a
structured assessment (p = 0.021) than those not trained with simulation.61
Other procedures and processes. Pediatrics interns (n = 38) were randomized to simulation
training on basic procedural skills or training as usual for bag-mask ventilation, venipuncture,
peripheral venous catheter placement, and lumbar puncture. Interns in the simulation group
exhibited a higher, but not significantly higher, rate of successful venipuncture (p = 0.08). No
significant differences were observed between groups on the other procedural skills, although
performance scores on all measures were higher in the simulation group.62 Simulation has shown
additional benefits over traditional training in other areas of patient safety. Bachelors level
nursing students randomized to traditional pharmacology coursework or coursework with
additional simulation training were observed for medication administration errors in subsequent
external training placements. Students who trained with simulation made fewer medication
administration errors (7 of 31 total errors observed, p < 0.05), and these results were consistent
across both maternal health (control group = 8 errors, simulation group = 0 errors) and medicalsurgical settings (control group = 16 errors, simulation group = 7 errors).63 In another study,
paramedic students who received simulation training for endotracheal intubation performed
similarly to traditionally-trained students on their first 15 intubations for overall success rate,
success rate on first attempt, and in complications resulting from the intubation procedure.64
Although central venous cathether (CVC) placement is a specific procedure, the in-depth
review of literature on CVC is reviewed below. Again, the physician is not only a technician but
an actor that must balance being cognitively engaged with direct patient care as well as the
444
system within which she or he works. These systems include interpersonal dynamics such as
dyadic relationships (e.g., doctor-nurse pairs), larger team relationships (e.g., cardiac
resuscitation), and the care culture and environment (e.g., patient safety culture, and available
resources). Simulation has the potential to impact care processes and relationships at each of
these levels.
Team and Systems Performance

A 1-day workshop and training program implemented simulation for seven common obstetric
emergencies in the third year of a 6-year study period: shoulder dystocia, postpartum
hemorrhage, eclampsia, delivery of twins, breech, adult resuscitation, and neonatal resuscitation.
The researchers compared live-term births in the period prior to simulation (n = 8,430) to liveterm births after implementing simulation (n = 11,030) into the training workshop. Births with 5minute APGAR less than or equal to 6 decreased from 73/10,000 births (SD = 86.6) presimulation to 49/10,000 births (SD = 44.4) after implementing simulation training (p < 0.05).
The rate of hypoxic-ischemic encephalopathy also decreased from 23/10,000 births (SD = 27.3)
to 15/10,000 (SD = 13.6, p <0 .05). The rate of moderate to severe hypoxic-ischemic
encephalopathy decreased, but not significantly.65
Primary care physicians (n = 51) in a large, multidisciplinary medical group were
randomized to simulation training alone, simulation training with physician-leader feedback, or a
control group. Medical records were evaluated for metformin prescriptions to patients with
diabetes who had regularly attended care for two years (n = 2,020) and that had either known
congestive heart failure or a laboratory result with elevated creatinine. The rate at which controlgroup physicians prescribed renal-contraindicated metformin was not statistically different
compared with either simulation group alone. However, when simulation groups were combined,
the simulation-trained physicians prescribed metformin in these unsafe scenarios significantly
less often (range from -3.8 to -10.3% across simulation groups, p = 0.03).66
Year-four postgraduate anesthesiology residents (n = 20) were randomized to a full missiontype simulation for patients weaning from cardiopulmonary bypass that included a complete
operating room environment and care team. Residents were scored during actual performance in
elective coronary artery bypass graft (CABG) with a structured anesthesiologists nontechnical
skills (ANTS) assessment before and after the simulation. The simulation-trained grouped
exhibited significantly increased scores after the simulation training over the increases realized in
the control group (control mean = 11.8, simulation mean = 14.3, p < 0.001) and at 5-week
followup (control mean = 11.7, simulation mean = 14.1, p < 0.001).67
Two studies reported patient outcomes after simulation-based training for resuscitation
teams. In one study, internal medicine residents were trained in team management for
resuscitation scenarios. These authors reported no differences attributable to simulation training
for actual team performance on ventilation rate, return of spontaneous blood circulation, or
survival to discharge rates. Of note, both of the latter rates were higher in the simulation group.68
However, another study examined resuscitation outcomes after implementing the TeamSTEPPS
team-building program coupled with simulation. This study reported a number of significant
communication improvements during observed resuscitations, such as leadership (p = 0.003),
situation monitoring (p = 0.009), mutual support (p = 0.004), and communication (p = 0.001).
Post-simulation reductions were also noted in average time to computed tomography (26.4 min
vs. 22.1 min, p = 0.005), to intubation (10.1 min vs. 6.6 min, p = 0.49), and to operating room
(130.1 min vs. 94.5 min, p = 0.021).69 A team-response and educational curriculum coupled with
445
debriefing was provided to second-year internal medicine residents (n = 38) whose performance
on subsequent resuscitations was compared with third-year residents (n = 40) who did not
receive simulation training. Based on American Heart Association standards, simulation-trained
residents made 68 percent correct responses compared with 44 percent in the non-simulationtrained residents (mean difference = 44%, p < 0.001).41
In-Depth Look at Simulation and Central Venous Catheterization

Central venous catheters (CVC) are used to obtain vascular access as well as hemodynamic
monitoring and are common fixtures in ICU settings. In fact, 48 percent of patients in the ICU
have indwelling CVC, which corresponds to 15 million CVC patient days per year.70,71 Despite
the broad presence of CVC in the ICU, significant morbidity and mortality are associated with
CVC.71,72 Although catheter-related blood stream infections (CRBSI) have been studied
extensively, and a separate companion evidence report is dedicated to Heathcare Associated
Infections, simulation promises to exert protective effects against risks involved with the
insertion process, including pneumothorax, arterial puncture, bleeding, and deep vein
thrombosis.73 Due to these risks, Federal agencies such as the National Quality Forum and
AHRQ have listed CVC practices as a top patient safety concern.74,75
Trainees in internal medicine, emergency medicine, and surgical specialties commonly insert
CVC in academic settings.76 Despite requirements to demonstrate knowledge about indications,
contraindications, complications, and sterile technique for CVC insertions,77,78 trainees remain
uncomfortable with performing the procedure.79 Training had previously consisted of the
apprenticeship model with learning at bedside on actual patients; however, this method is often
found to be inadequate.41,80 A recent meta-analysis has shown that simulation-based education in
CVC techniques improves both learner outcomes and performance during actual procedures:
fewer needle passes (standardized mean difference = -0.58, 95% CI -0.95 to -0.20) and reduced
pneumothoraces (relative risk = 0.62, 95% CI 0.40 to 0.97).81
In addition to the articles included in this meta-analysis, studies have demonstrated that
simulation improves learner outcomes such as knowledge, confidence, and performance on
simulators as well patient outcomes such as fewer needle passes, fewer pneumothoraces, and less
CRBSI (Table 1 below). These studies may not be direct replication of each other, but the
consistency in observed benefits of simulation across a variety of clinical specialties and care
settings is promising. Additionally, these studies have been conducted across numerous research
teams in academic or teaching hospitals. Many of these studies used a high-fidelity mannequin
partial-task trainer on CVC insertion. However, more involved simulations include sanitation
techniques and incorporate the entirety of the operation rather than any specific operative
technique. Additionally, once operational, these simulators have been utilized for differing
patient needs such as for pediatric medicine instead of adult medicine as well as other types of
CVC, including hemodialysis catheters.82,83 Future and ongoing investigations will give us
further insight into long term maintenance of CVC skills with simulation.84
Harms
Studies included in this review generally provided additive or supplemental interventions to
training as usual, and no study reported data indicating increased potential for or actual harm to
patients that resulted from implementing simulation techniques. However, it is conceivable that
simulation exercises would place demand on valuable resources that could be applied elsewhere
446
in patient safety efforts. Such considerations were not evaluated in literature captured for this
review.
How Has the Patient Safety Practice Been Implemented and in

What Context?
Context for Simulation
A meta-analysis of computer-aided simulation in education programs for health professionals
determined that 564 of the 609 studies included in the review (92.6%) examined effects of
simulation provided through dedicated simulation centers. Thirty-four additional studies (5.6%)
examined simulation provided in the clinical environment, and 11 studies (1.8%) reported from
both contexts.31 Across studies cited in this review that reported the context in which simulation
was implemented, academic medical systems and academically-affiliated hospitals
predominate.48,51,53-56,59-64,68,73,76,83,85-92 However, studies also reported outcomes specific to
implementing simulation in tertiary care facilities,52,68,85,87 in trauma centers,69,89 and in
multispecialty medical groups.65,66 Only one study in our review reported patient outcomes by
care setting63; however, these outcomes are not specific to the simulation training site but to a
subsequent external training placement.
447
Table 1, Chapter 38. Literature on simulation training for central venous catheterization
Study
Andreatta et
93
al. (2011)
Study Design
Randomized
controlled,
blinded, trial
Barsuk et al.
88
(2009)
Non-randomized
observational
cohort with
historical cohort
for controls
Barsuk et al.
76
(2009)
Pre-post, nonrandomized
observational
cohort
Barsuk et al.
73
(2009)
Observational
cohort study
Patient Safety Simulation

Practice
One hour minimum practice time
with high-fidelity partial task trainer
and ultrasound system, with
supervision and individualized
training program
Simulation
Participants
PGY-1 and PGY-2
radiology residents
(n = 32) performed
32 peripherally
inserted central
catheters
Minimum proficiency/mastery
model on standardized checklist of
CVC insertion with expert panel
review of performance using a
high-fidelity partial task trainer with
computer interface, simulation
exercises performed with deliberate
practice and individualized
feedback components
feedback components
feedback components
PGY-2 and PGY-3

internal medicine
and emergency
medicine residents
(n = 103) performed
407 internal jugular
and subclavian
CVCs
Setting, Results, and Patient-Related Outcomes

In interventional radiology at a single academic health system,
simulator-trained residents outperformed bedside-trained
residence on a number of performance criteria such as
ultrasound use, vein compressibility, needle localization and
guiding, and exchanging the needle/catheter via the guidewire (p
< 0.05 for all). All simulator-trained residents placed the catheter
successfully, whereas 4 of 16 bedside-trained residents were
unable to place the catheter independently within three attempts.
In the medical ICU of a single academic institution, there were no
significant differences in the outcomes of subclavian line
insertion; however, residents who were simulator trained reported
fewer needle passes (p < 0.0005), arterial punctures (p <
0.0005), catheter adjustments (p = 0.002), and higher success
rate (p = 0.005) in overall central venous catheter placement.
PGY-2 and PGY-3

internal medicine
and emergency
medicine residents
(n = 92), number of
CVCs performed NR
In the medical ICU of a single academic institution, fewer CRBSI

occurred after simulation, and the rate of CRBSI after simulation
training reduced from 3.2 per 1000 catheter-days to 0.5 per 1000
catheter-days (p = 0.001). Comparing the medical (MICU) to the
surgical ICU during the study period revealed a lower incidence
ratio for CRBSI in the MICU (0.16, 95% CI, 0.05 0.44, p =
0.001).
PGY-2 and PGY-3

internal medicine
residents (n = 41)
performed 46
internal jugular and
subclavian CVCs
In the medical ICU of a single academic institution, although the

groups did not differ significantly regarding complications
(pneumothorax, arterial puncture or CVC adjustment), the
simulator-trained group required fewer needle passes (M = 1.79,
SD = 1.0) than the traditionally trained group (M = 2.78, SD =
1.77, p = 0.04).
448
Table 1, Chapter 38. Literature on simulation training for central venous catheterization (continued)
Practice
Minimum proficiency model with
deliberate practice and
individualized feedback from
attending surgeon on a partial-task
high-fidelity trainer, added to
standard lecture on CVC technique
Study
Britt et al.
89
(2009)
Study Design
Randomized
controlled trial
Evans
85
(2010)
Randomized
controlled,
blinded, trial
Minimum proficiency model with

deliberate practice and
individualized feedback from
attending surgeon on a partial-task
high fidelity trainer, added to
standard lecture on CVC technique
Khouli, et al.
90
(2011)
Randomized
controlled,
blinded, trial
Full-immersion operation room

simulation, including partial task
trainer, in addition to video training
and structured assessment training
on sterile techniques.
Simulation
Participants
Junior surgical
residents (n = 34,
PGY-NR) that had
not completed a
trauma ICU rotation
performed 73
ultrasound-guided
and subclavian
CVCs
PGY-1 and PGY-2
residents from
emergency
medicine, internal
medicine, general
surgery, anesthesia,
and obstetricsgynecology (n =
115) performed a
total 495 internal
jugular, subclavian,
and femoral vein
CVCs
PGY-2 and PGY-3
internal medicine (n
= 47), and
subsequently
emergency residents
(n = 58 total) with
prior certification in
CVC placement,
number of CVCs
performed NR
449

At a single Level I trauma teaching hospital ICU, a higher level of
comfort and ability was found in the simulator-trained group (p =
0.03). The simulator-trained group outperformed the traditionallytrained group in 7 of 10 performance variables measured,
although none of these were statistically significant. More
complications (pneumothorax, arterial puncture, inability to
complete the procedure) were found in the traditional group than
in the simulator-trained group (p = 0.07).
In the emergency department, medical ICU, and surgical ICU at a
single teaching hospital, simulation-trained residents succeeded
in cannulation with first attempt more often (51% vs. 37%, p =
0.03) and successfully inserted the catheter on their first attempt
more often (78% vs. 67%, p = 0.02). No differences seen
between groups on measures of technical errors such as
standard precautions, insertion steps, and use of sterile
technique, or on measures of mechanical complications (e.g.,
pneumothorax, transient arrhythmia, catheter tip malposition, or
arterial puncture).
At a single academic hospital, the simulator plus video-trained

group had higher median scores than the video-only group on a
survey of preparedness and confidence (p < 0 .001). After
additional simulator training with both emergency and internal
medicine residents (n = 58), the rate of CRBSI among patients in
the MICU decreased from 3.5 per 1,000 catheter days prior to the
intervention to 1.0 per 1,000 catheter-days after the intervention.
The rate of CRBSI remained steady in the surgical ICU over that
same time period - 3.4 per 1,000, where surgical residents had
received traditional training during the study period.
Practice
Didactic training on anatomy and
CVC techniques coupled with
supervised practice on fresh
human cadavers, whereby
supervisors provided immediate
individualized feedback on
performance as well as video tape
review with debriefing
Simulation
Participants
Medical students
th
trained in their 4
year, performing
CVC during PGY-1,
number of total
CVCs performed by
health system
ranged from 1,682
1,884 annually
Internal medicine
residents (n = 150,
PGY-NR) performed
54 internal jugular
and subclavian
CVCs
Study
Martin et al.
(2003)
Study Design
Pre-post
observational
cohort design,
compared with
historical cohort
Miranda et
94
al. (2007)
Prospective
controlled cohort
design
Follow didactic training, practice in

placing CVC and sterile technique
was overseen by experienced
residents and attending internists
Sherertz et
91
al. (2000)
Pre-post, nonrandomized
observational
design
One hour course on basic infection

control, followed by 1 hour of
simulation with mannequins for
practicing blood stream access
techniques (including CVC) in the
presence of senior clinical staff
3 year medical
students and PGY-1
physicians (n > 100,
actual number NR),
total CVCs
performed at system
during study period
5,099
Smith et al.
87
(2010)
Randomized
controlled trial
Didactic training followed by partial

task trainer practice of entire CVC
placement under supervision with
immediate feedback, simulation
group was then allowed to continue
simulated practice unsupervised
Internal medicine
residents PGY-1 and
PGY-2 (n = 52),
number of CVCs
performed NR
rd
450

In a two-hospital academic health system, across the emergency,
surgery, and critical care, the incidence of pneumothorax
decreased significantly (p = 0.004) after implementing the
simulation-based training program. When limiting analyses to the
initial 3-month period that residents would begin the rotation, the
incidence of pneumothorax dropped from 34.0% to 3.8% after
implementing simulation.
In the general medical service at a single academic teaching

hospital, femoral vein catheterizations decreased nonsignificantly in the simulation-trained group. The simulation group
was more likely to use masks during the procedure (risk ratio 2.2,
95% CI, 1.3-2.7, p = 0.008), but there was no difference between
groups in the proper use of other sterile techniques. There was
no difference between groups in the rate of CRBSI per 1,000
catheter-days. The simulation group demonstrated significantly
increased knowledge of complications of femoral vein
catheterizations.
Medical students and physicians perceived need for full-sized
sterile drapes increased from 22% prior to the course to 73% 6
months after the course (p < 0.001). Across six ICUs and one
step-down unit at a single academic institution, the documented
use of full-size sterile drapes increased from 44% to 65% (p <
0.001), and the CRBSI rate decreased from 4.51 per 1,000
patient days prior to the course to 2.92 per 1,000 patient days 18
months after the course. The estimated cost savings of
decreases in CRBSI is between $63,000 and $800,000.
At medical ICU of a single teaching hospital, the simulation group
demonstrated a non-significantly improved performance on a
number of structured checklist items in initial simulation tasks
(13.2, SD = 4.9 vs. 9.7, SD = 5.0, p = 0.07). No differences
observed between groups in adverse outcomes or complications
of CVC placement.
Study
Velmahos et
92
al. (2004)
Study Design
Randomized
controlled trial

Practice
Three hour CVC surgical skills
course consisting of didactic
training, and small group practice
with high-fidelity mannequin and
supervising instructor; each intern
practiced CVC insertion a minimum
of 4 times in the exercise
Simulation
Participants
Surgical interns (n =
28), number of
CVCs performed NR

In a single academic surgery department, scores on a CVC
technical competency exam were similar between groups on the
pretest; however, the simulation-trained group scored
significantly higher on the posttest (p = 0.03). The simulation
group also achieved a significantly higher score (p < 0.001) on
structured checklist evaluation of performance during CVC
placement and required fewer attempts to find the vein (p =
0.046).
Catheter-related blood stream infection (CRBSI), Intensive care unit (ICU), Not reported (NR), Post-graduate year (PGY).
451
Implementing Simulation
Gaba conceptualized a framework that captures the versatility of simulation.24 In this
framework, each of 11 dimensions assist those looking to implement simulation to define their
needs and goals: (1) the purpose and aims of simulation activity; (2) the unit of participation in
the simulationindividuals or teams; (3) the experience level of simulation participants; (4) the
health care domain in which the simulation is applied; (5) the health care disciplines of personnel
participating in the simulation; (6) the type of knowledge, skill, attitudes, or behavior addressed
in simulation; (7) the age of the patient being simulated; (8) the technology applicable or
required for simulations; (9) the site of simulationin situ clinical setting versus dedicated
simulation center; (10) the extent of direct participation; and (11) the feedback method
accompanying simulation. Gaba further delineates implementation mechanisms for health care
settings as well as a variety of professional and governmental organizations.
In addition to the 11-dimensional framework by Gaba,24 which lays out implementable
aspects of types of simulation, a number of other factors must be considered before using
simulation:95
1. People: are both trainee and training participants available and appropriately trained in
simulation techniques?
2. Time: is sufficient time dedicated to meaningful simulations not only as adjunctive
training experiences provided in the flow of regular care activities?
3. Equipment: are simulation-specific materials (e.g., mannequin) and actual medical
equipment or devices available to recreate a desirable realism in simulated environments?
4. Space: is adequate space available for dedicating to simulated environments? Even in-situ
simulations will require storage and preparation of materials.
5. Supplies: if the simulation requires the use of real medical supplies, are these available?
6. Technical Support: implementation phases, especially of high-tech or complex
simulators, may require support, and there may be upkeep and maintenance to simulation
equipment.
People and time are likely to be the most expensive aspects of simulation in the long run.
However, startup costs vary substantially with the complexity of the simulation, which also
depends on the purpose. Large simulation centers may find financial support in philanthropic
sources or may be subsidized by participants and other organizational entities. Other expenses
will include the upkeep of simulation equipment and space dedicated to simulation.
Rosen and colleagues discussed the importance of considering components most likely to
enhance success in simulation techniques. For example, these authors differentiate the
importance of cognitive fidelity in a simulated exercise from physical fidelity. That is,
simulations that engage the participant in ways that, cognitively, most reflect the actual task are
likely to be more effective.25 Debriefing is considered crucial when implementing simulation,25,26
a recommended best practice according to one review of simulation-based medical education.96
Other national efforts have delineated future directions for simulation research.97 At an
international level, researchers have identified three primary foci to further understanding of best
practices in simulation: instructional design, outcomes measurement, and translational
activities.98
452

The cost of implementing simulation exercises varies from low to high depending on type of
exercise and the personnel and equipment resources involved. In addition, start-up costs for a
comprehensive simulation center may be accounted for differently than ongoing costs for
exercises, which complicates the ability to categorize the expected cost for simulation as a
patient safety practice. Future and ongoing investigations will give us further insight into the
cost-effectiveness of simulation exercises in general, and simulation that targets improvement in
patient safety. Unfortunately, research addressing cost savings attributable directly to simulation
remains sparse, but research has reported up to a 7-to-1 return on simulation costs through the
reduction in hospital days for blood stream infection.91,99

This review did not find published data about the effect of context on the effectiveness of
simulation exercises to improve patient safety. Therefore, this is an area for future research.

Simulation has continued to garner momentum in patient safety efforts in the past decade.
Various professional organizations have endorsed the use of simulation through accreditation
standards, and government agencies have continued to fund investigations into the use of
simulation for enhancing care. On a basic level, simulation allows for exercising and improving
aspects of health care without risk to patients. Simulation has been utilized in patient safety for
the purposes of education, assessment, research, and integrating system-level practices. These
efforts have been reported in the literature as research about simulation, that is, research that can
help elucidate and understand leverage points in enhancing patient safety. In the past decade,
researchers have also begun focusing on using simulation, which has increased our
understanding of benefits (especially with respect to patient outcomes) likely to be realized in the
translation of simulation techniques into practice.
The review found that studies reporting outcomes with actual patients, or systems of care,
have occurred primarily in academic settings, although researchers have reported on the use of
simulation for a variety of clinical specialties, experience levels, and care settings. These studies
varied in terms of individual quality, but the bulk of studies were randomized or
methodologically controlled designs. Researchers have replicated standardized simulation
training for CVC placement, and although promising for patient safety in CVC, we did not find
other examples of replication studies in our review. We did not find analysis of contextual effects
(e.g., MICU analyzed against SICU) on the validity of simulation to improve patient safety. Thus
the transferability of simulation techniques to increase patient safety similarly to that reported in
this review remains unknown. It is likely that these results will generalize to other settings, but
generalizability of any one technique is likely to vary depending on a number of factors such as
those in Gabas 11-dimensional framework24 and adequacy of resources dedicated to simulation
(e.g., debriefing).
At this juncture, simulation appears to have a favorable impact on quicker acquisition and
improved performance of technical skills. Although not yet thoroughly studied, simulating
complex or high-stakes procedures appears to be a promising technique to increase patient-safe
behavior at the clinician- and team-levels. Simulation has the potential to enhance patient safety
through structured assessment and debriefing in quality improvement initiatives. It has been used
453
to assess practices that would be difficult or unsafe to study empirically in real-time with actual
patients. Likewise, simulation has been endorsed for ongoing competency and continuing
education, as well as advancing to mastery-level practices.
Previous systematic reviews have reported that simulation contributes to enhanced
knowledge acquisition and improved clinical performance.29 Simulation techniques have been
utilized in translating results from within simulation lab to patient and health care system-level
outcomes.100 Protecting time for debriefing in a learning experience has been suggested as a
crucial component to simulation techniques in a systematic review.101 This review is the first to
examine effects that simulation exercises exert on patient safety outcomes, and in particular,
outcomes with patients outside of simulation laboratory settings (i.e., during clinical care).
Reviewing the literature on simulation approaches broadly resulted in several limitations.
First, it is possible broad search strategies missed studies that may otherwise be captured with
targeted and comprehensive strategies dedicated to each simulation technique, clinical specialty,
or application. Second, given relative infancy of the research into simulation exercises, the field
may be prone to selective reporting of studies with positive findings, leading to potential
publication bias in our review. Third, we limited our assessment of quality of evidence to study
design and did not perform structured assessment of the strength of evidence. Therefore, overall
strength of the evidence for simulation exercises to improve patient safety should be interpreted
with caution based on the current review.
Simulation is a versatile technique that continues to garner momentum in a variety of clinical
settings and applications, including patient safety strategies. Although evidence is largely
heterogeneous at this time, our review suggests potential for simulation exercises to contribute to
patient safety through increased technical and procedural performance, and through improved
team performance. Limited research using health system-level observations suggests that
simulation may enhance patient safety, although more research is needed on the potential for
simulation to contribute to system-level differences in patient safety outcomes. Systematic
reviews of simulation for specific procedures have begun reporting patient safety outcomes,57,102
and more reviews of this nature would enhance our understanding of the overall contribution of
simulation techniques to patient safety. Future systematic reviews would benefit from
investigators using a consistent framework to describe the intervention, its context and its
implementation, such as Gabas framework.
Simulation implies a spirit of innovation in a world with quickly evolving technologies. As a
specialty field, it may be in an end of the beginning phase, and some of the benefits of
simulation may not be realized from a short-term perspective.24 As a developing field, certain
aspects may remain under discussion, such as standardized definitions of validity in simulated
environments.103 However, the breadth of applications and even purposes has not yet been fully
realized.24 As only one further example of simulations potential role in making care better and
safer, a recent systematic review of simulation and video games for patients post-stroke reported
increased upper extremity functionality.104 A summary table is located in Table 2.
454

Scope of the
Strength of
the PSP
Effectiveness
Evidence or
Potential for
Harmful
Unintended
Consequences
Uncertain
Estimate of
Cost
How Much do We
Common/Moderate-to- Moderate to
Moderate
Moderate
high
high for
specific topics
*
Cost varies based on the simulation technique and resources involved. For example, exercises involving fully simulated
operating room environments may have higher costs to implement relative to those that require a simple mannequin or patient
actors.
References
1.
Chang Y, Mark B. Effects of learning

climate and registered nurse staffing on
medication errors. Nurs Res. 2011;60(1):329.21127452.
2.
Rogers SO, Jr., Gawande AA, Kwaan M, et

al. Analysis of surgical errors in closed
malpractice claims at 4 liability insurers.
Surgery. 2006;140(1):25-33.16857439.
3.
4.
5.
6.
7.
8.
Meltzer D, Manning WG, Morrison J, et al.

Effects of physician experience on costs and
outcomes on an academic general medicine
service: results of a trial of hospitalists. Ann
Intern Med. 2002;137(11):86674.12458986.
Singh H, Thomas EJ, Petersen LA, et al.
Medical errors involving trainees: a study of
closed malpractice claims from 5 insurers.
Arch Intern Med. 2007;167(19):20306.17954795.
Cooke M, Irby DM, Sullivan W, et al.
American medical education 100 years after
the Flexner report. N Engl J Med.
2006;355(13):1339-44.17005951.
Ziv A, Wolpe PR, Small SD, et al.
Simulation-based medical education: an
ethical imperative. Acad Med.
2003;78(8):783-8.12915366.
Ericsson KA. Deliberate practice and
acquisition of expert performance: a general
overview. Acad Emerg Med.
2008;15(11):988-94.18778378.
Ericsson KA. Deliberate practice and the
acquisition and maintenance of expert
performance in medicine and related
domains. Acad Med. 2004;79(10
Suppl):S70-81.15383395.
455
9.
Regenbogen SE, Greenberg CC, Studdert

DM, et al. Patterns of technical error among
surgical malpractice claims: an analysis of
strategies to prevent injury to surgical
patients. Ann Surg. 2007;246(5):70511.17968158.
10.
Semeraro F, Signore L, Cerchiari EL.

Retention of CPR performance in
anaesthetists. Resuscitation. 2006;68(1):1018.16325986.
11.
OSteen DS, Kee CC, Minick MP. The

retention of advanced cardiac life support
knowledge among registered nurses. J Nurs
Staff Dev. 1996;12(2):66-72.8715621.
12.
Park CS. Simulation and quality

improvement in anesthesiology. Anesthesiol
Clin. 2011;29(1):13-28.21295750.
13.
Gaba DM, Fish KJ, Howard SK. Crisis

management in anesthesiology. New York:
Churchill Livingstone 1994.
14.
Kohn LT, Corrigan J, Donaldson MS. To err

is human : building a safer health system.
Washington, D.C.: National Academy Press
2000.
15.
Institute of Medicine (U.S.). Committee on

Quality of Health Care in America. Crossing
the quality chasm : a new health system for
the 21st century. Washington, D.C.:
National Academy Press 2001.
16.
Improving Patient Safety Through

Simulation Research. June 2008. Agency for
MD. www.ahrq.gov/qual/simulproj.htm
17.
Improving Patient Safety Through

Simulation Research: Funded Projects.
AHRQ Publication No. 11-P012-EF,
September 2011. Agency for Healthcare
Research and Quality, Rockville, MD.
www.ahrq.gov/qual/simulproj11.htm
29.
McGaghie WC, Issenberg SB, Cohen ER, et

al. Does simulation-based medical education
with deliberate practice yield better results
than traditional clinical education? A metaanalytic comparative review of the evidence.
Acad Med. 2011;86(6):706-11.21512370.
18.
Johnson KA, Sachdeva AK, Pellegrini CA.

The critical role of accreditation in
establishing the ACS Education Institutes to
advance patient safety through simulation. J
Gastrointest Surg. 2008;12(2):2079.17972142.
30.
Cook DA, Erwin PJ, Triola MM.

Computerized virtual patients in health
professions education: a systematic review
and meta-analysis. Acad Med.
2010;85(10):1589-602.20703150.
31.
19.
Nehring WM. U.S. boards of nursing and

the use of high-fidelity patient simulators in
nursing education. J Prof Nurs.
2008;24(2):109-17.18358446.
Cook DA, Hatala R, Brydges R, et al.

Technology-enhanced simulation for health
professions education: a systematic review
and meta-analysis. JAMA. 2011;306(9):97888.21900138.
20.
Steadman RH. The American Society of

Anesthesiologists national endorsement
program for simulation centers. J Crit Care.
2008;23(2):203-6.18538213.
32.
21.
Buck GH. Development of simulators in

medical education. Gesnerus. 1991;48 Pt
1:7-28.1855669.
Kobayashi L, Shapiro MJ, Suner S, et al.

Disaster medicine: the potential role of high
fidelity medical simulation for mass casualty
incident training. Med Health R I.
2003;86(7):196-200.12973894.
33.
Mishra A, Catchpole K, Dale T, et al. The

influence of non-technical performance on
technical outcome in laparoscopic
cholecystectomy. Surg Endosc.
2008;22(1):68-73.17479324.
34.
Gaba DM, Howard SK, Fish KJ, et al.

Simulation-based training in Anesthesia
Crisis Resource Management (ACRM): A
decade of experience. Simul Gaming.
2001;32:175-93.
35.
Blum RH, Raemer DB, Carroll JS, et al. A

method for measuring the effectiveness of
simulation-based team training for
improving communication skills. Anesth
Analg. 2005;100(5):1375-80, table of
contents.15845689.
36.
Pian-Smith MC, Simon R, Minehart RD, et

al. Teaching residents the two-challenge
rule: a simulation-based approach to
improve education and patient safety. Simul
Healthc. 2009;4(2):84-91.19444045.
37.
Morgan PJ, Tarshis J, LeBlanc V, et al.

Efficacy of high-fidelity simulation
debriefing on the performance of practicing
anaesthetists in simulated scenarios. Br J
Anaesth. 2009;103(4):531-7.19703938.
22.
Nehring WM, Lashley FR. Nursing

simulation: A review of the past 40 years.
Simul Gaming. 2009;40:528-52.
23.
Abrahamson S, Denson JS, Wolf RM.

Effectiveness of a simulator in training
anesthesiology residents. J Med Educ.
1969;44(6):515-9.5789592.
24.
Gaba DM. The future vision of simulation in

healthcare. Simul Healthc. 2007;2(2):12635.19088617.
25.
Rosen MA, Salas E, Tannenbaum SI, et al.

Simulation-based training for teams in
health care: Designing scenarios, measuring
performance, and providing feedback. In:
Carayon P, ed. Handbook of HFE in health
care and patient safety, 2nd Ed. Boca Raton,
FL: CRC Press 2011:571-92.
26.
Fanning RM, Gaba DM. The role of

debriefing in simulation-based learning.
Simul Healthc. 2007;2(2):115-25.19088616.
27.
Rudolph JW, Simon R, Rivard P, et al.

Debriefing with good judgment: combining
rigorous feedback with genuine inquiry.
Anesthesiol Clin. 2007;25(2):36176.17574196.
28.
Gaba DM. The futures here. We are it.

Simul Healthc. 2006;1:1-2.
456
38.
39.
40.
41.
42.
Dalley P, Robinson B, Weller J, et al. The

use of high-fidelity human patient
simulation and the introduction of new
anesthesia delivery systems. Anesth Analg.
2004;99(6):1737-41, table of
contents.15562063.
Coopmans VC, Biddle C. CRNA
performance using a handheld,
computerized, decision-making aid during
critical events in a simulated environment: a
methodologic inquiry. AANA J.
2008;76(1):29-35.18323317.
Lipman S, Daniels K, Cohen SE, et al.
Labor room setting compared with the
operating room for simulated perimortem
cesarean delivery: a randomized controlled
trial. Obstet Gynecol. 2011;118(5):10904.22015877.
Wayne DB, Didwania A, Feinglass J, et al.
Simulation-based education improves
quality of care during cardiac arrest team
responses at an academic teaching hospital:
a case-control study. Chest. 2008;133(1):5661.17573509.
Wayne DB, Butter J, Siddall VJ, et al.
Mastery learning of advanced cardiac life
support skills by internal medicine residents
using simulation technology and deliberate
practice. J Gen Intern Med. 2006;21(3):2516.16637824.
43.
Wayne DB, Siddall VJ, Butter J, et al. A

longitudinal study of internal medicine
residents retention of advanced cardiac life
support skills. Acad Med. 2006;81(10
Suppl):S9-S12.17001145
44.
Daniels K, Lipman S, Harney K, et al. Use

of simulation based team training for
obstetric crises in resident education. Simul
Healthc. 2008;3(3):154-60.19088659.
45.
Lighthall GK, Poon T, Harrison TK. Using

in situ simulation to improve in-hospital
cardiopulmonary resuscitation. Jt Comm J
Qual Patient Saf. 2010;36(5):20916.20480753.
46.
Marshall S, Harrison J, Flanagan B. The

teaching of a structured tool improves the
clarity and content of interprofessional
clinical communication. Qual Saf Health
Care. 2009;18(2):137-40.19342529.
457
47.
McGaghie WC. Medical education research

as translational science. Sci Transl Med.
2010;2(19):19cm8.20371485.
48.
Cohen J, Cohen SA, Vora KC, et al.

Multicenter, randomized, controlled trial of
virtual-reality simulator training in
acquisition of competency in colonoscopy.
Gastrointest Endosc. 2006;64(3):3618.16923483.
49.
Sedlack RE, Kolars JC. Computer simulator

training enhances the competency of
gastroenterology fellows at colonoscopy:
results of a pilot study. Am J Gastroenterol.
2004;99(1):33-7.14687137.
50.
Ahlberg G, Hultcrantz R, Jaramillo E, et al.

Virtual reality colonoscopy simulation: a
compulsory practice for the future
colonoscopist? Endoscopy.
2005;37(12):1198-204.16329017.
51.
Park J, MacRae H, Musselman LJ, et al.

Randomized controlled trial of virtual reality
simulator training: transfer to live patients.
Am J Surg. 2007;194(2):205-11.17618805.
52.
Thomson M, Heuschkel R, Donaldson N, et

al. Acquisition of competence in paediatric
ileocolonoscopy with virtual endoscopy
training. J Pediatr Gastroenterol Nutr.
2006;43(5):699-701.17130753.
53.
Ferlitsch A, Schoefl R, Puespoek A, et al.

Effect of virtual endoscopy simulator
training on performance of upper
gastrointestinal endoscopy in patients: a
randomized controlled trial. Endoscopy.
2010;42(12):1049-56.20972956.
54.
Blum MG, Powers TW, Sundaresan S.

Bronchoscopy simulator effectively prepares
junior residents to competently perform
basic clinical bronchoscopy. Ann Thorac
Surg. 2004;78(1):287-91; discussion 91.15223446.
55.
Duncan DR, Morgenthaler TI, Ryu JH, et al.

Reducing iatrogenic risk in thoracentesis:
establishing best practice via experiential
training in a zero-risk environment. Chest.
2009;135(5):1315-20.19017865.
56.
Tongprasert F, Wanapirak C, Sirichotiyakul

S, et al. Training in cordocentesis: the first
50 case experience with and without a
cordocentesis training model. Prenat Diagn.
2010;30(5):467-70.20440735.
57.
Gurusamy KS, Aggarwal R, Palanivelu L, et

al. Virtual reality training for surgical
trainees in laparoscopic surgery. Cochrane
Database Syst Rev.
2009;(1):CD006575.19160288.
58.
Ahlberg G, Enochsson L, Gallagher AG, et

al. Proficiency-based virtual reality training
significantly reduces the error rate for
residents during their first 10 laparoscopic
cholecystectomies. Am J Surg.
2007;193(6):797-804.17512301.
59.
60.
61.
Calatayud D, Arora S, Aggarwal R, et al.

Warm-up in a virtual reality environment
improves performance in the operating
room. Ann Surg. 2010;251(6):11815.20485133.
Hamilton EC, Scott DJ, Kapoor A, et al.
Improving operative performance using a
laparoscopic hernia simulator. Am J Surg.
2001;182(6):725-8.11839347.
Kallstrom R, Hjertberg H, Svanvik J. Impact
of virtual reality-simulated training on
urology residents performance of
transurethral resection of the prostate. J
Endourol. 2010;24(9):1521-8.20677993.
62.
Gaies MG, Morris SA, Hafler JP, et al.

Reforming procedural skills training for
pediatric residents: a randomized,
interventional trial. Pediatrics.
2009;124(2):610-9.19651582.
63.
Sears K, Goldsworthy S, Goodman WM.

The relationship between simulation in
nursing education and medication safety. J
Nurs Educ. 2010;49(1):52-5.19810664.
64.
Hall RE, Plant JR, Bands CJ, et al. Human

patient simulation is effective for teaching
paramedic students endotracheal intubation.
Acad Emerg Med. 2005;12(9):8505.16141019.
65.
Draycott T, Sibanda T, Owen L, et al. Does

training in obstetric emergencies improve
neonatal outcome? BJOG. 2006;113(2):17782.16411995.
66.
OConnor PJ, Sperl-Hillen JM, Johnson PE,

et al. Simulated physician learning
intervention to improve safety and quality of
diabetes care: a randomized trial. Diabetes
Care. 2009;32(4):585-90.19171723.
458
67.
Bruppacher HR, Alam SK, LeBlanc VR, et

al. Simulation-based training improves
physicians performance in patient care in
high-stakes clinical setting of cardiac
surgery. Anesthesiology. 2010;112(4):98592.20234305.
68.
Weidman EK, Bell G, Walsh D, et al.

Assessing the impact of immersive
simulation on clinical performance during
actual in-hospital cardiac arrest with CPRsensing technology: A randomized
feasibility study. Resuscitation.
2010;81(11):1556-61.20724057.
69.
Capella J, Smith S, Philp A, et al. Teamwork

training improves the clinical care of trauma
patients. J Surg Educ. 2010;67(6):43943.21156305.
70.
Berenholtz SM, Pronovost PJ, Lipsett PA, et

al. Eliminating catheter-related bloodstream
infections in the intensive care unit. Crit
Care Med. 2004;32(10):2014-20.15483409.
71.
Mermel LA. Prevention of intravascular

catheter-related infections. Ann Intern Med.
2000;132(5):391-402.10691590.
72.
Pittet D, Tarara D, Wenzel RP. Nosocomial

bloodstream infection in critically ill
patients. Excess length of stay, extra costs,
and attributable mortality. JAMA.
1994;271(20):1598-601.8182812.
73.
Barsuk JH, McGaghie WC, Cohen ER, et al.

Use of simulation-based mastery learning to
improve the quality of central venous
catheter placement in a medical intensive
care unit. J Hosp Med. 2009;4(7):397403.19753568.
74.
National Quality Forum (NQF). Safe

practices for better healthcare - 2010 update:
A consensus report. Washington, D.C.:
NQF. 2010.
75.

Quality. Guide to the Patient Safety
Indicators v4.3. Rockville, MD. 2011.
76.
Barsuk JH, Cohen ER, Feinglass J, et al. Use

of simulation-based education to reduce
catheter-related bloodstream infections.
Arch Intern Med. 2009;169(15):14203.19667306.
77.
American Board of Internal Medicine.

Policies and procedures required for internal
medicine.
www.abim.org/certification/policies/imss/im
.aspx. Accessed December 15, 2011
78.
American Board of Emergency Medicine.

Model of the clinical practice of emergency
medicine.
www.abem.org/PUBLIC/portal/alias__Rain
bow/lang__enUS/tabID__4223/DesktopDefault.aspx.
Accessed December 15, 2011. 2009.
79.
Huang GC, Smith CC, Gordon CE, et al.

Beyond the comfort zone: residents assess
their comfort performing inpatient medical
procedures. Am J Med. 2006;119(1):71 e1724.16431194.
80.
Seymour NE, Gallagher AG, Roman SA, et

al. Virtual reality training improves
operating room performance: results of a
randomized, double-blinded study. Ann
Surg. 2002;236(4):458-63; discussion 634.12368674.
81.
82.
83.
84.
85.
Ma IW, Brindle ME, Ronksley PE, et al.

Use of simulation-based education to
improve outcomes of central venous
catheterization: a systematic review and
meta-analysis. Acad Med. 2011;86(9):113747.21785310.
Macnab AJ, Macnab M. Teaching pediatric
procedures: the Vancouver model for
instructing Seldingers technique of central
venous access via the femoral vein.
Pediatrics. 1999;103(1):E8.9917488.
Barsuk JH, Ahya SN, Cohen ER, et al.
Mastery learning of temporary hemodialysis
catheter insertion by nephrology fellows
using simulation technology and deliberate
practice. Am J Kidney Dis. 2009;54(1):706.19376620.
Barsuk JH, Cohen ER, Feinglass J, et al.
Unexpected Collateral Effects of
Simulation-Based Medical Education. Acad
Med. 2011.22030762.
Evans LV, Dodge KL, Shah TD, et al.
Simulation training in central venous
catheter insertion: improved performance in
clinical practice. Acad Med.
2010;85(9):1462-9.20736674.
459
86.
Griswold-Theodorson S, Hannan H, Handly

N, et al. Improving patient safety with
ultrasonography guidance during internal
jugular central venous catheter placement by
novice practitioners. Simul Healthc.
2009;4(4):212-6.21330794.
87.
Smith CC, Huang GC, Newman LR, et al.

Simulation training and its effect on longterm resident performance in central venous
catheterization. Simul Healthc.
2010;5(3):146-51.20651476.
88.
Barsuk JH, McGaghie WC, Cohen ER, et al.

Simulation-based mastery learning reduces
complications during central venous catheter
insertion in a medical intensive care unit.
Crit Care Med. 2009;37(10):2697701.19885989.
89.
Britt RC, Novosel TJ, Britt LD, et al. The

impact of central line simulation before the
ICU experience. Am J Surg.
2009;197(4):533-6.19249739.
90.
Khouli H, Jahnes K, Shapiro J, et al.

Performance of medical residents in sterile
techniques during central vein
catheterization: randomized trial of efficacy
of simulation-based training. Chest.
2011;139(1):80-7.20705795.
91.
Sherertz RJ, Ely EW, Westbrook DM, et al.

Education of physicians-in-training can
decrease the risk for vascular catheter
infection. Ann Intern Med.
2000;132(8):641-8.10766683.
92.
Velmahos GC, Toutouzas KG, Sillin LF, et

al. Cognitive task analysis for teaching
technical skills in an inanimate surgical
skills laboratory. Am J Surg.
2004;187(1):114-9.14706600
93.
Andreatta P, Chen Y, Marsh M, et al.

Simulation-based training improves applied
clinical placement of ultrasound-guided
PICCs. Support Care Cancer.
2011;19(4):539-43.20306091.
94.
Miranda JA, Trick WE, Evans AT, et al.

Firm-based trial to improve central venous
catheter insertion practices. J Hosp Med.
2007;2(3):135-42.17549773.
95.
David M. Gaba. Personal communication.

December 15, 2011.
96.
McGaghie WC, Issenberg SB, Petrusa ER,

et al. A critical review of simulation-based
medical education research: 2003-2009.
Med Educ. 2010;44(1):50-63.20078756.
97.
Van Hook JE. Building a national agenda

for simulation-based medical education.
U.S. Army Medical Research and Materiel
Command (Contract No. W81XWH-04-10583). Washington, D.C. 2004.
98.
99.
100.
Issenberg SB, Ringsted C, Ostergaard D, et

al. Setting a research agenda for simulationbased healthcare education: a synthesis of
the outcome from an Utstein style meeting.
Simul Healthc. 2011;6(3):155-67.21642804.
Cohen ER, Feinglass J, Barsuk JH, et al.
Cost savings from reduced catheter-related
bloodstream infection after simulation-based
education for residents in a medical
intensive care unit. Simul Healthc.
2010;5(2):98-102.20389233.
McGaghie WC, Draycott TJ, Dunn WF, et
al. Evaluating the impact of simulation on
translational patient outcomes. Simul
Healthc. 2011;6 Suppl:S42-7.21705966.
460
101.
McGaghie WC, Issenberg SB, Petrusa ER,

et al. Effect of practice on standardised
learning outcomes in simulation-based
medical education. Med Educ.
2006;40(8):792-7.16869926.
102.
Walsh CM, Sherlock ME, Ling SC, et al.

Virtual reality simulation training for health
professions trainees in gastrointestinal
endoscopy. Cochrane Database Syst Rev.
2012;6:CD008237.22696375.
103.
Schout BM, Hendrikx AJ, Scheele F, et al.

Validation and implementation of surgical
simulators: a critical review of present, past,
and future. Surg Endosc. 2010;24(3):53646.19633886.
104.
Laver KE, George S, Thomas S, et al.

Virtual reality for stroke rehabilitation.
2011;9:CD008349.21901720.
Chapter 39. Obtaining Informed Consent From Patients: Brief

Update Review
Kristina M. Cordasco, M.D., M.P.H., M.S.H.S.
Introduction
In health care, informed consent refers to the process whereby the patient and the health care
practitioner engage in a dialogue about a proposed medical treatments nature, consequences,
harms, benefits, risks, and alternatives.1 Informed consent is a fundamental principle of health
care.
The process of informed consent can be considered a patient safety issue from several
perspectives. At the extreme, performing a procedure on a patient without his or her consent has
been considered by the courts to be a form of battery.2 Informed consent may also be indirectly
related to patient safety in that, when done well, it opens a dialogue between the patient and
provider so that the patient can ask questions, knows what to expect during and after procedure,
and can at least theoretically help to avert medical errors.3
In general, studies have shown improved patient outcomes with effective physician-patient
communication and increased patient empowerment.4,5 Patient education has also been
associated with preventing medical errors.6
A review on this topic conducted in 2001 found few studies linking informed consent with
health outcomes and few studies on the impact of procedures used to obtain informed consent on
the quality of consent obtained; studies suggested that the value of informed consent might be
modestly enhanced by augmenting standard patient provider discussions with additional learning
and retention aids and that the process of consent can be modestly enhanced by using structured
interviews and asking patients to recall and restate the key elements of the discussion.
This update review focuses on what we have learned about the informed consent process and
the effectiveness of interventions that have been implemented to try to improve it. We conducted
a search of the health care and health services literature for the time interval 2001 to present and
reviewed all studies relevant to informed consent in the clinical setting.
What Is Informed Consent?

The document a patient signs to verify that he has engaged in a dialog with a health care
practitioner about a proposed medical treatment is commonly referred to as an informed
consent. However, it is the dialog itself that constitutes the actual informed consent process.3
Informed consent is used in both clinical and research settings; this review focuses primarily on
informed consent in the clinical setting.
Although no evidence currently links informed consent with improved adherence to
medication or other self-care procedures, to prevention of medical errors, or to improved overall
health outcomes, some evidence links increased patient-physician communication with more
realistic expectations, increased patient satisfaction, and fewer medical malpractice claims.7-9
461
How Has Informed Consent Been Implemented?

A complete informed consent process consists of seven elements: (1) Discussing the patients
role in the decision-making process; (2) Describing the clinical issue and suggested treatment;
(3) Discussing alternatives to the suggested treatment (including the option of no treatment);
(4) Discussing risks and benefits of the suggested treatment (and comparing them to the risks and
benefits of alternatives); (5) Discussing related uncertainties; (6) Assessing the patients
understanding of the information provided; and (7) Eliciting the patients preference (and thereby
consent).10 Not every detail needs to be discussed, but all details needed for a reasonable
person to make a decision must be provided.11 Therefore, all risks of serious complications,
even if they occur very rarely, need to be discussed. Less serious risks need to be discussed if
they occur more commonly.11 This process of informed consent may occur within one encounter,
or across multiple encounters.12
Although informed consent is often used prior to invasive procedures, designated radiologic
examinations, and other high-risk medical treatments (e.g., chemotherapy), the process of
informed consent, or informed decision-making, is applicable to all medical care decisions where
one or more alternatives exist (including the alternative of no treatment or procedure).13
Recently, there has also been increased attention to the importance of informed consent in
screening procedures and genetic testing.14,15 As such, the informed consent process has
considerable overlap with the principles of shared decisionmaking.13
What Have We Learned About Informed Consent?

Most Informed Consent Procedures Are Incomplete
Various studies have examined the completeness of informed consent procedures in various
settings and scenarios. In an examination of the informed decisionmaking process in 1057 audiorecorded outpatient encounters in the offices of primary care physicians and surgeons, regarding
mostly low-risk decisions, only 9% were deemed to contain all the elements of complete
informed decision-making.16 The most common element missing was an explicit assessment of
patient understanding. However, risks and benefits, and their associated uncertainties were also
commonly not included in the discussions. Among 141 discussions regarding orthopedic surgical
interventions, in no case were all elements fully discussed.17 Ninety-two percent had some
mention of the nature of the decision, 62% listed alternatives, 59% discussed pros and cons, 14%
discussed the patients role in the decisionmaking, and 12% of the time the patients
understanding was assessed. In an analysis of informed decision-making for 145 patients
considering high-risk elective major vascular surgery, audio-recorded discussions across multiple
visits for each patient contained all informed consent elements in 45% of the cases. 18 In 23% of
the cases, the surgeon failed to discuss one or more of the basic elements of consent: clarifying
the patients role in the decision-making; explaining the clinical condition; or eliciting the
patients preferences for treatment.
Reading Level of Informed Consent Documents Is Often Too High

A number of studies have examined the reading level of informed consent documents and
their utility for people with limited English proficiency. In a survey of informed consent forms
for iodinated contrast material from 160 academic and private United States (U.S.) hospitals,
average reading level exceeded 12th grade and only 5% had an 8th grade reading level or
below.19 Similarly, in a survey of surgery and other procedure informed consent forms from 616
462
U.S. hospitals, the mean reading level was 12.6 years and only 7% of the forms had an 8th grade
reading level or below.20 Regarding the content of informed consent documents, a separate
survey of a random sample of consent forms from 157 U.S. hospitals showed significant
variability in content and 74% omitted the nature of the procedure, risks, benefits, or
alternatives.21 Perhaps related to the readability of these documents, studies have shown patients
often do not read the consent forms provided to them and, in one study, patients who reported
reading the consent forms given to them were no better informed than those who did not. 22,23
Patients with limited English proficiency (LEP) are at particularly high risk for receiving
inadequate informed consent. In a study of 30 Latina women who were offered amniocentesis at
8 prenatal clinics without trained interpreters, the informed consent process contained all, or
nearly all, of the essential informed consent elements for 9% of the LEP compared with 68% of
the non-LEP women.24 When charts of 74 LEP Spanish and Chinese-speaking patients were
compared with those of 74 English-speakers, all of whom underwent thoracentesis, paracentesis,
or lumbar puncture at a teaching hospital where trained interpreters in Spanish and Chinese were
available, 28% of LEP patients had informed consent documented compared with 53% of
English speakers.25
Most Patients Are Unable To Recallor Dont UnderstandContent of

Informed Consent Documents
Multiple studies have shown that most patients are unable to recall or do not understand most
of the information that is presented to them in the informed consent process. Post-operative
interviews with patients 1 to 8 weeks after they underwent head and neck surgery revealed that,
on average, they could recall 48% of the main three or four complications (depending on the
surgery) they were counseled about pre-operatively.26 Interviews with 17 surrogates who
provided consent for surgery in pediatric patients, showed that 2 to 4 weeks after the surgery,
only three (18%) could recall any specifics of the procedure.27 Interviews 3 hours after consent in
100 patients scheduled for transurethral prostatectomy revealed that less than 50% of the patients
could accurately recall the risks of potential complications.28 Sixty five percent of 104 patients
consented for neurosurgery could remember no more than two of six major risks associated with
their surgery 2 hours after informed consent was obtained.29 Among 633 patients who were
offered coronary artery bypass grafting (CABG) or percutaneous coronary interventions (PCI),
there was very low concordance between what physicians reported telling the patients about
expected symptom benefits and what the patients reported as their expectations, and there was no
correlation between what physicians and patients expected regarding potential mortality benefit
(with patients believing there would be a survival benefit even when physicians reported telling
them there was not). 30
Lower levels of education are consistently associated with being less likely to recall
information in the informed consent process. Among 54 patients who underwent head and neck
surgery, 72% of those having a university education recalled more than 50% of the
complications, compared with 36% of those without a university education (p=0.04).26 In another
study of 200 patients with cancer, those who had completed high school had 35% higher scores
on tests asking them to recall, within 1 day of undergoing informed consent, written and oral
information provided to them in the informed consent process (p<0.001).22
Older age is also associated with being less likely to recall informed consent information.
Among 265 patients undergoing intrathoracic, intraperitoneal, and vascular surgery procedures,
patients over 60 years of age had less knowledge about their planned procedure immediately
463
after the informed consent process (median score on a knowledge test one point out six less at
both time periods, p<0.001).23 Among 54 patients who underwent head and neck surgery,
patients who recalled more than 50% of the complications they were told were, on average 7.6
years younger than those who recalled less than 50%.26 However, the association between older
age and less informed consent recall may be related to a lower average educational attainment
among older people. In a study of 200 patients with cancer who underwent informed consent for
radiation therapy, chemotherapy, or surgery, recall by age did not vary when adjusted for
educational attainment.22
Limited health literacy is also likely associated with less comprehension of informed consent.
Health literacy is the capacity to obtain, process, and understand basic health information and
services needed to make appropriate health decisions. In the 2003 National Assessment of Adult
Literacy (NAAL) 36% of Americans had basic or below basic health-related literacy.31 Older
age, lower educational attainment, and being African-American or Latino are all associated with
lower health literacy levels.32 In a study of men using a CD-ROM shared decisionmaking tool
about prostate cancer treatment options, lower levels of health literacy were associated with
lower prostate cancer knowledge after using the tool (Pearson correlation =0.65, p<0.001).33 In a
study of consent documents for a research study on chemotherapy agents, patients reading at or
below an 8th grade level had, on average, 28 percentage point lower comprehension scores
compared with participants with higher reading levels, even when the consent form was modified
to a 7th grade reading level.34 Similarly, in another study of a research consent form, modified to
a 6th grade reading level, patients with lower literacy were significantly less likely to respond
correctly to comprehension questions asked after a first reading of the consent form, adjusting
for other sociodemographic factors.35
Studies have also shown that minority race or ethnicity may be an independent risk factor for
having lower levels of comprehension in the informed consent process. Among 396 patients
being consented for surgery, African-American patients scored an adjusted average 9 percentage
points lower than white patients in a comprehension test administered immediately after the
consent process. This association was independent of education, age, and health literacy score.36
In the study of a research consent form cited above, in addition to those with lower health
literacy, patients who were Black or Asian/Pacific Islander were also less likely, when adjusting
for other factors, to respond correctly to comprehension questions.35
Other factors associated with lower informed consent recall are lower intelligence levels and
having cognitive dysfunction.23 Patients with cognitive dysfunction are particularly vulnerable in
the informed consent process. Cognitive dysfunction may be a long-term state (e.g., dementia) or
transient (e.g., after a medical procedure.) In one study of 302 acutely-ill medical inpatients, 48%
were estimated to have cognitive dysfunction such that they potentially lacked capacity to give
informed consent.37 However, not all patients with cognitive dysfunction lack informed consent
capacity: A structured assessment must be done to determine competence.38,39 If a patient does
not have capacity and his or her cognitive dysfunction is not expected to improve (or a decision
needs to be made prior to it improving), a surrogate decisionmaker must be established, except in
an emergency situation where the physician can determine the choice a reasonable person
would make.38
What Methods Have Been Used To Improve Informed Consent?

Multiple potential methods have been proposed for improving the informed consent process.
These methods include simplifying informed consent forms; providing supplemental written
464
materials; using decision aids; using video educational tools; using interactive computer-based
educational tools; having structured discussions; and using repeat back methods.
Simplifying Informed Consent Forms May Improve Satisfaction, if Not

Comprehension
Although the effects of simplifying informed consent forms in clinical settings have not been
studied, some studies have assessed simplified forms for consenting participants for research,
with mixed results. In a randomized-controlled study comparing a standard pharmaceutical
industry consent form to a simplified form, participants who read the modified form had a 23%
higher score on a 13-item multiple choice test about the study details (p<0.001).40 In another
randomized-control trial of 456 parents, comprehension was compared between parents
receiving a standard research consent form, written at an 11th grade level, and those who
received a consent form modified to be at an eighth grade reading level.41 Those who received
the simplified consent form demonstrated a 13% better overall understanding of the study
(p<0.001), as well as better specific understandings of the study protocol (33%, p<0.001),
duration (178%, p<0.001) and direct benefits (7%, p<0.001). There was a non-significant trend
to better understanding of the risks. These differences were seen across parents with high and
low reading abilities. Eighty-one percent of the parents reported preferring the simpler form. In a
third randomized trial that simulated research recruitment of 233 low-income children, there was
no difference ascertained in comprehension between parents receiving standard and simpler
forms; however, among the 124 parents with reading comprehension scores at or below 8th
grade, those who received the simpler form had nearly-significant higher comprehension scores
(p=0.06).42 In contrast to these findings, in a controlled (but not randomized) study comparing a
chemotherapy research consent form written at a 7th grade reading level to a standard one,
comprehension levels were similar for both forms, even among participants with reading levels
of 8th grade or less. However, literacy patients stated preference for receiving the simpler form.34
Another trial of a chemotherapy consent form simplified to the 8th grade level, with 44
institutions randomized to using the simplified or standard consent form, also showed no
difference in knowledge, but patients at the institutions using the simplified form had 9% higher
patient satisfaction scores (p=0.004).43
Providing Supplemental Written Materials Improves Recall and

Comprehension
Multiple randomized controlled trials have demonstrated that that providing patients with
supplemental written materials, in simplified language, results in higher patient recall of
informed consent information. In a study of 192 patients who underwent intraperitoneal,
intrathoracic or vascular surgery at a large teaching hospital, information cards which
explained in a simplified manner the procedure and what the patient could expect during and
after the surgery were given to half of the patients by random assignment. Both groups had the
same level of knowledge 1 hour after signing the consent form but those who received the
information cards had better information recall on the day of hospital discharge (p=0.04).23
Among 125 patients who underwent thyroidectomy or parotidectomy at an academic tertiary
center, those who were randomly assigned to receive a pamphlet with illustrations and written
information about the procedure were able to recall 50% of the risks compared with the control
group recalling 30% of the risks (p<0.001).44 In another randomized-controlled study of 126
patients who underwent total hip arthroplasty, a 1.5 page written information sheet in simplified
465
language with an illustration, given to patients at the pre-operative visit, resulted in patients
having 25% higher knowledge scores on admission for the procedure (p=0.004) compared with
patients who received a structured verbal discussion.45 Finally, in a randomized-controlled trial
of information leaflets describing risks and benefits, sent by mail 2 weeks prior to surgery to
patients scheduled for elective orthopedic procedures, the group receiving the leafelets had a
median comprehension score 30 percentage points higher than those who did not (p<0.001).46
Decision Aids Improve Knowledge and Participation in Shared

Decisionmaking
Decision aids are tools specifically designed to help patients make choices by having a
detailed, specific, and personalized focus on options and outcomes.47 For example, one
randomized-controlled study examined the effect of a touch-screen decision aid that provided
detailed information, including outcome probabilities, to the patient based on the information the
patient entered regarding his or her age and diagnosis.48 Patients had the option of getting more
detailed information, if desired, on pharmacologic and alternative medicine options. When this
decision aid was tested against an educational booklet, those who used the decision aid had an
adjusted six percentage point increase in knowledge compared with those who did not
(p=0.05).49 Another example, which is not technology-dependent, is an illustrated pamphlet
decision-aid for informed consent in prostate cancer screening which, in a randomized-controlled
trial, increased knowledge by 6% (p<0.01). A 2009 Cochrane review of 55 studies on the
efficacy of decision-aids for screening or treatment decisions found that, overall, they improve
knowledge scores by an average of 15 percentage points, improve patients participation in
decisionmaking, result in lower decisional conflict, and increase accuracy of risk perceptions.47
Video Educational Tools Also Improve Knowledge

Randomized-controlled trials of video educational tools (that are not also decision aids) have
also shown positive results. A randomized-controlled trial of an informational video for women
considering laparoscopic tubal ligation showed that women who watched the video, in addition
to standard consent procedures, demonstrated 56% higher knowledge scores than women who
were engaged in standard consent procedures alone (p<0.001).50 And, in a randomized-controlled
trial of an informational video on colonoscopy, those who watched the video in addition to
having a physician discussion had 19% higher knowledge scores than patients who had the
physician discussion alone (p<0.001).51 In a randomized-controlled trial of patients scheduled for
intravenous contrast for computed tomography, English and Spanish-speaking patients were
exposed to a low-literacy video in their preferred language. Participants who watched the video
displayed, in comparison to controls, 20 percentage point higher knowledge (95% CI 13-28%)
and 10 percentage point higher satisfaction scores. This result was consistent for both Spanish
and English speakers and Spanish and English speakers in the intervention group had similar
post-consent knowledge scores while Spanish speakers in the control group had significantly
lower post-consent knowledge scores than English-speaking controls.
Interactive Computer-Based Educational Tools Show Mixed Results

Limited studies of computer-based educational tools (that are not also decision aids) have
shown mixed results. In a randomized-controlled trial of a computer program that augments
practitioner-patient discussions with graphical content and illustrations, when used in patients
considering cardiology or endoscopy procedures, resulted in 43% higher patient knowledge
466
scores (p=0.006) as well as 34% higher satisfaction scores (p<0.001).52 A randomized-controlled

trial of an interactive computer program about colonoscopy indications, risks, and benefits,
tailored to an 8th-grade reading level, showed 16% higher knowledge scores among patients who
received the intervention.53 However, in a randomized-controlled trial of 101 patients consenting
to chemotherapy, recall of treatment information showed no difference in knowledge between
patients who received an interactive CD-ROM detailing treatment information and those who
received standard written information.54 In another randomized controlled trial of 44 patients
receiving standard genetic counseling versus education by an interactive computer program as
part of an informed consent process prior to cystic fibrosis carrier-status testing, both groups had
similar increases in knowledge.55
Structured Informed Consent Discussions Need Further Testing

Structured discussions for informed consent are those in which the practitioner engaging the
patient uses a written guide to structure the conversation. Two studies, limited by using nonrandomized designs, have examined structured discussions for informed consent. In a study of
patients considering cardiac catheterization, patients exposed to a half-hour structured informed
consent discussion had 29% higher knowledge scores compared with controls (p<0.001).56 In
another study of patients being consented for head and neck surgery, the group of patients for
which the provider used a structured interview guide were told about 65% more complications.57
However, in a study that did use a randomized design, and is mentioned above in the section on
supplemental written materials, structured verbal discussion compared with a 1.5 page
simplified and illustrated information sheet showed that patients had lower knowledge scores
with the structured discussion compared with the information sheet (p=0.004).45
Repeat Back Methods May Be Effective but Time Consuming

The repeat back method, also known as teach back, is an interactive communication
strategy in which the patient is asked to explain, in his or her own words, what has been told to
the patient. Then, as needed, the practitioner clarifies or tailors the explanation, serially
reassessing and re-explaining until the patient demonstrates recall and comprehension.58 In a
randomized-controlled study of 575 patients undergoing elective surgery, a computer prompted
and guided the practitioner in conducting the repeat-back procedure during the informed consent
discussion. Information comprehension, tested immediately, showed that patients receiving
repeat back comprehended 71% of the information while the control group comprehended
68% of the information (p=0.03). Discussions using repeat back took, on average, 2.6 minutes
longer.59 In a randomized-controlled trail of 20 patients who underwent repair of their anterior
cruciate ligament, 100% of patients whose discussions used repeat back, compared with 33%
of patients in the control group, were able to correctly answer a 3-item questionnaire about the
risks and benefits 1 month later (p=0.03).60

Informed consent is a process in which patients and health care practitioners dialogue about a
proposed medical treatments nature, consequences, harms, benefits, risks, and alternatives.
Although more evidence is needed on the potential specific association between informed
consent and patient safety, studies have shown that improved communication between
practitioners and patients leads to improved patient outcomes, less medical errors, and lower
rates of malpractice claims. Adequacy of the informed consent process has been more firmly
467
linked to patient satisfaction. Despite its importance, multiple studies have demonstrated that, in
practice, the informed consent process is often incomplete and patient recall and comprehension
of the discussion is usually low. Patients who are older, less educated, LEP, are of minority race,
or have cognitive dysfunction or low intelligence levels are particularly vulnerable in the
informed consent process.
Multiple methods have been proposed for improving the informed consent process. Studies
have shown that, in general, providing patients with simplified supplemental written materials,
using decision-aids, using video educational tools, and using the repeat back method improves
informed consent patient recall and comprehension. Studies using interactive computer programs
have had mixed results and further research is needed in this area. Studies using structured
informed consent discussion have also been limited. Studies of simplifying the informed consent
documents that have been done for research-related forms have shown mixed results on patient
recall and comprehension, but generally improve satisfaction; studies examining this effect
among informed consent documents in nonresearch clinical settings are lacking. A summary
table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Negligible
Estimate of
Cost
Low
How Much do We
References
1.
California Appellate Court, decided on Oct

22, 1957. Salgo v Leland Stanford Jr
University Board of Trustees. ;
2.
Faden RR, T.L. B, N.M.P. K. A History and

Theory of Informed Consent. New York:
Oxford University Press 1986.
3.
Jones JW, McCullough LB, Richman BW.

Informed consent: its not just signing a
form. Thorac Surg Clin. 2005;15(4):451-60,
v.16276809.
4.
Greenfield S, Kaplan S, Ware JE, Jr.

Expanding patient involvement in care.
Effects on patient outcomes. Ann Intern
Med. 1985;102(4):520-8.3977198.
5.
Stewart MA. Effective physician-patient

communication and health outcomes: a
review. CMAJ. 1995;152(9):142333.7728691.
6.
Davis RE, Jacklin R, Sevdalis N, et al.

Patient involvement in patient safety: what
factors influence patient participation and
engagement? Health Expect.
2007;10(3):259-67.17678514.
468
7.
Chen JY, Tao ML, Tisnado D, et al. Impact

of physician-patient discussions on patient
satisfaction. Med Care. 2008;46(11):115762.18953226.
8.
Iversen MD, Daltroy LH, Fossel AH, et al.

The prognostic importance of patient preoperative expectations of surgery for lumbar
spinal stenosis. Patient Educ Couns.
1998;34(2):169-78.9731176.
9.
Levinson W, Roter DL, Mullooly JP, et al.

Physician-patient communication. The
relationship with malpractice claims among
primary care physicians and surgeons.
JAMA. 1997;277(7):553-9.9032162.
10.
Braddock CH, 3rd, Fihn SD, Levinson W, et

al. How doctors and patients discuss routine
clinical decisions. Informed decision making
in the outpatient setting. J Gen Intern Med.
1997;12(6):339-45.9192250.
11.
Edwards WS, Yahne C. Thomas G. Orr

Memorial Lecture. Surgical informed
consent: what it is and is not. Am J Surg.
1987;154(6):574-8.3425796.
12.
Lidz CW, Appelbaum PS, Meisel A. Two

models of implementing informed consent.
Arch Intern Med. 1988;148(6):13859.3377623.
23.
Lavelle-Jones C, Byrne DJ, Rice P, et al.

Factors affecting quality of informed
consent. BMJ. 1993;306(6882):88590.8490411.
13.
Whitney SN, McGuire AL, McCullough LB.

A typology of shared decision making,
informed consent, and simple consent. Ann
Intern Med. 2004;140(1):54-9.14706973.
24.
Hunt LM, de Voogd KB. Are good

intentions good enough? Informed consent
without trained interpreters. J Gen Intern
Med. 2007;22(5):598-605.17443367.
14.
Geller G, Botkin JR, Green MJ, et al.

Genetic testing for susceptibility to adultonset cancer. The process and content of
informed consent. JAMA.
1997;277(18):1467-74.9145720.
25.
15.
Wolf AM, Nasser JF, Schorling JB. The

impact of informed consent on patient
interest in prostate-specific antigen
screening. Arch Intern Med.
1996;156(12):1333-6.8651843.
Schenker Y, Wang F, Selig SJ, et al. The

impact of language barriers on
documentation of informed consent at a
hospital with on-site interpreter services. J
Gen Intern Med. 2007;22 Suppl 2:2949.17957414.
26.
Hekkenberg RJ, Irish JC, Rotstein LE, et al.

Informed consent in head and neck surgery:
how much do patients actually remember? J
Otolaryngol. 1997;26(3):155-9.9176798.
16.
Braddock CH, 3rd, Edwards KA, Hasenberg

NM, et al. Informed decision making in
outpatient practice: time to get back to
basics. JAMA. 1999;282(24):231320.10612318.
27.
Lashley M, Talley W, Lands LC, et al.

Informed proxy consent: communication
between pediatric surgeons and surrogates
about surgery. Pediatrics. 2000;105(3 Pt
1):591-7.10699114.
17.
Braddock C, 3rd, Hudak PL, Feldman JJ, et

al. Surgery is certainly one good option:
quality and time-efficiency of informed
decision-making in surgery. J Bone Joint
Surg Am. 2008;90(9):1830-8.18762641.
28.
18.
Etchells E., Ferrari M., Kiss A., et al.

Informed decision-making in elective major
vascular surgery: analysis of 145 surgeonpatient consultations. Can J Surg.
2011;54(3):173-8.
Finch WJ, Rochester MA, Mills RD. A

randomised trial of conventional versus
BAUS procedure-specific consent forms for
transurethral resection of prostate. Ann R
Coll Surg Engl. 2009;91(3):2328.19220941.
29.
Krupp W, Spanehl O, Laubach W, et al.

Informed consent in neurosurgery: patients
recall of preoperative discussion. Acta
Neurochir (Wien). 2000;142(3):233-8;
discussion 8-9.10819252.
30.
Hopper KD, TenHave TR, Tully DA, et al.

The readability of currently used
surgical/procedure consent forms in the
United States. Surgery. 1998;123(5):496503.9591001.
Whittle J, Conigliaro J, Good CB, et al.

Understanding of the benefits of coronary
revascularization procedures among patients
who are offered such procedures. Am Heart
J. 2007;154(4):662-8.17892988.
31.
Bottrell MM, Alpert H, Fischbach RL, et al.

Hospital informed consent for procedure
forms: facilitating quality patient-physician
interaction. Arch Surg. 2000;135(1):2633.10636343.
Kutner M, Greenberg E, Jin Y, et al. The

Health Literacy of Americas Adults:
Results From The 2003 National
Assessment of Adult Literacy (NCES 206483). 2006.
32.
Gazmararian JA, Baker DW, Williams MV,

et al. Health literacy among Medicare
enrollees in a managed care organization.
JAMA. 1999;281(6):545-51.10022111.
19.
20.
21.
22.
Hopper KD, Lambe HA, Shirk SJ.

Readability of informed consent forms for
use with iodinated contrast media.
Radiology. 1993;187(1):279-83.8451429.
Cassileth BR, Zupkis RV, Sutton-Smith K,

et al. Informed consent -- why are its goals
imperfectly realized? N Engl J Med.
1980;302(16):896-900.7360175.
469
33.
Kim SP, Knight SJ, Tomori C, et al. Health

literacy and shared decision making for
prostate cancer patients with low
socioeconomic status. Cancer Invest.
2001;19(7):684-91.11577809.
34.
Davis TC, Holcombe RF, Berkel HJ, et al.

Informed consent for clinical trials: a
comparative study of standard versus
simplified forms. J Natl Cancer Inst.
1998;90(9):668-74.9586663.
43.
Coyne CA, Xu R, Raich P, et al.

Randomized, controlled trial of an easy-toread informed consent statement for clinical
trial participation: a study of the Eastern
Cooperative Oncology Group. J Clin Oncol.
2003;21(5):836-42.12610182.
44.
Chan Y, Irish JC, Wood SJ, et al. Patient

education and informed consent in head and
neck surgery. Arch Otolaryngol Head Neck
Surg. 2002;128(11):1269-74.12431168.
35.
Sudore RL, Landefeld CS, Williams BA, et

al. Use of a modified informed consent
process among vulnerable patients: a
descriptive study. J Gen Intern Med.
2006;21(8):867-73.16881949.
45.
Langdon IJ, Hardin R, Learmonth ID.

Informed consent for total hip arthroplasty:
does a written information sheet improve
recall by patients? Ann R Coll Surg Engl.
2002;84(6):404-8.12484580.
36.
Fink AS, Prochazka AV, Henderson WG, et

al. Predictors of comprehension during
surgical informed consent. J Am Coll Surg.
2010;210(6):919-26.20510800.
46.
37.
Raymont V, Bingley W, Buchanan A, et al.

Prevalence of mental incapacity in medical
inpatients and associated risk factors: crosssectional study. Lancet.
2004;364(9443):1421-7.15488217.
Ashraff S, Malawa G, Dolan T, et al.

Prospective randomised controlled trial on
the role of patient information leaflets in
obtaining informed consent. ANZ J Surg.
2006;76(3):139-41.16626351.
47.
OConnor AM, Bennett CL, Stacey D, et al.

Decision aids for people facing health
treatment or screening decisions. Cochrane
Database Syst Rev.
2009;(3):CD001431.19588325.
48.
Deyo RA, Cherkin DC, Weinstein J, et al.

Involving patients in clinical decisions:
impact of an interactive video program on
use of back surgery. Med Care.
2000;38(9):959-69.10982117.
49.
Phelan EA, Deyo RA, Cherkin DC, et al.

Helping patients decide about back surgery:
a randomized trial of an interactive video
program. Spine (Phila Pa 1976).
2001;26(2):206-11;discussion 12.11154542.
50.
Mason V, McEwan A, Walker D, et al. The

use of video information in obtaining
consent for female sterilisation: a
randomised study. BJOG.
2003;110(12):1062-71.14664877.
51.
Luck A, Pearson S, Maddern G, et al.

Effects of video information on
precolonoscopy anxiety and knowledge: a
randomised trial. Lancet.
1999;354(9195):2032-5.10636368.
52.
Enzenhofer M, Bludau HB, Komm N, et al.

Improvement of the educational process by
computer-based visualization of procedures:
randomized controlled trial. J Med Internet
Res. 2004;6(2):e16.15249265.
38.
39.
40.
41.
42.
Appelbaum PS. Clinical practice.

Assessment of patients competence to
consent to treatment. N Engl J Med.
2007;357(18):1834-40.17978292.
Etchells E, Darzins P, Silberfeld M, et al.
Assessment of patient capacity to consent to
treatment. J Gen Intern Med. 1999;14(1):2734.9893088.
Dresden GM, Levitt MA. Modifying a
standard industry clinical trial consent form
improves patient information retention as
part of the informed consent process. Acad
Emerg Med. 2001;8(3):246-52.11229946.
Tait AR, Voepel-Lewis T, Malviya S, et al.
Improving the readability and processability
of a pediatric informed consent document:
effects on parents understanding. Arch
Pediatr Adolesc Med. 2005;159(4):34752.15809387.
Campbell FA, Goldman BD, Boccia ML, et
al. The effect of format modifications and
reading comprehension on recall of
informed consent information by lowincome parents: a comparison of print,
video, and computer-based presentations.
Patient Educ Couns. 2004;53(2):20516.15140461.
470
53.
Shaw MJ, Beebe TJ, Tomshine PA, et al. A

randomized, controlled trial of interactive,
multimedia software for patient colonoscopy
education. J Clin Gastroenterol.
2001;32(2):142-7.11205650.
57.
Dawes PJ, OKeefe L, Adcock S. Informed

consent: the assessment of two structured
interview approaches compared to the
current approach. J Laryngol Otol.
1992;106(5):420-4.1613369.
54.
Olver IN, Whitford HS, Denson LA, et al.

Improving informed consent to
chemotherapy: a randomized controlled trial
of written information versus an interactive
multimedia CD-ROM. Patient Educ Couns.
2009;74(2):197-204.18945572.
58.
Schillinger D, Piette J, Grumbach K, et al.

Closing the loop: physician communication
with diabetic patients who have low health
literacy. Arch Intern Med. 2003;163(1):8390.12523921.
59.
55.
Castellani C, Perobelli S, Bianchi V, et al.

An interactive computer program can
effectively educate potential users of cystic
fibrosis carrier tests. Am J Med Genet A.
2011;155A(4):778-85.21416590.
Fink AS, Prochazka AV, Henderson WG, et

al. Enhancement of surgical informed
consent by addition of repeat back: a
multicenter, randomized controlled clinical
trial. Ann Surg. 2010;252(1):2736.20562609.
56.
Solomon J, Schwegman-Melton K.
Structured teaching and patient
understanding of informed consent. Crit
Care Nurse. 1987;7(3):74-9.3665530.
60.
Wadey V, Frank C. The effectiveness of

patient verbalization on informed consent.
Can J Surg. 1997;40(2):124-8.9126126.
471
Chapter 40. Team-Training in Health Care: Brief Update

Review
Sallie J. Weaver, Ph.D.; Michael A. Rosen, Ph.D.
Introduction
Deficiencies in communication and teamwork have long been cited as a frequent contributor
to adverse events. Precise estimates of the extent of the problem are difficult to make, given
definitional issues as well as reporting and measurement problems. However, a variety of studies
support the notion that teamwork and communication are critical components of safe health care
systems. Previous reviews have reported linkages between various aspects of teamwork (e.g.,
situational monitoring, communication, leadership, trust, shared mental models) and clinical
performance.1-3 For example, observational studies in the surgical domain have shown increased
odds of complications and death (odds ratio 4.82; 95% confidence interval, 1.30 17.87) when
surgical teams exhibit less frequent teamwork behaviors (e.g., less information sharing during
intraoperative and handoff phases, and less briefing during handoffs).4 Reviews of malpractice
claims indicate that communication problems are major contributing factors in 24% of cases that
result in such claims.5 Other studies using root cause analysis to examine contributing factors
have found teamwork and communication issues cited as root causes in 52% to 70% of adverse
events.6,7 Additionally, teamwork and communication dimensions of safety culture have been
significantly related to adverse clinical events.8,9
The 2001 Making Health Care Safer report reviewed the topic of team-training in a review
entitled, Crew Resource Management and Its Application in Medicine. This review discussed
early conceptualizations of team-training in other high reliability industries such as aviation and
summarized early studies attempting to translate team-training principles developed elsewhere
into health care settings. The development and implementation of team-training programs has
grown dramatically in the last decade with improvements in the content and methods of
training.10 Additionally, there is over 30 years of evidence examining team performance
processes and the impact of team-training across a wide variety of highly complex, high-risk
work environments.11 This review provides an update on the implementation and effectiveness of
team-training in health care.
While there has been no previous comprehensive formal systematic review dedicated
uniquely to team-training in health care to date, a systematic review of interventions to improve
team effectiveness in health care found that the majority involved some form of team-training
(42 of 48 reviewed studies).10 Several systematic reviews with narrowly defined foci have
investigated the effectiveness of team-training for obstetric emergencies,12 for enhancing
communication in surgery,13 and classroom-based team-training interventions13,14 for example.
Additionally, several narrative reviews have investigated the content, design, and delivery of
team-training and the impact of team processes in health care.2,15,16 We draw on results from
these previous reviews to describe articles on interventions involving team-training.
What Is Team-Training?
Team-training is defined as a constellation of content (i.e., the specific knowledge, skills, and
attitudes that underlie targeted teamwork competencies), tools (i.e., team task analysis,
472
performance measures), and delivery methods (i.e., information, demonstration, and practice
based learning methods) that together form an instructional strategy.17 In this sense, teamtraining is a systematic methodology for optimizing the communication, coordination, and
collaboration of health care teams that combines specific content with opportunities for practice,
formative feedback, and tools to support transfer of training to the daily care environment.
As described in the National Quality Forums 34 Safe Practices for Better Healthcare in the
2010 Update, teamwork training and skill building is defined as follows:18
Healthcare organizations must establish a proactive, systematic,
organization-wide approach to developing team-based care through
teamwork training, skill building, and team-led performance
improvement interventions that reduce preventable harm to
patientstraining programs should systematically address and
apply the principles of effective team leadership, team formation
[and team processes]
Borrowing from other high reliability communities, the concept of team-training in health
care originated in the form of Crew Resource Management (CRM), a specific team-training
strategy focused on developing a sub-set of teamwork competencies generally related to hazard
identification, assertive communication, and collective management of available resources (e.g.,
people, tools, and information).19-21 However, the practice of team-training has become much
more broadly conceptualized in health care as the science dedicated to understanding team
processes, and performance has grown. Today, team-training is an overarching term that
encompasses a broad range of learning and development strategies, methods, and teamwork
competencies. The critical element is that the learning activity focuses on developing, refining,
and reinforcing knowledge, skills, or attitudes that underlie effective teamwork. This
differentiates team-training activities from technical or procedural learning activities that are
focused on developing technical clinical skills (e.g., cognitive skills such as differential diagnosis
and procedural skills).22 Prior narrative reviews of team-training interventions in health care have
found that the most commonly targeted teamwork competencies include communication,
situational awareness, leadership, role clarity, and coordination.13-16,23,24
What Is the Context for the Use of Team-Training?

Previous reviews highlight that team-training has been implemented across a broad range of
contexts using a variety of implementation strategies and learning modalities.2,3,16,23,25 This
includes academic hospitals (e.g.,26) and community based hospitals (e.g.,27),28 as well as
medical centers affiliated with the VA and the Military Health System.29 Additionally, teamtraining programs have focused on a variety of audiences including both current practitioners
(e.g.,30-32) and trainees (e.g.,33).
In terms of implementation strategy, both train-the-trainer and direct train-the-staff strategies
have been utilized. For example, a train-the-trainer model formed the foundation for the National
Implementation of TeamSTEPPS Project,34 a collaborative effort of Department of Defense
(DoD), the Agency for Healthcare Research and Quality (AHRQ), and the American Institute for
Research (AIR) designed to create a national training and support infrastructure for health care
entities implementing team-training. Through a national network of five team resource centers,
individuals interested in leading the implementation of team-training within their organization
could become TeamSTEPPS Master Trainers by participating in an intensive 3-day training
473
session. Master Trainers then train administrators and frontline personnel within their own
organization using the customizable TeamSTEPPS curriculum. A slightly different approach was
utilized in the large-scale implementation of team-training throughout the Veterans
Administration (VA). As part of the VA National Center for Patient Safety Medical Team
Training (MTT) program learning sessions for participating VA medical centers were facilitated
directly by an interdisciplinary team (physician, nurse) of dedicated MTT faculty.35,36 Both
strategies, however, include local facility change teams, implementation of on-the-job tools (e.g.,
process checklists, scripts) to support training transfer, and measurement and evaluation
processes as integral implementation components.
What Have We Learned About Team-Training Effectiveness?

Team-training provides an opportunity for health care providers to learn, refine, and practice
different strategies for communication, leadership, coordination, and collaboration. A metaanalysis of team-training that included 93 effect sizes across a broad range of industries found
that participation in team-training can account for nearly 20% of the variance in team processes
( = 0.44) and outcomes ( = 0.39).17 Additionally, similar effect sizes were found for teams who
worked together on a regular basis (intact teams = 0.48) and teams who did not (ad-hoc teams
= 0.44). Previous reviews examining the relationship between teamwork and patient safety
reported significant relationships between both provider ratings and observer ratings of
teamwork, risk-adjusted mortality and length of stay.3
While no previous comprehensive systematic review has been dedicated uniquely to teamtraining in health care to date, a descriptive systematic review through April 2008 of
interventions to improve team effectiveness in health care found that the majority involved some
form of team-training (42 of 48 reviewed studies); other interventions focused on tools to support
team effectiveness (e.g., checklists, goal lists; 8 studies) and organizational interventions (e.g.,
redesign of care processes or team structures, 8 studies).10 This review included 32 studies
dedicated specifically to some form of team-training, including 7 studies of simulation-based
team-training, 8 studies of training based in CRM, 6 studies of interprofessional training, and 11
studies dedicated to other forms of team-training.10 The review found no studies that evaluated
exactly the same intervention. This lack of study homogeneity is an important consideration in
evaluating the evidence for such patient safety practices, given that local customization is a
common practice, and underscores the need for high quality implementation studies designed to
study variation in training design and implementation. Another descriptive systematic review
limited to classroom-based team-training interventions published through March 2010 included
18 studies.14 This review excluded web-based, simulation-based, mono-disciplinary studies as
well as studies conducted outside of the hospital setting. Based in Kirkpatricks four-level model
of evaluation,37 this review found 6 studies evaluated participant reactions to training, 9 studies
evaluated training effects on behavior change, 7 studies evaluated processes measures, and 4
studies evaluated the impact of team-training on patient outcomes.
Overall, prior reviews concluded that team-training interventions are effective in improving
teamwork and patient safety related attitudes, producing learning, and changing teamwork and
communication behaviors in a variety of clinical areas.12,14,28,38,39
More recently, some studies have shown a significant impact of teamwork training programs
on safety and quality metrics. An evaluation of the Veterans Affairs Medical Team-Training
program showed significant and sustained decreases in preoperative delays (from 16% to 7% of
cases, p = .004), increased antibiotic prophylaxis compliance (from 85% to 97%, p < .0001),
474
decreases in equipment issues/case delays (from 24% to 7% of cases, p < .0001), decreased
handoff issues (from 5.4% to 0.3% of cases, p < .0001), and most notably a reduction in
mortality (p = .01).36,40,41 Additionally, a dose-response relationship was established such that for
each quarter the program was in place at a facility, a decrease of 0.5 deaths per 1000 procedures
(p = .001) was observed. Implementation of a related team-training program jointly developed by
the Agency for Healthcare Research and Quality and the Department of Defense,
TeamSTEPPS, has been associated with increased efficiency in clinical processes for multidisciplinary trauma teams (e.g., decreased times from arrival to surgery from 130.1 to 94.5
minutes (p < .05), endotracheal intubation from 10.1 to 6.6 minutes (n.s.), and CT scan from 26.4
to 22.1 minutes (p < .01)42) as well as an 83% reduction in medication and transfusion errors
(p < .001) and a 70% reduction in needlestick injuries and exposures (p < .05) in a U.S. Combat
Support Hospital deployed in Iraq.29 Other studies have also reported significant reductions in
clinical decision time (p < .05)43 associated with team-training, as well as one study showing a
reduction in adverse clinical events and a 50% reduction in high severity malpractice claims
(pre-training 11 high severity claims, post-training 5 high severity claims, no statistics
reported).44
Overall, the systematic review by Buljac-Smardizic10 concluded that the majority of studies
reviewed were of low to moderate level quality; however, eight of the reviewed team-training
studies were categorized as high or moderate quality (i.e., RCT or high quality pre-post study).
In the review by Rabl38 of classroom-based team-training interventions 15 of the 18 reviewed
studies were uncontrolled and 17 studies were rated at a moderate or high risk for bias.
What Have We Learned About Team-Training Design and

Delivery?
Several narrative reviews of team-training and team processes in health care have also
examined how team-training curricula are being designed and delivered as described in the
published literature.1,13,15,23,24,38,45,46 These reviews find variation among team-training programs
in terms of how much time learners spend in training, how often clinicians and staff are
participating, and other details regarding content, delivery strategies, and evaluation efforts.
For example, programs vary in the instructional methods utilized. Instructional methods can
be conceptualized in terms of three broad categories: (1) information-based methods (e.g.,
didactic lecture), (2) demonstration-based methods (e.g., behavioral modeling, videos), and (3)
practice-based methods (e.g., simulation, role-playing). Previous reviews have found that the
majority (83%) of team-training programs integrated both information and practice-based
methods and that 68% reported using simulation-based learning in order to provide trainees with
the opportunity to practice and refine teamwork skills, as well as receive formative feedback.23
Only 35% of studies in this prior review, however, reported incorporating demonstration-based
learning opportunities.
Variation in program duration is an additional example. A review of 18 studies evaluating
classroom-based team-training interventions found course duration varied from 4 hour to 3 days
with several studies describing longer train-the-trainer programs.38 Another review found that
53% of 40 reviewed team-training programs were designed to last less than 1 day.14
In terms of structure, team-training in health care has been conducted with both in-tact (i.e.,
teams who have worked together currently) and ad-hoc teams (i.e., teams formed for training
purposes only). For example, Weaver14 found 8 studies reported training in-tact teams and 5
studies reported training in ad-hoc teams.
475
Overall, no comprehensive meta-analysis to date has directly examined training duration,

format, or other variations in design or delivery as boundary conditions influencing the
effectiveness of health care team-training programs. Multi-site studies such those of the VA
Medical Team Training program and TeamSTEPPS and comparative effectiveness studies are
important for establishing robust evidence regarding questions of how much, how often, and
through which modalities team-training is most effective for inpatient, outpatient, and long-term
care health care settings.

In summary, previous reviews of team-training in health care and more recent publications
have found that can improve teamwork processes (e.g., communication, coordination, and
cooperation), and that implementation of team-training programs has been associated with
improvements in patient safety outcomes (e.g., reductions in adverse events, reductions in
mortality). Several narrative reviews have examined how team-training is being developed and
delivered in health care.15,17,23,24 In terms of the strength of evidence, the previous systematic
review10 included several studies that utilized RCT or controlled pre-post designs and several
large-scale studies examining the impact of comprehensive team-training strategies have been
published since this review. However, it is important to also note that previous reviews reflect a
wide range in the quality of evidencewith several studies of team-training being limited due to
small sample sizes, weak study design, and limited detail regarding the team-training curriculum
or implementation strategy.10,23
Our non-systematic brief review included several studies that have been published since the
systematic review conducted by Buljac-Samardzic, as well as findings from previous narrative
reviews. Overall, there is some moderate to high quality evidence that team-training can
positively impact health care team processes and patient outcomes, as well as toolkits available
to support the development and implementation of team-training programs. For example, the
comprehensive TeamSTEPPS curriculum is available publically through AHRQ
(www.teamstepps.ahrq.gov) and there are five Team Resource Centers available nationally that
provide TeamSTEPPS Master Training. Additionally, the VA Medical Team Training program
is available to VA Medical Centers through the National Center for Patient Safety
(www.patientsafety.gov/mtt). There is also a large body of work dedicated to examining the
effectiveness of team-training interventions across a wide range of industries available to inform
training design and delivery decisions.47
To continue building this evidence base, future work should continue to evaluate teamtraining. This includes evaluating the impact of team-training on patient safety outcomes,
evaluating team-training in other settings (e.g., primary care, outpatient dialysis care settings),
examining the comparative effectiveness of different methods for delivering team-training, and
examining implementation methods to support sustainment of behavior changes achieved
through training. For example, there is little evidence available to date that provides insight into
the frequency of retraining or dedicated practice needed to develop and maintain effective
teamwork skills. Additionally, there is a need to examine how dynamic team composition (i.e.,
changes in team membership) moderate team processes and the effects of team-training.
Methodologically, robust validation studies are needed to strengthen the evidence surrounding
the indices used to measure teamwork processes within health care and more studies that utilize
robust experimental designs are needed. Finally, longitudinal studies and studies that address the
integration of team-training concepts throughout the career development of health care
476
professionals, from basic through continuing education, are needed to continue building this base
of evidence. A summary table is located in Table 1, Chapter 40.
Scope of the
Strength of
the PSP
Effectiveness
Common/High
Moderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Moderate
How Much do We
Moderate/Moderate-todifficult
References
1.
2.
3.
Baker DP, Gustafson S, Beaubien J, Salas E,

Barach P. Medical Teamwork and Patient
Safety: The Evidence-based Relation.
Literature Review. AHRQ Publication No.
05-0053, April 2005. Agency for Healthcare
Research and Quality, Rockville, MD.
www.ahrq.gov/qual/medteam.
Manser T. Teamwork and patient safety in
dynamic domains of healthcare: a review of
the literature. Acta Anaesthesiol Scand
2009; 53(2):143-51.
Sorbero ME, Farley DO, Mattke S, Lovejoy
SL. Outcome Measures for Effective
Teamwork in Inpatient Care: Final Report.
RAND. 2008.
4.
Mazzocco K, Petitti DB, Fong KT et al.

Surgical team behaviors and patient
outcomes. Am J Surg 2009; 197(5):678-85.
5.
Rogers SO Jr, Gawande AA, Kwaan M et al.

Analysis of surgical errors in closed
malpractice claims at 4 liability insurers.
Surgery 2006; 140(1):25-33.
6.
Rabol LI, Andersen ML, Ostergaard D,

Bjorn B, Lilja B, Mogensen T. Descriptions
of verbal communication errors between
staff. An analysis of 84 root cause analysisreports from Danish hospitals. BMJ Qual
Saf 2011; 20(3):268-74.
7.
Singh H, Thomas EJ, Petersen LA, Studdert

DM. Medical errors involving trainees: a
study of closed malpractice claims from 5
insurers. Arch Intern Med 2007;
167(19):2030-6.
8.
Mardon RE, Khanna K, Sorra J, Dyer N,

Famolaro T. Exploring relationships
between hospital patient safety culture and
adverse events. J Patient Saf 2010; 6(4):22632.
477
9.
Singer S, Lin S, Falwell A, Gaba D, Baker

L. Relationship of safety climate and safety
performance in hospitals. Health Serv Res
2009; 44(2 Pt 1):399-421.
10.
Buljac-Samardzic M, Dekker-van Doorn

CM, van Wijngaarden JD, van Wijk KP.
Interventions to improve team effectiveness:
a systematic review. Health Policy 2010;
94(3):183-95.
11.
Salas E, Cooke NJ, Gorman JC. The science

of team performance: progress and the need
for more.. Hum Factors 2010; 52(2):344-6.
12.
Merien AE, van de Ven J, Mol BW,

Houterman S, Oei SG. Multidisciplinary
team training in a simulation setting for
acute obstetric emergencies: a systematic
review. Obstet Gynecol 2010; 115(5):102131.
13.
Gillespie BM, Chaboyer W, Murray P.

Enhancing communication in surgery
through team training interventions: a
systematic literature review. AORN J 2010;
92(6):642-57.
14.
Rabol LI, Ostergaard D, Mogensen T.

Outcomes of classroom-based team training
interventions for multiprofessional hospital
staff. A systematic review. Qual Saf Health
Care 2010; 19(6):e27.
15.
Baker DP, Gustafson S, Beaubien JM, Salas

E, Barach P. Medical Team Training
Programs in Health Care. In: Henriksen K,
Battles JB, Marks ES, Lewin DI, editors.
SourceAdvances in Patient Safety: From
Research to Implementation (Volume 4:
Programs, Tools, and Products). Rockville
(MD): Agency for Healthcare Research and
Quality (U.S.); 2005 Feb.
16.
Gaba DM. Crisis resource management and

teamwork training in anaesthesia. Br J
Anaesth 2010; 105(1):3-6.
17.
Salas E, DiazGranados D, Klein C et al.

Does team training improve team
performance? A meta-analysis. Hum Factors
2008; 50(6):903-33.
18.

Better Healthcare 2010 Update: A
Consensus Report. qualityforum.org.
19.
Salas E, Burke CS, Bowers CA, Wilson KA.

Team training in the skies: does crew
resource management (CRM) training work?
Hum Factors 2001; 43(4):641-74.
20.
Howard SK, Gaba DM, Fish KJ, Yang G,

Sarnquist FH. Anesthesia crisis resource
management training: teaching
anesthesiologists to handle critical incidents.
Aviat Space Environ Med 1992; 63(9):76370.
21.
Reznek M, Smith-Coggins R, Howard S et

al. Emergency medicine crisis resource
management (EMCRM): pilot study of a
simulation-based crisis management course
for emergency medicine. Acad Emerg Med
2003; 10(4):386-9.
22.
Zeltser MV, Nash DB. Approaching the

evidence basis for aviation-derived
teamwork training in medicine. Am J Med
Qual 2010; 25(1):13-23.
23.
Weaver SJ, Lyons R, DiazGranados D et al.

The anatomy of health care team training
and the state of practice: a critical review.
Acad Med 2010; 85(11):1746-60.
24.
Weaver SJ, Salas E, Lyons R et al.

Simulation-based team training at the sharp
end: A qualitative study of simulation-based
team training design, implementation, and
evaluation in healthcare. J Emerg Trauma
Shock 2010; 3(4):369-77.
25.
26.
Thomas EJ. Improving teamwork in

healthcare: current approaches and the path
forward. BMJ Qual Saf 2011; 20(8):647-50.
Mayer CM, Cluff L, Lin WT et al.
Evaluating efforts to optimize TeamSTEPPS
implementation in surgical and pediatric
intensive care units. Jt Comm J Qual Patient
Saf 2011; 37(8):365-74.
478
27.
Riley W, Davis S, Miller K, Hansen H,

Sainfort F, Sweet R. Didactic and simulation
nontechnical skills team training to improve
perinatal patient outcomes in a community
hospital. Jt Comm J Qual Patient Saf 2011;
37(8):357-64.
28.
Weaver SJ, Rosen MA, DiazGranados D et

al. Does teamwork improve performance in
the operating room? A multilevel evaluation.
Jt Comm J Qual Patient Saf 2010;
36(3):133-42.
29.
Deering S, Rosen MA, Ludi V et al. On the

front lines of patient safety: implementation
and evaluation of team training in Iraq. Jt
Comm J Qual Patient Saf 2011; 37(8):350-6.
30.
Flin R, Maran N. Identifying and training

non-technical skills for teams in acute
medicine. Qual Saf Health Care 2004; 13
Suppl 1:i80-4.
31.
McCulloch P, Mishra A, Handa A, Dale T,

Hirst G, Catchpole K. The effects of
aviation-style non-technical skills training
on technical performance and outcome in
the operating theatre. Qual Saf Health Care
2009; 18(2):109-15.
32.
Morey JC, Simon R, Jay GD et al. Error

reduction and performance improvement in
the emergency department through formal
teamwork training: evaluation results of the
MedTeams project. Health Serv Res 2002;
37(6):1553-81.
33.
Thomas EJ, Williams AL, Reichman EF,

Lasky RE, Crandell S, Taggart WR. Team
training in the neonatal resuscitation
program for interns: teamwork and quality
of resuscitations. Pediatrics 2010;
125(3):539-46.
34.

Quality. TeamSTEPPSTM: National
implementation.
http://teamstepps.ahrq.gov/aboutnationalIP.h
tm. Accessed November 30, 2011.
35.
Dunn EJ, Mills PD, Neily J, Crittenden MD,

Carmack AL, Bagian JP. Medical team
training: applying crew resource
management in the Veterans Health
Administration. Jt Comm J Qual Patient Saf
2007; 33(6):317-25.
36.
Paull DE, Mazzia LM, Izu BS, Neily J,

Mills PD, Bagian JP. Predictors of
successful implementation of preoperative
briefings and postoperative debriefings after
medical team training. Am J Surg 2009;
198(5):675-8.
37.
Kirkpatrick DL, Kirkpatrick JD. Evaluating

Training Programs. the Four Levels. 3rd
edn. San Francisco: Berrett-Koehler
Publishers. Inc., 2006.
38.
Grogan EL, Stiles RA, France DJ et al. The

impact of aviation-based teamwork training
on the attitudes of health-care professionals.
J Am Coll Surg 2004; 199(6):843-8.
39.
Awad SS, Fagan SP, Bellows C et al.

Bridging the communication gap in the
operating room with medical team training.
Am J Surg 2005; 190(5):770-4.
40.
Wolf FA, Way LW, Stewart L. The efficacy

of medical team training: improved team
performance and decreased operating room
delays: a detailed analysis of 4863 cases.
Ann Surg 2010; 252(3):477-83; discussion
483-5.
41.
Neily J, Mills PD, Young-Xu Y et al.

Association between implementation of a
medical team training program and surgical
mortality. JAMA 2010; 304(15):1693-700.
42.
Capella J, Smith S, Philp A et al. Teamwork

training improves the clinical care of trauma
patients. J Surg Educ 2010; 67(6):439-43.
43.
Nielsen PE, Goldman MB, Mann S et al.

Effects of teamwork training on adverse
outcomes and process of care in labor and
delivery: a randomized controlled trial.
Obstet Gynecol 2007; 109(1):48-55.
44.
Mann S, Marcus R, Sachs B. Lessons from

the cockpit: How team training can reduce
errors on L&D. Contemporary OB Gyn.
2006;51:34-45.
45.
Salas E, Weaver SJ, DiazGranados D, Lyons

R, King H. Sounding the call for team
training in health care: some insights and
warnings. Acad Med 2009; 84(10
Suppl):S128-31.
46.
Salas E, Weaver SJ, Rosen MA, Gregory

ME. Team training for patient safety. In:
Carayon P (ed.). handbook of Human
Factors and Ergonomics in Healthcare (p.
627-648). Boca Raton, FL, CRC Press.
47.
479
Salas E, Almeida SA, Salisbury M et al.

What are the critical success factors for team
training in health care? Jt Comm J Qual
Patient Saf 2009; 35(8):398-405.
Chapter 41. Computerized Provider Order Entry With Clinical

Decision Support Systems: Brief Update Review
Sumant R. Ranji, M.D.; Stephanie Rennke, M.D.; Robert M. Wachter, M.D.
Introduction
Adverse drug events are one of the most common types of harmful errors in both hospitalized
and ambulatory patients. Studies have shown that preventable adverse drug events occur in 7 to
10 of every 100 hospital admissions,1-3 and may even occur more frequently in the ambulatory
setting.4 Prescribing errors are likely responsible for at least half of these events.5,6
What Are Computerized Provider Order Entry With Clinical

Decision Support Systems?
Computerized provider order entry (CPOE) refers to any system in which clinicians directly
enter orders for medications, tests, or procedures into an electronic system, which then transmits
the order directly to the recipient responsible for carrying out the order (e.g., the pharmacy,
laboratory, or radiology department). These systems were initially implemented in the inpatient
setting as a strategy to reduce medication errors, and their use is increasingly being broadened to
include entry of all types of orders in both the inpatient and outpatient settings. A CPOE system,
at a minimum, ensures standardized, legible, and complete orders and thus has the potential to
greatly reduce errors at the prescribing and transcribing stages.
How Have Computerized Provider Order Entry With Clinical

Decision Support Systems Been Implemented?
Clinical decision support systems (CDSS) are often integrated with CPOE systems. CDSS
provide clinicians with reminders or recommendations in order to optimize the safety and quality
of clinical decisions. For example, a medication CDSS may offer default values for doses, routes
of administration, and frequency for commonly used drugs. Such systems may also offer more
sophisticated drug safety features such as checking for drug allergies or drug-drug interactions,
providing reminders for appropriate laboratory monitoring (e.g. reminders to check coagulation
parameters if a patient is prescribed warfarin), or even suggesting appropriate orders based on
patient-specific factors (e.g., reminders to order prophylaxis against deep venous thrombosis in a
patient admitted with a hip fracture).
At the highest level of sophistication, the combination of CPOE and CDSS can therefore
prevent errors of commission and errors of omission. Optimal use of CPOE with CDSS in this
fashion requires integration across multiple hospital and ambulatory information systems,
including the medical record, clinical laboratory, radiology, and pharmacy.
Despite recommendations from a broad range of governmental and non-governmental
organizations, the pace of uptake of CPOE and CDSS has remained slow in both the inpatient
and outpatient environments.7,8 The use of CPOE and CDSS will likely increase with the
implementation of the Health Information Technology for Economic and Clinical Health
(HITECH) Act, part of the American Recovery and Reinvestment Act of 2009 (ARRA).
HITECH stipulates that health care providers must demonstrate the meaningful use of
480
electronic health records (EHR) by 2015, and will include penalties for failing to achieve that
standard by 2016. The meaningful use criteria requires in part that EHRs must include one
clinical decision support rule applied to a specialty or high-priority condition, as well as the
ability to track compliance with that rule.
Given that only CPOE systems with an integrated CDSS meet the HITECH criteria for
meaningful use, this brief update review will assess the state of the evidence regarding the
effectiveness, cost, and implementation issues related to CPOE systems with CDSS capabilities
(CPOE+CDSS).
The 2001 Making Health Care Safer report reviewed evidence on the effectiveness of
CPOE+CDSS, as well as isolated CDSS, at improving medication safety.9 The review defined
level 1 outcomes as adverse drug events (ADEs), and level 2 and 3 outcomes as medication
errors and change in prescribing practices, respectively. These definitions were used in order to
distinguish the effects of CPOE and CDSS on clinical outcomes (e.g., preventable ADEs) and
surrogate outcomes that may not have caused patient harm (e.g., medication errors).
The 2001 review included four studies of CPOE+CDSS, three of which were conducted at
the same academic medical center. These studies all found improvement in level 2 and 3
outcomes, but did not document a reduction in preventable ADEs. All of the studies included in
the report evaluated homegrown, institution-specific systems (as opposed to commercial
system purchased from vendors) and often focused on safety of a specific medication or
medication class (such as antibiotics). These factors limit the generalizability of these studies to
general ADE prevention and to other institutions. The review also noted the high cost and
complex implementation issues that accompany CPOE+CDSS, stating, CPOE requires a very
large up-front investment with more remote, albeit substantial returns. In addition, CPOE affects
clinicians and workflow substantially. Its complexity requires close integration with multiple
systems, such as the laboratory and pharmacy systems. Failure to attend to the impact of such a
large-scale effort on organizational culture and dynamics may result in implementation failure
(page 71).
The overall conclusion of the review was that CPOE+CDSS can lower the rates of
medication errors and can promote appropriate prescribing, but evidence of its impact on actual
patient-level harm was limited. This conclusion proved to be somewhat controversial. In
response, followup commentaries10 took issue with the fact that CPOE+CDSS received only a
medium strength of evidence recommendation in the report. The objection to this conclusion
centered around the argument that CPOE+CDSS are difficult and costly to evaluate in controlled
trials, particularly when evaluating a relatively infrequent single adverse event such as an ADE,
and that the face validity of such systems indicated that proof of clinical benefit should not be
required before wider adoption. The evidence reports authors responded that using evidence to
evaluate the effectiveness and generalizability of these patient safety practices was essential to
their appropriate prioritization and application.11
481
What Have We Learned About Computerized Provider Order

Entry and Clinical Decision Support Systems Since the Making
Health Care Safer Report?
Evidence for the Effectiveness of Computerized Provider Order Entry
With Clinical Decision Support Systems
Three systematic reviews published since 2008 have evaluated the effectiveness of
CPOE+CDSS at preventing ADEs. Wolfstadt and colleagues12 identified ten trials of
CPOE+CDSS, nine of which were conducted in the inpatient setting and one in the ambulatory
setting. The majority of these studies evaluated homegrown systems, and none were randomized
controlled trials. The review concluded that CPOE+CDSS are effective at reducing ADEs, with
five of the ten studies finding a statistically significant reduction in ADEs and four others
reporting a nonsignificant improvement.
Schedlbauer and colleagues13 identified 20 studies that evaluated a total of 27 forms of
electronic reminders and prompts embedded in CPOE systems. Only four of these studies were
randomized controlled trials (RCTs). The authors classified the alerts as basic (including only
information about allergies, drug-drug interactions, and default dosing), advanced (including
alerts targeting errors of omission and patient-specific dosing and safety guidelines), and
complex (including features of both basic and advanced systems). This review also found that
CPOE+CDSS are effective, with 23 of the 27 reminder types demonstrating improvement in
targeted outcomes. However, only four of these studies evaluated clinical adverse drug events;
three of them did find statistically significant reductions in preventable ADEs. Although the four
studies of complex alert systems all found significantly improved prescribing practices, only
one of these studies found a statistically significant improvement in preventable ADEs.
Van Rosse and colleagues review14 specifically focused on the effectiveness of
CPOE+CDSS in adult and pediatric intensive care units, where patients are particularly
vulnerable to ADEs. The 12 observational studies they identified collectively demonstrated
reductions in medication prescribing errors; however, no overall effect was found on ADEs or
mortality rates.
These three reviews almost exclusively identified studies conducted in the inpatient setting.
These studies generally included relatively small patient populations, often within a single
hospital or health system, and relatively short intervention periods. The use of CPOE+CDSS in
the ambulatory care setting is less extensively studied. Two recent studies15,16 conducted in large,
community-based practice settings found that mandatory use of CPOE+CDSS achieved
reductions in prescribing errors, but not clinical ADEsmirroring the evidence from the
inpatient setting.
Taken together, these reviews indicate that hospitals implementing CPOE+CDSS cannot
assume that these systems will reliably reduce clinical ADEs. Insight into the mechanism of this
(lack of) effect was provided by a systematic review by Shojania and colleagues17 that evaluated
the effect of electronic point-of-care reminders on changing physician behavior. This quantitative
review found that reminders overall resulted in only small changes in provider behavior, a degree
of behavior change that was generally insufficient to yield clinically significant improvement.
The authors further concluded that evidence was insufficient to identify key features of systems
that could result in clinically significant changes in provider behavior, as the subset of studies
reporting the largest effects all originated from a single hospital (Brigham and Womens
482
Hospital in Boston). The conclusions regarding CPOE+CDSS in the 2001 edition of Making
Health Care Safer thus appear to stand largely unchanged a decade later.
Computerized Provider Order Entry With Clinical Decision Support

Systems Can Affect Workflow and Patient Care Adversely
The growth in use of CPOE+CDSS has yielded a more nuanced appreciation of the
unintended consequences of the technology. These unintended consequences were classified in a
seminal 2006 article:18
More or new work for clinicians
Unfavorable workflow issues
Never-ending system demands
Problems related to persistence of paper orders
Unfavorable changes in communication patterns and practices
Negative feelings toward the new technology
Generation of new types of errors
Unexpected changes in an institutions power structure, organizational culture, or
professional roles
Overdependence on the technology
Surveys of clinicians in settings where CPOE was recently implemented have confirmed that
clinicians perceive these unintended consequences to be common and to affect patient care
adversely.19 An illustration of this phenomenon was provided in a recent study20 that evaluated
the effect of a hard-stop warning that essentially prevented co-prescribing of the anticoagulant
warfarin and the antibiotic trimethoprim-sulfamethoxazolea combination associated with
serious bleeding risks. The warning was abandoned after 6 months because four patients
experienced delays in needed treatment with one of the drugs. Another potential effect of these
electronic programs is the potential to create more workarounds, or bypassing a recognized
problem as a temporary solution, that may then lead to future systems failures.
One particular problem, alert fatigue, was discussed in the original Making Health Care
Safer report and has been further studied over the past decade. Alert fatigue refers to the
tendency of clinicians to ignore warnings that are not perceived as being clinically significant,
which may result in inappropriately ignoring critical alerts. Alert fatigue is now a welldocumented phenomenon in both the inpatient and ambulatory settings,21 as most existing
CPOE+CDSS systems lean toward providing comprehensive alerts for all potential drug safety
problems rather than focusing alerts on the most clinically significant problems. In one study of
an outpatient CPOE+CDSS system,22 more than 300 alerts were required to prevent one ADE,
and another study found that clinicians ignored 75 percent of even critical drug-drug
interaction alerts.23
CPOE+CDSS systems thus have the potential to affect clinician workflow and patient care
adversely. These unintended consequences have forced health care systems to pay very close
attention to how this technology is configured and implemented.
Implementation and Costs

Implementation issues around CPOE+CDSS chiefly involve two aspects: the technical
specifications of how the system is configured to minimize alert fatigue and other workflow-
483
related consequences, and how the transition from paper-based systems to an electronic system is
handled.
Some studies have successfully tailored alerts by incorporating patient-specific
characteristics into algorithms for displaying drug warnings. Seidling and colleagues24
implemented a tailored alert system at a German hospital and found a reduction in prescribing
errors; this study is notable because providers accepted nearly 25 percent of warnings, much
higher than rates generally reported in the literature. However, efforts to tailor drug warnings are
currently limited by the lack of standardized consensus definitions for drug-drug interactions that
are likely to lead to ADEs and unclear malpractice implications for users and manufacturers of
CDSS systems25 should patients be harmed if an alert is not provided. Recent commentaries25,26
have called for better guidance and legal protections to allow greater tailoring of alerts to
minimize alert fatigue and improve the safety performance of decision support systems, and a
recent consensus conference27 identified the key issues in developing more effective alert
mechanisms.
At the institutional level, it is clear that careful attention must be paid to the implementation
process of CPOE+CDSS, particularly with regard to how systems are integrated into existing
clinician workflow. Unfortunately, no clear consensus exists on the optimal implementation
methods in either the hospital or ambulatory setting. The CDSS five rights provides a
framework on implementation to improve medication management and outcomes by linking
each intervention with a specific objective. This framework includes each right be addressed to
ensure an optimal CDS program: right information, to the right person, in the right format,
through the right channel, at the right point in workflow.28 The Agency for Healthcare Research
and Quality has published the online Guide to reducing unintended consequences of electronic
health records (www.ucguide.org), and several case studies of implementation of commercial
CPOE+CDSS systems have also been published29-31 These reports likely provide the most useful
guides for decisionmakers regarding implementation issues.
We did not identify any formal cost-effectiveness analyses of CPOE+CDSS published in the
past 5 years. Individual institutions with homegrown CPOE+CDSS systems have estimated
considerable cost savings32 due to ADE prevention and optimizing medication use, but these data
may not be generalizable to other settings and systems. A 2009 review of the costs and benefits
of health information technology33 found a paucity of meaningful data on the cost-benefit
calculation of actual IT implementation, and concluded, although there is some empirical
evidence to support the positive economic value of an EHR system and the component parts of
EHRs, projections of large cost savings assume levels of health IT adoption and interoperability
that we are nowhere near achieving.

The 2001 Making Health Care Safer report concluded that evidence for the safety benefits
of CPOE (with or without CDSS) was only moderate. Unfortunately, a decade of wider
CPOE+CDSS implementation and intensive research does not appear to change that conclusion.
CPOE+CDSS appear to be effective at reducing medication prescribing errors, but there is no
clear evidence that these systems reduce clinical ADEs in either the inpatient or outpatient
setting. Reminder systems can stimulate provider behavior change to improve appropriate care,
although these benefits may be relatively small.
Significant progress has been made in understanding the unintended consequences and
potential for adverse events associated with CPOE+CDSS implementation, but a lack of
484
consensus exists on implementation processes, especially for health systems implementing

commercial applications. Therefore, while the HITECH act and related measures provide health
care organizations with considerable incentive to implement health IT, the actual process of
implementation may continue to consist of exercises in trial and error, and the return on
investment in health IT systems is not predictable. Health information technology certainly has
great potential to improve patient safety, but for the specific example of CPOE+CDSS, it appears
that potential remains unrealized. A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Low-tomoderate
Evidence or
Potential for
Harmful
Unintended
Consequences
Low-to-moderate
Estimate of
Cost
High
How Much do We
Moderate/Difficult
References
1.
Abramson EL, Bates DW, Jenter C, et al.

Ambulatory prescribing errors among
community-based providers in two states. J
Am Med Inform Assoc.22140209.
2.
Bates DW, Leape LL, Petrycki S. Incidence

and preventability of adverse drug events in
hospitalized adults. J Gen Intern Med.
1993;8(6):289-94.8320571.
3.
Lazarou J, Pomeranz BH, Corey PN.

Incidence of adverse drug reactions in
hospitalized patients: a meta-analysis of
prospective studies. JAMA.
1998;279(15):1200-5.9555760.
4.
Gandhi TK, Weingart SN, Borus J, et al.

Adverse drug events in ambulatory care. N
Engl J Med. 2003;348(16):155664.12700376.
5.
Gurwitz JH, Field TS, Harrold LR, et al.

Incidence and preventability of adverse drug
events among older persons in the
ambulatory setting. JAMA.
2003;289(9):1107-16.12622580.
6.
Nebeker JR, Hoffman JM, Weir CR, et al.

High rates of adverse drug events in a highly
computerized hospital. Arch Intern Med.
2005;165(10):1111-6.15911723.
7.
DesRoches CM, Rosenbaum SJ. Meaningful

use of health information technology in U.S.
hospitals. N Engl J Med.
2008;362(12):1153-5.20335600.
485
8.
Jha AK, DesRoches CM, Campbell EG, et

al. Use of electronic health records in U.S.
hospitals. N Engl J Med.
2009;360(16):1628-38.19321858.
9.
Kaushal R BD. Computerized Physician

Order Entry (CPOE) with Clinical Decision
Support Systems (CDSSs). Chapter 6 in
Shojania K, Duncan B, McDonald K,
Wachter RM, eds. Making Health Care
Practices. 2001;Evidence
Report/Technology Assessment No. 43;
AHRQ publication( 01-E058).
10.
Leape LL, Berwick DM, Bates DW. What

practices will most improve safety?
Evidence-based medicine meets patient
safety. JAMA. 2002;288(4):5017.12132984.
11.

et al. Safe but sound: patient safety meets
evidence-based medicine. JAMA.
2002;288(4):508-13.12132985.
12.
Wolfstadt JI, Gurwitz JH, Field TS, et al.

The effect of computerized physician order
entry with clinical decision support on the
rates of adverse drug events: a systematic
review. J Gen Intern Med. 2008;23(4):4518.18373144.
13.
Schedlbauer A, Prasad V, Mulvaney C, et al.

What evidence supports the use of
computerized alerts and prompts to improve
clinicians prescribing behavior? J Am Med
Inform Assoc. 2009;16(4):531-8.19390110.
14.
van Rosse F, Maat B, Rademaker CM, et al.

The effect of computerized physician order
entry on medication prescription errors and
clinical outcome in pediatric and intensive
care: a systematic review. Pediatrics.
2009;123(4):1184-90.19336379.
15.
Devine EB, Hansen RN, Wilson-Norton JL,

et al. The impact of computerized provider
order entry on medication errors in a
multispecialty group practice. J Am Med
Inform Assoc. 2010;17(1):78-84.20064806.
16.
Kaushal R, Kern LM, Barron Y, et al.

Electronic prescribing improves medication
safety in community-based office practices.
J Gen Intern Med. 2010;25(6):530-6.
20186499.
17.
Shojania KG, Jennings A, Mayhew A, et al.

Effect of point-of-care computer reminders
on physician behaviour: a systematic review.
Cmaj. 2010;182(5):E216-25.20212028.
18.
Campbell EM, Sittig DF, Ash JS, et al.

Types of unintended consequences related to
computerized provider order entry. J Am
Med Inform Assoc. 2006;13(5):54756.16799128.
19.
Ash JS, Sittig DF, Poon EG, et al. The

extent and importance of unintended
consequences related to computerized
provider order entry. J Am Med Inform
Assoc. 2007;14(4):415-23.17460127.
20.
Strom BL, Schinnar R, Aberra F, et al.

Unintended effects of a computerized
physician order entry nearly hard-stop alert
to prevent a drug interaction: a randomized
controlled trial. Arch Intern Med.
2010;170(17):1578-83.20876410.
21.
Isaac T, Weissman JS, Davis RB, et al.

Overrides of medication alerts in ambulatory
care. Arch Intern Med. 2009;169(3):30511.19204222.
22.
Weingart SN, Simchowitz B, Shiman L, et

al. Clinicians assessments of electronic
medication safety alerts in ambulatory care.
Arch Intern Med. 2009;169(17):162732.19786683.
23.
Payne TH, Nichol WP, Hoey P, et al.

Characteristics and override rates of order
checks in a practitioner order entry system.
Proc AMIA Symp. 2002:602-6.12463894.
486
24.
Seidling HM, Schmitt SP, Bruckner T, et al.

Patient-specific electronic decision support
reduces prescription of excessive doses.
Qual Saf Health Care. 2010;19(5):e15.
20427312.
25.
Greenberg M, Ridgely MS. Clinical decision

support and malpractice risk. JAMA.
2011;306(1):90-1.21730245.
26.
Kesselheim AS, Cresswell K, Phansalkar S,

et al. Clinical decision support systems
could be modified to reduce alert fatigue
while still minimizing the risk of litigation.
Health Aff (Millwood). 2011;30(12):23107.22147858.
27.
Riedmann D, Jung M, Hackl WO, et al.

How to improve the delivery of medication
alerts within computerized physician order
entry systems: an international Delphi study.
J Am Med Inform Assoc. 2011;18(6):7606.21697293.
28.
Osheroff J. Ed. Improving Medication Use

and Outcomes with Clinical Decision
Support: A Step-by-Step Guide. Chapter 1
Approaching Clinical Decision Support in
Medication Management. HIMSS Clinical
Decision Support Guidebook Series. United
States Pharmacopeial. 2009.
29.
Longhurst CA, Parast L, Sandborg CI, et al.

Decrease in hospital-wide mortality rate
after implementation of a commercially sold
computerized physician order entry system.
Pediatrics. 2010;126(1):14-21.20439590.
30.
Weir CR, McCarthy CA. Using

implementation safety indicators for CPOE
implementation. Jt Comm J Qual Patient
Saf. 2009;35(1):21-8.19213297.
31.
Wetterneck TB, Walker JM, Blosky MA, et

al. Factors contributing to an increase in
duplicate medication order errors after
CPOE implementation. J Am Med Inform
Assoc. 2011;18(6):774-82.21803925.
32.
Kuperman GJ, Gibson RF. Computer

physician order entry: benefits, costs, and
issues. Ann Intern Med. 2003;139(1):319.12834316.
33.
Goldzweig CL, Towfigh A, Maglione M, et

al. Costs and benefits of health information
technology: new trends from the literature.
Health Aff (Millwood). 2009;28(2):w28293.19174390.
Chapter 42. Tubing Misconnections: Brief Review (NEW)

Kelley Tipton, M.P.H.

Tubing lines connect patients to devices and allow for the delivery of medication or nutrition
therapy. Liquid-to-liquid misconnections can introduce fluids, medications, or nutritional
formulas into the wrong body part.1 Gas-to-liquid misconnections can deliver gas into the
vasculature or liquid into the respiratory tract.2 The consequences of tubing misconnections
include severe patient harm and death.
Although misconnections have been recognized as a serious problem for years, incidents are
still common.3 One of the first publications of a tubing misconnection (enteral) was in 1972
reporting the inadvertent intravenous (IV) administration of breast milk.2 From January 1, 2000
to December 31, 2006 the United States Pharmacopeia (USP) collected a total of 24 reports of
tubing misconnections of an enteral feeding formula, other solutions, or medications intended for
the feeding tube but administered via the wrong route.2 Eight (33%) of the reports resulted in
permanent injury, life threatening situation, and/or death.2 Between January 2008 and September
2009, the Pennsylvania Patient Safety Authority received 36 reports of tubing misconnections,
with the incidents ranging from near misses to serious events; 35 were liquid-to-liquid events and
one was liquid-to-gas.3
Factors that contribute to misconnections include lines that have been disconnected and need
to be reconnected, the use of adapters to permit connections that are meant to be impossible, luer
fittings (male and female components) allowing a variety of lines to be connected with no
indication that the connection might be inappropriate, and line connectors with similar features.1
Luer fittings have been listed by ECRI Institute as one of the top 10 technology hazards for
20111
Human error is one factor resulting in tubing misconnections. Clinicians are often under time
pressure, experience rotating shift work and fatigue, and attempt to use short-term recall for large
amounts of information.2 Inadequate training and lighting, moving patients from one setting or
service to another, and using tubes or catheters for unintended purposes (e.g., IV extension
tubing for epidurals, irrigation, etc.) can also result in tubing misconnections.2

Engineering (i.e., design) controls and the implementation of administrative controls
(hospital policies and work practices) are the two basic means that can minimize misconnections.
Human error is inevitable and may cause fatalities when tubing lines are misconnected.
Therefore, the Joint Commission has urged product manufacturers to implement appropriate
designed incompatibility to prevent dangerous misconnections of tubes and catheters.2
According to ECRI Institute, engineering controls to reduce misconnections fall into three
categories: connectors with physical incompatibilities (leaving users with little or no choice but
to make the correct connection), connectors with locking mechanisms (to prevent accidental
disconnection), and connectors with a distinct physical appearance (size, shape, or color).4 In
2000, The European Standards Organization created a standard using a graduated catheter tip on
the distal end of enteral tubing and standardized the proximal end of the tubing by replacing the
487
spike with a screw-type connection.5 In 2007, a catheter-adapter tip (i.e., Christmas-tree

connector) was standardized in the United States for the distal end of enteral feeding tubing to
prevent staff from plugging it into IV equipment.5
Until appropriately designed tubing is developed and consistently supplied, it is
recommended that hospitals incorporate both design solutions and work practices to decrease and
eliminate any tubing misconnections.1,6 Organizations such as The Joint Commission, The
International Organization for Standardization, and ECRI Institute have published tubing
misconnection risk reduction strategies for clinical and non-clinical staff (e.g.,
Clinical/biomedical engineering, risk management, purchasing, etc.), and specific to general
tubing and enteral tubing. We have listed the general tubing misconnection risk reduction
strategies below and organized the list by the target audience.
Clinical Staff1,7
Trace all lines back to their point of origin to verify that correct connections are made
Recheck connections and trace all lines to their point of origin after the patients arrival
to a new care area or as part of a handoff process
Do not force connections
Only use adapters in accordance with hospital policy for a specific indication
Label certain high-risk catheters as to the type of catheter (e.g., epidural, intrathecal)
Route lines with different purposes in unique and standardized directions (e.g., IV line
towards patients head, enteral feeding line towards patients feet)
Identify and manage conditions that may contribute to worker fatigue, which could result
in inattentiveness when making connections
Non-Clinical Staff1,7
Provide regular misconnection prevention education to all personnel working in the

patient care environments (e.g., explain the need to request help rather than attempting to
disconnect or reconnect lines).
Assess the need for adapters throughout the facility, and establish policies to limit or
restrict their routine use
Revise and/or establish purchasing policies that include, when possible, purchasing
equipment with misconnection safeguards (e.g., avoid purchasing nonintravenous
equipment)
The risk reduction strategies suggested for enteral feeding misconnections for clinical and
non-clinical staff include the following.
Clinical Staff1,7
Do not use standard luer syringes for oral medications or enteral feedings
Do not modify or adapt IV or enteral feeding devices
Route lines with different purposes in unique and standardized directions (e.g., route IV
lines towards patients head, route enteral feeding lines towards patients feet)
Identify and manage conditions that may contribute to worker fatigue
Review identification labels before administering solutions to ensure that the intended
delivery route is correct
488
Placing labels with warnings - WARNING for Enteral Use Only Not for IV Use
Non-Clinical Staff1,7
Ensure that an adequate number of distinctly labeled enteral pumps are purchased to
reduce or eliminate the use of infusion pumps for enteral administration to adult patients
Reinforce existing purchasing policies that mandate purchasing only enteral feeding sets
that are incompatible with female luer connectors
When possible purchase only non-IV compatible enteral feeding containers
Secure enteral administration sets with enteral feeding containers (e.g., with rubber band)
or pre-attached sets from the manufacturer before sending them to the patient care unit
Perform pre-purchase evaluations of enteral feeding systems under the guidance of a
multidisciplinary task force before purchasing decisions are made
According to ECRI Institute, the single most important work practice solution for clinicians
is to trace all lines back to their origin before connecting or disconnecting any devices and
infusions.8 Additional strategies that may be useful include ensuring proper lighting when
making connections, contacting manufacturers to determine if luer fittings can be replaced with
different connector types, storing medications for different delivery routes in different locations,
and using a color-code system.1
Why Should the Patient Safety Practice Work?

Theoretically, the combination of engineering controls and a change in work practices will
prevent any tubing misconnections. The engineering controls have varying levels of
effectiveness.4 The forcing function is the most reliable approach since it leaves the user with
little or no choice but to make the correct connection.4 Other solutions will prompt users to make
the correct connection by identifying the appropriate connector size, etc.
Work practice solutions such as tracing lines back to the point of origin verifies that the
correct lines will be connected and ultimately avoids errors. Trainings provided by
manufacturers will help users understand the equipment and its safeguards. While trainings
provided by the facility will increase the awareness of clinical and non-clinical staff of
appropriate tubing misconnection policies and procedures (e.g., potential consequences), only
trained staff should reconnect disconnected lines.
What Are the Benefits of the Patient Safety Practice?

The hospital environment is filled with lines and cables connecting medical devices with
patients and can cause confusion when patients are being connected or reconnected to the lines.1
The most important benefit of implementing equipment design solutions and changes in work
practices is the reduction and elimination of severe patient harm or death as a result of tubing
misconnections.
Searches performed for this report identify three studies that implemented work practice and
engineering controls. Each facilitys needs varied and the PSP implemented was based on the
specific needs. The multidisciplinary staff at Beaumont Commercialization Center in Royal Oak,
Michigan identified and examined potential connector hazards, produced educational materials,
provided hands-on training, and revised equipment purchasing procedures and staff guidelines.9
Woods and Schultz (2006)10 reported the standardization of labeling procedures and
489
nomenclature used by materials management and clinical staff at Columbus Childrens

Hospital.10 In-house equipment was also assessed for potential to contribute to misconnections
and revisions were made for future purchasing policies. Both studies report the elimination of
tubing misconnections. A third study by Lawton (2010)11 assessed the use of a new non-luer
device and a color-coding system.11 Clinicians involved in the study report that the device would
benefit patient safety and would be willing to adopt the non-luer device with resolutions to
design concerns.
Although the solutions implemented in these studies varied, each facility used a combination
of equipment and work practice solutions to address the tubing misconnection problem.

Labeling and color-coding tubing lines have been suggested as ways to reduce
misconnections. The Joint Commission and U.S. Pharmacopeia Medication Safety Forum have
acknowledged the potential in these methods. However, in the tubing misconnection Sentinel
Event Alert, The Joint Commission noted that users may rely on color-coding rather than
assuring a clear understanding of correct connections between tubes or catheters and body
inlets.7 Ongoing education and training about the color-coding system would be necessary for
staff on-site, as well as temporary or traveling staff. One study identified in our search results
mentioned a concern regarding the use of a color-coding system. Clinicians involved in a study
by Lawton (2010)11 were concerned with this system since visual discrimination is not possible
in poor lightning conditions.11 Another variable of consideration is the potential for various
facilities within the same geographic area to use different color-coding systems which could lead
to confusion, particularly for temporary or travel staff.7

What Contexts?
One of the first steps to implementing a risk reduction strategy is to formally assess the
current state of the work practices, equipment, and identify areas of improvement. One facility is
likely to have specific misconnection risks that require special attention compared with the needs
of another facility.4 A multidisciplinary task force should perform a formal risk assessment to
gauge the overall risks and the strategies that will reduce these risks.4
In 2004, the Beaumont Commercialization Center in Royal Oak, Michigan undertook a
program to address the problem of tubing misconnections.9 The task force collected all
equipment from the pediatric intensive care unit, developed human factors testing protocols, and
examined the connectors to identify potential misconnections and their severity.9 Along with
educational materials, employees had the opportunity to receive hands on training by performing
correct and incorrect connections via a training bear. This allowed staff to identify the right
and wrong way to connect tubing lines. As a result of this process, a corporate Misconnection
Prevention Policy was created and covered equipment purchases, technical and safety testing,
risk assessment, guidelines for clinical staff, and orientation and education.9 Also, Beaumont
reports that its misconnection rate dropped to zero.9
Lawton (2010) investigated the potential for and implementation of non-luer compatible
equipment for use in spinal procedures.11 The findings indicated that clinicians were enthusiastic
about the use of new well-designed devices for intrathecal chemotherapy, but not spinal (i.e.,
epidural) anesthesia as they were not convinced the devices will help tackle the problem of
490
spinal drug errors.11 The clinicians were also concerned with color-coding since visual
discrimination is not possible in poor lightning conditions, non-translucent devices preventing
the ability to see what a needle is doing and if it has reached the right place during injections, and
drug leakage.11 Overall, if the identified design issues are resolved, clinicians would be willing to
adopt the non-luer devices because they believe patient safety will benefit from
implementation.11
In 2004, the Columbus Childrens Hospital conducted a Health Failure Mode and Effects
Analysis (HFMEA) to identify the inherent risks of use and labeling of various enteral,
parenteral, and other tubing types in patient care and the potential for harm.10 Woods and Shultz
(2006) found three common themes causing all failure modes: non-standardized labeling of
tubing, lack of knowledge of nomenclature or alias, and inconsistent inventory.10 Labeling of
tubing with infused mechanisms happened 85% of the time in the pediatric intensive care unit,
53% of the time during surgery, and 93% of the time during interventional radiology. The risk
reduction methods subsequently implemented involved the standardization of the labeling
process throughout the organization (e.g., color of labels, content on label, size and placement of
tubing), and the development of an online pictorial catalog listing all available supplies by
category and the nomenclature used by materials management and common names used by
clinical staff.10 The third risk reduction method conducted an inventory with the help from
clinical staff and materials management to identify currently used tubing, connectors that fit
properly and those that needed to be removed from practice, and devices that needed to be
purchased. According to the authors, several recommendations have been implemented and no
tubing misconnections have been reported.10
Another point of consideration when implementing the PSP for tubing misconnections is
cost. In 2008, Peter Angood, the chief patient safety officer and vice president of The Joint
Commission stated that the cost of acquiring new devices, identifying risky connections and
practices, and implementing training and testing will no doubt impact hospitals.12 However,
the cost of not making such changes could, of course, be much greater in terms of lives lost
erroneous connections between tubes and catheters can create catastrophic outcomes, even
death says Angood.12
Conclusions and Comments

Ideally, the combination of engineering controls and a change in work practices should
eliminate all tubing misconnections. Organizations such as The Joint Commission, The
International Organization for Standardization, and ECRI Institute have been consistent with
suggestions for solutions to reduce the risks of tubing misconnections. In general, suggestions for
work practice solutions include tracing lines back to the point of origin, rechecking connections
and tracing lines when moving a patient or when work shifts change, not forcing connections,
using appropriate adapters, labeling high-risk catheters, routing lines in different directions, and
addressing and managing conditions contributing to worker fatigue. Equipment design solutions
involve assessing the need for adapters, revising purchasing procedures, and educating clinical
staff on correct equipment use. Angood states, this area of healthcare is difficult to get under
control because the issues cut across several sectors within the industry. It will take time before a
coordinated approach occurs to address these issues. In the interim, organizations and all
practitioners must be highly vigilant about preventing misconnections.12 Chapter 31 reviews the
evidence for Human Factors and Ergonomics (HFE). The goal of HFE is to address physical,
491
cognitive, and organizational issues of devices. The importance of designing devices that help
reduce human error is discussed further in this chapter.
As seen in the previously mentioned studies, each facility will have specific needs and
inherent risks that require the implementation of different risk reduction strategies. Again, one of
the most important work practice solutions involves the tracing of lines back to the point of
origin. Regardless of the differences between facilities, it is recommended that facilities perform
a risk assessment to determine their ultimate needs in equipment changes and work practice
policy updates. A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low
How Much do We
Moderate/Not Difficult
References
1.
ECRI Institute. Health devices. Health

Devices 2006 Mar;35(3):79-107.
2.
Guenter P, Hicks RW, Simmons D, et al.

Enteral feeding misconnections: a
consortium position statement. Jt Comm J
Qual Patient Saf 2008 May;34(5):285-92,
245. PMID: 18491692.
3.
4.
5.
6.
7.
Tubing misconnections: making the

connection to patient safety. PA Patient Saf
Advis 2010 Jun;7(2):41-5.
8.
ECRI Institute. Fixing bad links. Preventing

misconnections in your hospital. Health
Devices 2009 Jul;38(7):220-7.
Joint Commission on Accreditation of

Healthcare Organizations, USA. Tubing
misconnections--a persistent and potentially
deadly occurrence. Sentinel Event Alert
2006 Apr 3;(36):1-3. PMID: 16594109.
9.
ECRI Institute. Healthcare risk control: risk

analysis. Supplement A, Special Clinical
Services 14. Plymouth Meeting (PA):
ECRI Institute; 2007 Jul. Preventing
misconnections of lines and cables.
ECRI Institute. Fixing bad links. Preventing

misconnections in your hospital. Health
Devices 2009 Jul;38(7):220-7.
PMID: 20848951.
10.
Kimehi-Woods J, Shultz JP. Using HFMEA

to assess potential for patient harm from
tubing misconnections. Jt Comm J Qual
Patient Saf 2006 Jul;32(7):373-81. PMID:
16884124.
11.
Lawton R. Testing new devices to help

prevent misconnection errors in health
care. J Health Serv Res Policy 2010 Jan;15
Suppl 1:79-82. PMID: 20075137.
12.
Cruise C. Education, evaluation are key...

sentinel event alert compels hospitals to
address tubing and catheter misconnections.
Biomed Instrum Technol 2008 NovDec;42(6):469-70. PMID: 19012462
Pratt N. Tubing misconnections: a perilous

design flaw. Mater Manag Health Care 2006
Nov;15(11):36-9. PMID: 17191554.
Simmons D, Phillips MS, Grissinger M, et
al. Error-avoidance recommendations for
tubing misconnections when using Luer-tip
connectors: a statement by the USP Safe
Medication Use Expert Committee. Jt
May;34(5):293-6, 245. PMID: 18491693.
492
Chapter 43. Limiting Individual Providers Hours of Service:

Brief Update Review
Sumant R. Ranji, M.D.; Robert M. Wachter, M.D.
Introduction
Long and unpredictable work hours have been a staple of medical training for centuries, but
the effects of fatigue among residents on patient safety garnered little attention until March 1984,
when a young woman died at a teaching hospital in New York. Her death was attributed in part
to a medication prescribing error made by residents in the midst of a 36-hour shift. This seminal
event led to the passage of regulations in the State of New York limiting residents shift duration
to 24 consecutive hours and overall work week to 80 hours. However, at the time of the original
Making Health Care Safer1 report in 2001, widespread violations of the New York regulations
were common, and it was not unusual for residents still to work 36 hour shifts and over 100
hours per week in other states as well).
A considerable body of evidence from health care and other industries2 links acute and
chronic sleep deprivation to impaired cognitive performance. Some studies have also shown that
sleep deprivation can affect psychomotor skills.3 Working extended duration shifts can be
harmful for both clinicians and patients. Studies have shown that residents who work more than
16 consecutive hours have an increased risk of motor vehicle accidents after their shift4 and of
suffering a needlestick injury during their shift.5 Working over 16 consecutive hours in the
intensive care unit has been shown to result in interns committing more diagnostic and
therapeutic errors.6 Among nurses, shift duration of greater than 12 hours is also associated with
a significantly increased risk of committing errors.7 In response to these and other data, the
Accreditation Council for Graduate Medical Education (ACGME) implemented formal work
hour restrictions for resident physicians in 2003, and made these regulations even more stringent
in 2011.8
Despite these known risks, the extent to which patients are harmed by clinician fatigue is
difficult to determine. Fatigue on the part of an individual provider may only be one of several
latent causes of a preventable adverse event, especially in the complex hospital environment.
Few studies have attempted to directly address the connection between clinician fatigue and
adverse clinical outcomes. Two recent studies9,10 examined whether attending surgeons fatigue
was linked to an increased risk of complications, and reached conflicting results; one study10
found an increased risk of complications when the surgeon had the opportunity to sleep for less
than 6 hours the night prior to the procedure, but the other9 did not find the same association.
Despite the lack of hard data linking fatigue and complications, the traditional residency
work hours that existed prior to 2003 could not be justified from an educational, humanistic, or
patient safety standpoint. The implementation of regulations to reduce residents work hours
have resulted in fundamental changes to residency education over the past decade, and the
regulations effect on patient safety has been extensively studied.
The 2001 report reviewed the evidence linking sleep deprivation and fatigue to medical
errors, and reached four conclusions:
Sleep deprivation and disturbances of circadian rhythm lead to fatigue, decreased
alertness, and poor performance on standardized testing.
493
Although data from non-medical fields suggest that sleep deprivation leads to poor job
performance, this link has not yet been established in medicine.
Forward rather than backward shift rotation [i.e., progressing from day to evening to
night shifts, rather than the reverse], education about good sleep hygiene, and strategic
napping before or during shifts may reduce fatigue and improve performance. High face
validity, low likelihood of harm, and ease of implementation make these promising
strategies, although more evidence of their effectiveness in medicine is warranted.
Given that medical personnel, like all human beings, probably function suboptimally
when fatigued, efforts to reduce fatigue and sleepiness should be undertaken, and the
burden of proof should be in the hands of the advocates of the current system to
demonstrate that it is safe.
In this review, we assess the evidence that has accumulated since 2001 for the effect of
limiting individual providers hours of service on patient safety outcomes. Although other
countries have significantly more stringent regulations (for example, trainees in the European
Union are limited to 48 hours per week), the focus of this review will be on studies conducted in
the United States.
What Efforts Have Been Made To Reduce Clinician Work Hours?

Specific attempts have been made to reduce clinician work hours in order to improve safety
by minimizing fatigue. The vast majority of the research in this area pertains to resident
physicians in the United States. In 2003, the ACGME passed regulations intended to
significantly reduce work hours for trainees. These regulations included four principal
components:
A maximal limit of 80 hours worked per week
No more than 24 consecutive hours on duty (an additional 6 hours were allowed to ensure
safe transitions of care, meaning that residents could work a maximum of 30 consecutive
hours)
On-call frequency of no more than once every 3rd night
At least 4 days off per month
These regulations became effective on July 1, 2003 (some specialties received partial
exemption from the regulations). Since that date, the effect of the regulations has been
intensively studied, and forms the largest body of evidence specifically addressing the patient
safety effects of reducing individual providers hours of service.
What is the Context for Current Efforts To Reduce Work Hours?

The Institute of Medicine issued a report11 in 2008 that took into account the initial data on
the effect of the 2003 ACGME regulations, as well as the evolving evidence base in the area of
fatigue and performance. The IOMs recommendations included the following:
Continued maximal limit of 80 hours per week
No more than 16 consecutive hours on duty, after which residents must be off duty
completely or provided 5 hours of protected sleep time
No more than 4 consecutive night shifts
1 full day off per week, and 1 full weekend off per month
494
The revised resident duty hour regulations published by the ACGME in 2010, and
implemented July 1, 2011, did not incorporate all of the IOMs recommendations. The 2011
regulations have four key components:
Continued 80-hour work week limit
No more than 16 consecutive hours on duty for first-year residents only. Second-year and
more senior residents can work 24 hours on duty, with an additional 4 hours allowed for
transitions of care
No more than 6 consecutive night shifts
Continued minimum of 4 days off per month and on-call frequency of no more than once
every 3rd night
Although excessive work hours are linked to errors among nurses, regulation of nurses work
hours is less uniform. Currently, 16 states do restrict mandatory overtime for nurses, but many
nurses still routinely work more than 12 hours per shift.7 There are also no regulations on
working hours for practicing physicians, despite some data indicating that many practicing
physicians work schedules that would be prohibited were they still residents.12
What Have We Learned About Limiting Physician Work Hours?

Recent Reviews and Systematic Evaluations
Multiple systematic reviews have addressed the patient safety effects of reducing shift length
for residents. One systematic review that included only studies of the effect of the 2003 ACGME
regulations13 identified 20 studies that assessed mortality and 24 studies that assessed other
patient safety outcomes before and after implementation of the regulations. Meta-analysis of the
mortality studies did show a statistically significant decline in mortality after 2003 (OR 0.9, 95%
CI 0.84 0.95), which was consistent in studies examining either medical or surgical patient
populations. However, considerable unexplained heterogeneity was present (I2=83%), and the
subset of studies that used a contemporaneous control group of non-teaching hospitals generally
did not find a difference in mortality. The authors acknowledged that as they were unable to
control for secular trends or changes in patient characteristics, the mortality improvement could
be due to overall improvement in the quality of care during the time period studied. The studies
examining patient safety outcomes yielded mixed results, with no clear pattern of improvement
or worsening across studies.
Another review14 identified 36 studies that examined the association between reduced
trainees working hours and patient outcomes. This study also included studies performed
outside the U.S.. The conclusions were largely similar to that of the previous review: Mortality
and patient safety outcomes appeared unchanged after implementation of duty hour limits. Both
reviews found that limiting work hours appeared to improve residents quality of life.
The question of why patient safety outcomes have not improved after reducing resident work
hours is a subject of intense debate. Reduced shift length almost certainly led to greater
discontinuity among providers, and the resultant handoffs of care may have had deleterious
effects on patient safety. Adherence to work hour limitations was (and is) likely suboptimal.15,16
In addition, studies have shown that changing residents work schedules to meet the regulations
did not actually result in residents sleeping more.17 Finally, although resident quality of life
improved, in some studies objective measurements of burnout and depression among residents
495
did not change.18 Burnout and depression are themselves linked to impaired job performance,
independent of acute or chronic fatigue.19
The 2003 duty hour regulations still allowed all residents to work a maximum of 30
consecutive hours. These extended duration shifts are still longer than those allowed in virtually
any other industry, and studies have found an association between working more than 16
consecutive hours and an increased risk of self-reported errors and attentional failures.20 A 2010
systematic review21 that identified 13 studies in which shift length for clinicians was
purposefully reduced found consistent evidence among the higher-quality studies that both
objectively-measured and self-reported errors decreased after shift length reduction. One
particularly high quality study6 found a significant reduction in serious medical errors for
medical interns assigned to work a 16-hour shift in the intensive care unit, compared with interns
working a traditional 30-plus hour shift. However, the reviewers were unable to reach a firm
conclusion regarding the optimal shift length, due to heterogeneity between shift lengths used in
the primary literature.
Thus, the totality of the evidence on the 2003 ACGME duty hour regulations indicates that
reducing resident duty hours does not improveor worsenspatient safety or mortality. The
association between extended duration (>16 hour) shifts and adverse events ultimately was a
factor in the ACGMEs decision to enact a 16-hour shift length limit for first-year residents as
part of the 2011 regulations.
New Studies for Effectiveness of the Patient Safety Practices

As the ACGMEs latest regulations were implemented earlier this year, no further data are
available in addition to those summarized above.
Potential for Harm

The greatest potential harm of work hour regulations is an increase in adverse events due to
increased handoffs of care between providers. Although this association is certainly plausible,
and handoffs have unquestionably increased after both the 2003 and 2011 regulations, studies
have not specifically examined whether errors attributable to handoffs have increased after the
regulations were implemented.
The other oft-cited adverse consequence of duty hour reduction is decreased clinical
experience for trainees, limiting their ability to practice independently once training is
completed. Studies of the 2003 duty hour regulations generally did not find that objective clinical
experience worsened, when measured by criteria such as surgical case volume. However, both
faculty22,23 and residents24 have voiced concerns that duty hour regulations have actually
compromised their educational experience, and most residents25 appear unconvinced that further
duty hour reductions will improve either patient safety or their educational experience.
Costs and Implementation

Implementing the 2011 ACGME regulations is likely to be extremely costly for teaching
hospitals. A 2011 cost-effectiveness analysis26 estimated that implementing the new ACGME
regulations would cost teaching hospitals $1.6 billion if the decreased workload of interns was
replaced entirely by attending physicians, and $1.34 billion if interns were replaced by physician
extenders (nurse practitioners or physician assistants). A 7.2% decrease in preventable adverse
events would be required in order to make the regulations cost-neutral to society, but teaching
hospitals would still encounter considerable costs. The expenses associated with the need to
496
replace the housestaff workforce with alternative providers and the need to provide greater
supervision for residents by senior physicians is likely to be considerable.

Sleep deprivation and fatigue have clear deleterious consequences for patients and providers.
However, the most prominent effort to improve patient safety by reducing fatiguelimiting the
work hours of resident physicianshas not yielded the expected benefits. It is conceivable that
the 2003 ACGME duty hour regulations were simply not stringent enough, given that extended
duration shifts were still permitted and those shifts are associated with preventable adverse
events. Alternatively, it may be that advocates underestimated the complexity of the relationship
between duty hours and safety, or the detrimental impact of handoffs. The 2011 regulations
further restrict hours, particularly for first-year residents. The effects of these new regulations are
as yet unknown, and unfortunately, the existing evidence does not offer us great clarity regarding
the optimal work hour structure that would improve safety by decreasing clinician fatigue with
minimal potential for unintended consequences. A summary table is located below (Table 1).
Scope of the
Strength of
the PSP
Effectiveness
Common/Moderate
Low
Evidence or
Potential for
Harmful
Unintended
Consequences
Moderate (at
least)
Estimate of
Cost
High
How Much do We
Moderate/Difficult
Includes lack of
training time
References
1.

Wachter RM, Markowitz AJ. Making health
care safer: a critical analysis of patient
safety practices. Evid Rep Technol Assess
(Summ) 2001:i-x, 1-668.
2.
Philibert I. Sleep loss and performance in

residents and nonphysicians: a meta-analytic
examination. Sleep 2005;28:1392-402.
3.
Weinger MB, Ancoli-Israel S. Sleep

deprivation and clinical performance. In:
JAMA. United States; 2002:955-7.
4.
5.
Barger LK, Cade BE, Ayas NT, et al.

Extended work shifts and the risk of motor
vehicle crashes among interns. In: N Engl J
Med. United States: 2005 Massachusetts
Medical Society.; 2005:125-34.
Ayas NT, Barger LK, Cade BE, et al.
Extended work duration and the risk of selfreported percutaneous injuries in interns. In:
JAMA. United States; 2006:1055-62.
497
6.
Landrigan CP, Rothschild JM, Cronin JW,

et al. Effect of reducing interns work hours
on serious medical errors in intensive care
units. In: N Engl J Med. United States: 2004
Massachusetts Medical Society.; 2004:183848.
7.
Rogers AE, Hwang WT, Scott LD, Aiken

LH, Dinges DF. The working hours of
hospital staff nurses and patient safety.
Health Aff (Millwood) 2004;23:202-12.
8.
Nasca TJ, Day SH, Amis ES, Jr. The new

recommendations on duty hours from the
ACGME Task Force. In: N Engl J Med.
United States; 2010:e3.
9.
Chu MW, Stitt LW, Fox SA, et al.

Prospective evaluation of consultant surgeon
sleep deprivation and outcomes in more than
4000 consecutive cardiac surgical
procedures. In: Arch Surg. United States;
2011:1080-5.
10.
Rothschild JM, Keohane CA, Rogers S, et

al. Risks of complications by attending
physicians after performing nighttime
procedures. In: JAMA. United States;
2009:1565-72.
11.
Medicine. Io. Resident Duty Hours:

Enhancing Sleep, Supervision, and Safety.
Washington, DC: The National Academies
Press.; 2009.
12.
Anim M, Markert RJ, Wood VC, Schuster

BL. Physician practice patterns resemble
ACGME duty hours. In: Am J Med. United
States; 2009:587-93.
13.
Fletcher KE, Reed DA, Arora VM. Patient

safety, resident education and resident wellbeing following implementation of the 2003
ACGME duty hour rules. J Gen Intern Med
2011;26:907-19.
14.
15.
16.
17.
18.
19.
West CP, Tan AD, Habermann TM, Sloan

JA, Shanafelt TD. Association of resident
fatigue and distress with perceived medical
errors. JAMA 2009;302:1294-300.
20.
Barger LK, Ayas NT, Cade BE, et al. Impact

of extended-duration shifts on medical
errors, adverse events, and attentional
failures. In: PLoS Med. United States;
2006:e487.
21.
Reed DA, Fletcher KE, Arora VM.

Systematic review: association of shift
length, protected sleep time, and night float
with patient care, residents health, and
education. In: Ann Intern Med. United
States; 2010:829-42.
22.
Reed DA, Levine RB, Miller RG, et al.

Effect of residency duty-hour limits: views
of key clinical faculty. In: Arch Intern Med.
United States; 2007:1487-92.
23.
Cohen-Gadol AA, Piepgras DG,

Krishnamurthy S, Fessler RD. Resident duty
hours reform: results of a national survey of
the program directors and residents in
neurosurgery training programs.
Neurosurgery 2005;56:398-403; discussion
398-403.
24.
Landrigan CP, Barger LK, Cade BE, Ayas

NT, Czeisler CA. Interns compliance with
accreditation council for graduate medical
education work-hour limits. In: JAMA.
Vidyarthi AR, Katz PP, Wall SD, Wachter

RM, Auerbach AD. Impact of reduced duty
hours on residents educational satisfaction
at the University of California, San
Francisco. In: Acad Med. United States;
2006:76-81.
25.
Landrigan CP, Fahrenkopf AM, Lewin D, et

al. Effects of the accreditation council for
graduate medical education duty hour limits
on sleep, work hours, and safety. In:
Pediatrics. United States; 2008:250-8.
Drolet BC, Spalluto LB, Fischer SA.

Residents perspectives on ACGME
regulation of supervision and duty hours--a
national survey. N Engl J Med
2010;363:e34.
26.
Nuckols TK, Escarce JJ. Cost Implications

of ACGMEs 2011 Changes to Resident
Duty Hours and the Training Environment. J
Gen Intern Med 2011.
Moonesinghe SR, Lowery J, Shahi N,

Millen A, Beard JD. Impact of reduction in
working hours for doctors in training on
postgraduate medical education and
patients outcomes: systematic review. BMJ
2011;342:d1580.
Tabrizian P, Rajhbeharrysingh U, Khaitov S,
Divino CM. Persistent noncompliance with
the work-hour regulation. In: Arch Surg.
Fahrenkopf AM, Sectish TC, Barger LK, et

al. Rates of medication errors among
depressed and burnt out residents:
prospective cohort study. In: BMJ. England;
2008:488-91.
498
Part 3. Discussion
Chapter 44. Discussion
Introduction
Progress Since the 2001 Report
Over 2000 years ago, Hippocrates reminded physicians to, first, do no harm. In 1863,
Florence Nightingale wrote, It may seem a strange principle to enunciate as the very first
requirement in a hospital that it should do the sick no harm. Notwithstanding these
commonsensical admonitions, it was not until the turn of this century that a systematic effort to
improve patient safety began, catalyzed by the publication of the IOM report, To Err is
Human.1
The year following the publication of the IOM report, AHRQ commissioned a group of
investigators, led by the UCSF-Stanford EPC, to synthesize the worlds literature on PSPs, an
effort that culminated in the 2001 report, Making Health Care Safer.2 This report was widely
used by clinicians, safety workers, researchers, and policymakers, and it informed a variety of
other initiatives including the National Quality Forums Safe Practices list.
Since 2001, research in the patient safety field has exploded, with literally thousands of
published studies. In fact, some of todays popular safety practicesrapid response teams,
disclosure of errors to patients, or any of the checklistbased interventionshad barely been
invented at the time of the 2001 report.
In light of this maturation of the field, AHRQ asked a group of investigators, many of whom
were involved in producing the earlier report, to synthesize the vast amount of new information
on PSPs that has emerged since the release of Making Health Care Safer. Using a similar
method, which combined explicit criteria, detailed evidence reviews, and an international panel
of expert advisors, this report reviewed the evidence on 41 PSPs. In Table 1, below, we provide a
summary of this evidence, followed by a discussion of the evidence in the context of prior work
on patient safety, and then present priorities for adoption of PSPs
Table 1, Chapter 44. Summary table*
Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Practices Designed for a Specific Patient Safety Target
Adverse Drug Events
High-alert drugs: patient safety
Common/Moderate
practices for intravenous
anticoagulants;
in-depth review
Use of clinical pharmacists to
Common/Low
prevent adverse drug events;
brief review
The Joint Commissions Do Not
Common/Low
Use list; brief review
Smart infusion pumps; brief
Common/Low
review
Strength of
Evidence for
Effectiveness
of the PSPs
Evidence or
Potential for
Harmful
Unintended
Consequences
Estimate of
Cost
Implementation
Issues:
How Much do We
Know?/How Hard Is
it?
Low
Low-to-moderate
Low
Little/Moderate
Moderate-tohigh
Low
High
Little/Moderate
Low
Negligible
Low
Low
Low
Moderate
Little/Probably not
difficult
Moderate/Moderate
499
Table 1, Chapter 44. Summary table* (continued)

Infection Control
Barrier precautions, patient
isolation, and routine surveillance
for the prevention of healthcareassociated infections; brief
review
Interventions to improve hand
hygiene compliance; brief review
Reducing unnecessary urinary
catheter use and other strategies
to prevent catheter-associated
urinary tract infections; brief
review
Prevention of central lineassociated bloodstream
infections; brief review
Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Strength of
Evidence for
Effectiveness
of the PSPs
Evidence or
Potential for
Harmful
Unintended
Consequences
Estimate of
Cost
Implementation
Issues:
How Much do We
Know?/How Hard Is
it?
Common/Moderate
Moderate
Moderate
(isolation of
patients)
Moderate-tohigh
Moderate/Moderate
Common/Moderate
Low
Low
Low
Moderate/Moderate
Common/Moderate
Moderate-tohigh
Low
Low
Moderate/Moderate
Common/Moderate
Moderate-tohigh
Low
Low-tomoderate
Ventilator-associated
Common/High
pneumonia; brief review
Interventions to allow the reuse
Common/Low
of single use devices; brief
review
Surgery, Anesthesia, and Perioperative Medicine
Preoperative checklists and
Common/Moderate
anesthesia checklists to prevent
a number of operative safety
events, such as surgical site
infections and wrong site
surgeries; in-depth review
The use of ACS-NSQIP report
Common/High
cards and outcome
measurements to decrease
perioperative morbidity and
mortality; in-depth review
New interventions to prevent
Rare/Low
surgical items from being left
inside a patient; brief review
Moderate-tohigh
Low
Low
Low
Low-tomoderate
Low
Not difficult
bundle)-to-moderate
(understanding
organization culture
and context)
Moderate/Moderate
High
Negligible
Low
A lot/Moderate
Moderate-tohigh
Low
Moderate
Moderate/Moderate
Low
Negligible
Little
Operating room integration and

display systems, such as a
centralized display of
consolidated data; brief review
Low
Negligible
Low if it
simply
involves
more
frequent
manual
counting;
high if RFID
is used
Moderate
Common/Low-tohigh
500
A lot/Not difficult
Moderate/Moderate

Use of beta blockers to prevent

perioperative cardiac events;
brief review
Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Common/High
Use of real-time ultrasound

Common/Low-toguidance during central line
moderate
insertion to increase the
proportion correctly placed on
the first attempt; brief review
Safety Practices for Hospitalized Elders
Common/Low
prevent in-facility falls; in-depth
review
prevent in-facility delirium; indepth review
General Clinical Topics
Multicomponent initiatives to
prevent pressure ulcers; in-depth
review
Inpatient, intensive, glucose
control strategies to reduce
death and infection; in-depth
review
Interventions to prevent contrastinduced acute kidney injury; indepth review
Rapid-response systems to
prevent failure-to-rescue; indepth review
supported by clinical
pharmacists; in-depth review
Identifying patients at risk for
suicide; brief review
Strategies to prevent stressrelated gastrointestinal bleeding
(stress ulcer prophylaxis); brief
review
Strategies to increase
appropriate prophylaxis for
venous thromboembolism; brief
review
Preventing patient death or
serious injury associated with
radiation exposure from
fluoroscopy and computed
tomography through technical
interventions, appropriate
utilization, and use of algorithms
and protocols; brief review
Strength of
Evidence for
Effectiveness
of the PSPs
High evidence
harms may
equal or
exceed
benefits
High
High
Evidence or
Potential for
Harmful
Unintended
Consequences
High (death,
stroke,
hypotension, and
bradycardia)
Estimate of
Cost
Low
Implementation
Issues:
How Much do We
Know?/How Hard Is
it?
NA
Negligible
Low-tomoderate
A lot/Moderate
Moderate
Moderate/Moderate
Moderate
Moderate/Moderate
Common/Low
Moderate
Moderate
(increased use of
restraints and/or
sedation)
Low
Common/Moderate
Moderate
Negligible
Moderate
Moderate/Moderate
Common/Moderate
it doesnt help
High
(hypoglycemia)
Low-tomoderate
NA
Common/Low
Low
Negligible
Low
Common/High
Moderate
Low
Moderate
Moderate/Moderate
Common/Low
Moderate
Low
Moderate
Moderate/Moderate
Rare/High
Low
Low
Moderate
Little/Moderate
Rare/Moderate
Moderate
Moderate
(pneumonia)
Moderate
Common/Moderate
High
Moderate
(bleeding)
Low
Little/Moderate
Rare/High
Moderate
Negligible
Low
501

Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Common/Moderate
Strength of
Evidence for
Effectiveness
of the PSPs
Ensuring documentation of
Moderate
patient preferences for lifesustaining treatment, such as
advanced directives; brief review
Increasing nurse-to-patient
Common/High
Moderate
staffing ratios to prevent death;
in-depth review
Practices Designed To Improve Overall System/Multiple Targets
Increasing nurse-to-patient staff
Common/High
Low
ratios to prevent falls, pressure
ulcers, and other nursing
sensitive outcomes (other than
mortality); in-depth review
Incorporation of human factors
Not assessed
Potentially
and ergonomics in the design of
systematically,
applicable to all
health care practices by hiring an patient safety
but moderateexpert or training clinicians in
to-high
problems
human factors; in-depth review
evidence for
some specific
applications
Promoting engagement by
Common
Emerging
patients and families to reduce
practice (few
adverse events (such as patients
studies
encouraging providers to wash
available)
their hands); in-depth review
Interventions to promote a
Common/Low-toLow
culture of safety; in-depth review
high
Evidence or
Potential for
Harmful
Unintended
Consequences
Low
Estimate of
Cost
Low
Implementation
Issues:
How Much do We
Know?/How Hard Is
it?
Moderate/Moderate
Low
High
A lot/Not difficult
Low
High
A lot/Not difficult
Negligible
Moderate
A lot/Moderate
Uncertain
Low
Little/Moderate
Uncertain
Moderate/Not difficultto-moderate (varies

with intervention)
Varies
Emerging
practice (few
studies
available)
Low
Uncertain
Lowtomoderate
(varies)
Varies
Negligible
High
Moderate/Difficult
Low
Negligible
Moderate-tohigh
Little/Difficult
Uncertain
Moderate
Moderate
Common/Moderate
Moderate-tohigh for
specific topics
Moderate
Negligible
Low
Common/High
Moderate
Low
Moderate
Common/Moderate
Low-tomoderate
Low-to-moderate
High
Moderate/Moderateto-difficult
Moderate/Difficult
Common/Moderate
Low
Low
Low
Patient safety practices targeted

at diagnostic errors; in-depth
review
Common/High
Monitoring patient safety

problems; in-depth review
Interventions to improve care
transitions at hospital discharge;
in-depth review
Use of simulation-based training
and exercises; in-depth review
Common/Low-tohigh
Common/Moderate
Obtaining informed consent from

patients to improve patient
understanding of potential risks
of medical procedures; brief
review
Team-training in health care;
brief review
Computerized provider order
entry (CPOE) with clinical
decision support systems
(CDSS); brief review
Interventions to prevent tubing
misconnections; brief review
Common/Moderateto-high
502

Scope of the
Problem Targeted
by the PSP
(Frequency/
Severity)
Common/Moderate
Strength of
Evidence for
Effectiveness
of the PSPs
Evidence or
Estimate of
Implementation
Potential for
Cost
Issues:
How Much do We
Harmful
Know?/How Hard Is
Unintended
Consequences
it?
Limiting trainee work hours; brief
Low
Moderate (at
High
Moderate/Difficult
review
least); includes
lack of training
time
Abbreviations: ACS NSQIP=American College of Surgeons National Surgical Quality Improvement Program; NA = not
available; PSP: Patient Safety Practice; RFID = radio-frequency identification.
*In some cases, the text in the PSP column differs slightly from the chapter heading for that PSP. This difference is attributable
to our Technical Expert Panels desire to include the target safety problem (if the practice is in fact targeted at a specific safety
problem), more specification, or an example of the PSP (e.g., adding such as a centralized display of consolidated data to the
PSP designated as operating room integration and display systems).
Rating Scales:
Scope of the problem targeted by the PSP (frequency/severity): frequency = rare or common; severity = low, moderate, or high.
Strength of evidence for effectiveness of the PSPs: low, moderate, or high.
Evidence or potential for harmful unintended consequences: negligible, low, moderate, or high.
Estimate of cost: low, moderate, or high.
Implementation issues: How much do we know? = little, moderate, or a lot; How hard is it? = not difficult, moderate, or difficult.
One of the great challenges in measuring patient safety is determining whether to assess
primary outcomes (harms), intermediate outcomes (errors) or processes (such as adherence
to evidence-based safety practices). Each of these methods has advantages and disadvantages.
Over the past few years, the safety field has increasingly emphasized primary outcomes (namely,
harm measures), and the IHI Global Trigger Tool (GTT) has emerged as an increasingly popular
method for such assessment. In fact, several studies using the GTT3-5 have come to the same
disappointing conclusion: that rates of harm remain high and, at least in a group of North
Carolina hospitals, did not improve during the first several years of the patient safety movement.
Although the Global Trigger Tool has demonstrated better test characteristics than other
outcomes-oriented methods of measuring safety, such as voluntary incident reports and the
AHRQ Patient Safety Indicators,3 one of the main insights to have emerged from recent patient
safety research is that multiple lenses are needed to get a broad, and true, view of progress in
safety. Shojania has called this issue the elephant of patient safety, in that one gets a different
view depending on what part one is looking at.6
Because of the limitations of outcome measures in patient safety, it is important that we
continue to assess the degree to which we now understand and have implemented effective PSPs.
The present report, conducted by many of the same investigators, as 2001s Making Health
Care Safer illustrates both the progress and the challenges in this area of safety research.
Over the past decade, we have achieved greater agreement on what constitutes evidence of
effectiveness and the importance of implementation and context (this new understanding was
codified in a prior AHRQ report, Assessing the Evidence for Context-Sensitive Effectiveness
and Safety of Patient Safety Practices: Developing Criteria7 and in the peer-reviewed articles
that drew on this report).8-12 In the current review, 20, or about half of the PSPs reviewed, had
the strength of evidence for their effectiveness rated as at least moderate, which represents
significant progress since 2001. The evidence base supporting implementation strategies is also
improving. For 26 of the PSPs reviewed in the present report, we judged that there was at least
moderate evidence about how to implement the practice; the area of implementation was so
underdeveloped a decade ago that Making Health Care Safer did not even consider it.
503
However, for almost no PSPs do we understand with confidence the potential role that context
plays in effectiveness. This area remains a major gap in our knowledge base about how to select
and implement PSPs; and it is a particularly crucial gap as institutions and individuals try to
implement best practices, and policymakers, accreditors, and payers seek to create incentives
for implementation via transparency- or payment-related initiatives.
Priorities for Adoption of Patient Safety Practices

We identified sufficient evidence about effectiveness and implementation for our technical
experts to judge that some PSPs are ready to be strongly encouraged for adoption by health
care providers. Table 2 shows the strongly encouraged PSPs. For particular targets for which
we discussed multiple PSPs, (such as catheter-associated urinary tract infection), the table
describes a particular PSP or category of PSPs.
Table 2, Chapter 44. Strongly encouraged patient safety practices
Preoperative checklists and anesthesia checklists to prevent operative and post-operative events
Bundles that include checklists to prevent central line-associated bloodstream infections
Interventions to reduce urinary catheter use, including catheter reminders, stop orders, or nurse-initiated
removal protocols
Bundles that include head-of-bed elevation, sedation vacations, oral care with chlorhexidine, and
subglottic-suctioning endotracheal tubes to prevent ventilator-associated pneumonia
Hand hygiene
Do Not Use list for hazardous abbreviations
Multicomponent interventions to reduce pressure ulcers
Barrier precautions to prevent healthcare-associated infections
Use of real-time ultrasound for central line placement
Interventions to improve prophylaxis for venous thromboembolisms
The conclusions in this report explicitly represent a combination of the available evidence
with the judgment of our technical expert panelists interpreting that evidence.
Additional PSPs were judged by our technical experts as having sufficient evidence about
effectiveness and implementation that they should be encouraged for adoption. Table 3
presents the encouraged PSPs.
Table 3, Chapter 44. Encouraged patient safety practices

Use of clinical pharmacists to reduce adverse drug events
Documentation of patient preferences for life-sustaining treatment
Obtaining informed consent to improve patients understanding of the potential risks of procedures
Team training
Practices to reduce radiation exposure from fluoroscopy and computed tomography scans
Use of surgical outcome measurements and report cards, like the American College of Surgeons
National Surgical Quality Improvement Program
Rapid response systems
Utilization of complementary methods for detecting adverse events/medical errors to monitor for patient
safety problems
Computerized provider order entry
Use of simulation exercises in patient safety efforts
504
The 22 PSPs in Tables 2 and 3 represent practices that health care providers can consider for
adoption now. This recommendation particularly applies to the 10 strongly encouraged
practices in Table 2, which, at least in the judgment of our technical experts, providers have
sufficient knowledge to implement and doing so will likely result in safer care. And while future
evaluations will probably strengthen our knowledge base regarding how best to implement these
practices to make them most effective, our technical experts believe that providers should not
delay consideration of adopting these practices while waiting for more research: enough is
known now to permit health care systems to move ahead.
Limitations
Because of limited resources and time, the current report does not cover the entire patient
safety field, which has grown exponentially since the last report, both in the number of potential
PSPs and in the amount of data about individual PSPs). We used an explicit and transparent
process to select the PSPs we did evaluate, and our final list should include most PSPs of highest
priority to policymakers and providers.
Secondly, we did not do in-depth reviews of all the PSPs. Again, in order to make the best
use of the available time and resources, we tailored our methods to the needs of our stakeholders,
targeting those PSPs of greatest interest (or for which there was perceived to be the most new
information) for in-depth reviews; others received briefer reviews. It was crucial that the
decisions about which PSPs would receive in-depth review and which would receive brief
review were made by a broadly representative stakeholder committee. The in-depth reviews,
while thorough, did not conform to all of the criteria in the 2011 IOM report, Finding What
Works in Health Care: Standards for Systematic Reviews,13 nor all the criteria in the EPC
Methods Guide (for example, we did not publicly post a protocol for each individual review).
We used our collective experience as EPCs to adapt existing EPC methods that we judged best
preserved the essence of a systematic review while allowing us to complete 18 in-depth reviews
within 9 months and the available budget.
Additionally, over time, we will likely improve our methods for assessing evidence regarding
how patient safety interventions affect health care processes and outcomes. The methods we used
for this report incorporate new perspectives regarding the importance of implementation and
context, which was the focus of the Context Sensitivity report; likewise, in the future, we can
expect to increase our understanding of the interactions between multiple intervention,
implementation, and organizational variables and how these influence safety outcomes. If future
research reveals that these variables interact in ways that our current understanding of theory and
logic models cannot explain, we will need to modify the methods of evaluating PSPs again.
Lastly, we relied on the judgment of our technical experts at every important step of the
project: Therefore our results are as much a product of these judgments as of our systematic
review methods. Hence, our results might be sensitive to the selection of particular experts on
our technical expert panel. However, we mitigated this potential bias by including more than
double the number of experts on our technical expert panel as we typically would for an EPC
review, which allowed us to include a diverse set of stakeholders from the U.S., Canada, and the
United Kingdom; from PSP developers and evaluators to patient safety policymakers to experts
in design and evaluation methods. Rather than regarding the tight linkage between the needs of
the stakeholders and the work of the EPCs as a limitation, we view it as a strength that increases
the likelihood that the results of the review will be meaningful to providers, payors, and patients,
and that the reports results will lead to meaningful change.
505
Conclusions
In 2001, when we published Making Health Care Safer, the literature on PSPs was limited,
for several reasons.
First, the dominant cognitive model for patient safety was that errors represented human
lapses; thus, there seemed little to study. The key PSP, one might say with only slight hyperbole,
was to admonish caregivers to be more careful next time.
Moreover, no business case existed for institutions or individuals to focus on patient safety,
no public pressure was exerted to improve safety, and no research funding was available for
safety studies. The fact that the literature on safety was relatively primitive was anything but
surprising.
This picture changed completely over the ensuing decade. AHRQs Patient Safety Network,
the organizations main portal for safety literature, now lists more than 3000 research studies,
with 400 of these deemed by the editors as Classics. Safety research receives substantial
support from AHRQ and others, the business case for safety improvement has grown, and
policymakers are intensely interested in safety research as they consider what they can do to
promote safety.
We found evidence of this progress in our current review of the literature on patient safety.
There are now over a dozen practices for which the evidence of effectiveness is strong or very
strong, and data are emerging on the contextual factors that so often determine the outcome of
implementation. With the Federal investment in safety (under the Partnership for Patients) of
about $1 billion, and an investment of more than 20 times that amount in information technology
implementation, we are on the cusp of an exploding database of research on safety practices.
Yet recent studies of rates of harm have demonstrated how difficult improving safety really is
and have caused policymakers and researchers to redouble their efforts to identify and implement
safe practices in hospitals, nursing homes, and clinics. Individuals and institutions seeking to
improve safety would do well to scrutinize the practices described in this reportwidespread
implementation is likely to save hundreds, if not thousands, of lives. It will also help us continue
to refine our efforts to identify the factors associated with successful implementation of PSPs,
and to pinpoint, and hopefully prevent, any unintended consequences.

Our technical expert panel judged the following topics to be high priority for future research:
General issues:
Sufficient data about the costs of PSPs to support cost-effectiveness analyses or returnon-investment analyses
More patient safety measures for ambulatory care
Better measures of the major causes of harm
Specific PSPs that are the highest priority for future research:
Interventions to improve care transitions at hospital discharge
Simulation methods
Team training
Use of human factors engineering and ergonomics in the design of health care practices
506
Surgical outcome report cards

Systems and decision aids to reduce diagnostic errors
Measures to encourage a culture of patient engagement in patient safety
Some PSPs were not included in this review because they were not deemed sufficiently
developed, and new PSPs will subsequently be developed. Thus a strategy of surveillance should
be adopted regarding evidence on PSPs. Future research also requires advancing the basic
science of safety measurement such as standardized methods for rare events and for evaluating
studies that assess only process-related outcomes relative to those that assess patient outcomes.
Future Research Needs Specific to Context Sensitivity

As part of our project on developing criteria to assess context sensitive PSPs, we worked
with this same panel of technical experts to determine future research needs with respect to
context. They bear repeating here.
1. Developing and validating measures of patient safety culture. Discussion at the panel
meetings indicated that several technical experts considered patient safety culture to be the
overarching important construct. This view may explain why patient safety culture received
majority support as a high priority for future research, whereas research on leadership and
teamwork measures did not. Specific suggestions for future research included:
a. Developing validated measures of cultural adaptability to change.
b. Assessing the potential distinction between a culture of safety, a culture of excellence,
and organizational culture.
c. Establishing connections between aspects of patient safety culture and patient outcomes
or processes of care.
d. Assessing correlations between measures.
Additionally several TEP members commented that teamwork and leadership are important
concepts for which several measures are currently available. Several TEP members felt
researchers should use these measures working to mature and build the evidence about this
construct.
2.
Developing criteria and recommendations for what constitutes reporting the

intervention in sufficient detail that it can be replicated. More precise criteria for how
PSP interventions should be described warrant additional research. In particular, the guidance
described here, along with that provided by Standards for Quality Improvement Reporting
Excellence (SQUIRE) and the National Quality Forum (NQF), need to be evaluated. Doing
so will help determine which PSP elements need to be described and in what detail in order
to evaluate whether the PSP is truly effective. This also will help maximize the possibility of
successful PSP replication with similar outcomes. Further research could also evaluate the
effect of applying these draft criteria regarding PSP descriptions on the quality of PSP
projects and published articles. Thoroughly describing PSPs also can help readers determine
the relevance of an evaluation study to other PSPs or other contexts. For example, if a PSP
requires an individual behavior change such as hand-washing, then knowing intervention
details may help readers of the study assess whether the given results are relevant only to
hand-washing interventions or if they could be applied to other types of PSPs requiring
507
individual behavior change. Knowing the details of the intervention also could help readers
of the study determine how much the success of the PSP implementation depended on
contextual issues (e.g., organization or teamwork).
3. Understanding the important items to measure and report on for implementation.
Experts consider having comprehensive information about implementation key to being able
to replicate a PSP. However, little empirical evidence exists about what makes a description
of the PSP adequate for reporting. Assessing what implementers need to know, if they are to
be able to implement or adapt an intervention in their own settings, is critical. Most experts
considered understanding the important items to measure and report on for implementation
to be related to or even the same as reporting the intervention in sufficient detail that it can
be replicated. This view suggests that the distinction between the intervention and the
implementation may be an arbitrary line, and that ideal evaluations of PSP interventions
need to consider the implementation as part of the intervention.
4.
Developing a theory-based taxonomy or framework with which to describe and

evaluate key elements of interventions, contexts, and targeted behaviors. Although the
current project made a promising start on meeting this need, progress in this area will require
additional development to produce a taxonomy that would be both sufficiently broad based
and flexible enough to be widely useful. Issues to be considered include whether a taxonomy
is the preferable way to proceed, or whether a more useful strategy might be to create an
explicit methodology that researchers could apply to specific problems and contexts. Yet
another approach might be to devise an assessment framework. Some experts sounded
cautionary notes on this topic. They reported that outpatient PSP research may be too new to
apply a taxonomy at this stage. They also reported that a single unified taxonomy may not
be sufficiently flexible for diverse PSPs, and multiple taxonomies may be needed in any case.
The countervailing view to these cautionary notes was that the field would not be well-served
by having a proliferation of taxonomies. Instead, they reported, what is needed is a coherent,
sufficiently comprehensive taxonomy that can accommodate the challenges of the subject.
5. Refining a framework for assessing the strength of a body of evidence. The research team
did developmental work on an adaptation of the GRADE and Evidence-based Practice Center
(EPC) systems for assessing the strength of evidence across studies of a PSP. This work
warrants further development.
6. Generating empirical evidence that the contextual factors identified in this project
influence the success of the PSP. The research team acknowledges that most of the
recommendations in the report have a thin empirical evidence base, which simply reflects the
relatively immature state of research in this still relatively young field. Building a stronger
evidence base will help future efforts at refining the recommendations presented here.
Future Methodological Needs

Despite over a decade of effort, there is little evidence that patient outcomes (broadly
measured) have significantly improved. Yet there have been patches of success, generally
focused on efforts to reduce one type of harm, usually using one method of improvement. For
508
example, efforts have focused on reducing blood stream infections, improving teamwork, or
enhancing patient engagement.
If health care is to make significant improvements in patient safety, research should inform
and guide these efforts, just as it has done in every other field. We have learned much about how
to improve safety, yet we need to learn much more. Acquiring this knowledge will require
investments in patient safety research, including basic methodological research. To date,
investments in patient safety research have fallen far short of the magnitude of the problem.
To achieve progress in improving patient safety, research is needed in a number of areas:
basic patient safety research to develop new tools and measures and ensure that the tool
matches the problem;
a larger number of valid measures of patient safety;
better methods to measure context and how an intervention was implemented;
methods to identify and provide the necessary skills, resources, and accountability (i.e., a
safety management infrastructure) at each level of the health care system; and
more effective and less burdensome methods of improvement so that clinicians,
researchers, and administrators work on reducing all types of harms patients are at risk of
suffering rather than a select few. Below we briefly discuss each of these.
Basic patient safety research. Largely driven by an appropriate desire to reduce patient harm
from medical errors immediately, the field has often invested in quick fixes that may have lacked
sufficient theory or validated evaluation tools. For example, although the need to evaluate
context when implementing patient safety interventions is widely recognized, few validated
instruments have been developed to accomplish this task. Just as the hundreds of thousands of
deaths from heart disease or cancer each year inspire rather than obfuscate the need for basic
research, so too should the large number of deaths from preventable harm. To improve patient
safety, the Federal Government will need to invest in basic patient research, to diagnose
different types of safety problems, to match the theories and methods to the type of problem, to
better evaluate implementation efforts and their surrounding context, and to evaluate whether
patient safety is indeed improving
The future research needed to advance the science of basic patient safety research was
covered in Context Sensitivity,7 where this technical expert panel rated as highest priority
topics such as development of a theory-based taxonomy with which to describe and evaluate
key elements of interventions, contexts, and targeted behaviors and understanding the
important items to measure and report in implementation.
Larger number and more valid tools to measure safety. Despite more than a decade of effort,
the health care system remains unable to quantify the magnitude of preventable harm or to use
valid tools to evaluate progress in improving patient safety over time or among provider
organizations. Moreover, for most of the patient safety harms discussed in this report, the field
lacks valid, broadly accepted definitions forand the mechanisms to monitorprogress in
reducing patient harm. Thus, despite significant efforts to improve patient safety over the last
decade, no oneincluding patients, providers, researchers, payors, and policymakersknows
whether care is safer. The need for evaluation tools is urgent.
Better methods to measure context and to describe an intervention. In contrast to most
clinical research interventions, patient safety interventions are iterative, evolve over time, are
509
context dependent, and are strongly influenced by the organizations in which they are
implemented and the personnel involved. Both this report and Context Sensitivity have
demonstrated that too often, the intervention is insufficiently described and the context is barely
mentioned; the result is stalled learning and reduced generalizability. Research is needed to
better understand how an intervention was implemented over time, the most salient iterations the
intervention has undergone, and the critical contextual domains that may have supported or
mitigated the improvement effort. For example, while leadership and teamwork are widely
regarded as important in implementing patient safety interventions, the field lacks consensus
both on how best to measure these domains and on a theory that explains how various domains
of context support or hinder the success of an intervention.
The topics more patient safety measures for ambulatory care and better measures of the
major causes of harm, which were judged as high priority by our technical expert panel, fall into
this domain.
Methods to build a safety management infrastructure. Health care is largely organized
around the care of individual patients, yet patient safety requires the management of populations
of patients and accountability for complications. While physicians have profound individual
accountability for their patients, especially for complications that are directly related to their
care, their accountability often diminishes for complications less directly related to the care they
provide, or complications that are influenced by the care of a care team or how care is organized;
infections are an example of such complications.
If patient safety requires the management of a population of patients, an infrastructure should
be in place to help monitor risks and prioritize interventions, to implement interventions, and to
monitor progress. At multiple levels of a provider organization (clinic or unit; department or
region; hospital or group practice), an infrastructure is needed to ensure that safety leaders have
sufficient resources, skills, and accountabilities to improve safety. Little is known regarding the
specifics of the infrastructure that is needed. Nevertheless, the existing infrastructure is largely
underdeveloped.
Specifically, researchers and managers must determine how much physician, nurse, and other
staff support are needed at each unit/clinic, department/groups of clinics, hospital/health system
levels to ensure patient safety; what skills they need, and how they should be held accountable
for the safety of the care provided. For example, if an employees job is to evaluate progress in
patient safety, does he require training in clinical epidemiology? If an employees job is to
implement interventions, does she require training in human factors engineering and
implementation science? Finally, how will managers and researchers create a cascading
accountability system in which unit and clinical leaders hold individual clinicians accountable,
department or regional leaders hold unit/clinical leaders accountable, and hospital or health
systems leaders hold department and regional leaders accountable. Although this type of
infrastructure exists in other industries, little is known about its benefits and costs. Investments
are needed to understand, implement, and evaluate a safety management infrastructure in health
care.
The topic, more data about the costs of patient safety practices, which was judged high
priority by our technical expert panel, falls into this domain.
More effective and efficient system interventions to reduce multiple, rather than single
types of harm. Our technical expert panel strongly advocated for a systematic approach to
510
preventing harm and exploring the concept of mutually reinforcing practices that reduce many
kinds of harm.
The Center for Medicare and Medicaid Services (CMS) is now embarking on an ambitious
national effort to reduce 10 types of preventable harm. However, because of the burden of
implementing the interventions, hospitals are generally selecting interventions to reduce only a
subset of the harms. Yet most patients are at risk for all 10 harms and many others as well.14
In health care, too many improvement efforts rely on the heroism of clinicians rather than
safely designed systems. In other industries, as the amount of information has increased,
technological improvements are implemented, with the result that productivity and safety
increase. Health care productivity remains flat and clinicians use technologies that generally do
not talk to each other. For example, the infusion pump does not talk to a respiratory monitor. If
such intercommunication occurred, a respiratory monitor would automatically shut off an
infusion pump if a patient developed a dangerously low respiratory rate from an infusion of
narcotics, a common cause of respiratory arrest. Preciously few examples exist of safe design in
health care. Such an approach will require close collaboration with systems engineers.
Patient harms do not occur in isolation, and they are not independent. Rather, they are
interdependent; thus, the solutions must be as well. Hospitalized patients are at risk for multiple
complications. For example, a patient on a breathing machine after surgery is at risk for 9 of the
10 complications targeted for reduction by CMS (the tenth, which is an obstetric complication,
would not apply in this case).
Complex patients suffer the same fate. Because these patients have a variety of chronic
diseases, they are at risk for a variety of harms, yet few efforts have been undertaken to
systematically reduce the risk for all types of harm.
Yet there is an alternative. Health care could more fully embrace systems engineering. The
need to develop and demonstrate a system framework that addresses the universal and
fundamental challenges in contemporary health care delivery remains a critical challenge. In the
2005 report, Building a Better Delivery SystemA New Engineering/Health Care Partnership,
the National Academy of Engineering (NAE) and the IOM noted:
a systems approach to healthcare delivery could transform the U.S.
health care sector from an underperforming conglomerate of
independent entities into a high-performance system in which
every participating unit recognizes its interdependence and
influence on every other unit
Ideally, the new and improved health care delivery system should include: 1) an integrated,
ubiquitous, distributed, responsive, expansive, flexible, affordable and resilient system; 2)
personalized delivery facilitated by secure information flow and optimized information that runs
smoothly, efficiently, and safely.
The IOM/NAE report highlighted the observation that health care is significantly underengineered and called for greater input from systems engineers to make health care safer. Future
research in patient safety needs to take a systems approach, focusing on all the harms a patient
suffers, clarifying the therapies that may reduce harm, and ensuring that patients always receive
them. This research could include three key areas: a focus on engaging patients and their
families, ensuring patients receive therapies to reduce harm, and creating a learning and
accountability system. Few examples of successful collaboration between engineering and
medicine currently exist. One example of such a collaboration is the Systems Engineering
511
Initiative for Patient Safety (SEIPS) program at the University of Wisconsin-Madison, which
brings human factors and systems engineers together with clinicians to work on solving complex
patient safety problems.15 This model, developed by Carayon, views the work system as an
interaction among people, tasks, tools and technologies, organization, and environment. Another
example, occurring at Johns Hopkins University, involves a collaboration with the Johns
Hopkins University Whiting School of Engineering, Applied Physics Laboratory, the School of
Medicine, the Bloomberg School of Public Health, the Carey Business School, the School of
Nursing, and the private sector. The purpose of this collaboration is to develop a model to
eliminate preventable harm.
As a part of these collaborations, clinicians and researchers are exploring ways to apply
systems engineering to patient safety. For example, providers might consider all potential harms
of being hooked up to a breathing machine. The resulting list includes 9 of the 10 harms
enumerated by CMS and several more, including patient-centered harms from loss of dignity,
autonomy, and respect. The clinicians and engineers would then consider the tasks or treatments
to prevent those harms and the barriers to performing those tasks and ultimately design a system
that ensures patients receive the recommend therapies and monitors and improve performance.
For example, clinicians could be provided visual displays indicating when a treatment is due
and when it has been completed. As discussed above, at present, few of the technologies
responsible for care processes are integrated: these technologies, which include the medical
devices including the bed, the ventilator, the infusion pump, the monitors, and the electronic
health record, do not communicate. Research into how to better integrate systems engineering
into health care could help improve safety and allow clinicians and managers to work on
preventing all types of patient harm.
A part of adopting a systems approach dictates including patients and their families. Patients
and their families are an integral part of the health care system, yet often, are not adequately
engaged or provided with sufficient information. Future research needs to explore how best to
engage and activate patients to help improve safety and how best to view patients as an integral
part of the health care system.
Efforts to improve patient safety are often fragmented into specific clinical or academic
disciplines. While each approach and method is necessary, it is unlikely any one, by itself, will
be sufficient to address the entire patient safety problem and reduce all types of harms. These
systems approaches that link clinicians, human factors and systems engineers, social scientists,
health services researchers, informatics specialists, economists, and biostatisticians, offer hope
for realizing broad improvements in patient safety. Although such diverse groups pose
management challenges, they have enormous potential to rebuild the health care chasse, which
remains largely broken. If we are to make progress, science must guide us, an endeavor that will
require investments in patient safety research. The research agenda outlined above may point
health care in the right direction.
References
1.
2.
Kohn L, Corrigan J, Donaldson M, et al. To

System Committee on Quality of Health
Care in America, Institute of Medicine.
Washington, DC: Press NA; 2000.
512

analysis of patient safety practices. Evid Rep
Technol Assess (Summ). 2001(43):i-x, 1668. PMID: 11510252.
3.
Classen DC, Resar R, Griffin F, et al.

Global trigger tool shows that adverse
(Millwood). 2011 Apr;30(4):581-9. PMID:
21471476.
4.

resulting from medical care. N Engl J Med.
2010 Nov 25;363(22):2124-34. PMID:
21105794.
9.
Ovretveit JC, Shekelle PG, Dy SM, et al.

How does context affect interventions to
improve patient safety? An assessment of
evidence from studies of five patient safety
practices and proposals for research. BMJ
Qual Saf. 2011 Jul;20(7):604-10. PMID:
21493589.
10.

al. Advancing the science of patient safety.
Ann Intern Med. 2011 May 17;154(10):6936. PMID: 21576538.
5.
Levinson DR, Office of Inspector General.

Adverse Events in Hospitals: National
Incidence Among Medicare Beneficiaries
November 2010.
11.
Dy SM, Taylor SL, Carr LH, et al. A

framework for classifying patient safety
practices: results from an expert consensus
process. BMJ Qual Saf. 2011 Jul;20(7):61824. PMID: 21610267.
6.
Shojania KG. The elephant of patient safety:

what you see depends on how you look. Jt
Sep;36(9):399-401. PMID: 20873672.
12.
7.

Contract Final Report. AHRQ Publication
No. 11-0006-EF. December 2010. Prepared
under Contract No. HHSA-290-200910001C. Agency for Healthcare Research
and Quality, Rockville, MD.
www.ahrq.gov/qual/contextsensitive/

interventions? BMJ Qual Saf. 2011
Jul;20(7):611-7. PMID: 21617166.
13.
Committee on Standards for Systematic

Reviews of Comparative Effectiveness
Research, Institute of Medicine. Finding
What Works in Health Care: Standards for
Systematic Reviews: The National
Academies Press; 2011.
14.
Pronovost PJ, Bo-Linn GW. Preventing

patient harms through systems of care.
JAMA. 2012 Aug 22;308(8):769-70. PMID:
22910751.
15.
Carayon P, Schoofs Hundt A, Karsh BT, et

al. Work system design for patient safety:
the SEIPS model. Qual Saf Health Care.
2006 Dec;15 Suppl 1:i50-8. PMID:
17142610.
8.
Foy R, Ovretveit J, Shekelle PG, et al. The

role of theory in research to develop and
evaluate the implementation of patient
safety practices. BMJ Qual Saf. 2011
May;20(5):453-9. PMID: 21317181.
513
Abbreviations/Acronyms
AAGBI
Association of Anaesthetists of Great Britain and Ireland
AANA
American Association of Nurse Anesthetists
AAPM
American Association of Physicists in Medicine
ACCP
American College of Chest Physicians
ACGME
Accreditation Council for Graduate Medical Education
ACR
American College of Radiology
ACS NSQIP
American College of Surgeons National Surgical Quality Improvement
Project
ADE
Adverse Drug Event
ADL
Activities of Daily Living
ADR
Adverse Drug Reaction
AE
Adverse Event
AHRQ
AIR
American Institute for Research
AMI
Acute Myocardial Infarction
AMSTAR
A measurement tool to assess systematic reviews
AORN
Association of periOperative Registered Nurses
APN
Advanced Practice Nurses
APOE4
Apolipoprotein E4
ARRA
American Recovery and Reinvestment
ASA
American Society of Anesthesiologists
ASATT
American Society of Anesthesia Technicians and Technologists
ASRT
American Society of Radiologic Technologists
AV
Audio/Video
BATS
Brain Attack Teams
BC/BS
Blue Cross/Blue Shield
BG
Blood Glucose
BIS
Bispectral Index
BMI
Body Mass Index
BPMH
Best Possible Medication History
BPOC
Barcode-Enabled Point of Care
C
Comparator
CAN-NSQIP
Canadian National Surgical Quality Improvement Collaborative
CAUTI
Catheter-Associated Urinary Tract Infections
cc
Milliliter
CCAs
Critical Care Areas
CCOT
Critical Care Outreach Team
CCT
Clinical Controlled Trial
CCTA
Cardiac Computed Tomography Angiography
CCU
Coronary Care Unit
CDC
Centers for Disease Control
CDC HICPAC
Centers for Disease Control and Prevention Healthcare Infection Control
Practices Advisory Committee
CDSS
Clinical Decision Support Systems
CHF
Congestive Heart Failure
CHG
Chlorhexidine
514
CI
CI-AKI
CKD
CLABSI
CMS
CNA
CNS
CPOE
CPRS
CPSI
CRBSI
CRCPD
CRM
CSO
CSRS
CSS
CT
CTA
CTI
CTPA
CTSQC
CUSP
CVA
CVC
CWOCN
DEER
DERS
DM
DNR
DoD
DRG
DVT
EAST
EBA
EBP
ED
eMARs
EMR
EPC
EPOC
ESA
FDA
FHA
FMEA
FPTK
FSCI
FTE
FTE
Confidence Interval
Contrast-induced Acute Kidney Injury
Chronic Kidney Disease
Central Line-Associated Bloodstream Infections
Centers for Medicare & Medicaid Services
Certified Nursing Assistant
Clinical Nurse Specialist
Computerized Provider Order Entry
Computerized Patient Record System
Canadian Patient Safety Institute
Catheter-related Bloodstream Infection
Conference of Radiation Control Program Directors
Crew Resource Management
Constant Special Observation
Cardiac Surgery Reporting System
Computerized Surveillance System
Computed Tomography
Computed Tomography Angiography
Care Transitions Intervention
Computed Tomographic Pulmonary Angiography
Connecticut Surgical Quality Coalition
Comprehensive Unit Based Safety Program
Cerebrovascular Accident
Central Venous Catheters
Certified Wound Ostomy Continence Nurse
Diagnostic Error Evaluation and Research
Dose Error Reduction System
Diabetes Mellitus
Do Not Resuscitate
Department of Defense
Diagnosis-Related Group
Deep Venous Thrombosis
Eastern Association for the Surgery of Trauma
European Board of Anesthesiology
Evidence-based Practices
Emergency Department
Electronic Medication Administration Records
Electronic Medical Record
Evidence-Based Practice Center
Effective Practice and Organization of Care
European Society of Anesthesiology
Food and Drug Administration
Florida Hospital Association
Failure Mode and Effects Analysis
Fall Prevention Tool Kit
Florida Surgical Care Initiative
Full Time Employment
Full-time Equivalent
515
GAO
GI
GRAM
HAI
HAPU
HAT
HELP
HFE
HFMEA
HH
HICPAC
HITEC
HR
HS
HSOPS
ICDSC
ICU
IEA
IHI
IIDs
IIT
INR
IOM
IOCM
ISMP
ISQIC
IT
IV
JC
JCAHO
KPNCRNC
kV
LEB
LEP
LMWH
LVN
MCR
MCSQC
MDS
MET
MHCS
MICU
MRSA
mSv
MS
MTT
NAAL
NAC
Government Accountability Office

Gastrointestinal
Geriatric Risk Assessment Medguide
Hospital-associated Infections
Hospital-acquired Pressure Ulcer
Heart Attack Team
Hospital Elder Life Program
Human Factors and Ergonomics
Healthcare Failure Mode and Effects Analysis
Hand Hygiene
Healthcare Infection Control Practices Advisory Committee
Health Information Technology for Economic and Clinical Health
Hazard Ratio
At Bedtime
Hospital Survey on Patient Safety
Intensive Care Delirium Screening Checklist
Intensive Care Unit
International Ergonomics Association
Institute for Healthcare Improvement
Intelligent Infusion Devices
Intensive Insulin Therapy
International Normalized Ratio
Institute of Medicine
Iso-osmolar Contrast Media
Institute for Safe Medication Practices
Illinois Surgical Quality Improvement Collaborative
Information Technology
Intravenous
Joint Commission
Joint Commission on Accreditation of Healthcare Organizations
Kaiser Permanente Northern California Regional NSQIP Collaborative
Kilovolts
Lower Extremity Bypass
Limited English Proficiency
Low Molecular Weight Heparins
Licensed Vocational Nurses
Manual Chart Review
Mayo Clinic Surgical Quality Consortium
Minimum Data Set
Medical Emergency Team
Making Health Care Safer
Medical Intensive Care Unit
Methicillin-Resistant Staphylococcus Aureus
Millisievert
Morphine Sulfate
Medical Team Training
National Assessment of Adult Literacy
N-acetylcysteine
516
NCHS
NCSQC
NGC
NHSN
NPSA
NPSG
NPUAP
NQF
NR
NS
NSAIDs
NVASRS
OEMs
OR
OR
OSATS
PAC
PACS
PAE
PCA
PDA
PDSA
PE
PGY
PICC
POLST
PPACA
PPV
ProFaNE
PSC
PSCHO
PSP(s)
PTT
PU
QA
QD
QI
QM
QS
RCA
RCT
RED
RFID
RN
RR
RR
RRS
RRT
National Center for Health Statistics

Northern California Surgical Quality Collaborative
National Guideline Clearinghouse
National Patient Safety Agency
National Patient Safety Goal
National Pressure Ulcer Advisory Panel
National Quality Forum
Not Reported
Not Statistically Significant
Non-steroidal Anti-inflammatory Drugs
National VA Surgical Risk Study
Original Equipment Manufacturers
Operating Room
Odds Ratio
Observed Structured Assessment of Technical Skills
Pre-Anesthesia Checkout
Picture Archiving and Communication Systems
Potential Adverse Event
Patient-Controlled Analgesia
Personal Digital Assistant
Plan-Do-Study-Act
Pulmonary Embolism
Post-graduate Year
Peripherally Inserted Central Catheters
Physician Orders for Life-Sustaining Treatment
Patient Protection and Affordable Care Act
Positive Predictive Value
Prevention of Falls Network Europe
Patient Safety Culture
Patient Safety Climate in Healthcare Organizations
Patient Safety Practice(s)
Partial Thromboplastin Time
Pressure Ulcer
Quality Assurance
Once Daily
Quality Improvement
Quality Measures
Quality System
Root Cause Analysis
Randomized Controlled Trial
ReEngineered Discharge Program
Radio-frequency Identification
Registered Nurse
Relative Risk
Rate Ratio
Rapid Response System
Renal Replacement Therapy
517
RSNA
SAQ
SCHA
SCIP
SCOAP
SCR
SCS
SHEA
SHS
SICU
SMD
SQAN
SSI
STS
SUD
SURPASS
T
TBD
TCPS
TEP
TIA
TIW
TSA
TSQC
TURP
U
U.K.
U.S.
UCSF
UHC
UP
VA
VAP
VRE
VRSA
VTBI
VTE
WHO
WOCN
WON
Radiological Society of North America

Safety Attitudes Questionnaire
The South Carolina Hospital Association
Surgical Care Improvement Project
Surgical Care and Outcome Assessment Program
Surgical Clinical Reviewer
Safety Climate Scale
Society for Healthcare Epidemiology of America
Summa Health System
Surgical Intensive Care Unit
Standardized Mean Difference
Surgical Quality Action Network
Sliding Scale Insulin
Society of Thoracic Surgery
Single-Use Devices
SURgical PAtient Safety System
Treatment
To Be Determined
Tennessee Center for Patient Safety
Technical Expert Panel
Transient Ischemic Attack
Three Times A Week
Team Situation Awarness
Tennessee Surgical Quality Collaborative
Transurethral Resection of the Prostate
Units
United Kingdom
United States
University Health System Consortium
Universal Protocol
Veterans Affairs
Ventilator Associated Pneumonia
Vancomycin-Resistant Enterococci
Vancomycin-Resistant S. Aureus
Volume-To-Be-Infused
Venous Thromboembolism
World Health Organization
Wound, Ostomy and Continence Nurses Society
Wound Ostomy Nurse
518
Appendix A. Original List of Patient Safety Practices

PSPs From Making Health Care Safer (MHCS2001 AHRQ
Report)
Computerized physician order entry (CPOE) with clinical decision support system (CDSS)
(Medication errors and adverse drug events (ADEs) primarily related to ordering process)
Clinical pharmacist consultation services (Medication errors and ADEs related to ordering and
monitoring)
Use of computer monitoring for potential ADEs (ADEs related to targeted classes (analgesics,
KCl, antibiotics, heparin) (focus on detection))
Monitoring for patient safety problems (more general topic of monitoring)
Protocols for high risk drugs: nomograms for heparin (Adverse events related to anticoagulation)
Anticoagulation services and clinics for coumadin8 (Adverse events related to anticoagulation)
Patient self- management using home monitoring devices (Adverse events related to chronic
anticoagulation with warfarin)
Unit-dosing distribution system (ADEs in dispensing medications)
Use of automated medication dispensing devices (ADEs in drug dispensing and/or
administration)
Improved hand washing compliance (via education/behavior change; sink technology and
placement; washing substance) (Hospital-acquired infections)
Barrier precautions (via gowns & gloves; dedicated equipment; dedicated personnel) (Serious
nosocomial infections (e.g., vancomycin- resistant enterococcus, C. difficile))
Hospital-acquired infections (overall topic)
Limitations placed on antibiotic use (Hospital-acquired infections due to antibiotic-resistant
organisms)
Use of silver alloy- coated catheters (Hospital-acquired urinary tract infection)
Use of suprapubic catheters (Hospital-acquired urinary tract infection)
Bundles for central venous catheter-related blood infections (overall topic)
Use of maximum sterile barriers during catheter insertion (Central venous catheter-related blood
infections)
Antibiotic- impregnated catheters (Central venous catheter-related blood infections)
Cleaning site (povidone-iodine to chlorhexidine) (Central venous catheter-related blood
infections)
Changing catheters routinely (Central venous catheter-related blood infections)
Use of heparin (Central venous catheter-related blood infections)
Tunneling short- term central venous catheters (Central venous catheter-related blood infections)
Routine antibiotic prophylaxis (Central venous catheter-related blood infections)
Bundle for ventilator-associated pneumonia (overall topic)
Semi-recumbent positioning (Ventilator- associated pneumonia)
Continuous oscillation (Ventilator- associated pneumonia)
Continuous aspiration of subglottic secretions (CASS) (Ventilator- associated pneumonia)
Selective decontamination of digestive tract (Ventilator- associated pneumonia)
Sucralfate (Ventilator- associated pneumonia)
Localizing specific surgeries and procedures to high volume centers (Mortality associated with
surgical procedures)
A-1
Surgical checklists (overall topic)

Appropriate use of antibiotic prophylaxis (Surgical site infections)
Maintenance of perioperative normothermia (Surgical site infections)
Use of supplemental perioperative oxygen (Surgical site infections)
Perioperative glucose control (Surgical site infections)
Use of real-time ultrasound guidance during central line insertion (Morbidity due to central
venous catheter insertion)
Counting sharps, instruments, sponges (Surgical items left inside patient)
Use of preoperative anesthesia checklists (Complications due to anesthesia equipment failures)
Intraoperative monitoring of vital signs and oxygenation (Critical events in anesthesia)
Use of perioperative beta- blockers (Perioperative cardiac events in patients undergoing
noncardiac surgery)
Fall prevention (overall topic)
Use of identification bracelets (Falls)
Interventions to reduce the use of physical restraints safely (Restraint-related injuries; Falls)
Use of bed alarms (Falls)
Use of special flooring material in patient care areas (Falls and fall- related injuries)
Use of hip protectors (Falls and fall injuries)
Use of pressure relieving bedding materials (Pressure ulcers)
Multi-component delirium prevention program (Hospital-related delirium)
Geriatric consultation services (Hospital-acquired complications (e.g., falls, delirium, functional
decline, mortality))
Geriatric evaluation and management unit (Hospital-acquired complications (functional decline,
mortality))
Appropriate VTE prophylaxis and methods for implementation (broader topic)
Appropriate VTE prophylaxis (Venous thromboembolism (VTE))
Risk assessment and prevention of contrast-induced renal failure (overall topic)
Use of low osmolar contrast media (Contrast-induced renal failure)
Hydration protocols with theophylline (Contrast-induced renal failure)
Hydration protocols with acetylcysteine (Contrast-induced renal failure)
Various nutritional strategies (Morbidity and mortality in postsurgical and critically ill patients)
H2-antagonists (Stress-related gastrointestinal bleeding)
Education interventions and continuous quality improvement strategies (Clinically significant
misread radiographs and CT scans by non- radiologists)
Methods to increase pneumococcal vaccination rate (Pneumococcal pneumonia)
Use of analgesics in patients with acute abdomen without compromising diagnostic accuracy
(Inadequate pain relief in hospital patients with abdominal pain)
Pain management (overall topic)
Acute pain service (Inadequate pain relief)
Non- pharmacologic interventions (e.g., relaxation, distraction) (Inadequate postoperative pain
management)
Change in ICU structureactive management by intensivist (Morbidity and mortality in ICU
patients)
Changes in nursing staffing (Morbidity and mortality)
Promoting a culture of safety (Any safety problem amenable to culture)
A-2
Use of human factors principles in evaluation of medical devices (Medical device related adverse
events)
Refining performance of medical device alarms (e.g., balancing sensitivity and specificity of
alarms, ergonomic design) (Adverse events)
Transitions in care (broader topic)
Information transfer between inpatient and outpatient pharmacy (Adverse events related to
discontinuities in care)
Handoff protocols (broader topic)
Standardized, structured sign- outs for physicians (Adverse events during cross- coverage)
Use of structured discharge summaries (Adverse events related to information loss at discharge)
Protocols for notification of test results to patients (Failures to communicate significant abnormal
results (e.g., pap smears))
Use of bar coding (Adverse events due to patient misidentification)
Sign your site protocols (Performance of invasive diagnostic or therapeutic procedure on
wrong body part)
Team training (broader topic)
Application of aviation style crew resource management (e.g., Anesthesia Crisis Management;
MedTeams) (Adverse events related to team performance issues)
Simulator-based training (Adverse events due to provider inexperience or unfamiliarity with
certain procedures and situations)
Limiting individual providers hours of service (Adverse events related to fatigue in health care
workers)
Fixed shifts or forward shift rotations (Adverse events related to fatigue in health care workers)
Napping strategies (Adverse events related to fatigue in health care workers)
Specialized teams for inter-hospital transport (Adverse events due to transportation of critically
ill patients between health care facilities)
Mechanical ventilation (Adverse events due to transportation of critically ill patients within a
hospital)
Asking that patients recall and restate what they have been told during informed consent
(Missed, incomplete or not fully comprehended informed consent)
Use of video or audio stimuli (Missed, incomplete or not fully comprehended informed consent)
Provision of written informed consent information (Missed, incomplete or not fully
comprehended informed consent)
Computer- generated reminders to discuss advanced directives (Failure to honor patient
preferences for end-of-life care)
Use of physician order form for life- sustaining treatment (POLST) (Failure to honor patient
preferences for end-of-life care)
Additional PSPs From Our Prior Project and Updated Review

of NQF, Joint Commission, IHI, Leapfrog, PSNet Taxonomy,
Other Suggestions From Team
Universal protocol/preoperative checklist (Wrong-site surgery, perioperative infections)
Rapid response teams
Medication reconciliation and process redesign (Medication errors- wrong medication or dose)
Non-reimbursable serious reportable events (i.e., do not pay for never events) (CMS)
A-3
Do not use abbreviations, acronyms, symbols, and dose designation campaign

(education/campaigns, removal from forms, audit/feedback) (Medication errors wrong
medication) (Joint Commission)
Read back or computerized system (verbal or telephone orders or critical test results)
(Medication errors,
Implement a standardized process to ensure that critical results are communicated quickly to a
licensed healthcare provider so that action can be taken. (NQF)
Adverse event reporting
Periodic inspection of medication storage areas (Medication errors use of contaminated drugs)
Drug labeling (Medication errors dispensing)
Institute protocols for managing Look Alike, Sound Alike Medications; standard methods for
labeling and packaging medications (Medication errors dispensing, administration) (NQF)
Identify all high-alert drugs, and establish policies and processes to minimize the risks associated
with the use of these drugs. (NQF)
Identifying patients at risk for suicide (Patient suicide or attempted suicide) (NQF)
Immunize healthcare workers and patients who should be immunized against influenza
(Nosocomial influenza)
Following serious unanticipated outcomes, including those that are clearly caused by systems
failures, the patient and, as appropriate, the family should receive timely, transparent, and clear
communication concerning what is known about the event (NQF)
Ensure that written documentation of the patients preferences for life-sustaining treatments is
prominently displayed in his or her chart (NQF)
Implement standardized policies, processes, and systems to ensure accurate labeling of
radiographs, laboratory specimens, or other diagnostic studies, so that the right study is labeled
for the right patient at the right time. (NQF)
Take actions to improve glycemic control by implementing evidence-based intervention
practices that prevent hypoglycemia and optimize the care of patients with hyperglycemia and
diabetes. (NQF) protocols and order sets
When CT imaging studies are undertaken on children, child-size techniques should be used to
reduce unnecessary exposure to ionizing radiation (NQF)
Institutional safety plan (NQF, PSNet)
Health literacy improvement (PSNet)
Hospitalists (PSNet)
Discharge interventions (care transition interventions, Project Red, calling patients after
discharge, etc) (Preventable readmissions)
Techniques to prevent diagnostic errors (teaching heuristics/meta-cognition; artificial
intelligence programs)
Red Rules/Stop the Line (Rules that must be followed to the letter- any deviation from a red rule
will bring work to a halt until compliance is achieved)
Environmental modifications for health care workers, e.g., quiet place for nurses to mix meds
(Medication errors administration)
Patient engagement strategies (patients questioning their providers; patients on safety
committees)
Unit based safety teams
Executive walk rounds
Bundles and checklists as a general strategy (not just for specific indications)
A-4
Methods for reducing inappropriate prescribing in the elderly

Cognitive aids as more general strategy simulations, debriefings
Protocols for standardizing/improving patient transitions/handoffs as a broader category
CT dosage adjustments for height/weight/sex
(Excessive diagnostic imaging increasing lifetime cancer risks)
Evaluating whether diagnostic imaging studies are actually warranted or can be done through
non-radiation-based modality (Excessive diagnostic imaging increasing lifetime cancer risks)
Public health messages about harms of over diagnosis
Physician-patient discussion/education about appropriate scenarios for testing (Risks from
unnecessary cancer screening)
Institutional algorithms to ensure testing occurs in patients with risk factors for disease (to
prevent high number of false positives) (Risks from unnecessary cancer screening)
Review of hospital staffing patterns, nurse-to-patient ratios, physician handovers (Increased
morbidity and mortality associated with hospital care on weekends and in evenings) (may be
related to work hours, shift work)
Education of hospital staff to be aware of possible changes in care during these time periods
(Increased morbidity and mortality associated with hospital care on weekends and in evenings)
Algorithms to determine if patients truly require prophylaxis on admission
Reducing non-indicated prescribing prior to discharge (Harms of inappropriate use of acidsuppressing medications)
Protocols and order sets (Risks from inappropriately dosed chemotherapy)
New Potential Device-Related Technologies (Some Overlap

With List Above)
Free-flow protection in IVs (Medication error- administration prevent overdose)
Smart pumps (Medication errors wrong dose, wrong drug)
Radiofrequency identification (RFID) tags (Retained foreign bodies following surgery)
Dose reduction technologies for CT systems to prevent unnecessary radiation exposure
Processes related to reprocessing single-use medical devices (Healthcare associated infections)
*1.6
Remote monitoring of ICU patients by critical care physicians (Reduce in-hospital mortality
and/or complications from cardiac events)
Operating room (OR) data integration and display systems (Surgical adverse events resulting
from lack of availability of critical patient information and access to intraoperative consults from
remote providers)
Robot assisted surgery (Reduce surgical complications)
Color-coded patient wristbands (Apprise staff of patient risk factors for adverse events and to
reduce risk of inappropriate care)
Device-related strategies for preventing tubing misconnections (e.g., labeling lines, color coding)
(Adverse events related to tubing misconnections (e.g., connecting drains to nasogastric tubes))
IV infiltration alarms to prevent infiltration/extravasations (Complications from intravenous
therapy)
Patient lift devices (Falls and caregiver injury)
Environmental modifications to prevent patient self-harm (e.g., hinge less door systems) (Reduce
suicide or other self-harm)
Active electrode monitoring for laparoscopic electro surgery (Perioperative burns)
A-5
Air embolism detection devices for CT contrast injectors (Pulmonary emboli)

Alarm integration systems (Adverse events related to caregiver response time to patients in need
of assistance)
Electro surgery return electrode contact quality monitors (Perioperative burns)
Endoscope reprocessors (Healthcare-associated infections)
Ferromagnetic detectors in MR suites (Patient and provider injury from metal objects being
drawn into the MRI bore)
Laser resistant endotracheal tubes (Surgical fire)
Surgical and exam gloves (i.e., to prevent infection from clinician to patient)
RFID-type tracking of patient location (e.g., for wandering) (Wandering and elopement in
patients/residents with dementia, or infant abduction)
Treatment planning systems for radiation therapy (Radiation under/overdoses)
Use of Vocera-style communication devices for alarm notification (Adverse events related to
caregiver response time to patients in need of assistance)
A-6
Appendix B. AMSTAR: A Measurement Tool To

Assess Systematic Reviews
Additional File 1 AMSTAR*
Yes
1. Was an a priori design provided?
The research question and inclusion criteria should be established before No
the conduct of the review.
Cant answer
Not applicable
Yes
2. Was there duplicate study selection and data extraction?
There should be at least two independent data extractors and a
No
consensus procedure for disagreements should be in place.
Cant answer
Not applicable
Yes
3. Was a comprehensive literature search performed?
At least two electronic sources should be searched. The report must
No
include years and databases used (e.g. Central, EMBASE, and
Cant answer
MEDLINE). Key words and/or MESH terms must be stated and where
Not applicable
feasible the search strategy should be provided. All searches should be
supplemented by consulting current contents, reviews, textbooks,
specialized registers, or experts in the particular field of study, and by
reviewing the references in the studies found.
Yes
4. Was the status of publication (i.e. grey literature) used as an
No
inclusion criterion?
The authors should state that they searched for reports regardless of
Cant answer
their publication type. The authors should state whether or not they
Not applicable
excluded any reports (from the systematic review), based on their
publication status, language etc.
Yes
5. Was a list of studies (included and excluded) provided?
A list of included and excluded studies should be provided.
No
Cant answer
Not applicable
Yes
6. Were the characteristics of the included studies provided?
In an aggregated form such as a table, data from the original studies
No
should be provided on the participants, interventions and outcomes. The Cant answer
ranges of characteristics in all the studies analyzed e.g. age, race, sex,
Not applicable
relevant socioeconomic data, disease status, duration, severity, or other
diseases should be reported.
Yes
7. Was the scientific quality of the included studies assessed and
No
documented?
A priori methods of assessment should be provided (e.g., for
Cant answer
effectiveness studies if the author(s) chose to include only randomized,
Not applicable
double-blind, placebo controlled studies, or allocation concealment as
inclusion criteria); for other types of studies alternative items will be
relevant.
B-1
8. Was the scientific quality of the included studies used

appropriately in formulating conclusions?
The results of the methodological rigor and scientific quality should be
considered in the analysis and the conclusions of the review, and
explicitly stated in formulating recommendations.
9. Were the methods used to combine the findings of studies
appropriate?
For the pooled results, a test should be done to ensure the studies were
combinable, to assess their homogeneity (i.e. Chi-squared test for
homogeneity, I). If heterogeneity exists a random effects model should
be used and/or the clinical appropriateness of combining should be
taken into consideration (i.e. is it sensible to combine?).
10. Was the likelihood of publication bias assessed?
An assessment of publication bias should include a combination of
graphical aids (e.g., funnel plot, other available tests) and/or statistical
tests (e.g., Egger regression test).
Yes
No
Cant answer
Not applicable
11. Was the conflict of interest stated?

Potential sources of support should be clearly acknowledged in both the
systematic review and the included studies.
Yes
No
Cant answer
Not applicable
Yes
No
Cant answer
Not
applicable
Yes
No
Cant answer
Not applicable
*The AMSTAR criteria was taken from Shea et al, 2007 Development of AMSTAR: a measurement tool to assess the
methodological quality of systematic reviews. BMC Med Res Methodol 7: 10.
B-2
Appendix C. Literature Searches and Topic-Specific

Methods
Chapter 3. High-Alert Drugs: Patient Safety Practices for
Intravenous Anticoagulants
SECTION A. Literature Search
SEARCH METHODOLOGY
DATABASE SEARCHED & TIME PERIOD COVERED:
PubMed 2000-11/4/2011
LANGUAGE:
English
SEARCH STRATEGY #1:
Heparin
AND
intravenous OR infusion
AND
adverse events OR Iatrogenic Disease/prevention and control OR Medical
Errors/prevention and control OR medication errors/prevention and control OR Medical
Errors/adverse effects OR Safety Management OR Cross Infection/prevention and control
OR infection control OR error*[tiab] OR safe*[tiab] OR overdos* OR adverse[tiab] OR
((infection OR infections OR iatrogenic) AND (prevent OR prevention OR preventive OR
preventing)) OR protocol* OR nomogram* OR inpatient coagulation service OR inpatient
coagulation services OR human factors OR decision support
NUMBER OF RESULTS: 908
SEARCH STRATEGY #2:
Heparin/adverse effects[Mesh] OR Heparin/contraindications[Mesh] OR
Heparin/injuries[Mesh] OR Heparin/poisoning[Mesh] OR Heparin/toxicity[Mesh]
AND
intravenous OR infusion*
NOT
Results of Search #1
===================================================================
Embase 2000-11/18/2011
C-1
LANGUAGE:
English
SEARCH STRATEGY:
intravenous heparin OR heparin NEAR/5 infus*
AND
error* OR cross infection OR infection control OR safe* OR overdos* OR adverse OR
(infection OR infections OR iatrogenic AND (prevent OR prevention OR preventive OR
preventing)) OR protocol* OR nomogram* OR inpatient coagulation service OR inpatient
coagulation services OR human factors OR decision support
AND
article/it OR article in press/it OR conference abstract/it OR conference paper/it OR
review/it
====================================================================
Cochrane 2000-11/21/2011
LANGUAGE:
English
SEARCH STRATEGY:
heparin AND (intravenous OR infusion) in Title, Abstract or Keywords
AND
adverse OR Error* OR Safe* OR overdos* OR ((infection OR infections OR iatrogenic) AND
(prevent OR prevention OR preventive OR preventing)) OR protocol* OR nomogram* OR
inpatient coagulation service OR inpatient coagulation services OR human factors OR
decision support in Title, Abstract or Keywords
NUMBER OF RESULTS: 803 (Cochrane Reviews [11], Other Reviews [4], Clinical Trials
[781], Methods Studies [1], Economic Evaluations [6]
====================================================================
CINAHL 2000-11/21/2011
LANGUAGE:
English
SEARCH STRATEGY:
heparin AND (intravenous OR infusion)
AND
C-2
adverse OR Error* OR Safe* OR overdos* OR ((infection OR infections OR iatrogenic) AND

(prevent OR prevention OR preventive OR preventing)) OR protocol* OR nomogram* OR
inpatient coagulation service OR inpatient coagulation services OR human factors OR
decision support
SECTION B. Methods
PICOTS
Elements
Population
Patients in inpatient healthcare settings (adult and pediatric)
Intervention
Any intervention with a goal to improve safety of intravenous heparin administration
Comparator
Usual practice
Outcomes
Effectiveness of the intervention
Timing
Before and after the intervention

Any inpatient setting
Settings
Inclusion/exclusion criteria:
No restrictions were made by language, country of study, or indication for use of heparin.
Chapter 4. Clinical Pharmacists Role in Preventing Adverse

Drug Events: Brief Update Review
SEARCH METHODOLOGY
PubMed 2000-10/19/2011
LANGUAGE:
English
SEARCH STRATEGY:
clinical pharmacist*
AND
adverse OR harm* OR side effect* OR safe* OR reaction*
Cochrane Databases 2000-10/24/2011
C-3
SEARCH STRATEGY:
clinical pharmacist*
AND
adverse OR harm* OR side effect* OR safe* OR reaction*
SECTION B. Methods
Titles and abstracts were reviewed by a physician health services researcher with experience in
both systematic reviews and in clinical pharmacist services. Included studies were those most
relevant to clinical pharmacist interventions on medication errors and adverse drug events in
various health care settings. The focus was on studies that addressed the possible association
between clinical pharmacist activities and improved prescribing practices and/or assessed
whether such activities might lead to reduced medication errors and adverse drug events.
Included studies were narratively summarized by the author.
Chapter 5. The Joint Commissions Do Not Use List: Brief

Review (NEW)
SEARCH METHODOLOGY
PubMed 2000-10/21/2011
LANGUAGE:
English
SEARCH STRATEGY:
abbreviation*
AND
safe* OR unsafe* OR adverse OR harm*
NUMBER OF RESULTS AFTER FILTERING FOR HUMAN ONLY AND REMOVING
OTHER NON-RELEVANT REFERENCES: 71
SECTION B. Methods
both systematic reviews and in prescribing errors. The search was expanded by using Google to
search for possibly pertinent articles and links; additional articles were identified by reference
mining. The focus was on United States-based studies, since the Do Not Use list is a US
regulatory issue. Clinical trials, observational studies, reviews, and anecdotal reports on
C-4
implementation were the primary resources and given priority in the order above. The synthesis
was narrative.
Chapter 6. Smart Pumps and Other Protocols for Infusion

Pumps: Brief Review (NEW)
Electronic Database Searches
The following databases have been searched for relevant information:
Name
Date limits
Platform/provider
PubMed
2000 November 8, 2011
National Library of Medicine
Cochrane Library
2000- November 9, 2011
Wiley
ECRI Institute website
2000- November 8, 2011
ECRI Institute
Health Devices
2000- November 9, 2011
ECRI Institute
Institute for Healthcare

Improvement
2000-November 9, 2011
http://www.ihi.org
Joint Commission
2000-November 10, 2011
http://www.jointcommission.org/
Patient Safety Network
2000-November 9, 2011

(AHRQ): http://psnet.ahrq.gov/
Pennsylvania Patient Safety

Authority PASR (PA Safety
Authority-patient safety
reporting system)
2000-November 9, 2011
http://patientsafetyauthority.org/Pages/Default.aspx#
Hand Searches of Journal and Nonjournal Literature

Journals and supplements maintained in ECRI Institutes collections were routinely reviewed.
Nonjournal publications and conference proceedings from professional organizations, private
agencies, and government agencies were also screened. Other mechanisms used to retrieve
additional relevant information included review of bibliographies/reference lists from peerreviewed and gray literature. (Gray literature consists of reports, studies, articles, and
monographs produced by federal and local government agencies, private organizations,
educational facilities, consulting firms, and corporations. These documents do not appear in the
peer-reviewed journal literature.)
The search strategies employed combinations of freetext keywords as well as controlled
vocabulary terms including (but not limited to) the following concepts. The strategy below is
presented in PubMed syntax. A parallel strategy was used to search the databases comprising the
Cochrane Library.
Medical Subject Headings (MeSH), and Keywords
C-5
Conventions:
PubMed
[mh] =
[majr] =
[pt]
=
[sb] =
[sh] =
[tiab] =
[ti]
=
MeSH heading
MeSH heading designated as major topic
publication type
subset of PubMed database (PreMEDLINE, Systematic, OldMEDLINE)
MeSH subheading (qualifiers used in conjunction with MeSH headings)
keyword in title or abstract
keyword in title
Topic-Specific Search Terms

Concept
Infusion
pumps
Safety
Controlled
Vocabulary
Infusion pumps[majr]
medication errors[mh]
safety[mh]
safety
management[mh]
risk management[mh])
Keywords
infusion pump*
smart pump*
IV pump*
drug delivery system
drug infusion system
(infusion OR medication OR intravenous] OR IV OR drug OR smart) AND
(pump] OR pumps])
error*
mistake*
safe
safety
risk*
malfunction*
overdos*
wrong
C-6
PUBMED
English language, human, remove overlap
Set
Number
Concept
Search statement
Infusion pumps
infusion pumps[majr]
infusion pump* OR smart pump* OR IV pump* OR drug

delivery system OR drug infusion system
infusion[ti] OR medication[ti] OR intravenous[ti] OR IV[ti] OR

drug[ti] OR smart[ti]) AND (pump[ti] OR pumps[ti])
medication errors[mh] OR safety[mh] OR safety

management[mh] OR risk management[mh]
Safety issues
(error*[tiab] OR mistake*[tiab] OR safe[tiab] OR safety[tiab] OR

risk*[tiab] OR malfunction*[tiab] OR overdos*[tiab] OR
wrong[tiab])
Combine sets - Mesh

headings
1 AND 5
Combine sets
keywords for in
process citations (i.e.
not yet indexed with
MeSh headings and
therefore not captured
in set 7)
2 AND 6 AND (in process[sb] OR publisher[sb])
Combine sets
keywords in title (to
ensure retrieval of any
relevant records that
were not picked up in
the previous sets)
3 AND 6
10
Combine sets
7 OR 8 OR 9
11
Eliminate records on
proton pump inhibitors
10 NOT proton*
12
Eliminate case reports,

comments, editorials,
letters, and news items
11 NOT (case reports[pt] OR comment[pt] OR editorial[pt] OR

letter[pt] OR news[pt])
13
Apply limits
Human; English; with Abstract
Total Downloaded
Total Retrieved
Total Included
53
150
18
SECTION B. Methods
Titles and abstracts were reviewed by a health services research methodologist with experience
in both systematic reviews and medical devices. Included studies were those most relevant to
evaluation of smart pumps and related protocols for reduction of medication errors and adverse
drug events in various health care settings. The focus was on studies that compared medication
C-7
error rates and adverse drug events following implementation of these technologies in hospitals
compared to a control period when the technologies were not active or in place. Potential barriers
to implementation (e.g. user compliance) were also assessed. Included studies were narratively
summarized by the author.
Chapter 7. Barrier Precautions, Patient Isolation, and Routine

Surveillance for Prevention of Healthcare-Associated
Infections: Brief Update Review
A structured search of the PubMed database and review of the bibliographies of relevant articles
identified 158 articles published from 2001 to 2011 that assessed barrier precautions for the
prevention of health care-associated pathogen transmission. Search terms included active
surveillance, active detection, and contact precautions. Few of these studies utilized the
cluster randomized trial design and most were quasi-experimental studies. Low quality studies
that did not include a control group were excluded from this review.
SECTION B. Methods
Titles and abstracts were reviewed by an epidemiologist with special expertise in health careassociated infections. An evidence table was constructed that included study design, population,
setting, and the principal outcomes. The synthesis was narrative.
Chapter 8. Interventions To Improve Hand Hygiene

Compliance: Brief Update Review
For this topic we did not do a formal literature search, as the principal reviews and trials were
already known to the authors as part of their work on recent a recent report, where previous
comprehensive searches had been performed to identify the most pertinent and up to date
literature.
SECTION B. Methods
These reviews and studies were reviewed by a health services researcher and epidemiologist with
expertise in hand hygiene quality improvement. The synthesis was narrative.
Chapter 9. Reducing Unnecessary Urinary Catheter Use and

Other Strategies To Prevent Catheter-Associated Urinary
Tract Infection: Brief Update Review
The 14 studies for the previously published systematic review and meta-analysis (Meddings et al,
Clin Infect Dis, 2010) were obtained from a comprehensive search of the worlds literature for
C-8
interventions from 1950 to 2008 to decrease catheter-associated urinary tract infections by means
of the MEDLINE and Cochrane databases (using Ovid), the PubMed Journals and Medical
Subject Heading (MeSH) datasets, the ISI knowledge databases (Web of Science and Biosis
Previews) and the CINAHL and EMBASE databases. The MEDLINE and Cochrane database
searches were conducted by exploding and combining the following Medical Subject Heading
(MeSH) terms: urinary tract infection, urinary catheterization, indwelling catheter, inpatient,
reminder system, device removal, intervention studies. The MeSH reminder system was also
searched separately. We included the following terms in a keyword search (with wildcard
indicated with *): urinary tract infection; ((urin* or uret*) and cath*)) or catheter*; nosocomial
or inpatient or hospital*; reminder, removal, and intervention. We used similar strategies with
the other databases. A research librarian provided guidance to improve search completeness.
This search yielded 6679 citations, including many duplicate citations. As our initial search was
broad and yielded many guidelines and reviews published regarding prevention of catheterassociated urinary tract infection, we also evaluated these articles reference lists for additional
studies; 1 additional reference was located in this manner. More detailed review was required for
118 articles to determine whether they met inclusion criteria. After applying inclusion and
exclusion criteria to focus on human studies of adults admitted to acute care hospitals reporting
at least one outcome involving catheter use or CAUTI events as a result of the intervention, and
with a comparison group (either pre- versus post-intervention or a separate control group); this
yielded 16 studies for further review. Two authors of the systematic review (J.M. and M.M.)
independently reviewed and abstracted data from the 16 articles that appeared to meet inclusion
criteria, including setting, study population, inclusion/exclusion criteria, definitions used, health
outcomes, and quality issues. A third investigator (S.S.) resolved any differences in abstraction
and reviewed the joint decisions made to exclude 2 of the 16 articles that no longer met inclusion
criteria after further review. As a result, this systematic search in 2008 yielded the 14 articles
reviewed in the previously published meta-analysis.
To update the prior literature search for Chapter 9, a search was performed of MEDLINE and
Cochrane databases (using Ovid) and PubMed for intervention studies (published from August
2008 to February 2012) to reduce use of unnecessary urinary catheters in the acute care of adults,
using the same detailed search strategy as employed in the 2008 search. Yet, unlike the 2008
search which was focused on removal of recently placed indwelling catheters (which excluded
emergency environments), the patient population for the 2012 search was expanded to include
emergency department patients because use of interventions to restrict initial placement was an
additional topic of interest for Chapter 9. The 2012 search results were also supplemented with
prior lists of articles excluded from the prior 2008 search that were focused on emergency
department interventions. A secondary evaluation of the CINAHL database was also performed
for interventions developed and implemented by nurses related to urinary catheter use. In light of
the somewhat different terminology on the topic found in the nursing literature, we searched
CINAHL using variations of the following terms: reminder, removal, urinary catheter, nurse
empowered, nurse directed, nurse protocol. No date limits were employed in the CINAHL
search, which retrieved 5 records. Overall, the MEDLINE and CINAHL searches yielded 479
citations, including 353 from MEDLINE through Ovid, 9 additional from PubMed, 117 from the
Cochrane EBM databases, and 7 duplicates. Studies were included if at least one outcome
involving catheter use or CAUTI events (Table 1 in Chapter 9) was reported as a result of the
intervention with a comparison group. A review of reference lists for additional studies was also
C-9
performed, yielding one additional study. After applying inclusion and exclusion criteria to focus
on human studies of adult patients with at least one outcome involving catheter use or CAUTI
events reported as a result of the intervention, and with a comparison group, this updated search
yielded 12 intervention studies published since the prior meta-analysis.
SECTION B. Methods
As summarized in the previously published meta-analysis for the 14 selected studies from 2008
or earlier, a systematic review process was performed. Correspondence with 24 authors was
initiated to clarify details regarding the interventions and outcomes with responses received from
11 authors, and 4 authors provided unpublished numeric data necessary for statistical pooling.
Two physician reviewers performed a detailed abstraction of the 14 studies. Details of the
statistical analyses for obtaining the pooled effects are detailed in the prior published analyses,
and were not replicated or expanded for writing Chapter 9.
A similar review and abstraction process was performed by one physician (J.M.) for the 12
recent articles in the updated search. No contact was initiated with authors, and theses articles
were analyzed and compared in a narrative process rather than a meta-analysis.
Details of the 14 prior and 12 recent studies are summarized in the Appendix Table for Chapter
9, regarding study design, patient population size and care environment, and details of the
interventions used to either avoid inappropriate placement or to prompt removal of unnecessary
catheters. Other important interventions that could possibly influence the outcomes of the studies
were also summarized in this table. Important outcomes of the 14 prior studies as previously
published in the meta-analyses were summarized in Figure 2; similar outcomes for the 12 recent
studies were summarized in Figure 3.
Chapter 10. Prevention of Central Line-Associated

Bloodstream Infections: Brief Update Review
DATABASES SEARCHED:
Medline Via Ovid
Cochrane Central Register of Controlled Trials Databases
Cochrane Database of Systematic Reviews
Cochrane Database of Abstracts of Reviews of Effects
Cochrane Methodology Register
Health Technology Assessment
NHS Economic Evaluation Database
ACP Journal Club
TIME PERIOD COVERED:
January 1, 2000 January 1, 2012
LANGUAGE:
English language articles only
C-10
SEARCH STRATEGY:
Medical Subject Headings (MeSH) Bacteremia and Catheterization, Central Venous, and the
MeSH subheadings Prevention & control and Adverse effects, as well as the keywords
central line-associated bloodstream infection, central line, and central venous catheter.
Search terms included variations of the keywords bacteremia, bloodstream infection,
central line, central venous catheter, prophylaxis, and prevention, using wildcards and
truncation to capture alternate spellings and endings.
NUMBER OF RESULTS:
1,087 unique manuscripts were retrieved by the search of which 337 articles were relevant for
this report.
SECTION B. Methods
All relevant titles and abstracts were reviewed by a physician health services researcher (VC)
with experience in both systematic reviews and in the topic of central line associated
bloodstream infections (CLABSI). Studies included were those most relevant to prevention of
CLABSI; in addition, studies that reported on local and national policies, economic impact and
interventions associated with CLABSI reduction were included in this report.
Chapter 11. Ventilator-Associated Pneumonia: Brief Update

Review
already known to the authors as part of their quality improvement work where previous
literature.
SECTION B. Methods
These reviews and studies were reviewed by an intensive care unit physician health services
researcher with clinical and quality improvement experience with ventilator-acquired
pneumonia. The synthesis was narrative.
C-11
Chapter 12. Interventions To Allow the Reuse of Single-Use

Devices: Brief Review (NEW)
Name
Date limits
Platform/provider
ECRI Institute members website
2001-November 3, 2011
ECRI Institute
2001-November 3, 2011
PSNet
2001-November 3, 2011
Agency for Healthcare Research

and Quality
PubMed
2001-November 2, 2011
FDA
2001-November 8, 2011
The Food and Drug Administration
JCAHO
2001-November 3, 2011
Joint Commission (JCAHO)
NCQA
2001-November 3, 2011
National Committee for Quality

Assurance (NCQA)

Non-journal publications and conference proceedings from professional organizations, private
additional relevant information included review of bibliographies/reference lists from peerreviewed and gray literature as well as related citation searches using the Scopus database. (Gray
literature consists of reports, studies, articles, and monographs produced by federal and local
government agencies, private organizations, educational facilities, consulting firms, and
corporations. These documents do not appear in the peer-reviewed journal literature.) A number
of organization websites were searched for relevant information, including: ECRI Institute
members website, the Institute for Healthcare Improvement (ISI), and the Agency for Healthcare
Research and Qualitys Patient Safety Network (PSNet).
The search strategies employed combinations of free text keywords as well as controlled
presented in PubMed syntax.
Medical Subject Headings (MeSH) and Keywords
Conventions:
PubMed
[mh] =
[majr] =
MeSH heading
C-12
[pt]
[sb]
[sh]
[tiab]
=
=
=
=
publication type
Topic Specific Search Terms

Concept
Controlled Vocabulary
Keywords
Equipment
surgical instruments
equipment and supplies
disposable equipment
surgical
medical
cutting
instrument*
tool*
equipment*
device*
trocar*
scalpel*
shaver*
raz*
drill*
catheter*
syringe*
needle*
mask*
gown
glove*
endoscop*
instrumentation
Reprocessing
equipment reuse
reuse
reusing
reus*
reprocess
reprocessing
reprocessed
reprocess*
recycled
recycling
recycle*
repurposed
repurposing
repurpose*
Single-Use
single use
single-use
disposable
one-time
one time
C-13
PubMed
English language, human, date limit: January 1, 2001-November 2, 2011
Set
Number
Concept
Search statement
Equipment
Reprocessing
surgical instruments[mh] OR equipment and supplies[mh] OR

disposable equipment[mh] OR ((surgical[tiab] OR medical[tiab] OR
cutting[tiab]) AND (instrument*[tiab] OR tool*[tiab] OR equipment* OR
device*)) OR trocar*[tiab] OR scalpel*[tiab] OR shaver*[tiab] OR
razor*[tiab] OR drill*[tiab] OR catheter*[tiab] OR syringe*[tiab] OR
needle*[tiab] OR mask*[tiab] OR gown[tiab] OR glove*[tiab] OR
endoscop*[ti] OR instrumentation[sh]
equipment reuse[mh] OR reuse[tiab] OR reusing[tiab] OR reus*[tiab]
OR reprocess[tiab] OR reprocessing[tiab] OR reprocessed[tiab] OR
reprocess*[tiab] OR recycled[tiab] OR recycling[tiab] OR recycle*[tiab]
OR repurposed[tiab] OR repurposing OR repurpose*[tiab]
Single Use
single use[tiab] OR single-use[tiab] OR disposable[tiab] OR onetime[tiab] OR one time[tiab]
Combine
S1 AND S2 AND S3
Total Downloaded
Total Retrieved
Total Included
75
11
15
SECTION B. Methods
in both systematic reviews and medical devices. Included studies consisted of systematic reviews
and clinical studies that compared patient outcomes following use of new versus reprocessed
single-use devices as well as laboratory studies that tested an array of reprocessed single-use
devices for microbiological contamination. Data regarding potential cost-savings of reprocessed
single-use devices was also presented. Included studies were narratively summarized by the
author.
Chapter 13. Preoperative Checklists and Anesthesia

Checklists
Name
Date limits
Platform/provider
EMBASE (Excerpta Medica)
1996 August 23, 2011
OVID SP
MEDLINE
1996 August 23, 2011
OVID SP
PreMEDLINE
1990 August 23, 2011
OVID SP
C-14
Name
Date limits
Platform/provider
CINAHL (Cumulative Index to

Nursing and Allied Health Literature)
Searched June 10, 2011
EBSCOHost
Cochrane Library
Wiley
PubMed
Searched August 17, 2011
Scopus
Searched October 24, 2011
Science Direct

corporations. These documents do not appear in the peer-reviewed journal literature.)
presented in OVID syntax; the search was simultaneously conducted across EMBASE and
MEDLINE. A parallel strategy was used to search the databases comprising the Cochrane
Library.
Medical Subject Headings (MeSH), Emtree and Keywords
Conventions:
OVID
$
=
truncation character (wildcard)
exp
=
explodes controlled vocabulary term (e.g., expands search to all more specific
related terms in the vocabularys hierarchy)
.de. or
/
=
limit controlled vocabulary heading
.fs.
=
floating subheading
.hw. =
limit to heading word
.md. =
type of methodology (PsycINFO)
.mp. =
combined search fields (default if no fields are specified)
.pt.
=
publication type
.ti.
=
limit to title
.tw.
=
limit to title and abstract fields
PubMed
[mh] =
MeSH heading
[majr] =
[pt]
=
publication type
[sb] =
[sh] =
C-15
[tiab] =

Concept
Anesthesia
Anesthesia/
Anesthesiology/
Exp perioperative care/in, ae, mt, st
Checklist
Checklist/
World Health Organization/
Context/Setting
Exp general surgery/

Exp perioperative care/
Exp surgical procedures, operative/
Operating room/
Operating rooms/
Equipment
Electronics, medical/
Equipment failure/
Equipment failure analysis/
is.fs.
Hospital
Procedures/Administration/Protocol
Anesthesia department, hospital/og,

st
Surgery Department, Hospital/og, st
Operating Rooms/og, st
Exp patient care/
Incidence
Exp incidence/
C-16
Keywords
Analges$
Anaesthe$
Anesthe$
Sedat$
Checklist$
safety checklist$
Checkout
Checkt-out
WHO
world health
*world health organization
Intraoperat$
Operat$
Patient$
operating room$
intraoperat$
Perioperat$
Preoperat$ or pre-operat$
Perioperat$ or peri-operat$
Postoperat$ or post-operat$
Surg$
Surgical suite$
Alarm$
Apparatus
Check-out or checkout
Electronic checklist$
Equipment
Error$
Failure
fault
Machine$
Monitor$
System$
Administrat$
Organization$
Patient care
Patient safety
Polic$
Protocol$
Standard$
Decrease$
Incidence
Prevalence
Reduc$
Number
Concept
Interventions
Checklist/
Intraoperative complications/pc
Medical errors/pc
Postoperative complications/pc
Nursing staff, hospital/og, st, ut

Exp Perioperative care/mt, og, st
Exp Preoperative care/mt, st
Obstacles
Attitude/
Attitude of health personnel/
Exp Medical staff, hospital/
Nursing staff, hospital/og, st, ut
Organizational Culture
Health care delivery/

Organizational culture/
C-17
Keywords
An?esthesia adj2 check$
An?esthesia safety
checklist
Checklist$
Communicat$
Document$
Guideline$
Implement$
Information adj shar$
Instrument$
Knowledge
Missed step$
Precaution$
Practice$
Preoperative
Pre-op$
proactive
Postoperative procedure$
Safety measure$
Strateg$
Surg$ adj2 check$
Surgical safety checklist
Team briefing$
Tool$
Anesthesiologist$
Anesthetist$
Attitude$
Barrier$
Competen$
Educat$
Gap$
Knowledge
Nurse$
Obstacle$
overload
Resident$
Resource$
Surgeon$
time
Train$
Change$
Context$
Culture$
Direct$
Hospital$
Manage$
Organization$
Staff
Team$
Unit$
Concept
Outcomes
Safety management/
Quality Management
Health care quality/

Total quality management/
Surgery
Exp surgical procedures, operative/

Exp general surgery/
Operating rooms/
su.fs.
Keywords
Adverse event$
Adverse effect$
Cardiac arrest
coma
Complication$
CPR
Culture change$
Death$
Heart attack
Infection$
Injur$
Myocardial infarction of MI
Outcome$
Patient safety
Perform$
Renal failure
Resuscitat$
Risk management
sepsis
Transfusion$
Ventilat$
Health$
Healthcare$
Hospital$
Quality
TQM
Total quality management
Operat$
operating room$
Surg$
Surgical suite$
Embase/Medline/Premedline
Set
Number
1
Concept
Anesthesia
Context/Setting
5
6
Combine sets for

Anesthesia and
Context/Setting
Patient Safety
Combine sets for Patient

Safety
Search statement
(anesthesia/ OR anesthesiology/ OR (anaesthes$ OR
anesthes$).ti,ab.) OR anesthesia department, hospital/og, st OR exp
patient care/og, st
operating rooms/ OR operating room/ OR exp perioperative care/in, ae,
mt, st, og OR exp surgical procedures, operative/ OR exp general
surgery/ OR su.fs. OR surgery department, hospital/og, st OR
operating rooms/og, st OR exp patient care/og, st
((OR OR operating room$ OR operat$ OR surg$ OR surgical
suite$).ti,ab.) OR ((pre?operative OR pre?op OR peri?operative OR
pre?surgical OR intra?op$).ti,ab.)
1 OR 2 OR 3
patient safety/ OR (patient adj safety).ti,ab.

(safe$ AND ((policy OR policies OR protocol$ OR standard$ OR
administration OR organization) AND (anesthes$ OR
anaesthes$))).ti,ab.
(safe$ AND ((policy OR policies OR protocol$ OR standard$ OR
administration OR organization) AND (surg$ OR operate OR operating
OR operative))).ti,ab.
5 OR 6 OR 7
C-18
Set
Number
9
Concept
Equipment
10
11
12
Combine sets for Equipment

Incidence
13
Medical errors
14
Staff
15
Safety checklists
16
17
18
Combine for Checklists

Safety management
19
20
Combine sets for Safety

Management and Patient
Safety
Total Quality Management
21
Organizational culture
22
Combine sets for TQM and

Staff attitudes and
Organizational Culutre
World Health Organization
checklist
Combine Context/setting
and Checklists
Combine Context/setting
and Checklists with Medical
Errors/pc
Combine Context/Setting
and Checklists with
Incidence
and Checklists with Safety
Combine
Combine Checklists and
Equipment
Combine with medical
errors/pc and safety
and Checklists with TQM
Combine concepts for
Checklists
23
24
25
26
27
28
29
30
31
32
Search statement
((exp equipment failure/ OR electronics, medical/ OR is.fs.) AND
check$.ti,ab.) OR ((equipment or machine$ OR apparatus OR system$
OR monitor$) AND (check$ OR checklist OR check-out OR
checkout)).ti,ab.
((anesthesia OR anaesthesia) AND simulation$ AND check$).ti,ab.
9 OR 10
exp incidence/ OR (decrease$ OR incidence OR prevalence OR
reduc$ OR number$).ti,ab. OR ((before and after) OR (preintervention
OR preintervention OR pretest OR pre-test OR postintervention OR
postintervention OR posttest OR post-test)).ti,ab.
medical errors/pc OR postoperative complications/pc OR ((error$ OR
complication$ OR adverse event$ OR intraoperative awareness OR
wrong$) AND (prevent$ OR control)).ti,ab.
exp medical staff, hospital/ OR nursing staff hospital/og, st, ut OR
attitude of health personnel/ OR attitude/ OR ((nurse$ OR anesthetist$
OR anesthesiologist$ OR resident$ OR surgeon$) AND (knowledge
OR attitude$ OR competen$ OR train$ OR educat$)).ti.
(safety checklist$ OR ((an?esthesia OR surg$) adj2 check$) OR
((surg$ OR pre?surg$ OR pre?op$ OR peri?op$ OR intra?op$) AND
(checkout$ OR checkout$))).ti,ab.
Checklist/
15 OR 16
safety/ OR safety management/ OR safety.ti,ab. OR ((preop$ OR preop$ OR periop$ OR peri-op$ OR pre?surg$) AND (safety OR
precaution$)).ti,ab.
8 OR 18
total quality management/ OR health care quality/ OR ((health$ OR
healthcare OR hospital$) AND (quality OR TQM OR total quality
management)).ti,ab.
health care delivery/ OR organizational culture/ OR ((organization$ OR
hospital$ OR unit$ OR team$ OR staff) AND (culture$ OR change$ OR
manage$ OR direct$ OR context$)).ti,ab.
14 OR 20 OR 21
((*world health organization/ OR World Health Organization.ti,ab. OR

world health.ti,ab.) AND checklist$.mp.)
4 AND 17
24 AND 13
24 AND 12
24 AND 18
25 OR 26 OR 27
11 AND 15
29 AND (13 OR 19)
24 AND 22
28 OR 30 OR 31
C-19
Set
Number
33
34
35
36
37
Concept
Combine with WHO
checklist
Combine for final set
Apply limits
Remove duplicates
Total Downloaded
Search statement
32 OR 23
32 OR 33
Limit 34 to yr=2000-2011
Limit 35 to English language
Remove duplicates from 36
Total Retrieved
Total Included
459
SECTION B. Methods
Patient safety problem: Preoperative checklists can help prevent errors and complications
related to surgery. Checklists are often implemented within a multifactorial strategy of
interventions, therefore they cannot be judged alone as a patient safety practice. The World
Health Organization Surgical Safety Checklist is a prominent example of a preoperative checklist
intended to ensure safe surgery and minimize complications; it has been translated into at least
six languages.1 One family of errors involves wrong site surgery (such as wrong procedure,
wrong site, wrong person), and in 2004, the Joint Commission created the Universal Protocol for
Preventing Wrong Site, Wrong Procedure, Wrong Person Surgery.2 It comprises three sets of
steps: pre-operative verification process, marking the operative site, and a time out
immediately before the operation. A checklist can be used to clarify the details of these three
steps. For anesthesia checklists, in 2008 the American Society of Anesthesiologists provided
general guidelines that should be checked before surgery, and institutions can implement the
guidelines to tailor the checklist to their specific equipment and clinical setting.3
Proposed key questions
1. What is the evidence on the context and implementation of preoperative checklists in
healthcare facilities?
2. What is the evidence on the adoption and diffusion of preoperative checklists in
3. What is the evidence on the effectiveness of preoperative anesthesia checklists in
4. What is the evidence on the context and implementation of preoperative anesthesia
checklists in healthcare facilities?
5. What is the evidence on the adoption and diffusion of preoperative anesthesia checklists
in healthcare facilities?
C-20
PICOTS
Elements
Population
Intervention
Comparison
Outcomes
Timing
Settings
Comments
KQ1 and KQ2: Patients undergoing any surgery.
KQ3, KQ4, and KQ5: Patients undergoing any surgery involving general anesthesia.
Preoperative checklist, either electronic or hard-copy.
KQ1 and KQ2: For a preoperative checklist addressing surgical safety in general, we
examined in detail the World Health Organization Surgical Safety Checklist. For
preoperative checklist specifically designed to implement the Universal Protocol and
prevent wrong-site surgery, any checklist.
KQ3, KQ4, and KQ5: For anesthesia, it must have been an equipment checklist prior
to administering general anesthesia before surgery
KQ1 and KQ2: No comparison required to be reported, but we extracted information
on comparisons that were made.
KQ3, KQ4, and KQ5: Not using a checklist, or a different checklist.
KQ1 and KQ2: No health outcomes necessary to be reported (because these
questions do not involve effectiveness), but we extracted information on outcomes that
were reported
KQ3, KQ4, and KQ5: Rates of intraoperative awareness, any equipment
complications, intraoperative patient complications, postoperative patient
complications
Only examined postoperative events within one month of surgery, because later
events are less likely to have been caused by the surgery itself.
Hospitals and surgical centers
Inclusion criteria:
General inclusion criteria: Full article published in a peer-reviewed journal, Abstracts will be
excluded, English language publications only, published in 2000 or later, preoperative checklist
(either electronic or hard-copy), surgery at either a hospital or a surgical center.
Inclusion criteria for Key Questions 1 and 2:
Patients undergoing any surgery
For preoperative checklists primarily designed to implement the Universal Protocol and
prevent wrong-site surgery: At least 20,000 procedures. This number may change
depending on the size of the literature that meets the inclusion criteria.
For a preoperative checklist addressing surgical safety in general, we examined in detail
the World Health Organization Surgical Safety Checklist.
For preoperative checklist specifically designed to implement the Universal Protocol and
prevent wrong-site surgery, any checklist.
Any study design included, because these questions involve issues of implementation and
adoption, which do not require a comparison set of procedures.
Inclusion criteria for Key Questions 3, 4, and 5:
Patients undergoing any surgery involving general anesthesia
At least 100 procedures. This number may change depending on the size of the literature
that meets the inclusion criteria.
Equipment checklist prior to administering general anesthesia before surgery
Study must either compare the use of a checklist to not using a checklist, or study must
compare checklists. We included any design that made such as comparison (e.g., beforeafter, interrupted time series, or time series with concurrent control group, etc).
C-21
Reported at least one of the outcomes of interest (rates of intraoperative awareness,

equipment complications, intraoperative patient complications, postoperative patient
complications) within one month of the operation
References
1.

Safe Surgery Saves Lives Study Group. A
J Med 2009;360(5):491-9.
Msa0810119. PMID: 19144931
2.
Facts about the universal protocol for

preventing wrong site, wrong procedure and
wrong person surgery. Oakbrook Terrace,
IL: The Joint Commission; 2004.
www.jointcommission.org/PatientSafety/Un
iversalProtocol/up_facts.htm. Accessed
January 9, 2008.
3.

(IL): American Society of Anesthesiologists;
2008.
Chapter 14. Use of Report Cards and Outcome

Measurements To Improve the Safety of Surgical Care:
American College of Surgeons National Quality Improvement
Program (NEW)
PubMed 2000-11/26/2011
LANGUAGE:
English
SEARCH STRATEGY:
American College Surgeon* AND National Surgical Improvement Program
In addition to searching the published literature, this topic relied on evidence available on the
American College of Surgeons NSQIP website at http://www.acsnsqip.org/. Interviews with
leadership and administrators in ACS NSQIP were performed. Surgeon champions were
questioned.
In addition to searching the published literature, this topic relied on evidence available on the
American College of Surgeons NSQIP website at http://www.acsnsqip.org/.
C-22
SECTION B. Methods
Evidence from the literature and the ACS NSQIP website was reviewed by a general surgeon
health services researcher with experience in systematic reviews. The synthesis was narrative.
Chapter 15. Prevention of Surgical Items Being Left Inside a

Patient: Brief Update Review
Name
Date limits
Platform/provider
Cochrane Library
Searched November 3, 2011
Wiley
ECRI Institute
PSNet

and Quality
PubMed

Cochrane Library.
Conventions:
OVID
$
= truncation character (wildcard)
C-23
exp
= explodes controlled vocabulary term (e.g., expands search to all more specific
.de. or
/
=
.fs.
=
.hw. =
.md. =
.mp. =
.pt.
=
.ti.
=
.tw.
=
PubMed
[mh] =
[majr] =
[pt]
=
[sb] =
[sh] =
[tiab] =

floating subheading
publication type
limit to title
MeSH heading
publication type

Concept
Foreign bodies
Foreign bodies[mh]
Retained item
Medical errors
Medical errors[mh]
Surgical
Surgical instrument[mh]
Technology
Keywords
Foreign
Gossypiboma
nothing left behind
Retained
Body
Bodies
Instrument*
Sponge*
Tool*
Error*
Medical
never event
Prevent*
Surgical
Surgery
Surgical
bar code
Bar-code
count
RFID
Tag*
Set
Number
Concept
Search statement
Foreign bodies
Foreign bodies[mh] OR ((foreign OR retained) AND (instrument* OR

sponge* OR body OR bodies)) OR gossypiboma[tiab]
Surgery
Surgical[tiab] OR surgery[tiab]
C-24
Set
Number
Concept
Search statement
Medical errors
Medical errors[mh] OR ((medical OR surgical) AND (error* OR never

event))
Surgical tools
Surgical instruments[mh] OR (surgical AND (tool* OR instrument*))
Technology
RFID OR tag* OR bar code OR bar-code OR count
Term
nothing left behind
Combine
S1 AND S2 AND S3 AND S4
Combine
S1 AND S5
S6 OR S7 OR S*
10
Apply date limit
2000-2011
Total Downloaded
Total Retrieved
Total Included
70
20
13
SECTION B. Methods
in both systematic reviews and medical devices. Included studies were those most relevant to the
risk and prevention of retained foreign objects as a result of surgery. We examined studies on
manual counting, as well as those using various forms of radiofrequency identification. Potential
barriers to implementation (e.g. user compliance) and the costs of various technologies were also
assessed. Included studies were narratively summarized by the author.
Chapter 16. Operating Room Integration and Display

Systems: Brief Review (NEW)
Name
Date Limits
Platform/Provider
PubMed
2000-November 28, 2011
CINAHL (Cumulative
Index to Nursing and
Allied Health Literature)
2000-November 29, 2011
EBSCOhost
ECRI Institute website
2000-November 29, 2011
ECRI Institute
Health Devices
2000-November 29, 2011
ECRI Institute
C-25
Name
Date Limits
Platform/Provider
Institute for Healthcare

Improvement
2000-November 29, 2011
http://www.ihi.org
Joint Commission
2000-November 29, 2011
http://www.jointcommission.org/
Patient Safety Network
2000-November 30, 2011

(AHRQ): http://psnet.ahrq.gov/
Pennsylvania Patient
Safety Authority PASR
(PA Safety Authoritypatient safety reporting
system)
2000-November 30, 2011
http://patientsafetyauthority.org/Pages/Default.aspx#
Google
2000-November 30, 2011
http://www.google.com

Cochrane Library.
Medical Subject Headings (MeSH), and Keywords
Conventions:
PubMed
*
=
[mh] =
[majr] =
[pt]
=
[sb] =
[sh] =
[tiab] =
[ti]
=

MeSH heading
publication type
keyword in title
C-26

Concept
Operating rooms
Controlled Vocabulary*
operating rooms[majr]
Integration
operating room information

systems[mh]
systems integration[mh]
video-assisted surgery[majr]
Information itself
Transmission of
information
computer communication
networks[majr]
monitoring,
intraoperative/instrumentation[majr]
videorecording[majr]
C-27
Keywords
operating room/s
operating suite/s
surgery suite/s
surgical suite/s
OR/s
C-suite/s
hybrid operating suite/s
supersuite/s
super suite/s
VIOR/s (visually integrated operating room)
surgical field/s
integration
integrated
central control
centralized control
centralised control
common location
single location
plug and play
router
routing
audio
AV
camera
data
EMR
image/s
imaging
PACS
picture archiving
video
communication/s
digital
Concept
Display of
information
Controlled Vocabulary*
data display[mj]
Keywords
boom
cart-mounted
dashboard/s
display/s
flat panel
high definition
HD
LCD/s
monitor/s
television/s
touch screen/s
touchscreen/s
TV/s
screen/s
surgical display/s
streaming
video
VCR
wall-mounted
workstation/s
*Note: none of the MeSH terms were specific enough to retrieve results relevant to the topic and were not incorporated into the
main search strategy.
PubMED main search

Set Number
Concept
Search Statement
Main concept
integrated operating room[tiab] OR integrated

operating rooms[tiab]
Setting
operating rooms[mh] OR operating room[tiab] OR

operating rooms[ti] OR operating suite[tiab] OR
operating suites[tiab] OR surgical suite[tiab] OR
surgical suites[tiab] OR surgery suite[tiab] OR
surgery suites[tiab]
Descriptive concept
integrat*[ti]
Additional descriptive concepts
central control[tiab] OR centralized control[tiab] OR

centralised control[tiab] OR dashboard*[tiab] OR
digital[ti] OR display*[ti] OR high definition[tiab] OR
PACS[tiab] OR plug and play[tiab] OR screen[ti] OR
screens[ti] OR video[ti]
Combine sets
2 AND (3 OR 4)
Combine sets
1 OR 5*
*Note: no publication limits or safety concepts were applied to the final set of search results because there were few results.
C-28
PUBMED ADDITIONAL TERMS THAT WERE BROWSED*

Set Number
1
Concept
Setting
Search Statement
C-suite OR C-suites OR hybrid operating OR
supersuite OR super suite
OR
ORs
operating room AND future
VIOR*
2
Concepts
integrat*[tiab]
central[ti] OR central location OR common location
OR single location OR (visually[ti] AND
integrated[tiab])
router*[tiab] OR routing[tiab]
audio* OR AV
camera*[ti] OR image[ti] OR images[ti] OR imaging[ti]
OR picture archiving OR video*[ti] OR (visually[ti]
AND integrated[tiab])
EMR[tiab]
cart mounted OR touch screen OR touch screens
OR touchscreen* OR workstation*
boom* OR flat panel OR TV* OR television* OR
wall-mounted OR HD OR LCD* OR VCR* OR
streaming
communication[ti] OR communications[ti] OR data[ti]
OR data integration[tiab] OR digital integration[tiab]
monitor[ti] OR monitors[ti] OR surgical display OR
surgical displays OR (surgical[ti] AND display*[ti]) OR
(surgical field AND display[tiab]) OR (surgical field
AND monitor*[tiab])
3
Mesh concepts
Computer communication networks[mj] OR data
display[mj] OR monitoring,
intraoperative/instrumentation[mj] OR systems
integration[mj] OR video-assisted surgery[mj] OR
video recording[mj]
*Note: The terms in this table mainly yielded no results, irrelevant results, or large sets of results with few relevant. Some
references were identified and kept during searches with these terms, but the terms were not useful enough to include in the main
search strategy above.
CINAHL
Set Number
Concept
Search Statement
Setting
operating room OR operating room
Main concept
integrat*
Combine
1 AND 2
Total Downloaded
Total Retrieved
Total Included
140
44 instructions
18
SECTION B. Methods
systematic review. Only full published studies were considered for review (meeting abstracts
were excluded). Only English-language publications were eligible for inclusion. For the
effectiveness and harms of the PSP, we considered including studies of any design (e.g.,
C-29
randomized controlled trials, non-randomized controlled trials, prospective and retrospective

observational studies, surveys) that may provide relevant data. For the implementation of the
PSP, we considered including any qualitative or quantitative research that addressed the
implementation issues. Included studies were narratively summarized by the author. As this PSP
project team has agreed, we did not assess risk of bias of included individual studies or the
overall strength of evidence for this brief PSP review.
Chapter 17. Use of Beta Blockers To Prevent Perioperative

Cardiac Events: Brief Update Review
For this topic we did not do a formal literature search. Rather the principal meta-analyses and
trials were already known to the authors as part of their clinical work. A related articles search
was done on these to look for any additional relevant publications.
SECTION B. Methods
The meta-analyses, trials, and related articles search were reviewed by a physician health
services researcher with experience in cardiovascular systematic reviews. The synthesis was
narrative.
Chapter 18. Use of Real-Time Ultrasound Guidance During

Central Line Insertion: Brief Update Review
PubMed 2000-12/15/2011
SEARCH #1:
ultrasound OR ultrasonograph*
AND
guided OR guidance OR ultrasound-guided OR doppler-guided OR ultrasound-assisted
AND
catheter* OR cannulat*
AND
vein OR veins OR venous OR vascular
MANUALLY FILTERED IN ENDNOTE FOR THE FOLLOWING JOURNALS:
JAMA
New England Journal of Medicine
British Medical Journal
Lancet
Annals of Internal Medicine
Critical Care Medicine
Journal of Clinical Monitoring
C-30
Anaesthesia
Circulation
Chest
Anesthesia & Analgesia
Annals of Emergency Medicine
Anesthesiology
Archives of Surgery
NUMBER OF RESULTS AFTER FILTERING: 92
============================================================
SEARCH #2 (RCTS OR META-ANALYSES):
PubMed 2000-12/16/2011
LANGUAGE:
English
SEARCH STRATEGY:
AND
AND
AND
AND
randomized controlled trial* OR randomized controlled trial[pt] OR rct* OR double-blind* OR
single-blind* OR double blind OR single blind OR systematic review OR meta-analy* OR
metaanaly* OR meta analy* OR Meta-Analysis[pt]
NUMBER AFTER REMOVING REFERENCES FROM SPECIFIED JOURNALS
LIST: 74
====================================================================
SEARCH #3
SCOPUS 2000-12/16/2011
LANGUAGE:
English
SEARCH STRATEGY:
C-31
TITLE-ABS-KEY((ultrasound OR ultrasonograph*)
AND
AND
TITLE-ABS-KEY-AUTH(catheter* OR cannulat*)
AND
AND
SUBJAREA(mult OR medi OR nurs OR vete OR dent OR heal OR mult OR arts OR busi OR
deci OR econ OR psyc OR soci)
MANUALLY FILTERED IN ENDNOTE FOR RCTS OR META-ANALYSES
====================================================================
SEARCH #4
Web of Science 2000-12/16/2011
LANGUAGE:
English
SEARCH STRATEGY:
AND
AND
AND
MANUALLY FILTERED IN ENDNOTE FOR RCTS OR META-ANALYSES
B. Methods
Titles and abstracts were reviewed by a general internist with experience in systematic reviews
(Figure 1). Relevant articles were narratively summarized. This summary was reviewed by the
second author, a general internist experienced in the implementation and use of ultrasound for
central-line placement, who suggested several additional references and described program
implementation at one health care site.
C-32
Figure 1, Chapter 18. Literature flow diagram
* A meta-analysis is also included in these totals, hence numbers sum to more than the total.
Three of these studies were also included in a meta-analysis on the topic.
Chapter 19. Preventing In-Facility Falls

DATABASE SEARCHED AND TIME PERIOD COVERED:
PubMed: 2005-8/1/2011
LANGUAGE:
English
SEARCH STRATEGY:
Accidental Falls[Mesh] OR fallers[tiab] OR falls per[tiab] OR falls rate[tiab] OR falls
incidence[tiab] OR falls prevention[tiab] OR fall prevention[tiab] OR prevention of
falls[tiab] OR prevent falls[tiab] OR prevents falls[tiab] OR prevent patient falls[tiab]
OR prevents patient falls[tiab] OR preventing fall[tiab] OR preventing falls[tiab] OR
falls reduction[tiab] OR fall reduction[tiab] OR reduction of falls[tiab] OR reduce
falls[tiab] OR reduces falls[tiab] OR reducing fall[tiab] OR reducing falls[tiab] OR
C-33
improve fall[tiab] OR improve falls[tiab] OR improves fall[tiab] OR improves

falls[tiab] OR improving fall[tiab] OR improving falls[tiab]
AND
hospital OR hospitals OR hospitali*
NOT
Publication Type:Meta-Analysis, Review

CINAHL: 2005-8/2/2011
SEARCH STRATEGY:
Accidental Falls OR fallers OR falls per OR falls rate OR falls incidence OR falls
prevention OR fall prevention OR prevention of falls OR prevent falls OR prevents
falls OR prevent patient falls OR prevents patient falls OR preventing fall OR
preventing falls OR falls reduction OR fall reduction OR reduction of falls OR reduce
falls OR reduces falls OR reducing fall OR reducing falls OR improve fall OR
improve falls OR improves fall OR improves falls OR improving fall OR improving
falls
AND
hospital OR hospitals OR hospitali*
NUMBER AFTER REMOVAL OF DUPLICATES: 524
====================================================================
WEB OF SCIENCE Science Citation Index, Social Science Citation Index, Arts &
Humanities Index, Conference Proceedings Science Index, Conference Proceedings Social
Science Index: 2005-8/5/2011
SEARCH STRATEGY:
Topic=(Accidental Falls OR fallers OR falls per OR falls rate OR falls incidence OR
falls prevention OR fall prevention OR prevention of falls OR prevent falls OR
prevents falls OR prevent patient falls OR prevents patient falls OR preventing fall OR
preventing falls OR falls reduction OR fall reduction OR reduction of falls OR reduce
falls OR reduces falls OR reducing fall OR reducing falls OR improve fall OR
improve falls OR improves fall OR improves falls OR improving fall OR improving
falls)
AND
Topic=(hospital OR hospitals OR hospitali*)
C-34

SECTION B. Methods
Articles identified by Hempel and colleagues using the above process were then reviewed by us
using the following criteria:
Acute care hospitals
With large sample sizes (at least N=1,000)
General population or older adult population
From the Prevention of Falls Newtork Europe (ProFANE) Manual for the fall prevention
classification system: Domain 3: Components (Combination sub-section)1
Combination
Most interventions fall under the following sub-domains (detailed under Domain 4: Descriptors of the intervention).
Exercises (supervised and/or unsupervised)
Medication (drug target)
Surgery
Management of urinary incontinence
Fluid or nutrition therapy
Psychological
Environment/Assistive technology
Social environment
Knowledge/education interventions
Other interventions/procedures
Combination refers to how many sub-domains are delivered to the participants of an intervention, and importantly, the
manner in which these sub-domains are combined.
1
Lamb SE, Hauer K, Becker C. Manual for the fall prevention classification system. 2007.
www.profane.eu.org/documents/Falls_Taxonomy.pdf.
Chapter 20. Preventing In-Facility Delirium

Name
Date Limits
Platform/Provider

EBSCOHost
Cochrane Library
Wiley
1996 August 23, 2011
OVID SP
MEDLINE
1996 August 23, 2011
OVID SP
PreMEDLINE
1990 August 23, 2011
OVID SP
PubMed
C-35

presented in OVID syntax; the search was simultaneously conducted across Embase and
Medline. A parallel strategy was used to search the databases comprising the Cochrane Library.
Medical Subject Headings (MeSH), EMTREE and Keywords
Conventions:
OVID
$
=
exp
=
.de. or
/
=
.fs.
=
floating subheading
.hw. =
.md. =
.mp. =
.pt.
=
publication type
.ti.
=
limit to title
.tw.
=
PubMed
[mh] =
MeSH heading
[majr] =
[pt]
=
publication type
[sb] =
[sh] =
[tiab] =
C-36

Concept
Anesthesia
Pre/Peri/Intra/Postoperative
period
Surgery
Disease/Condition
Intervention program
Analgesic agent/adverse drug reaction,
drug interaction, drug toxicity,
pharmacokinetics, pharmacology
Analgesics, opioid/ae
Exp analgesics/adverse effects,
pharmacokinetics, poisoning, toxicity
Exp perioperative care/
Exp Postoperative complication/
Postoperative care/
Delirium/
Delirium/et
Delirium/pc
Delirium/prevention and control

Dt.fs.
Tu.fs.
C-37
Keywords
Anaesthe$
Analgesic$
Anesthe$
Complicat$
Opioid$
Postop$
Sedat$
Acute confusional state

Cause$
Complicat$
Control$
Delirium
Develop$
Effect$
etiology
Event$
Outcome$
Prevent$
Result$
Sundown syndrome
Approach$
Barrier$
barrier*
Checklist$
Collaborat$
control
Delirium
Exercise$
Families
Family
hospital elder life
Implement$
Initiative$
Intervention$
Monitor$
movement
non-pharma$
non-pharmacolog$
obstacle$
occupational therapy
Pharma
Pharmacologic$
Physical therap$
plan
Prevention
Program$
Project$
prophylactic$
protocol
reduc$
screen$
sleep
strateg$
volunteer$
walk*
Concept
Risk/Screening
Delirium/et
Exp Risk/ or Risk Assessment/
Setting
Intensive care units/
Keywords
Assess
Assessment
Checklist
Examination
Examine
History
Interview
Predict
Prediction
Predictor
Risk
Survey
admission
Hospital$
Hospital-acquired
Iatrogenic
Inpatient
Intensive care
ICU
nosocomial
patient
Set
Number
1
2
Concept
Disease/Condition
3
4
Combine sets
Risk
Risk of developing delirium
Combine sets for risk of

developing delirium
Setting/Context
8
9
10
11
Combine sets for Setting

Combine for final set of risk of
developing delirium in hospital
settings
Postoperative complications
12
13
14
15
16
17
Combine sets for Postoperative

complications
Sedation
Search Statement
*Delirium/
(deliriuim or sundown syndrome or acute confusional
state).ti,ab.
1 or 2
exp risk/ or risk assessment/ or (epidemiology or etiology or
prevention).fs. or (avoid$ or caus$ or risk$ or predict$ or
prevent$).ti,ab.
(delirium/et or (delirium and (cause$ or result$ or outcome$ or
complicat$ or etiology or develop$ or effect$ or event$)).ti,ab.)
(3 AND 4) OR 5
(hospital or hospitals or hospitaliz* or hospitalis* or inpatient$ or
iatrogenic or admission or admitted or ICU or intensive care
or post anesthesia or post anaesthesia or post surgery or
post surgical or postoperative or post operative).ti.
exp hospitalization/ or exp intensive care units/
7 OR 8
6 AND 9
exp postoperative complication/ or (postop$ adj2

complication$).ti,ab.
exp perioperative care/ or (sedat$ or analgesic$ or anesthe$ or
anaesthe$ or opioid$).ti,ab.
exp surgical procedures, operative/ae or (surgery or surgical or
intraoperative or intra-operative).ti,ab.
11 OR 12 OR 13
analgesics, opioid/ae
analgesic agent/ae, it, to, pk, pd [Adverse Drug Reaction, Drug
Interaction, Drug Toxicity, Pharmacokinetics, Pharmacology]
exp Analgesics/ae, pk, po, to [Adverse Effects,
Pharmacokinetics, Poisoning, Toxicity]
C-38
Set
Number
18
19
20
21
22
Concept
Combine sets for Sedation
or Sedation Complications
or Sedation Complications and
Risk of Delirium
Disease/Condition prevention
and control
Interventions
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Combine sets for Interventions

Combine sets for Disease control
and prevention and Interventions
Incidence
Barriers
Combine sets for Barriers to
Disease control and prevention
and Interventions and Incidence
of delirium
Combine Disease/Condition and
Disease prevention and control
Quality improvement hedge
Combine Disease and Quality
Improvement hedge
Combine final sets for review
Limit
Limit
Apply Systematic Review narrow
hedge
Remove duplicates
Search Statement
15 or 16 OR 17
14 OR 18
19 AND 6
*Delirium/pc or (delirium and (prevent$ or control$)).ti,ab.

(interven$ or initiative$ or program$ or project$ or plan$ or
protocol$ or monitor$ or checklist$ or collabor$ or approach$ or
screen$ or strateg$).ti,ab.
(exercise$ or walk or family or families or movement or nonpharma$ or occupational therap$ or physical therap$ or sleep or
hydrat$ or volunteer$).ti,ab. OR hospital elder life.mp.
(pharma$ or drug$ or medication$ or prophylactic$ or
therap$).ti,ab. or dt.fs. or tu.fs.
program evaluation/ or program development/ or safety
management/methods or models, organizational/ or clinical
effectiveness/evaluation or quality assurance, health care/ or
((clinical or medical) adj (protocol* or checklist* or
documentation*)).ti,ab.
22 OR 23 OR 24 OR 25
21 AND 26
exp incidence/ or (incidence or prevalence or rate or increase or
decrease or reduc$ or number).mp.
(barrier$ or obstacle$ or resource$ or cost$ or time).ti,ab.
27 AND 28 AND 29
3 OR 21
(quality and improv$ and intervention$).mp.
31 AND 32
10 OR 20 OR 30 OR 33
Limit 34 to yr=2000-2011
36 AND ((research synthesis or pooled).mp. or systematic
review/ or meta analysis/ or meta-analysis/ or ((evidence base$
or methodol$ or systematic or quantitative$ or studies or
search$).mp. and (review/ or review.pt.)))
Total Downloaded
Total Retrieved
Total Included
587
301
85
SECTION B. Methods
Inclusion/Exclusion Criteria
General criteria: Only full published studies were considered for review (meeting abstracts were
excluded). Only English-language publications were eligible for inclusion.
C-39
Risk factors:
Included RCTs comparing groups with different risk factors; also prospective and
retrospective cohort studies that perform multivariate analyses of factors associated with
incidence of delirium.
Comparative studies must have at least 20 patients in each arm, while cohort studies must
have at least 20 patients overall.
Effectiveness and harms:
Included RCTs, controlled clinical trials (CCTs), interrupted time series, and controlled
before-after studies (CBAs) where at least the after-intervention portion is prospective;
CBAs are necessary to look at implementation (KQ6).
Studies must have at least 20 patients in each arm.
Implementation and Context:
Abstracted information on implementation and context from effectiveness studies, and
descriptive studies of implementation with an associated effectiveness study
Qualitative research studies addressing implementation of delirium prevention
interventions
Quantitative research studies on implementation of delirium prevention interventions
Chapter 21. Preventing In-Facility Pressure Ulcers

Name
Date Limits
Platform/Provider

1981 June 9, 2011
EBSCOHost
Cochrane Library
Wiley
1996 September 15, 2011
OVID SP
MEDLINE
1996 September 15, 2011
OVID SP
PreMEDLINE
OVID SP

C-40

MEDLINE. A parallel strategy was used to search the databases comprising the Cochrane
Library.
Conventions:
OVID
$
=
exp
=
.de. or
/
=
.fs.
=
floating subheading
.hw. =
.md. =
.mp. =
.pt.
=
publication type
.ti.
=
limit to title
.tw.
=
PubMed
[mh] =
MeSH heading
[majr] =
[pt]
=
publication type
[sb] =
[sh] =
[tiab] =
Concept
Pressure ulcers
Pressure ulcer/
Skin ulcer/ or skin ulcers/
Decubitus/ or decubitus ulcer/
C-41
Keywords
bed sore$
bedsore$
decubitus adj ulcer$
pressure sore$
Pressure ulcer$
pressure ulcer$
pressure wound$
Skin ulcer$
wound
Concept
Intervention program
clinical effectiveness/evaluation
exp health care quality/
models, organizational/
program development/
program evaluation/
quality assurance, health care/
quality of health care or quality
assurance,health care or quality
indicators, health care or health
plan implementation
safety management/methods
Barriers
attitude of health personnel/

clinical competence/
education, medical/
health knowledge, attitudes,
practice/
physician practice patterns/
staff, hospital/education
Setting
exp health care organization/

exp health planning organizations/
Effectiveness/Measure
Exp incidence/
Exp prevalence/
Exp vital statistics/
C-42
Keywords
Checklist
checklist*
Clinical checklist
Clinical documentation
clinical protocol
Implement*
implement*
Initiative
initiative*
Intervention
intervention
Medical checklist*
Medical documentation*
Medical protocol*
Program
program*
Protocol
protocol
quality and improvement and
intervention$
Standard
standard*
train*
Training
barrier$
Barriers
Compliance
compliance$
imped$
(nurse$ or physician$ or staff or
employee) and (educat$ or
train$ or knowledge or attitude$
or competen$ or time)
obstacle$
outcome$
Outcomes
alliance$
coalition$
collaborat$
health care or healthcare medical
health system$
hospital$
network$
decrease
incidence
increase
number
prevalence
rate
Set
Number
1
2
3
4
5
6
7
Concept
Disease/Condition
Combine sets for Disease

Intervention/Program
Implementation/Quality
improvement
Search Statement
*pressure ulcer/ or pressure ulcer*.ti,ab. or ((skin ulcers/ or skin ulcer/)
and pressure.ti,ab.)
(exp decubitus/ or exp decubitus ulcer/) and skin.mp.
(pressure adj2 (sore$ or ulcer$ or wound$)).ti,ab.
(bedsore$ or (bed adj2 sore$)).ti,ab.
(decubitus adj ulcer$).ti,ab.
1 or 2 or 3 or 4 or 5
program evaluation/ or program development/ or safety
management/methods or models, organizational/ or clinical
effectiveness/evaluation or quality assurance, health care/ or ((clinical or
medical) adj (protocol* or checklist* or documentation*)).ti,ab.
(quality and improv$ and intervention$).mp.
(implement* or initiative* or program* or intervention or train* or
checklist* or standard* or protocol).mp.
exp health care quality/ or (quality of health care or quality
assurance,health care or quality indicators, health care or health plan
implementation).sh.
Combine sets for

improvement
7 or 8 or 9 or 10
8
9
10
11
12
Obstacles/Barriers
13
14
15
Combine sets for

Obstacles
Context/Setting
16
17
18
19
20
21
22
23
24
25
Combine sets for Setting

Combine
Disease/Condition and
improvement
Combine Disease and
improvement
Combine Obstacles and
Disease and
improvement and Setting
Remove duplicates
Limit by date
Incidence or prevalence
Combine for Incidence or

prevalence
attitude of health personnel/ or education, medical/ or staff,

hospital/education or clinical competence/ or health knowledge,
attitudes, practice/ or physician practice patterns/
(imped$ or obstacle$ or barrier$ or outcome$ or compliance$ or
((nurse$ or physician$ or staff or employee) and (educat$ or train$ or
knowledge or attitude$ or competen$ or time))).mp.
12 or 13
(hospital$ or hospital-acquired or inpatient$ or patient$ or acute care or
long term care or long-term care).ti,ab.
exp health care organization/ or exp health planning organizations/ or
((hospital$ or health system$ or health care or healthcare or medical)
and (collaborat$ or alliance$ or coalition$ or network$)).mp.
15 or 16
6 and 11
14 and 18
19 and 17

Limit 21 to yr=2000-2011
exp incidence/ or exp prevalence/ or (incidence or prevalence or rate or
increase or decrease or number).mp.
exp Demography/sn [Statistics & Numerical Data]
23 or 24
C-43
Set
Number
26
Concept
Combine Obstacles and
Disease and
improvement and Setting
and Incidence
Limit
Limit
Combine Disease and
Quality Improvement
Limit
Limit
Limit
27
28
29
30
31
32
33
Search Statement
22 and 25

Limit 27 to human
6 and 8
22 or 28 or 29
Limit 30 to yr=2000-2011
Limit 32 to human
Total Downloaded
Total Retrieved
Total Included
454
87
47
CINAHL
Set
Number
Concept
Search Statement
S1
Disease/Condition
(MM Pressure Ulcer) OR (MH Pressure Ulcer Care (Saba
#
Downloaded
CCC)) OR (MH Pressure Ulcer Stage 1 Care (Saba CCC))

OR
(MH Pressure Ulcer Stage 2 Care (Saba CCC)) OR (MH
Pressure Ulcer Stage 3 Care (Saba CCC)) OR (MH Pressure
Ulcer Stage 4 Care (Saba CCC)) OR (MH Pressure Ulcer
Care (Iowa NIC)) OR (MH Pressure Ulcer Prevention (Iowa
NIC)) OR TI pressure ulcers
S2
Program
implement* OR program* OR initiative* OR protocol* OR

checklist* OR train* OR standard*
S3
Obstacles
Barrier* OR outcome* OR compliance*
S4
Protocols/Guidelines
(MH Nursing Protocols) OR (MH Research Protocols) OR

(MH Protocols) OR (MH Guideline Adherence) OR protocol
S5
Combine Programs
and Protocols
S2 OR S4
S6
Combine Condition
and Obstacles AND
Programs or
Protocols
S1 AND S3 AND S5
S7
Limit
Limit S6 to 2000-2011
350
From S7 keep
64
C-44
SECTION B. Methods
Inclusion criteria:
Experimental research studies including randomized controlled trials, non-randomized
controlled trials, pre-post studies (or before and after studies), and cohort studies that
evaluated the implementation of a multicomponent pressure ulcer (PU) prevention
programs
Published post-2000 and conducted in the U.S.
Study must report on PU rate (incidence/prevalence)
Studies must report PU rate for at least 6 months postimplementation of prevention
program
Exclusion criteria:
Studies that did not report a baseline (pre-prevention program implementation) PU rate
Studies with less than 50% of patient population at study end
Studies focused on PU risk assessment or singular interventions that prevent PUs
(e.g., special mattresses, skin care items, etc.). These topics are currently covered in a
separate comparative effectiveness review.
C-45
Figure 1, Chapter 21. Study attrition diagram
Chapter 22. Inpatient Intensive Glucose Control Strategies To

Reduce Death and Infection (NEW)
Search for studies about cost
Database: Ovid MEDLINE(R) and Ovid OLDMEDLINE(R) <1948 to January Week 2 2010>
Search Strategy: yield through January 2010, updated October 2011
-------------------------------------------------------------------------------1 exp insulin/ (135826)
2 exp hypoglycemic agents/ (159801)
3 exp Blood Glucose/ (105380)
4 (insulin or hypoglycemic agent$ or hypoglycaemic agent$ or glycemic control or glycaemic
control).mp. {mp=title, original title, abstract, name of substance word, subject heading word,
unique identifier} (264080)
5 1 or 2 or 3 or 4 (316621)
6 Critical Illness/ (10449)
7 critical care/ or intensive care/ (31575)
8 exp Perioperative Care/ (65391)
9 exp Postoperative Period/ (30372)
10 ((critical$ adj6 ill$) or critical care or icu or intensive care or burn unit$ or coronary
care).mp. {mp=title, original title, abstract, name of substance word, subject heading word,
unique identifier} (118455)
C-46
11 intensive care units/ or burn units/ or coronary care units/ or recovery room/ (31083)
12 postoperative complications/ or prosthesis-related infections/ or surgical wound dehiscence/
or surgical wound infection/ (269644)
13 (postoperative$ or post operative$).mp. {mp=title, original title, abstract, name of substance
word, subject heading word, unique identifier} (501444)
14 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 (643228)
15 5 and 14 (6711)
16 randomized controlled trial.pt. (278973)
17 controlled clinical trial.pt. (79853)
18 randomized controlled trials.sh. (0)
19 random allocation.sh. (66268)
20 double blind method.sh. (103038)
21 single blind method.sh. (13368)
22 16 or 17 or 18 or 19 or 20 or 21 (416935)
23 (animals not human).sh. (4467853)
24 22 not 23 (374747)
25 clinical trial.pt. (452229)
26 exp clinical trials/ (0)
27 (clin$ adj25 trial$).ti,ab. (166370)
28 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. (103187)
29 placebos.sh. (28486)
30 placebo$.ti,ab. (118661)
31 random$.ti,ab. (460194)
32 research design.sh. (57708)
33 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 (924508)
34 33 not 23 (807020)
35 34 or 24 (836845)
36 15 and 35 (1138)
37 exp Myocardial Infarction/ (124416)
38 exp Hospitalization/ (120073)
39 exp Inpatients/ (8203)
40 exp Cerebrovascular Accident/ (55676)
41 cerebrovascular disorders/ or brain ischemia/ or exp intracranial embolism and
thrombosis/ or exp intracranial hemorrhages/ (120133)
42 exp myocardial revascularization/ or exp coronary artery bypass/ (64411)
43 37 or 40 or 41 or 42 (330035)
44 5 and 43 (4681)
45 35 and 44 (689)
46 45 not 36 (556)
47 38 or 39 (126877)
48 5 and 47 (1329)
49 35 and 48 (243)
50 49 not (36 or 46) (130)
51 exp Hypoglycemia/ci, ep, et {Chemically Induced, Epidemiology, Etiology} (9091)
52 1 or 2 or 4 (271238)
53 51 and 52 (5827)
C-47
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
14 and 53 (251)
43 and 53 (52)
47 and 53 (89)
54 or 55 or 56 (362)
57 not (36 or 46 or 49) (312)
exp Hypoglycemia/ (18273)
52 and 59 (10136)
14 and 60 (400)
43 and 60 (101)
47 and 60 (126)
61 or 62 or 63 (580)
64 not 57 (218)
65 not (36 or 46 or 49) (193)
exp Costs and Cost Analysis/ (145993)
15 and 67 (51)
exp Economics/ (413412)
ec.fs. (261917)
69 or 70 (488778)
15 and 71 (82)
72 not 68 (31)
from 73 keep 1-31 (31)
Search for trials

Database: Ovid MEDLINE(R) <1950 to November Week 2 2007>
1 exp insulin/ (130835)
2 exp hypoglycemic agents/ (151706)
3 exp Blood Glucose/ (98489)
control).mp. {mp=title, original title, abstract, name of substance word, subject heading word}
(243159)
5 1 or 2 or 3 or 4 (292506)
6 Critical Illness/ (8301)
7 critical care/ or intensive care/ (28092)
8 exp Perioperative Care/ (60582)
9 exp Postoperative Period/ (28181)
10 ((critical$ adj6 ill$) or critical care or icu or intensive care or burn unit$ or coronary
care).mp. {mp=title, original title, abstract, name of substance word, subject heading word}
(103498)
11 intensive care units/ or burn units/ or coronary care units/ or recovery room/ (27247)
12 postoperative complications/ or prosthesis-related infections/ or surgical wound dehiscence/
or surgical wound infection/ (252519)
13 (postoperative$ or post operative$).mp. {mp=title, original title, abstract, name of substance
word, subject heading word} (457854)
14 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 (582795)
C-48
15 5 and 14 (5822)
16 randomized controlled trial.pt. (246761)
17 controlled clinical trial.pt. (77022)
18 randomized controlled trials.sh. (52472)
19 random allocation.sh. (59778)
20 double blind method.sh. (94781)
21 single blind method.sh. (11591)
22 16 or 17 or 18 or 19 or 20 or 21 (418296)
23 (animals not human).sh. (4261058)
24 22 not 23 (382274)
25 clinical trial.pt. (444490)
26 exp clinical trials/ (199910)
27 (clin$ adj25 trial$).ti,ab. (139332)
28 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. (94254)
29 placebos.sh. (26956)
30 placebo$.ti,ab. (106977)
31 random$.ti,ab. (394441)
32 research design.sh. (50582)
33 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 (887876)
34 33 not 23 (778635)
35 34 or 24 (798240)
36 15 and 35 (979)
37 exp Myocardial Infarction/ (115916)
38 exp Hospitalization/ (107713)
39 exp Inpatients/ (6673)
40 exp Cerebrovascular Accident/ (44100)
41 cerebrovascular disorders/ or brain ischemia/ or exp intracranial embolism and
thrombosis/ or exp intracranial hemorrhages/ (112871)
42 exp myocardial revascularization/ or exp coronary artery bypass/ (56866)
43 37 or 40 or 41 or 42 (300510)
44 5 and 43 (4061)
45 35 and 44 (657)
46 45 not 36 (544)
47 38 or 39 (113294)
48 5 and 47 (1078)
49 35 and 48 (202)
50 49 not (36 or 46) (114)
51 exp Hypoglycemia/ci, ep, et {Chemically Induced, Epidemiology, Etiology} (8651)
52 1 or 2 or 4 (250082)
53 51 and 52 (5520)
54 14 and 53 (180)
55 43 and 53 (41)
56 47 and 53 (65)
57 54 or 55 or 56 (276)
58 57 not (36 or 46 or 49) (254)
59 exp Hypoglycemia/ (17277)
C-49
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
52 and 59 (9545)
14 and 60 (285)
43 and 60 (86)
47 and 60 (97)
61 or 62 or 63 (445)
64 not 57 (169)
65 not (36 or 46 or 49) (152)
limit 36 to english language (865)
from 67 keep 1-865 (865)
from 68 keep 1-476 (476)
from 69 keep 1-104 (104)
from 70 keep 1-215 (215)
from 71 keep 1-113 (113)
An additional search for adverse effects used the above strategy through line 71, followed
by:
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
(ae or po or to).fs. (1254721)

exp Drug Toxicity/ (15829)
medical errors/ or medication errors/ (13158)
exp Drug Interactions/ (116890)
72 or 73 or 74 or 75 (1359022)
1 or 3 (186918)
6 or 7 or 8 or 9 or 11 or 12 (379861)
77 and 78 (2545)
76 and 79 (364)
limit 81 to humans (276)
15 and 76 (871)
limit 84 to humans (668)
85 not 82 (392)
from 82 keep 1-276 (276)
from 86 keep 1-392 (392)
Database: EBM Reviews - Database of Abstracts of Reviews of Effects <3rd Quarter 2008>
control).mp. {mp=title, full text, keywords} (163)
2 ((critical$ adj6 ill$) or critical care or icu or intensive care or burn unit$ or coronary care).mp.
{mp=title, full text, keywords} (327)
3 (postoperative$ or post operative$).mp. {mp=title, full text, keywords} (705)
C-50
4 2 or 3 (973)
5 1 and 4 (6)
6 from 5 keep 1-6 (6)
Database: EBM Reviews - Cochrane Central Register of Controlled Trials <3rd Quarter
2008>
control).mp. {mp=title, original title, abstract, mesh headings, heading words, keyword} (14093)
2 ((critical$ adj6 ill$) or critical care or icu or intensive care or burn unit$ or coronary care).mp.
{mp=title, original title, abstract, mesh headings, heading words, keyword} (6526)
3 (postoperative$ or post operative$).mp. {mp=title, original title, abstract, mesh headings,
heading words, keyword} (36957)
4 2 or 3 (42208)
5 1 and 4 (541)
6 from 5 keep 1-541 (541)
SECTION B. Methods
Search strategy
We searched MEDLINE and the Cochrane Database of Systematic Reviews for literature
published from database inception through January 2010 and obtained additional articles from
consultation with experts and from reference lists of pertinent studies, reviews, and editorials.
We updated this search for the purposes of this report in October 2011. Appendix Table 1
provides the search strategies in detail. We searched clinicaltrials.gov for information about
unpublished studies. All citations were imported into an electronic database (EndNote X2,
Thomson Reuters, New York, NY).
Study selection
Three investigators reviewed the abstracts of citations identified from literature searches. Fulltext articles of potentially relevant abstracts were retrieved for further review. Eligible articles
were published in English and provided primary data on the use of IIT in hospitalized patients.
We excluded studies that evaluated fixed-dose insulin and glucose-insulin-potassium (GIK)
infusions.
To evaluate the efficacy of and hypoglycemia risk associated with IIT in hospitalized patients,
we considered randomized controlled trials that reported at least one of the following
prespecified outcomes: mortality, cardiovascular events, congestive heart failure, disability,
wound infection, sepsis, or renal failure requiring hemodialysis. We defined perioperative trials
as those in which IIT was begun pre-, intra-, or immediately post-operatively and discontinued
less than 24 hours post-operatively.
C-51
Because the safety of IIT may vary based on intervention and implementation characteristics, we
evaluated hypoglycemia rates in controlled and uncontrolled studies of IIT protocols, even if
they did not report other health outcomes (study selection details in Appendix Table 1).
To assess the risk of hypoglycemia associated with IIT, we included controlled and uncontrolled
studies that evaluated IIT protocols in hospitalized patients, even if they did not report health
outcomes. We excluded IIT studies that did not report rates of hypoglycemia [1-12]. In order to
avoid studies with potential selection bias, we excluded prospective cohort studies in which
patients were not consecutively enrolled or in which there was excessive loss to follow-up [1221]. Because tight glycemic control strategies require personnel training and institutional
acceptance, we excluded studies in which the intervention was evaluated over a short period of
time (defined as 6 months or less) as we felt these studies were less likely to provide externally
valid results [22-27].
Data extraction and quality assessment
From each study, we abstracted the following: study design, objectives, setting, demographics
(sex, age, baseline morbidity), subject eligibility and exclusion criteria, number of subjects, years
of enrollment, duration of follow-up, study and comparator interventions, method used to
monitor blood glucose, target range for blood glucose control, outcomes measured, analytic
method used, variables adjusted in the analysis, results of the study and mean blood glucose
achieved in each group, information on concomitant therapy/nutrition, occurrence of
hypoglycemia in each group, and any other adverse events.
The quality of each study was rated as good, fair, or poor based on U.S. Preventive Task Force
Service criteria [28]. When reviewers disagreed about quality rating, consensus was reached
through discussion with all authors.
Meta-analysis
We conducted meta-analyses using IIT studies identified in our original search through January
2010. Studies identified from the update search through October 2011 were described, but not
included in these meta-analyses. The primary outcome of interest was short-term mortality,
defined as mortality occurring within 28 days or during the ICU or hospital stay. If studies
reported more than one of these outcomes, we preferentially used 28-day mortality for the
analysis, followed by hospital- or ICU-mortality. We conducted a sensitivity analysis based on
short-term mortality definition. Secondary outcomes included 90- or 180-day mortality,
infection, length of stay, and hypoglycemia. For each outcome, we abstracted the number of
events and total subjects from each treatment arm and obtained a pooled estimate of relative risk
(RR) using a random effects model [29]. Statistical heterogeneity was assessed by Cochrans Q
test and I2 statistic [30]. All analyses were performed using Stata 10.0 (StataCorp, College
Station, TX, 2007).
We conducted prespecified subgroup analyses comparing ICU with non-ICU studies, and
sensitivity analyses on the following aspects: 1) the proportion of diabetic patients included,
using 25% as a cut-point based on a natural division in the included studies; 2) mean blood
C-52
glucose achieved in the intervention group, using 6.7 mmol/L (120 mg/dL) as the cut-point since
a lower threshold (6.1 mmol/L, 110 mg/dL) would have yielded only one study; and 3) study
quality.
Study yield
From our initial search through January 2010, we identified 3,055 titles and abstracts of which
461 articles selected for full-text review. We included 31 trials conducted among critically ill
patients, patients with acute MI or stroke, or perioperative patients. We also found 29 insulin
protocol studies not reporting health outcomes. Our update search through October 2011
identified an additional 331 titles and abstracts of which 40 articles were selected for full-text
review. We included 2 trials conducted in neurologic intensive care units, 1 trial in gastrectomy
patients, and 1 trial of a subcutaneous insulin regimen in general surgical ward patients. We also
found 10 insulin protocol studies not reporting health outcomes. The yield is summarized in
Figure 1.
C-53
Figure 1, Chapter 22. Management of inpatient hyperglycemia literature flow
C-54
References
1.
Furnary AP et al. Continuous intravenous

insulin infusion reduces the incidence of
deep sternal wound infection in diabetic
patients after cardiac surgical
procedures.[see comment]. Annals of
Thoracic Surgery 1999. 67(2):352-60;
discussion 360-2.
2.
Scalea TM et al. Tight glycemic control in

critically injured trauma patients. Annals of
Surgery 2007;246(4):605-10; discussion
610-2.
3.
Beilman GJ, Joseph JI. Practical

considerations for glucose control in
hospitalized patients. Diabetes Technology
& Therapeutics 2005;7(5):823-30.
4.
Button E, Keaton P. Glycemic control after

coronary bypass graft: using intravenous
insulin regulated by a computerized system.
Critical Care Nursing Clinics of North
America 2006;18(2):257-65.
5.
6.
Davis ED et al., Implementation of a new

intravenous insulin method on intermediatecare units in hospitalized patients. Diabetes
Educator 2005;31(6):818-21.
Gonzalez-Michaca L, Ahumada M, Poncede-Leon S. Insulin subcutaneous application
vs. continuous infusion for postoperative
blood glucose control in patients with noninsulin-dependent diabetes mellitus.
Archives of Medical Research
2002;33(1):48-52.
7.
Markovitz LJ et al. Description and

evaluation of a glycemic management
protocol for patients with diabetes
undergoing heart surgery [see comment].
Endocrine Practice 2002;8(1):10-8.
8.
Raucoules-Aime, M., et al., Use of i.v.

insulin in well-controlled non-insulindependent diabetics undergoing major
surgery. British Journal of Anaesthesia,
1996. 76(2): p. 198-202.
9.
Rood, E., et al., Use of a computerized

guideline for glucose regulation in the
intensive care unit improved both guideline
adherence and glucose regulation. Journal of
the American Medical Informatics
Association, 2005. 12(2): p. 172-80.
C-55
10.
Simmons, D., et al., A comparison of two

intravenous insulin regimens among surgical
patients with insulin-dependent diabetes
mellitus. Diabetes Educator, 1994. 20(5): p.
422-7.
11.
Eigsti, J. and K. Henke, Innovative

solutions: development and implementation
of a tight blood glucose management
protocol: one community hospitals
experience. DCCN - Dimensions of Critical
Care Nursing, 2006. 25(2): p. 62-5.
12.
Reed, C.C., et al., Intensive insulin protocol

improves glucose control and is associated
with a reduction in intensive care unit
mortality. Journal of the American College
of Surgeons, 2007. 204(5): p. 1048-54;
discussion 1054-5.
13.
Chaney, M.A., et al., Attempting to maintain

normoglycemia during cardiopulmonary
bypass with insulin may initiate
postoperative hypoglycemia. Anesthesia &
Analgesia, 1999. 89(5): p. 1091-5.
14.
Carvalho, G., et al., Maintenance of

normoglycemia during cardiac surgery.[see
comment]. Anesthesia & Analgesia, 2004.
99(2): p. 319-24.
15.
Chee, F., et al., Expert PID control system

for blood glucose control in critically ill
patients.[erratum appears in IEEE Trans Inf
Technol Biomed. 2004 Jun;8(2):228]. IEEE
Transactions on Information Technology in
Biomedicine, 2003. 7(4): p. 419-25.
16.
Wong, X.W., et al., Model predictive

glycaemic regulation in critical illness using
insulin and nutrition input: a pilot study.
Medical Engineering & Physics, 2006.
28(7): p. 665-81.
17.
Hemmerling, T.M., et al., Comparison of a

continuous glucose-insulin-potassium
infusion versus intermittent bolus
application of insulin on perioperative
glucose control and hormone status in
insulin-treated type 2 diabetics. Journal of
Clinical Anesthesia, 2001. 13(4): p. 293300.
18.
Quinn, J.A., et al., A practical approach to

hyperglycemia management in the intensive
care unit: evaluation of an intensive insulin
infusion protocol. Pharmacotherapy, 2006.
26(10): p. 1410-20.
19.
Shaw, G.M., et al., Rethinking glycaemic

control in critical illness--from concept to
clinical practice change. Critical Care &
Resuscitation, 2006. 8(2): p. 90-9.
20.
Rowen, M., et al., On rendering continuous

glucose monitoring ready for prime time in
the cardiac care unit. Coronary Artery
Disease, 2007. 18(5): p. 405-9.
21.
22.
23.
24.
Osburne, R.C., et al., Improving

hyperglycemia management in the intensive
care unit: preliminary report of a nursedriven quality improvement project using a
redesigned insulin infusion algorithm.
Diabetes Educator, 2006. 32(3): p. 394-403.
Elinav, H., et al., In-hospital treatment of
hyperglycemia: effects of intensified
subcutaneous insulin treatment. Current
Medical Research & Opinion, 2007. 23(4):
p. 757-65.
Ku, S.Y., et al., New insulin infusion
protocol Improves blood glucose control in
hospitalized patients without increasing
hypoglycemia. Joint Commission Journal on
Quality & Patient Safety, 2005. 31(3): p.
141-7.
Lien, L.F., et al., Optimizing hospital use of
intravenous insulin therapy: improved
management of hyperglycemia and error
reduction with a new nomogram. Endocrine
Practice, 2005. 11(4): p. 240-53.
25.
Vogelzang, M., F. Zijlstra, and M.W.N.

Nijsten, Design and implementation of
GRIP: a computerized glucose control
system at a surgical intensive care unit.
BMC Medical Informatics & Decision
Making, 2005. 5: p. 38.
26.
Kanji, S., et al., Standardization of

intravenous insulin therapy improves the
efficiency and safety of blood glucose
control in critically ill adults. Intensive Care
Medicine, 2004. 30(5): p. 804-10.
27.
Bland, D.K., et al., Intensive versus

modified conventional control of blood
glucose level in medical intensive care
patients: a pilot study. American Journal of
Critical Care, 2005. 14(5): p. 370-6.
28.
Harris, R.P., et al., Current methods of the

US Preventive Services Task Force. A
review of the process. American Journal of
Preventive Medicine, 2001. 30(3S): p. 2135.
29.
DerSimonian, R. and N. Laird, Metaanalysis in clinical trials. Controlled Clinical

Trials, 1986. 7(3): p. 177-88.
30.
Higgins, J.P.T., et al., Measuring

inconsistency in meta-analyses. BMJ, 2003.
327(7414): p. 557-60.
Chapter 23. Interventions To Prevent Contrast-Induced Acute

Kidney Injury
SEARCH METHODOLOGY
We conducted a structured search of PubMed using a search strategy developed by a medical
librarian. The search strategy was last updated on December 6, 2011 and was as follows:
C-56
#1
#2
#3
#4
Search string
((contrast medium OR contrast media OR contrast dye OR radiographic
contrast OR radiocontrast media OR radiocontrast medium OR contrast
agent*) AND (kidney diseases/ci OR kidney/ae OR kidney/de OR nephritis OR
nephropath* OR nephrotox* OR renal insufficiencies[mh] OR renal insufficienc*
OR diabetic nephropathies[mh] OR creatinine OR kidney injury OR kidney
dysfunction OR renal dysfunction)) OR (contrast induced OR contrast
associated AND (renal OR kidney OR nephropath* OR nephrotox*)) OR
(contrast media/ae AND (kidney diseases[majr] OR kidney[majr] OR diabetic
nephropathy[majr]))
#1 limited to Randomized Controlled Trials, English language, publication date
since 1/1/2001
#1 limited to Systematic Reviews, English language, publication date since
1/1/2001
#3 limited to publication date since 1/1/2007
Results
5217
193
53
32
SECTION B. Methods
Based on the large number of systematic reviews identified by the above search, we opted to
perform a systematic meta-review of the existing systematic reviews. We included only
systematic reviews published since January 1, 2007. Two authors independently reviewed the 32
reviews identified through the PubMed search to identify systematic reviews and meta-analyses.
All of the systematic reviews identified (N=20) were assessed for methodologic quality by two
reviewers who independently completed the AMSTAR checklist. Disagreements in this process
were resolved by consensus. The included systematic reviews were grouped according to the
specific CI-AKI preventive intervention studied, and were summarized narratively.
Chapter 24. Rapid Response Systems (NEW)

PubMed: 2000-8/2/2011
LANGUAGE:
English
SEARCH STRATEGY:
Hospital Rapid Response Team[Mesh] OR rapid response team OR rapid response teams
OR rapid response system OR rapid response systems OR medical emergency team OR
medical emergency teams OR critical care outreach team OR critical care outreach teams
OR patient at-risk team OR patient at-risk teams OR patient at risk team OR patient at
risk teams OR emergency medical team OR emergency medical teams
AND
effectiv* OR implement* OR success* OR fail* OR utiliz* OR adopt*
OR
patient care AND team* AND (emergency OR emergencies OR rapid OR critical care)
C-57
AND
====================================================================
CINAHL: 2000-8/16/2011
SEARCH STRATEGY:
rapid response team OR rapid response teams OR rapid response system OR rapid
response systems OR medical emergency team OR medical emergency teams OR critical
care outreach team OR critical care outreach teams OR patient at-risk team OR patient atrisk teams OR patient at risk team OR patient at risk teams OR emergency medical team
OR emergency medical teams
AND
OR
patient care AND team* AND (emergency OR emergencies OR rapid OR critical care)
AND
Search modes - Phrase Searching (Boolean)
====================================================================
EMBASE: 2000-11/4/2011
SEARCH STRATEGY:
rapid response team/exp OR rapid response team OR rapid response teams/exp OR rapid
response teams OR rapid response system/exp OR rapid response system OR rapid response
systems/exp OR rapid response systems OR medical emergency team/exp OR medical
emergency team OR medical emergency teams/exp OR medical emergency teams OR
critical care outreach team OR critical care outreach teams OR patient at-risk team OR
patient at-risk teams OR patient at risk team OR patient at risk teams OR emergency
medical team OR emergency medical teams OR ((patient care NEAR/3 team*) AND
(emergency OR emergencies OR rapid OR critical care))
AND
C-58
====================================================================
Cochrane: 2000-11/4/2011
SEARCH STRATEGY:
rapid response team OR rapid response teams OR rapid response system OR rapid
response systems OR medical emergency team OR medical emergency teams OR critical
care outreach team OR critical care outreach teams OR patient at-risk team OR patient atrisk teams OR patient at risk team OR patient at risk teams OR emergency medical team
OR emergency medical teams:ti,ab,kw or patient care AND team* AND (emergency OR
emergencies OR rapid OR critical care) :ti,ab,kw
NUMBER OF RESULTS: 72 (Syst Revs 4, Other Revs 7, Clin Trials 55, Econ 6)
SECTION B. Methods
PICOTS
Elements
Population
Intervention
Comparator
Outcomes
Timing
Settings
Patients on general hospital wards - Adult and pediatric

Rapid Response Systems
Effectiveness: Usual practice
Implementation:
Technology/tools: criteria for activating team (extended vs restricted criteria),
investment in human resources (team availability)
Staff selection/ training: Physician on team (MET model) vs. Nurse led (RRT
model); investment in team; education/training of team and floor staff
Identifying/addressing barriers/facilitators: Reluctance to call team, nursing
workload, availability of team to respond
Mortality (total or preventable)
Incidence of cardio-respiratory arrest
Unanticipated intensive care unit admission

Before and after intervention
Hospitals
Studies from all countries and languages were included
Effectiveness: Included all studies with a comparison group and at least some component of an
RRS. Critical Care Outreach Team studies were included if they also included a pre-intensive
care unit RRS component (general response to all ward patients). Effectiveness studies were only
included after November 2008, the end date for a high-quality systematic review and metaanalysis.
Implementation: Included qualitative and quantitative studies addressing implementation.
Studies were defined as qualitative research studies if they used a formal qualitative
methodology such as interviews, focus groups, or ethnography
C-59
Studies were defined as quantitative implementation studies if they evaluated the impact of a
change or difference in implementation strategy on utilization of the RRS and/or patient
outcomes.
Chapter 25. Medication Reconciliation Supported by Clinical

Pharmacists (NEW)
A search strategy comprising multiple terms was developed by a library scientist with extensive
experience in conducting systematic reviews in collaboration with a physician health services
researcher also very experienced in literature searching and with content expertise in patient
safety. As noted in the detailed description of the search, databases covered included MEDLINE,
EMBASE, and the Cochrane library.
Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations and Ovid MEDLINE(R)
<1946 to July 16, 2012>
1 ((reconcil* adj3 medicat*) or (med* reconcil* or medrec)).mp.
2 patient admission/ or patient discharge/ or patient readmission/ or patient transfer/ or
Continuity of Patient Care/ or transition.ti.
3 Medication Errors/ or ((medication or discrepanc* or discontinuit* or reconciliation).ti. or
(medication adj8 discrepanc*).ti,ab.)
4 (2 and 3) or 1
5 limit 4 to english
Embase <1980 to 2011 Week 18>
1 ((reconcil* adj3 medicat*) or (med* reconcil* or medrec)).mp.
2 Medication Errors/ or ((medication or discrepanc* or discontinuit* or reconciliation).ti. or
(medication adj8 discrepanc*).ti,ab.)
3 hospital admission/ or hospital discharge/ or hospital readmission/ or patient transfer*.mp. or
(continuity adj3 care).mp. or transition.ti.
4 1 or (3 and 2)
5 limit 4 to english
Cochrane CENTRAL (Central Register of ControlLed Trials) <June 2009>
#1 (medication reconciliation):ti,ab,kw
#2 medication reconciliation:ti,ab,kw
#3 (reconcil* near/3 medicat*):ti,ab,kw
#4 (medrec):ti,ab,kw
#5 (#1 OR #2 OR #3 OR #4)
SECTION B. Methods
Titles and abstracts were independently reviewed by 2 of 3 individuals, including a library
scientist who has conducted numerous systematic reviews, a masters level research assistant
with content experience in patient safety, a physician trainee with health services research
C-60
experience. Disagreements were resolved by discussion between reviewers as well as and

consultation with a physician health services researcher with expertise in patient safety and
extensive experience with systematic reviews.
C-61
Chapter 26. Identifying Patients at Risk for Suicide: Brief

Review (NEW)
To conduct the review, we searched PubMed in October 2011 using major heading search terms
Suicide, and Hospital or Inpatient or Safety Management, for English language articles published
starting in the year 2000. We expanded the search using the PubMed related citations feature,
and Google Scholar to search for citing articles of those retained for review; we identified
additional relevant articles by reference mining. We also searched PSNet. Clinical trials, large
observational studies, reviews, and reports on implementations were given priority. Systematic
reviews were scored for methodologic quality using the 11-point AMSTAR scale; items rated
Not Applicable were not counted towards either the score or the total.
SECTION B. Methods
Titles, abstracts and articles were reviewed by a psychiatrist health services researcher with
extensive experience in systematic reviews, including a prior review of suicide prevention
programs. The synthesis was narrative.
Chapter 27. Strategies To Prevent Stress-Related

Gastrointestinal Bleeding (Stress Ulcer Prophylaxis): Brief
Update Review
We searched PubMed for relevant articles using the search terms stress ulcer and stress ulcer
prophylaxis, limited to systematic reviews published in the past 5 years. This search identified
19 articles
SECTION B. Methods
Articles identified using the above strategies were reviewed by two practicing hospitalists, one of
whom has prior expertise in conducting and analyzing systematic reviews. The systematic
reviews identified through this search form the basis of this review. These systematic reviews
were summarized narratively, and their reference lists were reviewed by hand to identify other
key articles on costs and implementation.
Chapter 28. Prevention of Venous Thromboembolism: Brief

Update Review
already known to the authors as part of their quality improvement work where previous
literature.
C-62
SECTION B. Methods
These reviews and studies were reviewed by a surgeon health services researcher with clinical
and quality improvement experience with venous thromboembolism. The synthesis was
narrative.
Chapter 29. Preventing Patient Death or Serious Injury

Associated With Radiation Exposure from Fluoroscopy and
Computed Tomography: Brief Review (NEW)
Name
PSNet
Date limits
PubMed
Platform/provider
ECRI Institute
and Quality

Cochrane Library.
Medical Subject Headings (MeSH)
Conventions:
PubMed
[mh] =
MeSH heading
[majr] =
C-63
[pt]
[sb]
[sh]
[tiab]
=
=
=
=
publication type

Concept
Adverse effects of radiation
therapy
Programs
Study design
radiation injuries/prevention and
control[majr]
radiation-protective agents[majr]
radiation monitoring[majr]
radiation dosage[majr]
dose-response relationship,
radiation[majr]
radiation protection[majr]
fluoroscopy/adverse effects[majr]
radiography, interventional/adverse
effects[majr]
radiotherapy/adverse effects[majr]
radiation[majr]
death[majr]
mortality[majr]
wounds and injuries[majr]
burns[mesh]
skin transplantation[mesh]
outcome and process assessment
(health care)[majr]
safety management[majr]
risk assessment[majr]
secondary prevention[mesh]
program development[mesh]
program evaluation[mesh]
health plan implementation[mesh]
randomized controlled trial[pt]
controlled clinical trial[pt]
randomized controlled trials[mh]
random allocation[mh]
double-blind method[mh]
single-blind method[mh]
clinical trial[pt]
clinical trials[mh]
research design[mh:noexp]
comparative study[pt]
evaluation studies[pt]
evaluation studies as topic[mh]
follow-up studies[mh]
prospective studies[mh]
cross-over studies[mh]
meta-analysis[mh]
meta-analysis[pt]
outcomes research[mh]
multicenter study[pt]
C-64
Keywords
radiation
fluoroscop*
injuri*
injury*
death
mortality
injur*
harm
burn*
skin graft
skin transplant
grade 3
grade 4
prevent*
reduc*
initiative*
program*
implement*
clinical trial
clinical trials
comparative study
comparative studies
evaluation study
evaluation studies
Pubmed
English language, human
Set
Number
Concept
Search statement

therapy, controlled
vocabulary only
radiation injuries/prevention and control[majr] OR radiation-protective

agents[majr] OR radiation monitoring[majr] OR radiation
dosage[majr] OR dose-response relationship, radiation[majr] OR
radiation protection[majr] OR fluoroscopy/adverse effects[majr] OR
radiography, interventional/adverse effects[majr] OR
radiotherapy/adverse effects[majr]

therapy, controlled
vocabulary and title/abstract
(radiation[majr] OR radiation[ti] OR fluoroscop*[ti]) AND (injuri*[ti] OR

injury*[ti] OR death[majr] OR death[ti] OR mortality[majr] OR mortality[ti]
OR wounds and injuries[majr] OR injur*[ti] OR harm[ti] OR
burns[mesh] OR burn*[ti] OR skin transplantation[mesh] OR skin
graft[tiab] OR skin transplant[tiab] OR grade 3[tiab] OR grade 4[tiab])
Programs, preventive or
assessment
outcome and process assessment (health care)[majr] OR safety

management[majr] OR risk assessment[majr] OR prevent*[ti] OR
reduc*[ti] OR secondary prevention[mesh] OR prevention[tiab] OR
initiative*[tiab] OR program development[mesh] OR program
evaluation[mesh] OR program*[tiab] OR health plan
implementation[mesh] OR implement*[tiab]
Study design
randomized controlled trial[pt] OR controlled clinical trial[pt] OR

randomized controlled trials[mh] OR random allocation[mh] OR doubleblind method[mh] OR single-blind method[mh] OR clinical trial[pt] OR
clinical trials[mh] OR research design[mh:noexp] OR comparative
study[pt] OR evaluation studies[pt] OR evaluation studies as topic[MH]
OR follow-up studies[mh] OR prospective studies[mh] OR cross-over
studies[mh] OR meta-analysis[mh] OR meta-analysis[pt] OR outcomes
research[mh] OR multicenter study[pt] OR clinical trial[tw] OR clinical
trials[tw] OR comparative study[tw] OR comparative studies[tw] OR
evaluation study[tw] OR evaluation studies[tw]
Combine
(S1 OR S2) AND S3 AND S4
Apply date limit
2000-2011
Total Downloaded
Total Retrieved
Total Included
80
56
10
SECTION B. Methods
in both systematic reviews and radiation therapy. Our research was limited to studies
implementing practices (e.g., protocols, technical measures) to reduce radiation exposure to
patients from fluoroscopy and computed tomography-guided diagnostic and interventional
procedures. The focus was on studies published from 2005 to the present that compared
outcomes (e.g., radiation dose, imaging time) following implementation of these practices
compared to a control period when the technologies were not in place. Potential barriers to
implementation, technical difficulty of practices and reported harms from patient safety practices
were also assessed. Included studies were narratively summarized by the author.
C-65
Chapter 30. Ensuring Documentation of Patients

Preferences for Life-Sustaining Treatment: Brief Update
Review
For this topic, since we had just completed a 2011 Agency for Healthcare Research and Quality
systematic review on the topic of interventions to improve end-of-life care, we used syntheses
and articles identified in this search, and did not conduct any additional literature searches.
SECTION B. Methods
Relevant reviews and studies were reviewed by a palliative care physician health services
researcher with clinical and quality improvement experience with end-of-life care. The synthesis
was narrative.
Chapter 31. Human Factors and Ergonomics

SECTION A and B. Literature Search and Methods
For this topic we determined that a systematic review of Human Factors and Ergonomics
would be too diffuse to be useful to readers. Therefore this topic uses exemplars to illustrate
different ways that human factors and ergonomics can be useful in patient safety.
Chapter 32. Promoting Engagement by Patients and Families

To Reduce Adverse Events
SEARCH METHODOLOGIES
SEARCH #1:
PubMed 2006-9/15/2011
LANGUAGE:
English
SEARCH STRATEGY:
PhysicianPatient Relations OR NursePatient Relations OR Patient Participation OR
Patient Education as Topic OR Social Responsibility OR Patient-Centered Care OR
informed consent OR chronic disease
AND
Errors/prevention and control or Medical Errors/adverse effects or Safety Management or
Cross Infection/prevention and control
C-66

=====================================================================
SEARCH #2:
PubMed 2006-9/15/2011
SEARCH STRATEGY:
Patient participation OR patient role OR patients role OR patient complain* OR patients
complain*
AND
Cross Infection/prevention and control OR safe* OR medical error OR medical errors OR
mistake* OR medication error OR medication errors
AND
patient participation[All Fields]
=====================================================================
SEARCH #3:
PubMed 2006-9/15/2011
SEARCH STRATEGY:
PhysicianPatient Relations OR NursePatient Relations OR Patient Participation[mh]
OR Patient Education as Topic OR informed consent OR patient reporting[tiab] OR
patient reports[tiab] OR patients reporting[tiab] OR patients reports[tiab] OR
complain*[ti] OR patient participa* OR patients participa* OR patient education[tiab] OR
education of patients[tiab]) OR patient role OR patients role OR patients role* OR
health literacy
AND
mistake* OR medication error OR medication errors OR adverse OR dangerous
AND
systematic[sb]
=====================================================================
C-67
SEARCH #4:
PubMed 2006-9/15/2011
SEARCH STRATEGY:
patient participation[mh] OR patient participation[tiab] OR patient role OR patients role
OR patient complain* OR patients complain* OR patient reporting[tiab] OR patient
reports[tiab] OR patients reporting[tiab] OR patients reports[tiab]
AND
mistake* OR medication error OR medication errors
=====================================================================
SEARCH #5:
Cochrane Database of Systematic Reviews 2006-9/15/2011
SEARCH STRATEGY:
Patient Education OR informed consent OR patient reporting OR patient reports OR
patients reporting OR patients reports OR complain* OR patient participa* OR patients
participa* OR patient education OR education of patients OR patient role OR patients
role OR patients role* OR health literacy in Title, Abstract or Keywords
AND
mistake* OR medication error OR medication errors OR adverse OR dangerous in Title,
Abstract or Keywords
=====================================================================
SEARCH #6:
PubMed 2006-9/19/2011
SEARCH STRATEGY:
C-68

Patient Education as Topic OR Social Responsibility OR Patient-Centered Care OR
informed consent OR chronic disease
AND
Cross Infection/prevention and control
=====================================================================
SEARCH #7:
Medline on OVID 2000-9/19/2011
SEARCH STRATEGY:
Physician Patient Relations/ OR Nurse Patient Relations/ OR Patient Participation/ OR Patient
Education as Topic/ or Social Responsibility.mp. OR Patient Centered Care/ OR informed
consent.mp. OR chronic disease$.mp.
AND
(adverse event$ OR iatrogenic disease OR medical adj error$ OR safety adj2 manag$ OR cross
adj2 Infection$ adj2 prevent$ OR adverse effect$).mp. [mp=protocol supplementary concept,
rare disease supplementary concept, title, original title, abstract, name of substance word, subject
heading word, unique identifier]
=====================================================================
SEARCH #8:
Medline on OVID 2000-9/19/2011
SEARCH STRATEGY:
(Patient adj2 participa$ or patient$ adj5 role or patient adj5 complain4 or patient$ adj3 involv$
or patient$ adj3 engag$ or patient$ adj3 report$).mp.
AND
(adverse adj2 event$).mp. or iatrogenic Disease/pc or Medical Errors/pc or Medical Errors/ae or
Cross Infection/pc or safe$.mp. or unsaf$.mp. or medical.mp. adj3 error$.mp. or mistake$.mp. or
medication.mp. adj3 error$.mp.
=====================================================================
C-69
SEARCH #9:
PubMed 2000-10/5/2011
SEARCH STRATEGY:
Errors/prevention and control OR medication errors/prevention and control OR Medical
Errors/adverse effects OR Safety Management OR Cross Infection/prevention and control
OR infection control
AND
Physician-Patient Relations OR Nurse-Patient Relations OR Patient Participation OR
Patient Education as Topic OR Patient-Centered Care OR patient reporting OR patientempowering OR patient empowerment OR patient partnership OR patient activation or
patient self-effectiveness or patient involvement
=====================================================================
SEARCH #10:
CINAHL 2000-10/24/2011
SEARCH STRATEGY:
(TI (Physician n3 Patient n3 relation*) OR (Nurse n3 Patient n3 relation*) OR Patient n3
Participat* OR Patient n3 Education OR Patient-Centered Care OR patient reporting OR
patient n2 empower* OR patient n3 partner* OR patient activation OR patient selfeffectiveness OR patient n3 involv*
OR AB (Physician n3 Patient n3 relation*) OR (Nurse n3 Patient n3 relation*) OR Patient n3
patient n2 empower* OR patient n3 partner* OR patient activation OR patient selfeffectiveness OR patient n3 involv*
OR MW (Physician n3 Patient n3 relation*) OR (Nurse n3 Patient n3 relation*) OR Patient n3
patient n2 empower* OR patient n3 partner* OR patient activation OR patient selfeffectiveness OR patient n3 involv*)
AND
adverse events OR iatrogenic disease OR medical errors OR medication errors OR
medication error OR medical error OR (cross infection AND prevent*) OR safety
management OR infection control
=====================================================================
SEARCH #11:
C-70

Embase 2000-10/27/2011
SEARCH STRATEGY:
physician-patient relations OR nurse-patient relations OR patient participation OR patient
education OR patient-centered care OR patient reporting OR patient-empowering OR
patient empowerment OR patient partnership OR patient activation OR patient selfeffectiveness OR patient involvement OR doctor patient relation
AND
adverse events OR iatrogenic disease OR medical errors OR medication errors OR
medication error OR medical error OR cross infection AND prevent* OR safety
management OR infection control
=====================================================================
SEARCH #12a:
SEARCH STRATEGY:
physician-patient relations OR nurse-patient relations OR patient participation OR patient
education OR patient-centered care OR patient reporting OR patient-empowering OR
patient empowerment OR patient partnership OR patient activation OR patient selfeffectiveness OR patient involvement OR doctor patient relation OR physician-patient
relation in Title, Abstract or Keywords
AND
adverse OR iatrogenic OR error* OR harm* OR safe* OR (infection* AND prevent*) OR
(infection* AND control*) in Title, Abstract or Keywords
NUMBER OF RESULTS: 7 (Cochrane Reviews [1] Other Reviews [0] | Clinical Trials [5] |
Methods Studies [1] | Technology Assessments [0] | Economic Evaluations [0] Cochrane
Groups [0])
=====================================================================
SEARCH #12b:
SEARCH STRATEGY:
MeSH descriptor Physician-Patient Relations, this term only OR MeSH descriptor Patient
Education as Topic, this term only OR MeSH descriptor Patient Participation explode all trees
AND
C-71
MeSH descriptor Safety Management, this term only OR drug toxicity OR MeSH descriptor
Medical Errors explode all trees OR MeSH descriptor Infection explode all trees with qualifier:
PC (Prevention & Control)
NUMBER OF RESULTS: 81 (Cochrane Reviews [5] | Other Reviews [4] | Clinical Trials
[71] | Methods Studies [0] | Technology Assessments [0] | Economic Evaluations [1] |
Cochrane Groups [0])
SECTION B. Methods
PICOTS
Elements
Population
Intervention
Comparator
Outcomes
Timing
Settings
Patients in inpatient healthcare settings (adult and pediatric) and their family members
Any intervention to encourage patient involvement in safety, including reporting
adverse outcomes or errors
Usual practice
Effectiveness of the intervention
Hospitals
- Only English-language studies from the US, UK, Canada, and Australia were included in the
present review, due to potentially significantly different cultural issues in patient engagement in
their health care outside of these countries, as well as potential differences in tools for promoting
engagement.
-Included studies were required to focus on hospital care settings (e.g., intensive care units)
patient engagement in safety in the home setting would be difficult to differentiate from patient
self-management of their medications and care, when providers are not present.
-Only systematic reviews focusing on effectiveness and prospective, controlled studies were
included.
Chapter 33. Promoting a Culture of Safety

Search Methodology
PubMed:
Limit from 2000
Final Search Strategy:
patient safety culture OR safety culture survey OR safety attitude questionnaire
OR safety attitudes questionnaire OR safety attitude OR patient safety practice
OR (Hospital Survey AND patient safety culture) OR Manchester Patient Safety
Framework OR (Patient Safety Culture AND survey) OR patient safety climate
OR ((safety culture OR safety practice OR safety climate OR high reliability)
AND
(rehabilitation OR snf OR nursing home OR skilled nursing facility OR hospital OR
hospitals OR ICU OR intensive care unit OR emergency room OR attitude OR attitudes OR
assisted living OR long term care OR resident OR residents OR health center OR
C-72
healthcare OR health care OR patients OR patient OR intervention OR improvement OR scale

OR primary care)) OR hospital patient climate safety scale
OR culture of safety OR culture of trust OR (culture[ti] reliability[ti])
Total Retrieved: 1637
------------------------------------------------------------------------------------------------------------------Patient Safety Practices/Culture PTN: HQ208-1000
search by J Larkin 9/19/2011
CINAHL:
Limit from 2000
Search Strategy:
patient safety culture OR safety attitude questionnaire OR patient safety practice OR
hospital survey on patient safety OR manchester patient safety framework OR hospital
patient climate safety scale OR culture of safety OR culture of reliability OR culture of
trust
OR
(safety culture OR safety practice OR safety climate OR high reliability
AND
skilled nursing facility OR hospital OR hospitals OR ICU OR intensive care unit OR
emergency room OR attitude OR attitudes OR assisted living OR long term care OR
resident OR residents OR health center OR healthcare OR health care OR patients OR
patient OR intervention OR improvement OR scale OR primary care)
802 results (NOT deduped with PubMed or any other database)
Results sent in .txt file(this is the generic bibliographic software file option) (unable to save to an
.ris file from Ebsco)
PSC_Cinahl.txt
-----------------------------------------------------------------------------------------------------------------Patient Safety Practices/Culture PTN: HQ208-1000
Cochrane:
Limit from 2000
Search Strategy:
trust
OR
AND
C-73

PSC_cochrane.ATXT
-------------------------------------------------------------------------------------------------------------------Patient Safety Practices/Culture PTN: HQ208-1000
embase Search:
Limit from 2000
no mapping or exploding of terms and unchecked medline
Search Strategy:
trust
OR
AND
In this instance, the combination of terms and this database in general tend to produce higher
numbers of results. I do believe there will be a decent amount of overlap with the other databases
though.
PSC_embase.ris
------------------------------------------------------------------------------------------------------------------Patient Safety Practices/Culture PTN: HQ208-1000
PsycInfo Search:
Limit from 2000
Search Strategy:
trust
OR
AND
C-74

Results sent in .txt file(this is the generic bibliographic software file option) (unable to save to an
.ris file from Ebsco)
PSC_Psycinfo.txt
SECTION B. Methods
PICOTS
Elements
Population
Intervention
Comparator
Outcomes
Timing
Settings
Patients in inpatient healthcare settings

Adult and pediatric
Promoting a culture of safety
Usual practice
Overall
Change in patient safety culture/climate
Clinical indicators of safety/harm where available
Hospitals
-Only English-language studies from the US, UK, Canada, and Australia were included in the
present review. While there are a growing number of studies that have translated Englishlanguage surveys of culture into other languages, there is still limited evidence that construct
validity of such measures is comparable across samples.
-Included studies were required to focus on in-patient units within hospital care settings (e.g.,
intensive care units). Studies in the operating room, radiology, and other non-inpatient units were
not included in this review.
- Included studies had to report on measures of culture over at least two points in time.
- Included studies were also required to use a psychometrically valid measure of safety culture
that is present in the peer-reviewed literature. As a guideline, studies that utilized one of the
following measures of patient safety culture were included: hospital survey on patient safety
culture (HSOPSC) / HSOPS, Hospital Survey on Patient Safety Patient Safety Climate/ patient
safety climate in health care organizations (PSCHO), Safety Climate Survey (SCS) Safety
Attitudes Questionnaire (SAQ) Hospital safety climate scale (HSC) Operating Room
Management Attitudes Questionnaire (ORMAQ) Hospital Survey on Patient Safety
Stanford/PSCI Culture Survey Safety Climate Scale MSSA, Medication Safety Self Assessment
HTSSCS, Hospital Transfusion Service Safety Culture Survey Manchester Patient Safety
Framework.
- Interventions had to target practicing health care professionals or para-professionals. Studies
examining PSPs aimed at medical or nursing students, or otherwise including only education
outside of a clinical setting, were not included.
- The stated purpose of the PSP described had to specifically include aims to improve culture.
C-75
-Studies specifically targeting patient safety culture were included. Studies of other types of
culture (e.g., general organizational culture) were not included.
-Only prospective studies were included.
Chapter 34. Effect of Nurse-to-Patient Staffing Ratios on

Patient Morbidity and Mortality
Web of Science 2007-9/12/2012
SEARCH STRATEGY:
Forward searches on the following 4 source publications:
Kane RL, Shamliyan T, Mueller C, Duval S, Wilt TJ.
Nursing Staffing and Quality of Patient Care, Evid Rep Technol Assess. 2007 Mar;(151):1115.
Robert L. Kane, MD,* Tatyana A. Shamliyan, Christine Mueller, Sue Duval, and Timothy J.
Wilt
The Association of Registered Nurse Staffing Levels and Patient Outcomes: Systematic Review
and Meta-Analysis, Medical Care 2007 Dec;45(12):1195-1204.
Aiken LH, Clarke SP, Cheung RB, Sloane DM, Silber JH.
Educational levels of hospital nurses and surgical patient mortality.
JAMA. 2003 Sep 24;290(12):1617-23.
Needleman J, Buerhaus P, Mattke S, Stewart M, Zelevinsky K.
Nurse-staffing levels and the quality of care in hospitals.
N Engl J Med. 2002 May 30;346(22):1715-22.
TOTAL NUMBER AFTER REMOVAL OF DUPLICATES: 546
SECTION B. Methods
C-76
Figure 1, Chapter 34. Literature flow diagram
*1-4The figure shows the flow of articles through the review process. With one exception we did not include any additional crosssectional studies of association. The one exception is detailed in the text.
References
1.
Stone PW, Pogorzelska M, Kunches L, et al.

Hospital staffing and health care-associated
infections: A systematic review of the
literature. Clinical Infectious Diseases.
2008;47(7):937-44.WOS:000259038400015
3.
Butler M, Collins R, Drennan J, et al.

Hospital nurse staffing models and patient
and staff-related outcomes. Cochrane
Database of Systematic Reviews.
2011;(7).WOS:000292554300006
2.
Cummings GG, MacGregor T, Davey M, et

al. Leadership styles and outcome patterns
for the nursing workforce and work
environment: A systematic review.
International Journal of Nursing Studies.
2010;47(3):363-85.WOS:000275611400012
4.
Flynn M, McKeown M. Nurse staffing

levels revisited: a consideration of key
issues in nurse staffing levels and skill mix
research. Journal of Nursing Management.
2009;17(6):75966.WOS:000283227200011.
C-77
Chapter 35. Patient Safety Practices Targeted at Diagnostic

Errors (NEW)
We designed a four-pronged literature search strategy to identify a broad range of interventional
studies with implications for errors in clinical diagnosis. In the first mechanism of our overall
strategy, we utilized the Agency for Healthcare Research and Qualitys Patient Safety Network
(PSNet, see psnet.ahrq.gov). The PSNet website contains regularly-updated resources related to
patient safety, and therefore, the PSNet online database was searched for articles classified under
the safety target diagnostic error. We used articles from this search in the final review, and also
to test search terms for the PubMed MEDLINE database search. In the second search
mechanism, we hand-screened two previous systematic reviews1,2 related to diagnostic errors,
and adapted their search strategies for our review. In the third mechanism, a structured search
was built to identify published literature indexed to the PubMed MEDLINE database (see Table
1). In the fourth mechanism, we reviewed reference lists of articles flagged in the earlier search
phases for the purpose of identifying further eligible studies.
During the first review phase, every article identified through the first three mechanisms (n =
1,389) was screened by two independent reviewers who recommended the study for inclusion or
exclusion based upon title and abstract (if available). To make these recommendations, reviewers
excluded those studies that: (1) did not contain an intervention component, (2) contained an
intervention component unrelated to diagnostic errors, or (3) did not report patient-related
outcomes. Discrepancies between reviewers recommendations in this phase were evaluated by
the entire team until a consensus was achieved. Articles that met our inclusion criteria proceeded
to the second review phase, a full text review (n = 269). During the full text review, our final
inclusion criteria required that studies reported: (1) results from an intervention related to
diagnostic errors, and (2) relevant patient outcomes, or proxy measures of patient outcomes,
indicating that (3) the study was conducted with real patients. Again, two independent reviewers
evaluated full-text articles and recommended them for inclusion in the final report. The fourth
search mechanismthe references reviewwas conducted by one researcher who screened
2,332 article titles. For all relevant articles, the abstracts were reviewed (n = 115) using the same
inclusion criteria as above. Once an abstract was considered relevant, it entered data abstraction
review by two separate reviewers. Discrepancies between reviewer recommendations in the
second full-text phase were again addressed by the team consensus method. Studies that met the
all aforementioned requirements in the second phase were included within the chapter (n = 91).
Data were then abstracted from the final set of articles by two independent abstractors, and
discrepancies in the data abstraction phase were evaluated by the team consensus method.
Figure 2 summarizes details number of studies identified initially by the four search mechanisms,
and the number of included and excluded studies from each review phase.
Search Limitation: Due to the variety of potential topics related to diagnostic errors, the search
strategy was built to identify a broad base of literature addressing potential contributors to errors
across clinical domains and care settings. However, the current strategy did not include
additional searches to directly target interventional types with lower yields as might be expected
C-78
if focusing on a specific clinical specialty or care setting. For example, the major subheading
search for Delayed Diagnosis in MEDLINE would likely capture studies with a primary focus
related to the current review, but may not capture a study primarily focusing on particular
clinical processes such as x-ray review for injuries. The extensive number of possible search
terminology combinations using specific clinical domains (e.g., radiology), care settings (e.g.,
critical care), and intervention types (e.g., double review) where diagnostic errors may occur
was beyond the scope of the current review. However, with our extensive reference review we
expect that our included studies represent at a minimum a reasonable probe of the literature for
certain clinical specialties, care settings, and intervention types likely to be of importance to
reducing diagnostic errors (e.g., evaluations of an additional reviewer of radiology reports added
to the diagnostic pathway; laboratory-focused interventions).
Table 1, Chapter 35. MEDLINE search strategy
The final search was performed on October 10, 2011. The search was conducted using PubMed MEDLINE
and was limited to English language publications.
The following two search strategies were used in conjunction with one another to identify articles with
diagnostic error reduction interventions published between 1980 and 2011.
Search Strategy A: (((Diagnostic Errors[Majr]) OR Delayed Diagnosis[Majr])) AND (interven*[tiab] OR
intervention studies[mh] OR model*[ti] OR strateg* OR improv*[ti] OR implementation*[tiab] OR practices
OR random* OR controlled clinical trial[pt] OR program[tiab] OR programs[tiab] OR programme[tiab] OR
programmes[tiab] OR (program evaluation[mh] AND treatment outcome[mh]) OR systematic review*)
Search Strategy B: (((Diagnostic Errors[Majr]) OR Delayed Diagnosis[Majr])) AND (Affect[Mesh] OR
Clinical Competence[Mesh] OR Communication[Mesh] OR Continuity of Patient Care[Mesh] OR
Decision Making[Mesh] OR Decision Making, Organizational[Mesh] OR Decision Support Systems,
Clinical[Mesh] OR Decision Support Techniques[Mesh] OR Human Engineering[Mesh] OR
Judgment[Mesh] OR Medical Informatics[Mesh] OR Medical Records Systems, Computerized[Mesh]
OR Mental Recall[Mesh] OR Organizational Culture[Mesh] OR Patient Access to Records[Mesh] OR
Patient Participation[Mesh] OR Feedback[Mesh] OR Forms and Records Control/standards[Mesh] OR
Guidelines as Topic[Mesh] OR Health Knowledge, Attitudes, Practice[Mesh] OR Health Literacy[Mesh]
OR Health Records, Personal[Mesh] OR Physician-Patient Relations[Mesh] OR Physicians Practice
Patterns[Mesh] OR Problem Solving[Mesh] OR Professional-Patient Relations[Mesh] OR Reminder
Systems[Mesh] OR Systems Analysis[Mesh] OR Time Factors[Mesh] OR Truth Disclosure[Mesh] OR
Knowledge Bases[Mesh] OR cognitive error OR bias OR metacognition)
2
Search Strategy B replicates that used by Singh et al , though for a wider date range. Citations from their
search dates (2000-2009) were removed from both Search A and B, so as not to duplicate their work.
References
1.
Ioannidis JP, Lau J. Evidence on

interventions to reduce medical errors: an
overview and recommendations for future
research. J Gen Intern Med. 2001;16(5):32534.11359552
2.
C-79
Singh H, Graber ML, Kissam SM, et al.

System-related interventions to reduce
diagnostic errors: a narrative review. BMJ
Qual Saf. 2012;21(2):160-70.22129930
SECTION B. Methods
Figure 2, Chapter 35. Diagnostic errors systematic review flow chart
Chapter 36. Monitoring Patient Safety Problems (NEW)

Name
MEDLINE
PreMEDLINE
PSNet
Date limits
1996 September 21, 2011
1996 September 21, 2011
Searched September 22, 2011
Searched September 13, 2011
Platform/provider
OVID SP
OVID SP
PubMed
AHRQ

C-80

MEDLINE.
Conventions:
OVID
$
=
exp
=
.de. or
/
=
.fs.
=
floating subheading
.hw. =
.md. =
.mp. =
.pt.
=
publication type
.ti.
=
limit to title
.tw.
=
PubMed
[mh] =
MeSH heading
[majr] =
[pt]
=
publication type
[sb] =
[sh] =
[tiab] =
Concept
Adverse events
ae.fs.
Adverse outcome/
exp Cross infection/
Hospital infection/
Iatrogenic disease/
exp Medical errors/
Chart review
Concurrent review/
Documentation/
Drug utilization review/
Medical audit/
Medical records/
Medical record review/
Patient safety/
Safety/
Safety management
Patient safety
C-81
Keywords
Administration
Adverse events
Diagnostic
Error$
Iatrogenic
Medical
Medication
Nosocomial
Chart review
Case finding
Computerized surveillance
Patient safety
Patient Safety Organization
PSO
Concept
Reporting Systems
Medline/Embase
Keywords
Adverse drug reaction reporting systems/

Population surveillance/
exp Product surveillance, postmarketing/
exp Postmarketing surveillance/
Sentinel surveillance/
PSNet
Trigger tools
Error reporting
Government reporting
Institutional reporting
Reporting
Medline/Embase
Electronic medical record/
Hospital information systems/
PSNet
Computerized adverse event detection
Electronic health records
Automated
Automatic
Computer-based detection
Data mining
Electronic adj2 screen$
Surveillance
Trigger tool$
Set
Number
1
Concept
Adverse events
2
3
4
5
6
Combine sets
Direct observation
Chart review
Combine sets chart

review
Trigger tools
8
9
10
11
12
13
Combine sets
Combine sets trigger
tools
Search statement
Ae.fs. or exp cross infection/ or iatrogenic disease/ or exp
medical errors/ or adverse outcome/ or hospital infection/ or
iatrogenic disease/
Iatrogenic or nosocomial or (hospital adj acquired) or ((medical
or medication or diagnostic or administration) adj2 error$)
1 or 2
Direct observation or (executive adj walk$)
Chart review or chart$.ti. or case finding
Concurrent review/ or documentation or drug utilization review/
or medical audit/ or medical records/ or medical record review/
or utilization review/
5 or 6
(electronic medical record/ or hospital information systems/ )
and (data mining or automated or automatic or surveillance)
Electronic adj2 screen$
Computer-based detection
Trigger tool$
8 or 9 or 10 or 11
3 and (4 or 7 or 12)
C-82
Patient Safety Net (PSNET)

Set
Number
Concept
#
downloaded
Search statement
Browsed categories:
Computerized adverse event detection Governmental reporting
Institutional reporting
Non-governmental reporting
Patient safety indicators
Total Downloaded
Total Retrieved
250
176
Total Included
SECTION B. Methods
Inclusion/Exclusion Criteria
Healthcare organizations have been using a wide array of methods to detect patient safety
problems. These methods include incident reporting, direct observation of patient care, chart
review, analysis of malpractice claims, patient complaints and reports to risk management,
executive walk rounds, trigger-tool use, patient interviews, morbidity and mortality conferences,
autopsy, and clinical surveillance. Many of these methods (e.g., trigger tools) can be further
categorized by the targeted problems (e.g., medication-related medical errors or iatrogenic
infections), tools, algorithms, and data source used. Given the limited time frame for this review,
we focus this chapter on general approaches to detecting patient safety problems that involve
using multiple methods (e.g., incident reporting, executive walk rounds, clinical surveillance,
chart review, and trigger tools) to collect data. We primarily reviewed studies that compared the
utilities of different methods, because we believe that understanding the strengths and
weaknesses of different methods is most relevant to decision makers who need to form an
effective strategy for monitoring patient safety problems for their organizations. Comparison
studies that used one method (e.g., chart review) as a gold standard to validate another method
(e.g., incident reporting) were not included for this chapter, because, in essence, these studies
still focused on one individual method (i.e., the method being validated).
Only full published studies were considered for review (meeting abstracts were excluded). Only
English-language publications were eligible for inclusion. For the effectiveness and harms of the
PSP, we considered including studies of any design (e.g., systematic reviews, randomized
controlled trials, non-randomized controlled trials, prospective and retrospective observational
studies, surveys) that may provide relevant data. For the implementation and context of the PSP,
we primarily abstracted data from the effectiveness or safety studies being reviewed.
Risk of Bias and Strength of Evidence
We did not assess the risk of bias of the included studies or the overall strength of evidence. The
body of evidence consists of one systematic review and several studies that compared the types
and numbers of patient safety problems identified using different methods. No adequately
C-83
validated tool is available for assessing this type of comparison studies or for assessing the
overall strength of evidence that mixes such comparison studies with a systematic review.
Chapter 37. Interventions To Improve Care Transitions at

PubMed: 1990-7/29/2011
LANGUAGE: English
SEARCH #1:
transitional care OR readmission OR readmit* OR patient discharg* OR (length of stay[mh]
AND discharge plan*) OR (emergency service,hospital[mh] OR emergency service*[tiab] OR
emergency room*) AND (discharg* OR home visits[mh] OR home visit*[tiab] OR house
calls[mh] OR house call*)
AND
interven*[ti] OR intervention studies[mh] OR model*[ti] OR strateg* OR improv*[ti] OR
practices OR random*
SEARCH #2:
transitional care OR care transition* OR patient discharge[mh] OR patient discharge[tiab] OR
discharge plan*[tiab] OR hospital discharg*[tiab] OR patient readmission[mh] OR
readmission[tiab] OR readmit*[tiab] OR rehospital*[tiab]
OR
(emergency service*[tiab] OR emergency room* OR emergency service, hospital[mh]) AND
(post-discharg* OR postdischarg* OR home visit[mh] OR home visit*[tiab] OR house calls[mh]
OR house call*[tiab])
AND
home OR homes OR continuity of patient care OR continuity of care OR outpatient* OR
ambulatory care OR self care[mh] OR multidisciplinary care
AND
systematic OR observation* OR prospective OR retrospective OR cohort OR intervention stud*
OR evaluation stud* OR comparative stud* OR design* OR model[tiab] OR models[tiab] OR
program[tiab] OR programs[tiab] OR program evaluation[mh] OR ((single OR double OR triple
OR treble) AND (blind* OR mask*)) OR random*
===================================================================
C-84

Cochrane Central Register of Controlled Trials 1990-8/12/2011
LANGUAGE:
English
SEARCH STRATEGY:
(transitional care OR readmission OR readmit* OR patient discharg* ):ti,ab,kw or (length of
stay OR length-of-stay) AND discharge plan*:ti,ab,kw or (emergency service,hospital OR
emergency service* OR emergency room* ):ti,ab,kw
AND
(discharg* OR home visits OR home visit* OR house calls OR house call*):ti,ab,kw
AND
(interven* OR model* OR strateg* OR improv* OR practices OR random*):ti,ab,kw
AND
home OR house
=====================================================================
EconLit 1990-8/15//2011
LANGUAGE:
English
SEARCH STRATEGY:
EconLit 1990-8/15/2011
SEARCH STRATEGY:
transitional care OR readmission OR readmit* OR patient discharg* OR ( (length of stay OR
length-of-stay) AND discharge plan* ) OR emergency service,hospital OR emergency
service* OR emergency room*
AND
discharg* OR home visits OR home visit* OR house calls OR house call*
OR
readmit* OR readmission* OR rehospitali* OR re-hospitali*
AND
home OR house
C-85
=====================================================================
PsycINFO 1990-8/15/2011
SEARCH STRATEGY:
transitional care OR readmission OR readmit* OR patient discharg* OR ( (length of stay OR
length-of-stay) AND discharge plan* ) OR emergency service,hospital OR emergency
service* OR emergency room*
AND
discharg* OR home visits OR home visit* OR house calls OR house call*
AND
home OR house
Search modes - Phrase Searching (Boolean)
OR
transitional care
OR
readmit* OR readmission* OR rehospitali* OR re-hospitali*
AND
home OR house
=====================================================================
Embase 1990-9/12/2011
LANGUAGE:
English
SEARCH STRATEGIES:
SEARCH #1 (S. Rennke Strategy)
transitional care OR care NEAR/3 transition* OR discharge NEAR/3 plan* OR patient
NEAR/3 discharg* AND [humans]/lim AND [english]/lim AND [1990-2012]/py 8,919 View |
Edit
#1transitional AND care OR care AND transition* OR transition* AND in AND care OR
transition* OR discharge AND plan* OR patient AND discharg*
AND
C-86
readmi* OR rehospital* OR (avoid* OR reduc* AND admission*) OR medical NEAR/2 error*

OR medical NEAR/2 mistake* OR medication NEAR/2 error* OR adverse NEAR/2 event* OR
continuity of patient care OR home NEAR/2 visit* OR house NEAR/2 call* OR aftercare OR
team*
AND
interven* OR model*:ti OR strateg* OR improv*:ti OR implement* OR practices OR random*
OR controlled clinical trial OR controlled clinical trials OR program OR programs OR
programme OR programmes OR (program evaluation AND outcome*) OR systematic NEAR/2
review*
AND
human
SEARCH #2 (Revision 1):
transitional care OR rehospitali* OR discharge NEAR/2 plan* OR (emergency NEAR/2
service* OR emergency NEAR/2 room* AND hospital readmission) OR (medical NEAR/2
error* OR medical NEAR/2 mistake* OR medication NEAR/2 error* OR adverse NEAR/2
event* OR patient care team/exp OR patient care team OR patient care teams OR
multidisciplinary NEAR/2 team* AND hospital NEAR/2 readmi*)
AND
readmi* OR patient NEAR/2 discharg* OR (avoid* OR reduc* AND admission*) OR
continuity of care OR continuity of patient care OR home NEAR/2 visit* OR house NEAR/2
call* OR aftercare
AND
OR clinical NEAR/2 trial* OR (program* NEAR/2 evaluation* AND treatment NEAR/2
outcome*) OR systematic NEAR/2 review*
SEARCH #3 (Revision 2):
transitional care OR rehospitali* OR (discharge NEAR/2 plan* OR emergency NEAR/2
service* OR emergency NEAR/2 room* OR medical NEAR/2 error* OR medical NEAR/2
mistake* OR medication NEAR/2 error* OR adverse NEAR/2 event* OR patient care
team/exp OR patient care team OR patient care teams OR multidisciplinary NEAR/2 team*
AND readmi*)
AND
readmi* OR postdischarge OR post-discharge OR patient NEAR/2 discharg* OR (avoid* OR
reduc* AND (admission* OR admit*)) OR continuity of care OR continuity of patient care
OR home NEAR/2 visit* OR house NEAR/2 call* OR aftercare
AND
OR clinical NEAR/2 trial* OR (program* NEAR/2 evaluation* AND treatment NEAR/2
outcome*) OR systematic NEAR/2 review*
ALL THREE SEARCHES WERE COMBINED (ORED TOGETHER) &
DUPLICATES WERE REMOVED
TOTAL NUMBER OF RESULTS: 1427
C-87
=====================================================================
CINAHL 1990-9/12/2011
LANGUAGE:
English
SEARCH STRATEGY:
transitional care OR care transition* OR rehospital* OR discharge plan* OR patient care team
OR multidisciplinary team OR ( (emergency service, hospital OR emergency service* OR
emergency room*) AND readmi* ) OR medical error OR medical errors OR medical mistake*
OR medication errors OR adverse event*
AND
postdischarge OR post-discharge OR readmi* OR patient discharg* OR continuity of patient
care OR continuity of care OR home visits OR home visit OR house calls OR house
call OR aftercare OR ( (avoid* OR reduc*) AND (admit* OR admission ) )
AND
interven* OR strateg* OR implement* OR practices OR random* OR controlled clinical trial
OR controlled clinical trials OR program* OR systematic review* OR ( program evaluation
AND outcome* ) OR TI ( model* OR improv* )
NUMBER OF RESULTS AFTER REMOVING DUPLICATES: 588
SECTION B. Methods
C-88
Figure 1, Chapter 37. Theoretical model for the effectiveness of patient safety practices for
transitional care
C-89
Figure 2, Chapter 37. Trial flow diagram
C-90
Chapter 38. Use of Simulation Exercises in Patient Safety

Efforts
Search Strategies and Citation Results
PubMed Search Strategies
The first PubMed search, run November 14, 2011, yielded 42 titles. Full-text copies of 31
articles (74%) were reviewed, of which none reported T2 or T3 outcomes. Six articles from this
search were included in the chapter at the recommendation of experts or to provide supplemental
information about simulation. The first strategy was:
(((patient simulation[all fields] OR patient simulator[all fields]) AND (safety[all fields]
OR accident prevention[mesh])) AND (Clinical Trial[ptyp] OR Meta-Analysis[ptyp] OR
Practice Guideline[ptyp] OR Randomized Controlled Trial[ptyp])) OR (Models,
Theoretical[Mesh] AND accident prevention[mesh] AND Meta-Analysis[ptyp])
The second PubMed search, run November 15, 2011, yielded 304 titles. Full-text copies of 64
articles (21%) were reviewed, of which 11 studies reported T2 or T3 outcomes that were
included in the review. Three additional articles from this search were included in the chapter to
provide supplemental information about simulation. The second strategy was:
((patient simulation[all fields] OR patient simulator[all fields]) AND (Clinical Trial[ptyp]
OR Meta-Analysis[ptyp] OR Practice Guideline[ptyp] OR Randomized Controlled Trial[ptyp]))
OR (Models, Theoretical[Mesh] AND accident prevention[mesh] AND MetaAnalysis[ptyp])
The third PubMed search, run November 29, 2011, yielded 142 titles. Full-text copies of 12
articles (8%) were reviewed, of which 4 studies reported T2 or T3 outcomes and were included
in the chapter. One additional article from this search was included to provide supplemental
information about simulation. The third strategy was:
((Health Care Category[Mesh])) AND (simulation[All Fields]) AND (Meta-Analysis[ptyp])
Cochrane Library Databases Search Strategies
The Cochrane Database of Systematic Reviews was searched November 16, 2011 for simulat*
in title, abstract, or keyword fields. This search yielded 17 results, of which full-text copies of 6
articles (35%) were reviewed. Two of these studies reported T2 or T3 outcomes in combinations
with T1 outcomes and were included in the review of empirical literature. A third study was
included for an example of other uses of simulation that were not a focus of the current review.
The Cochrane Central Register of Clinical Trials was searched November 16, 2011 for
simulat* and safe* across any field. This search yielded 328 results, of which full-text copies
C-91
of 66 articles (20%) were reviewed. Thirteen of these studies reported T2 or T3 outcomes and
were included for review of empirical literature.
Other Literature Capture Methods
Secondary or chain-method literature prompted full-text review of an additional 15 empirical
articles, 13 of which are presented in the empirical review. This search method also resulted in
16 articles that were used to provide supplemental information about simulation. Experts
recommended an additional 22 articles, all of which are presented in the review. Many of these
articles included important theoretical work or important resources for those looking to
implement simulation. Experts also recommended important T1 studies for inclusion in the
review (n = 9).
Literature Totals
The empirical literature search resulted in 833 titles, of which 174 (21%) were reviewed in fulltext for inclusion in the review. The final reference list in Simulation and Patient Safety was
ultimately comprised of 45% (n = 40) literature directly captured by database searches, 32%
secondary literature (n = 27), and 25% literature recommended by practitioners with expertise in
simulation (n = 25). Eight additional articles served an explanatory function (e.g., clinical
rationale for placing a central venous catheter), and these articles were retrieved from free-text
searches in PubMed and Google Scholar.
SECTION B. Methods
The methodology for identifying empirical literature in this review involved three primary
mechanisms. In the first mechanism, structured search strategies for PubMed and the Cochrane
Library Databases provided the initial capture of simulation references. These searches were
limited to meta-analyses or systematic reviews, and to studies that were empirical in nature.
Theoretical pieces and commentary publications were not excluded in these search strategies, but
these publication types were not a focus of this mechanism to capture literature. The search
strategies were limited to general terminology (e.g., simulation) rather than specific terms that
might be required if one wished to perform a systematic review of simulation practices. Specific
simulation search terms might include the clinical specialty under investigation (e.g.,
anesthesiology), the procedure under investigation (e.g., laparoscopic cholecystectomy), the
purpose of the simulation (e.g., curriculum), or the fidelity and specific simulation exercise
(e.g., mannequin). Due to the brief nature of the current review, and to the extensive possible
combinations of these specific terms, this search mechanism identified literature through general
terms rather than exhaust these combinations. In the second mechanism, practitioners with
expertise in simulation were asked to provide recommendations on seminal work in simulation,
current key articles, empirical research on simulation and patient safety, areas of focus most
pertinent to implementing simulation, and guidance in terms of implementing simulation.
Secondary or chain-method capture of references provided the third mechanism to inform this
review. That is, reference lists in articles captured from the first two mechanisms provided
additional literature. Specifics and resulting citations are provided below on each of these search
mechanisms.
C-92
Empirical articles were held to a translational science paradigm for inclusion. Articles were
given priority if they reported outcomes from care provided to actual patients, or from actual care
system interventions. In terms of translational science, these are T2 or T3 simulation studies,
respectively. Due to the nature of simulation, selected studies that did not report outcomes from
care provided to actual patients (i.e., T1 or within the lab studies) were included if experts
recommended their inclusion to adequately represent the applications of simulation. All efforts
were made to remain inclusive across clinical specialties, no preference was assigned to specific
procedures or care practices. There is a section on central venous catheter placement that
provided an in-depth look at simulation to improve patient safety. This literature was selected
for in-depth presentation because (1) as a specific topic it had the greatest number of articles
captured in our review with outcomes reported at both the T2 and T3 level, and (2) this particular
line of research included analyses of costs for those looking to implement simulation.
Chapter 39. Obtaining Informed Consent From Patients: Brief

Update Review
Pubmed was searched for review articles with the MeSH term of Informed Consent that were
published since 2001. This was supplemented with a Google search for informed consent and
patient safety, informed consent and health literacy, simplified informed consent; written
educational materials and informed consent, decision aids and informed consent informed
consent and reading comprehension informed consent and Limited English Proficiency,
informed consent and patient comprehension, informed consent and teach back, informed
consent and structured interview, informed consent and computer and video informed
consent. Forward citation searches using Google Scholar were also done for included original
studies. A search on PSNet was also performed.
SECTION B. Methods
Titles and abstracts were reviewed by a physician health services researcher with expertise in
informed consent. Inclusion criteria were informed consent in a clinical setting; articles about
informed consent in the research setting were excluded. The citations from relevant reviews were
searched for original studies, and full text articles of potentially relevant original studies were
reviewed. The synthesis of included studies was narrative.
Chapter 40. Team Training in Health Care: Brief Update

Review
Given the presence of recent reports and systematic reviews in this area, we did not conduct a
systematic literature search for this topic. Key information was compiled from previous reports
and from articles identified by experts in this area.
C-93
SECTION B. Methods
Systematic reviews and articles were abstracted by health services researchers with expertise in
the topic area, and the results were narratively synthesized.
Chapter 41. Computerized Provider Order Entry With Clinical

Decision Support Systems: Brief Update Review
We searched the specialized database AHRQ Patient Safety Net (PSNet) using the search terms
computerized provider order entry, computerized physician order entry, clinician decision
support systems, clinical decision support systems, electronic medical records, and health
information technology. We also manually reviewed the reference lists of the articles identified
through this search.
SECTION B. Methods
We evaluated the effectiveness of this PSP by identifying and narratively summarizing the
systematic reviews of this topic that have been published since 2007, as well as identifying and
summarizing additional original research studies that were published in 2011 (and thus are too
recent to have been included in systematic reviews). Data regarding cost, implementation issues,
and potential for harm associated with this PSP were summarized narratively.
Chapter 42. Tubing Misconnections: Brief Review (NEW)

Name
PubMed
ECRI Institute:
Clinical Risk
Management
Health Devices Alerts
Health Devices
Healthcare Risk
Control
Health Technology
Assessment
Information Service
Operating Room Risk

Management
Date limits
2005 November 9, 2011
2005 November 9, 2011
C-94
Platform/providers
www.pubmed.gov
www.ecri.org
PUBMED English language, human

Set Number
1
2
3
4
5
6
7
Concept
Tubing
misconnections
Reviewed related
citations for PMIDs
Search statement
Equipment design[mh] OR equipment safety[mh]
#1 AND (tube* OR tubing OR catheter*)
#2 AND (connector* OR connection* OR misconnect*)
Misconnections[ti]
(luer*[tw] OR tubing[tw]) AND misconnection*
#2 OR #3 OR #4 OR #5
16610452
17090266
18491692
16739386
ECRI Institute Resources

Set Number
1
2
Concept
Tubing
Misconnections
Relevant UMDNS
codes
Combine sets
Search statement
Misconnection OR connectors
17-501 Intravenous Line Connectors
OR
11-726 Fittings/Adapters
OR
16-795 Fittings/Adapters, Pin-indexed
OR
11-729 Fittings/Adapters, Luer
1 OR 2
Other mechanisms used to retrieve additional relevant information included review of

bibliographies/reference lists from peer-reviewed and gray literature. (Gray literature consists of
reports, studies, articles, and monographs produced by federal and local government agencies,
private organizations, educational facilities, consulting firms, and corporations. These documents
do not appear in the peer-reviewed journal literature.)
Sites viewed for this topic include:
Joint Commission www.jointcommission.org
U.S. Food and Drug Administration www.fda.gov
Pennsylvania Patient Safety Authority www.patientsafetyauthority.org
PSNet- www.psnet.ahrq.gov
SECTION B. Methods
in both systematic reviews and medical devices. Included studies consisted of guidance
documents and clinical studies that discussed engineering controls and work practice changes to
prevent tubing misconnections. Potential barriers to implementation, and reported benefits and
harms from the patient safety practices were also assessed. Included guidance documents and
studies were narratively summarized by the author.
C-95
Chapter 43. Limiting Individual Providers Hours of Service:

Brief Update Review
We searched the specialized database AHRQ Patient Safety Net (PSNet) using the search terms
work hours, duty hours, hours of service, fatigue, sleep deprivation, and burnout,
and manually reviewed the reference lists of the articles and reports identified through this
search.
SECTION B. Methods
We evaluated the effectiveness of this PSP by identifying and narratively summarizing the
systematic reviews and original research studies of this topic that have been published since
2004. Data regarding cost, implementation issues, and potential for harm of this PSP were also
summarized narratively.
C-96
Appendix D. Supplementary Evidence Tables
D-1
Evidence Tables for Chapter 3. High-Alert Drugs: Patient Safety Practices for Intravenous
Anticoagulants
Table 1, Chapter 3. Evidence table
Author, year
Description of
PSP
Study Design
Theory or Description of
Logic Model Organization
Contexts
Sample Size
1
Baird, 2001
Fanikos,
2
2007
Fraipont,
3
2003
Multi-component
A single protocol
for heparin
administration was
developed by a
team of doctors,
nurses and a
pharmacists.
Pre-post
Not reported large tertiary

care hospital58 patients on 5
intensive care
physicianunits, 115 beds
specific
protocols; 10
patients on new
protocols.
Smart pump; drug Pre-post

Not reported Brigham and
library with point-ofWomens
7,395
care decision
Hospital
support for high or medication
low infusion rates; alerts from a
possible 14,012
can infusing 4
administered
drugs
heparin doses
simultaneously;
programmable hard in 3,674
drug alerts
patients
smart infusion
device with a
hospitaldetermined drug
library and software
Nurse-directed
Pre-post
Not reported
weight-based
nomogram
19 nomogram,
19 not
Leadership :
Protocol
development
team
Implementation Outcomes: Benefits Outcomes: Influence of

Details
Harms
Contexts on
Outcomes
None.
Implementation None stated.

tools : Est. hard
limits for rates
outside the
defined
guardrails & softlimits for
anticoagulants
8-bed Intensive Implementation

care unit in 635- tools : Raschke
bed university
nomogram
hospital in
Belgium
Received optimal
Not reported Not reported
bolus dose Results: 5
(8.6%) pre vs: 10
(90%) post
Statistics: NR
Mean time to
anticoagulation
Results: 34 hrs vs 63
+- 49 hours
Statistics: NR
Results:
Anticoagulation
medication errors: 49
before; 48 after
Statistics: NS
Time to therapeutic Not reported Not reported

anticoagulation: 13.5
hours standard vs 9.5
hours nomogram, NS
Complications: 2
standard vs 1
nomogram, NS
D-2
Comments
Results post
implementation
only: Prevented 10fold overdose in 40
patients; 100-fold
overdose in 40
patients; and >100fold overdose in 10
patients; similar
results for under
doses; heparin was
#4 most common
drug generating
alerts
Author, year
Oyen, 2005
Description of
PSP
Study Design
Sample Size
Multi-component
Computerized
Pre-post
Logic model
nomogram for
419 nomogram,
acute coronary
98 comparison
syndromes
Cardiovascular
services (88
beds) at a 1300bed teaching
hospital
Contexts

Details
Harms
Contexts on
Outcomes
Implementation Ot described
tools: Dosing
based on US
organization
guidelines
Percentage aPTT in Not reported Not reported

goal range
Results: 44%
nomogram vs 27%
not
Statistic: p<0.01
Time to goal aPTT
Result: 0.42 days
nomogram, 1.6 days
not
Statistic: p<0.01
Prusch, 2011 Intelligent infusion Pre-post

Model for
devices (IIDs), barhow IID
code-assisted
16,533
works
medication
opportunities
administration
pre and 16,833
system, and
opportunities
electronic
postmedication
implementation
administration
record systemintegrated to
populate providerordered,
pharmacistvalidated infusion
parameters on IIDs
IV interoperability
538-bed
Organizational
community
characteristics :
teaching hospital multidisciplinary
- expanded to all team and
units
relationship with
BCMA and IID
vendors to
develop
interoperability
between
systems
Leadership :
Executive
sponsorship,
Direction and
support of
pharmacy and
therapeutics
committee
Implementation
tools : Nurse
education
D-3
preparation,
pilot, validation,
and expansion;
extensive
software design
and testing
before
introduction to
patient care
Telemetry drug
library monthly
compliance
Results: 56.5 pre to
72.1 post
Statistics: p<0.001
Number of telemetry
manual pump edits
Results: 56.9 to 14.7
Statistics: p<0.001
Comments
Complications not
reported;
discussion that on
prior paper
nomogram,
clinicians deviated
over 50% of the
time by adjusting
doses; program
provided feedback
and performed
calculations;
computerization
allowed
individualized
protocol for acute
coronary
syndromes
similar decrease in
medical-surgical
drug library results;
reduction in
monthly reported
intravenous
heparin errors (28
to 17, NS); cost:
24.8% reduction
(23.4 sec onds) in
the mean nursing
time for pump
programming; 90%
compliance
Author, year
Toth, 2002
Description of
PSP
Study Design
Contexts
Sample Size
Multi-component
Weight-based
RCT
Not reported Neurology ward,
nomogram for
Canada
heparin dosing in 206 patients
TIA and/or stroke.

Details
Harms
Contexts on
Outcomes
Results: Total
complications: 9 pre
(8.5%) vs 2 post (2%)
Statistics: p=0.04
Supratherapeutic
aPTT
Results: 1.1
nomogram vs. 1.6 no
nomogram
Statistics: p=0.01
Time to therapeuticrange aPTT
Results: 13
nomogram, 18 no
nomogram
Statistics: p<0.01
Zimmermann, Weight-based
7
2003
heparin nomogram
for patients with
acute coronary
syndromes
Pre-post
Not reported Public hospital
Weight-based
nomogram was
based on other
nomograms in
literature;
dosage based
on absolute
weight. Weight
and aPTT
determined later
adjustment in
infusion rate.
84 patients
weight-based,
89 patients in
non-weightbased
Results: Time to first Not reported Not reported

therapeutic aPTT:
Nomogram median
8.75 vs >24 hours
Statistics: (p<0.001)
Mean number of
aPTT determinations
Results: 3.62(.85) (no
nomogram) vs 4.15
(.83) (nomogram)
Statistics: (p=0.002)
Major hemorrhagic
events
Results: 4 (4.5%)
non-weight-based, vs
2 (2.4%) weightbased, NS
D-4
Comments
Doctor completed
nomogram; bolus
provided if
indicated.Initial
heparin found by
nomogram. Nurses
changed heparin
from aPTT results
by following
nomogram. Also,
significantly fewer
calls to house staff
and mistakes made
in nomogram
group. Time to
discontinue
heparin:4 02.8 vs.
4.63.8; P=0.33;
94% of staff
preferred use of
nomogram
Adherence to
nomograms was
good (not
described in detail)
References
1.
Baird RW. Quality improvement efforts in

the intensive care unit: development of a
new heparin protocol. Proc (Bayl Univ Med
Cent) 2001; 14(3):294-6; discussion 296-8.
2.
Fanikos J, Fiumara K, Baroletti S et al.

Impact of smart infusion technology on
2007; 99(7):1002-5.
3.
Fraipont V, Lambermont B, Moonen M,

DOrio V. Annales Francaises dAnesthesie
et de Reanimation: Comparison of a nursedirected weight-based heparin nomogram
with standard empirical doctor-based
heparin dosage. 2003; 22:591-4.
4.
Oyen LJ, Nishimura RA, Ou NN, Armon JJ,

Zhou M. Effectiveness of a computerized
system for intravenous heparin
administration: using information
technology to improve patient care and
patient safety. Am Heart Hosp J 2005;
3(2):75-81.
5.
Prusch AE, Suess TM, Paoletti RD, Olin ST,

Watts SD. Integrating technology to
improve medication administration. Am J
Health Syst Pharm 2011; 68(9):835-42.
6.
Toth C, Voll C. Validation of a weightbased nomogram for the use of intravenous

heparin in transient ischemic attack or
stroke. Stroke 2002; 33(3):670-4.
7.
Zimmermann AT, Jeffries WS, McElroy H,

Horowitz JD. Utility of a weight-based
heparin nomogram for patients with acute
coronary syndromes. Intern Med J 2003;
33(1-2):18-25.
Evidence Tables for Chapter 4. The Clinical Pharmacists

Role in Preventing Adverse Drug Events: Brief Update
Review
This brief review had no additional evidence tables. There is one table in the text.
Evidence Tables for Chapter 5. The Joint Commissions Do

Not Use List: Brief Review (NEW)
This brief review had no additional evidence tables.
Evidence Tables for Chapter 6. Smart Pumps and Other

Protocols for Infusion Pumps: Brief Review (NEW)
D-5
Evidence Tables for Chapter 7. Barrier Precautions, Patient Isolation, and Routine
Surveillance for the Prevention of Healthcare-Associated Infections: Brief Update
Review
Reference
McGinigle KL,
2008(33)
Study Population/Setting
Systematic literature review evaluated
studies assessing the use of active
surveillance to reduce MRSA-related
morbidity, mortality and costs
Systematic literature
review
Backman C, 2011(35)
Systematic literature review evaluated

articles published on infection prevention
and control programs for multidrug-resistant
organisms in acute care hospitals
review
Cooper BS, 2003(10)
Systematic literature review assessed the

quality of the literature regarding the
effectiveness of different isolation policies
and screening practices in reducing the
incidence of MRSA colonization and
infection among hospitalized patients.
review
D-6
Summary/Main Contribution
The investigators did not identify any randomized, controlled
trials. They reviewed 16 observational studies and 4 economic
analyses. Only 2 studies included control groups. None of the
studies were of good quality. Thirteen studies reported
decreases in the incidence of MRSA infections associated with
the use of active surveillance
Existing evidence may favor the use of active surveillance, but
the evidence was of poor quality and the investigators could not
make definitive recommendations.
32 articles were assessed, of which 53% used surveillance; 75%
implemented infection control precautions to prevent
transmission; 22% introduced environmental measures; 28%
used patient decolonization; 56% had an administrative measure
as an intervention; 63% had education and training of healthcare
personnel; and 25% had judicious use of antimicrobial agents.
Although the evidence of the relationship between infection
prevention and multidrug-resistant infection rates was weak; the
overall evidence supported use of multiple interventions to
reduce the rates of multidrug-resistant organisms
46 studies were evaluated, of which 18 assessed isolation, 9
assessed cohorting nurses, and 19 assessed other isolation
policies. Few were planned prospective studies and all but one
included multiple interventions. Investigators for most studies did
not consider potential confounders, implement measures to
prevent bias, or use appropriate statistical analysis.
The studies were limited by major methodological weaknesses
and inadequate reporting. Thus, Cooper et al, could not exclude
plausible alternative explanations for the decreased incidence of
MRSA acquisition. The investigators conducting the
metaanalysis did not identify any well-designed studies that
allowed them to assess the role of isolation measures alone.
Active Surveillance
Reference
Robicsek A, 2008(36)
In a 3-hospital, 850-bed organization with
approximately 40,000 admissions each year,
a 3-phase quasi-experimental study
compared:
Phase 1: Baseline,
Phase 2: Universal surveillance for MRSA
among all patients admitted to the ICU and
contact isolation for patients who carried
MRSA,
Phase 3: Universal surveillance for MRSA
among all patients admitted to the hospital
and contact isolation plus decolonization for
patients who carried MRSA.
The investigators implemented an MRSA
bundle, including active surveillance and
contact isolation for MRSA in 153 acute care
Veterans Affairs hospitals nationwide to
decrease healthcare-associated MRSA.
21,754 surgical ICU patients admitted to a
Swiss teaching hospital were included in a
crossover study comparing rapid screening
on admission to detect MRSA colonization
plus standard infection control measures vs
standard infection control measures alone.
Active Surveillance
Jain R, 2011(24)
Active Surveillance
Harbarth S, 2008(37)
Active Surveillance
Huskins WC,
2011(13)
This cluster-randomized trial included 5,434

admissions to 10 intervention ICUs, and
3,705 admissions to 8 control ICUs.
Intervention ICUs performed surveillance for
MRSA and VRE colonization and expanded
use of barrier precautions. Control ICUs
continued to use existing practice.
Active Surveillance
Siddiqui AH, 2002

(70)
Four time periods (pre-active surveillance,

first period of active surveillance for VRE,
period without active surveillance, and
second period of active surveillance for VRE)
were compared to determine the effect of
active surveillance for VRE in two ICUs.
D-7
The absolute change between baseline and ICU surveillance
was -1.5 infections per 10,000 patient-days (p = 0.15), and the
absolute change between baseline and universal surveillance of
all patients admitted to the hospital was -5.0 infections per
10,000 patient-days (p < 0.01)
The investigators concluded that universal surveillance for MRSA
of all patients on admission was associated with a reduction in
MRSA infections during admission and within 30 days after
discharge.
After the VA system implemented the bundle, the rates of

healthcare-associated MRSA decreased by 45% in the non-ICUs
and 62% in the ICUs compared with baseline rates (p< 0.001 for
trend).
During the intervention periods, 1.11 patients per 1,000 patientdays acquired healthcare-associated MRSA compared with 0.91
per 1,000 patient-days during the control periods (adjusted
incidence rate ratio=1.20; 95% CI: 0.85,1.69; p= 0.29).
Universal, rapid MRSA screening on admission was not
associated with decreased rates of healthcare-associated MRSA
among patients admitted to a surgical department where MRSA
carriage was endemic but where rates of healthcare-associated
MRSA was relatively low.
The mean ( standard error) ICU-level incidence of events of
colonization or infection with MRSA or VRE per 1,000 patientdays at risk, adjusted for baseline incidence, did not differ
significantly between the intervention and control ICUs (40.4
3.3 and 35.6 3.7 in the two groups, respectively; p = 0.35).
Surveillance for MRSA and VRE colonization and expanded use
of barrier precautions did not reduce transmission of MRSA or
VRE. However, the results may have been affected by
suboptimal adherence to standard precautions.
Incidence of VRE decreased during the periods of active
surveillance demonstrating reductions in VRE ranging from 32%
to 64%.
Active Surveillance
Reference
Price CS, 2003(71)
Isolation of high-risk
patients
Matsushima A,
2011(52)
Universal Glove
Bearman G, 2007
(49)
Universal Glove
Bearman G, 2010
(50)
This quasi-experimental study compared
hospital A, which did not routinely screen for
VRE colonization, to hospital B, which
actively screened high-risk patients for VRE
and placed VRE colonized or infected
patients under contact isolation. High-risk
patients were those admitted to the
hematology-oncology, transplant, or
intensive care units
This single ICU, quasi-experimental study to
assessed an intervention where all intubated
patients were placed under pre-emptive
contact precautions. In the first phase of the
study (415 patients), active surveillance for
MRSA was performed at ICU admission and
weekly with contact precautions for MRSA
positive patients. In the second phase of the
study (1280 patients), active surveillance
and contact precautions for MRSA remained,
however all intubated patients were also
placed on contact precautions.
6 month, single ICU study in which phase 1
(3 months) consisted of VRE and MRSA
surveillance cultures on admission and every
4 days with contact precautions for patients
colonized or infected with VRE or MRSA;
and phase 2 (3 months) consisted of
universal gloving only. In phase 1 there were
1090 patient-days and in phase 2 there were
1377 patient-days.
12 month single ICU study in which phase 1
(6 months) consisted of VRE and MRSA
surveillance cultures on admission and every
4 days with contact precautions for patients
colonized or infected with VRE or MRSA;
and phase 2 (6 months) consisted of
universal gloving and staff education. In
phase 1 there were 3,486 patient days and
in phase 2 there were 2,946 patient days.
D-8
When the analysis was corrected for patient-days, the rate of
VRE bacteremia was 2.1-fold higher in hospital A compared to
hospital B. The majority of VRE isolates in hospital A were
clonally related.
This study found a decrease in healthcare-associated MRSA

infections for both intubated patients (p=0.02) and in all ICU
patients (p<0.05) in the second phase of the study.
No difference was seen in VRE or MRSA acquisition in the 2

study phases. The total nosocomial infection rate was higher in
phase 2 compared to phase 1.
No difference was seen in VRE or MRSA acquisition in the 2

study phases.
Universal Gown and

Glove
Reference
Wright MO, 2004 (50)
A single ICU, quasi-experimental study in
which phase 1 assessed active surveillance
and contact precautions for MRSA and VRE
while phase 2 included active surveillance
for MRSA and VRE but also implemented a
bundle to stop a multidrug-resistant
Acinetobacter baumannii outbreak. The
bundle included contact isolation for all
patients in the ICU regardless of culture
positivity, supervised terminal cleaning,
education sessions, and ban on artificial
fingernails.
D-9
The intervention bundle controlled the outbreak of multidrugresistant Acinetobacter baumannii. In additions the bundle in
phase 2 led to a decrease in MRSA acquisition from 14% to 10%
(p=0.5), and VRE acquisition from 21% to 9% (p=0.05)
References
1.
Klevens RM, Edwards JR, Richards CL,Jr,

Horan TC, Gaynes RP, Pollock DA, et al.
Estimating health care-associated infections
and deaths in U.S. hospitals, 2002. Public
Health Rep. 2007 Mar-Apr;122(2):160-6.
2.
Klevens RM, Morrison MA, Nadle J, Petit

S, Gershman K, Ray S, et al. Invasive
infections in the united states. JAMA. 2007
Oct 17;298(15):1763-71.
3.
Xu JQ, Kochanek KD, Murphy SL, TejadaVera B. Deaths: Final data for 2007.
National vital statistics reports. Hyattsville,
MD: National Center for Health Statistics;
2010.
4.
Edwards JR, Peterson KD, Mu Y, Banerjee

S, Allen-Bridson K, Morrell G, et al.
National healthcare safety network (NHSN)
report: Data summary for 2006 through
2008, issued December 2009. Am J Infect
Control. 2009 Dec;37(10):783-805.
5.
6.
7.

(CDC). Staphylococcus aureus resistant to
vancomycin--United States, 2002. MMWR
Morb Mortal Wkly Rep. 2002 Jul
5;51(26):565-7.
(CDC). Vancomycin-resistant
Staphylococcus aureus--Pennsylvania, 2002.
MMWR Morb Mortal Wkly Rep. 2002 Oct
11;51(40):902.
Muto CA, Jernigan JA, Ostrowsky BE,
Richet HM, Jarvis WR, Boyce JM, et al.
SHEA guideline for preventing nosocomial
transmission of multidrug-resistant strains of
Staphylococcus aureus and enterococcus.
May;24(5):362-86.
8.
Wenzel RP, Edmond MB. Infection control:

The case for horizontal rather than vertical
interventional programs. Int J Infect Dis.
2010 Oct;14 Suppl 4:S3-5.
9.
Diekema DJ, Edmond MB. Look before you

leap: Active surveillance for multidrugresistant organisms. Clin Infect Dis. 2007
Apr 15;44(8):1101-7.
D-10
10.
Cooper BS, Stone SP, Kibbler CC, Cookson

BD, Roberts JA, Medley GF, et al.
Systematic review of isolation policies in the
hospital management of methicillin-resistant
Staphylococcus aureus: A review of the
literature with epidemiological and
economic modelling. Health Technol
Assess. 2003;7(39):1-194.
11.
Wright MO, Hebden JN, Harris AD,

Shanholtz CB, Standiford HC, Furuno JP, et
al. Aggressive control measures for resistant
Acinetobacter baumannii and the impact on
acquisition of methicillin-resistant
Staphylococcus aureus and vancomycinresistant enterococcus in a medical intensive
2004 Feb;25(2):167-8.
12.
ClinicalTrials.gov [Internet]; 2010.

Available from: www.clinicaltrials.gov.
13.
Huskins WC, Huckabee CM, OGrady NP,

Murray P, Kopetskie H, Zimmer L, et al.
Intervention to reduce transmission of
resistant bacteria in intensive care. N Engl J
Med. 2011 Apr 14;364(15):1407-18.
14.
Kallen AJ, Mu Y, Bulens S, Reingold A,

Petit S, Gershman K, et al. Health careassociated invasive MRSA infections, 20052008. JAMA. 2010 Aug 11;304(6):641-8.
15.
Chambers HF, Deleo FR. Waves of

resistance: Staphylococcus aureus in the
antibiotic era. Nat Rev Microbiol. 2009
Sep;7(9):629-41.
16.
Johnson JK, Khoie T, Shurland S, Kreisel K,

Stine OC, Roghmann MC. Skin and soft
tissue infections caused by methicillinresistant Staphylococcus aureus USA300
clone. Emerg Infect Dis. 2007
Aug;13(8):1195-200.
17.
Barbut F, Jones G, Eckert C. Epidemiology

and control of Clostridium difficile
infections in healthcare settings: An update.
Curr Opin Infect Dis. 2011 Aug;24(4):3706.
18.

(CDC). Surveillance for communityassociated Clostridium difficile-Connecticut, 2006. MMWR Morb Mortal
Wkly Rep. 2008 Apr 4;57(13):340-3.
19.

(CDC). Severe Clostridium difficileassociated disease in populations previously
at low risk--four states, 2005. MMWR Morb
Mortal Wkly Rep. 2005 Dec 2;54(47):12015.
29.
Siegel JD, Rhinehart E, Jackson M,

Chiarello L, Healthcare Infection Control
Practices Advisory Committee. Management
of multidrug-resistant organisms in health
care settings, 2006. Am J Infect Control.
2007 Dec;35(10 Suppl 2):S165-93.
20.
Farr BM, Jarvis WR. What works and what

doesnt for the control of methicillinresistant Staphylococcus aureus infection:
Dogma and data. Clin Infect Dis. 2009 Sep
15;49(6):987,8; author reply 989-90.
30.
Australian guidelines for the prevention and

control of infection in healthcare.
Commonwealth of Australia; 2010.
Available from: www.nhmrc.gov.au.
31.
21.
Vriens MR, Fluit AC, Troelstra A, Verhoef

J, van der Werken C. Is methicillin-resistant
Staphylococcus aureus more contagious than
methicillin-susceptible S. aureus in a
surgical intensive care unit? Infect Control
Hosp Epidemiol. 2002 Sep;23(9):491-4.
22.
Farr BM. Doing the right thing (and figuring

out what that is). Infect Control Hosp
Epidemiol. 2006 Oct;27(10):999-1003.
Wertheim HF, Ammerlaan HS, Bonten MJ,

van den Broek PJ, Troelstra A,
Vandenbroucke-Grauls CM, et al.
Optimisation of the antibiotic policy in the
Netherlands. XII. the SWAB guideline for
antimicrobial eradication of MRSA in
carriers. Ned Tijdschr Geneeskd. 2008 Dec
6;152(49):2667-71.
32.
Coia JE, Duckworth GJ, Edwards DI,

Farrington M, Fry C, Humphreys H, et al.
Guidelines for the control and prevention of
(MRSA) in healthcare facilities. J Hosp
Infect. 2006 May;63 Suppl 1:S1-44.
33.
McGinigle KL, Gourlay ML, Buchanan IB.

The use of active surveillance cultures in
adult intensive care units to reduce
methicillin-resistant Staphylococcus aureusrelated morbidity, mortality, and costs: A
systematic review. Clin Infect Dis. 2008 Jun
1;46(11):1717-25.
34.
Loveday HP, Pellowe CM, Jones SR, Pratt

RJ. A systematic review of the evidence for
interventions for the prevention and control
of meticillin-resistant Staphylococcus aureus
(1996-2004): Report to the joint MRSA
working party (subgroup A). J Hosp Infect.
2006 May;63 Suppl 1:S45-70.
35.
Backman C, Taylor G, Sales A, Marck PB.

An integrative review of infection
prevention and control programs for
multidrug-resistant organisms in acute care
hospitals: A socio-ecological perspective.
Am J Infect Control. 2011 Jun;39(5):368-78.
36.
Robicsek A, Beaumont JL, Paule SM,

Hacek DM, Thomson RB,Jr, Kaul KL, et al.
Universal surveillance for methicillinresistant Staphylococcus aureus in 3
affiliated hospitals. Ann Intern Med. 2008
Mar 18;148(6):409-18.
23.
Calfee DP, Giannetta ET, Durbin LJ,

Germanson TP, Farr BM. Control of
endemic vancomycin-resistant enterococcus
among inpatients at a university hospital.
Clin Infect Dis. 2003 Aug 1;37(3):326-32.
24.
Jain R, Kralovic SM, Evans ME, Ambrose

M, Simbartl LA, Obrosky DS, et al.
Veterans Affairs initiative to prevent
infections. N Engl J Med. 2011 Apr
14;364(15):1419-30.
25.
Kirkland KB. Taking off the gloves: Toward

a less dogmatic approach to the use of
contact isolation. Clin Infect Dis. 2009 Mar
15;48(6):766-71.
26.
Wenzel RP, Bearman G, Edmond MB.

Screening for MRSA: A flawed hospital
infection control intervention. Infect Control
27.
28.
Morgan DJ, Liang SY, Smith CL, Johnson

JK, Harris AD, Furuno JP, et al. Frequent
multidrug-resistant Acinetobacter baumannii
contamination of gloves, gowns, and hands
of healthcare workers. Infect Control Hosp
Epidemiol. 2010 Jul;31(7):716-21.
Marchaim D, Navon-Venezia S, Schwartz
D, Tarabeia J, Fefer I, Schwaber MJ, et al.
Surveillance cultures and duration of
carriage of multidrug-resistant
Acinetobacter baumannii. J Clin Microbiol.
2007 May;45(5):1551-5.
D-11
37.
Harbarth S, Fankhauser C, Schrenzel J,

Christenson J, Gervaz P, Bandiera-Clerc C,
et al. Universal screening for methicillinresistant Staphylococcus aureus at hospital
admission and nosocomial infection in
surgical patients. JAMA. 2008 Mar
12;299(10):1149-57.
38.
Tacconelli E, Karchmer AW, Yokoe D,

DAgata EM. Preventing the influx of
vancomycin-resistant enterococci into health
care institutions, by use of a simple
validated prediction rule. Clin Infect Dis.
2004 Oct 1;39(7):964-70.
39.
Edmond M. Cost-effectiveness of perirectal

surveillance cultures for controlling
vancomycin-resistant enterococcus. Infect
May;24(5):309,10; author reply 310-2.
40.
Riedel S, Von Stein D, Richardson K, Page

J, Miller S, Winokur P, et al. Development
of a prediction rule for methicillin-resistant
Staphylococcus aureus and vancomycinresistant enterococcus carriage in a Veterans
Affairs medical center population. Infect
Oct;29(10):969-71.
41.
Robicsek A, Beaumont JL, Wright MO,

Thomson RB,Jr, Kaul KL, Peterson LR.
Electronic prediction rules for methicillinresistant Staphylococcus aureus
colonization. Infect Control Hosp
Epidemiol. 2011 Jan;32(1):9-19.
42.
Wibbenmeyer L, Williams I, Ward M, Xiao

X, Light T, Latenser B, et al. Risk factors
for acquiring vancomycin-resistant
enterococcus and methicillin-resistant
Staphylococcus aureus on a burn surgery
step-down unit. J Burn Care Res. 2010 MarApr;31(2):269-79.
43.
Morgan DJ, Day HR, Furuno JP, Young A,

Johnson JK, Bradham DD, et al. Improving
efficiency in active surveillance for
or vancomycin-resistant enterococcus at
hospital admission. Infect Control Hosp
Epidemiol. 2010 Dec;31(12):1230-5.
44.
Furuno JP, McGregor JC, Harris AD,

Johnson JA, Johnson JK, Langenberg P, et
al. Identifying groups at high risk for
carriage of antibiotic-resistant bacteria. Arch
Intern Med. 2006 Mar 13;166(5):580-5.
D-12
45.
Harbarth S, Sax H, Fankhauser-Rodriguez

C, Schrenzel J, Agostinho A, Pittet D.
Evaluating the probability of previously
unknown carriage of MRSA at hospital
admission. Am J Med. 2006
Mar;119(3):275.e15,275.e23.
46.
Dubberke ER, Yan Y, Reske KA, Butler

AM, Doherty J, Pham V, et al. Development
and validation of a Clostridium difficile
infection risk prediction model. Infect
Apr;32(4):360-6.
47.
Tanner J, Khan D, Anthony D, Paton J.

Waterlow score to predict patients at risk of
developing Clostridium difficile-associated
disease. J Hosp Infect. 2009 Mar;71(3):23944.
48.
Garey KW, Dao-Tran TK, Jiang ZD, Price

MP, Gentry LO, Dupont HL. A clinical risk
index for Clostridium difficile infection in
hospitalised patients receiving broadspectrum antibiotics. J Hosp Infect. 2008
Oct;70(2):142-7.
49.
Bearman GM, Marra AR, Sessler CN, Smith

WR, Rosato A, Laplante JK, et al. A
controlled trial of universal gloving versus
contact precautions for preventing the
transmission of multidrug-resistant
organisms. Am J Infect Control. 2007
Dec;35(10):650-5.
50.
Bearman G, Rosato AE, Duane TM, Elam

K, Sanogo K, Haner C, et al. Trial of
universal gloving with emollientimpregnated gloves to promote skin health
and prevent the transmission of multidrugresistant organisms in a surgical intensive
2010 May;31(5):491-7.
51.
Tacconelli E. Screening and isolation for

infection control. J Hosp Infect. 2009
Dec;73(4):371-7.
52.
Matsushima A, Tasaki O, Tomono K, Ogura

H, Kuwagata Y, Sugimoto H, et al. Preemptive contact precautions for intubated
patients reduced healthcare-associated
transmission and infection in an intensive
care unit. J Hosp Infect. 2011 Jun;78(2):97101.
53.
Dubberke ER, Gerding DN, Classen D,

Arias KM, Podgorny K, Anderson DJ, et al.
Strategies to prevent Clostridium difficile
Control Hosp Epidemiol. 2008 Oct;29 Suppl
1:S81-92.
54.
Cohen SH, Gerding DN, Johnson S, Kelly

CP, Loo VG, McDonald LC, et al. Clinical
practice guidelines for Clostridium difficile
infection in adults: 2010 update by the
Society for Healthcare Epidemiology of
America (SHEA) and the Infectious
Diseases Society of America (IDSA). Infect
May;31(5):431-55.
61.
Filice GA, Nyman JA, Lexau C, Lees CH,

Bockstedt LA, Como-Sabetti K, et al.
Excess costs and utilization associated with
methicillin resistance for patients with
Staphylococcus aureus infection. Infect
Apr;31(4):365-73.
62.
Cosgrove SE, Qi Y, Kaye KS, Harbarth S,

Karchmer AW, Carmeli Y. The impact of
methicillin resistance in Staphylococcus
aureus bacteremia on patient outcomes:
Mortality, length of stay, and hospital
charges. Infect Control Hosp Epidemiol.
2005 Feb;26(2):166-74.
63.
Shadel BN, Puzniak LA, Gillespie KN,

Lawrence SJ, Kollef M, Mundy LM.
Surveillance for vancomycin-resistant
enterococci: Type, rates, costs, and
implications. Infect Control Hosp
Epidemiol. 2006 Oct;27(10):1068-75.
55.
Morgan DJ, Diekema DJ, Sepkowitz K,

Perencevich EN. Adverse outcomes
associated with contact precautions: A
review of the literature. Am J Infect Control.
2009 Mar;37(2):85-93.
56.
Day HR, Morgan DJ, Himelhoch S, Young

A, Perencevich EN. Association between
depression and contact precautions in
veterans at hospital admission. Am J Infect
Control. 2011 Mar;39(2):163-5.
64.
Stosor V, Peterson LR, Postelnick M,

Noskin GA. Enterococcus faecium
bacteremia: Does vancomycin resistance
make a difference? Arch Intern Med. 1998
Mar 9;158(5):522-7.
57.
Day HR, Perencevich EN, Harris AD,

Himelhoch SS, Brown CH, Gruber-Baldini
AL, et al. Do contact precautions cause
depression? A two-year study at a tertiary
care medical centre. J Hosp Infect. 2011 Jun
8.
65.
58.
Tarzi S, Kennedy P, Stone S, Evans M.

Methicillin-resistant Staphylococcus aureus:
Psychological impact of hospitalization and
isolation in an older adult population. J Hosp
Infect. 2001 Dec;49(4):250-4.
Clancy M, Graepler A, Wilson M, Douglas

I, Johnson J, Price CS. Active screening in
high-risk units is an effective and costavoidant method to reduce the rate of
infection in the hospital. Infect Control Hosp
Epidemiol. 2006 Oct;27(10):1009-17.
66.
Muto CA, Giannetta ET, Durbin LJ,

Simonton BM, Farr BM. Cost-effectiveness
of perirectal surveillance cultures for
controlling vancomycin-resistant
enterococcus. Infect Control Hosp
Epidemiol. 2002 Aug;23(8):429-35.
67.
Hubben G, Bootsma M, Luteijn M, Glynn

D, Bishai D, Bonten M, et al. Modelling the
costs and effects of selective and universal
hospital admission screening for methicillinresistant Staphylococcus aureus. PLoS One.
2011 Mar 31;6(3):e14783.
68.
Platt R, Takvorian SU, Septimus E, Hickok

J, Moody J, Perlin J, et al. Cluster
randomized trials in comparative
effectiveness research: Randomizing
hospitals to test methods for prevention of
healthcare-associated infections. Med Care.
2010 Jun;48(6 Suppl):S52-7.
59.
60.
Catalano G, Houston SH, Catalano MC,

Butera AS, Jennings SM, Hakala SM, et al.
Anxiety and depression in hospitalized
patients in resistant organism isolation.
South Med J. 2003 Feb;96(2):141-5.
Perencevich EN, Stone PW, Wright SB,
Carmeli Y, Fisman DN, Cosgrove SE, et al.
Raising standards while watching the
bottom line: Making a business case for
infection control. Infect Control Hosp
Epidemiol. 2007 Oct;28(10):1121-33.
D-13
69.
Gasink LB, Brennan PJ. Isolation

precautions for antibiotic-resistant bacteria
in healthcare settings. Curr Opin Infect Dis.
2009 Aug;22(4):339-44.
70.
Siddiqui AH, Harris AD, Hebden J, Wilson

PD, Morris JG,Jr, Roghmann MC. The
effect of active surveillance for vancomycinresistant enterococci in high-risk units on
vancomycin-resistant enterococci incidence
hospital-wide. Am J Infect Control. 2002
Feb;30(1):40-3.
71.
Price CS, Paule S, Noskin GA, Peterson LR.

Active surveillance reduces the incidence of
vancomycin-resistant enterococcal
bacteremia. Clin Infect Dis. 2003 Oct
1;37(7):921-8.
Evidence Tables for Chapter 8. Interventions To Improve

Hand Hygiene Compliance: Brief Update Review
This brief review had no additional evidence tables
D-14
Evidence Tables for Chapter 9. Reducing Unnecessary Urinary Catheter Use and
Other Strategies To Prevent Catheter-Associated Urinary Tract Infection: Brief
Update Review
Table 1, Chapter 9. Characteristics of studies with interventions to avoid unnecessary urinary catheter use
Study
(Country)
Study Design
Population,
Total N
Interventions to avoid
unnecessary catheter
PLACEMENT
None
Interventions to prompt REMOVAL of

unnecessary catheters
Other
Interventions
Apisarnthanarak et
1
al, 2007
(Thailand)
Pre-Post
All Inpatients,
N=2412 patients
Reminder: Nurse-generated daily

bedside verbal reminders to encourage
physicians to remove unnecessary UC.
None
Bruminhent et al,
2
2010 (USA)
Pre-Post
Med-Surg: Ward +
ICU, N=400 patients
None
Reminder: Sticker applied to medical

record to remind physicians to
discontinue unnecessary UCs.
None
Nonrandomized
crossover trial
Medical (non-ICU),
N=70 patients
Computerized UC order
required selection of an
appropriate UC indication
UC care
education
Crouzet et al,
4
2007 (France)
Pre-Post
All Inpatients,
N=234 patients
None
Dumigan et al,
5
1998 (USA)
Pre-Post
ICU: Med-Surg,
N=27103 patientdays
Guideline for appropriate UC

indications
Pre-Post
ICU: Medical,
N=337 patients
Pre-Post with
concurrent
controls
Med-Surg (non-ICU)
N=3736 intervention
patient-days, and
4041 control patientdays
Appropriate indications for UC

insertion were emphasized, and
list of inappropriate reasons to
insert was provided.
None
Stop order: Computer-generated stop

order for physicians to
discontinue/renew UC order 72 hours
after placement.
Reminder: Daily reminders from nurses
to physicians to remove unnecessary
UC >=4 days after insertion.
Stop order, nurse-empowered: Daily
use of UC indication protocol by nurse
empowered to remove UC no longer
meeting criteria without requesting
physician order.
Reminder: Daily review by nurses for
UC indication to make recommendations
for removal; removal required physician
order.
Reminder: Nurse generated reminder to
physician to remove UC when no
appropriate indication.
Cornia et al, 2003

(USA)
Elpern et al, 2009

(USA)
7
Fakih et al, 2008

(USA)
D-15
None
UC care
education
None
None
Study
(Country)
Study Design
Population,
Total N
Pre-Post
Pre-Post
ED,
N=322 patients had
UCs placed, of 2517
ED patients in
sample
Statewide, N=163
inpatient units in 71
hospitals
ICU: Med-Surg,
N=not provided
Fakih et al, 2010

(USA)
Fakih et al, 2012

(USA)
Fuchs et al, 2011

(USA)
10
Gokula et al,
11
2007 (USA)
Pre-Post
Gotelli et al, 2008

(USA)
Loeb et al, 2008

(Canada)
Other
Interventions
None
None
Education intervention to
promote adherence to
appropriate UC indications
Urinary retention protocol,
including use of bladder
scanner
None
None
Stop order: Daily checklist for

evaluating UCs; when not indicated,
physician order was requested for
removal.
None
Procedure-specific protocols for

appropriate indications for UC
placement
UC indication checklist attached
to UC kits
Pre-Post
13
Pre-Post
ICU: Med-Surg,
N=6297 patients
None
Pre-Post
ICU: Med-Surg,
N=13471
catheter-days
Pre-Post
All Inpatients,
N=112,140
patient-days
14
Knoll et al, 2011

(USA)
ED,
N=200 patients with
UCs placed in ED
Medical (not ICU),
N=not provided

12
Huang et al, 2004

(Taiwan)
Jain et al, 2006
(USA)
Pre-Post
PLACEMENT
Institutional guidelines for
appropriate UC placement, ED
physician education regarding
UC utilization
15
16
RCT
Medical (non-ICU),
N=692 patients
Stop order: Procedure-specific

protocols for UC removal.
None
None
Stop order, nurse-empowered: Nurses

were empowered to assess UC need by
protocol and remove if not indicated.
Reminder: Nurse generated daily
reminder to physician to remove
unnecessary UC 5 days after insertion.
None
None
Reminder: Daily use of checklist in

multidisciplinary rounds to determine if
UC still indicated, then nurse contacted
physician for order to removal UC if no
longer indicated.
Bladder Bundle:
UC care steps,
selected use of
silver-alloy UC.
Education interventions about

an approved hospital list of UC
indications
Stop order: Computerized order for UC

with indications and 72 h default stop
date.
Reminder: ICU daily checklist for UC
necessity.
Bundle: UC care
education,
dedicated UC
nurse.
Stop order, nurse-empowered: Prewritten in chart for nurses empowered to

discontinue UC based on criteria without
an additional physician order.
None
None
Computer UC order template

with indication
None
D-16
None
Study
(Country)
Study Design
Population,
Total N
Murphy et al,
17
2007 (USA)
Pre-Post
Not explained,
N=Not provided
Patrizzi et al,
18
2009 (USA)
Pre-Post
ED,
N=Not provided
Reilly et al, 2008

(USA)
Pre-Post
ICU: Med-Surg,
N=207 patients
Robinson et al,
20
2007 (USA)
Pre-Post
Med-Surg (non-ICU),
N=69 patients
Rothfield et al,
21
2010 (USA)
Pre-Post
Medical ICU stepdown unit,

N=99 patients
Intervention Group:
Medical,
Control Group:
Surgery.
N=3027 patients
Developed list of appropriate

indications for which UCs could
be requested by nurses
None
Urinary retention protocol,

including use of bladder
scanner
None
19
Saint et al, 2005

(USA)
22
Pre-Post with
concurrent
nonequivalent
controls
Schultz et al,
23
2011 (USA)
Pre-Post
ICU: unclear type,

N=Not provided
Seguin et al,
24
2010 (France)
Pre-Post
ICU: Surgical,
N=1271 patients
PLACEMENT
None

Other
Interventions
Reminder: Foley bag sticker with

time/date of insertion to remind to nurse
to notify physician when Foley in place
>48h in order to request removal.
UC care
education
Computerized ED UC order
with indications, UC alternatives
promoted, urinary retention
protocol with bladder scanner
use
Developed criteria for
appropriate UC placement in
ICU, implemented with
educational interventions
regarding UC indications, and
urinary retention protocol
None
None
Reminder: Daily use of checklist of

appropriate UC indications by nurse,
reminding nurse to contact physician to
recommend UC removal.
UC care
education
Stop order: Nurse identified patients

without appropriate indications, then
requested removal order from
physicians
Stop order: Nurses asked physicians
for order to remove UCs when not
indicated.
Reminder: Study nurse generated
sticker placed in chart reminding
physician to generate stop order after 48
hours of UC use if no longer needed
None

were empowered to insert and remove
UCs by protocol.
Stop order: Daily assessment required
by physicians to assess if UC is needed
or not; when categorized as not
indicated, then removed by nurses.
None
D-17
None
None
None
Study
(Country)
Study Design
Population,
Total N
Stephan et al,
25
2006
(Switzerland)
Pre-Post with
concurrent
nonequivalent
controls
Surgery: Ward+ICU
Intervention:
Orthopedic, N=539
Control:
Abdominal, N=489
Pre-Post
Medical (non-ICU),
N = 245 patients
Topal et al, 2005

(USA)
26
van den Broek et

27
al, 2011
(The Netherlands)
Voss, 2009
(USA)
28
Weitzel, 2008
(USA)
29
Wenger, 2010
(USA)
30
Pre-Post
All Inpatients, in 5
hospitals.
N=2943 patients
PLACEMENT
UC placement restrictions,
urinary retention protocol

Other
Interventions
Stop order: Pre-operative written order

to remove UC on post-operative day 1 or
2, depending on surgery.
UC care
education
Urinary retention protocol

including bladder scanner
Stop order: Computerized order entry

system order to prompt physicians to
remove/re-order UC if placed in ED or in
place >48 hours.
UC care
education
Bladder scanner protocol in 2

hospitals
Pre-Post
Medical (non-ICU),
N=187 patients age
65 or older
None
Pre-Post
Medical (unclear if
ICU), N=50 patients
None
Pre-Post
All Inpatients,
N=Not provided
None

were also empowered to remove UCs
no longer needed by protocol criteria.
Intervention varied by hospital:
Reminders: Used by 4 hospitals, placed
in patients record.
Stop order: Fixed order for removal,
employed by 1 hospital.
assessment by nurse for UC indications,
with authority for nurse to remove if not
indicated.
Reminder: Daily use of protocol by
nurse to review if UC still indicated,
unclear if protocol allowed for UC
removal without physician order.
assessment by nurse of UC necessity,
with authority to remove if not indicated.
D-18
Specially trained
UC nurse
None
None
UC care
education,
silver-alloy UC
References
1.
2.
3.
Apisarnthanarak A, Thongphubeth K,
Sirinvaravong S, et al. Effectiveness of
multifaceted hospitalwide quality
improvement programs featuring an
intervention to remove unnecessary urinary
catheters at a tertiary care center in
Thailand. Infect Control Hosp Epidemiol.
2007;28(7):791-8. PMID 17564980.
Bruminhent J, Keegan M, Lakhani A, et al.
Effectiveness of a simple intervention for
prevention of catheter-associated urinary
tract infections in a community teaching
hospital. Am J Infect Control.
2010;38(9):689-93. PMID 21034979.
Cornia PB, Amory JK, Fraser S, et al.
Computer-based order entry decreases
duration of indwelling urinary
catheterization in hospitalized patients. Am J
Med. 2003;114(5):404-7. PMID 12714131.
10.
Fuchs MA, Sexton DJ, Thornlow DK, et al.

Evaluation of an evidence-based, nursedriven checklist to prevent hospital-acquired
catheter-associated urinary tract infections in
intensive care units. J Nurs Care Qual.
2011;26(2):101-9. PMID 21037484.
11.
Gokula RM, Smith MA, Hickner J.

Emergency room staff education and use of
a urinary catheter indication sheet improves
appropriate use of foley catheters. Am J
17980237.
12.
Gotelli JM, Merryman P, Carr C, et al. A

quality improvement project to reduce the
complications associated with indwelling
urinary catheters. Urol Nurs.
2008;28(6):465-7, 73. PMID 19241786.
13.
Huang WC, Wann SR, Lin SL, et al.

Catheter-associated urinary tract infections
in intensive care units can be reduced by
prompting physicians to remove
unnecessary catheters. Infect Control Hosp
Epidemiol. 2004;25(11):974-8. PMID
15566033.
4.
Crouzet J, Bertrand X, Venier AG, et al.

Control of the duration of urinary
catheterization: impact on catheterassociated urinary tract infection. J Hosp
Infect. 2007;67(3):253-7. PMID 17949851.
5.
Dumigan DG, Kohan CA, Reed CR, et al.

Utilizing national nosocomial infection
surveillance system data to improve urinary
tract infection rates in three intensive-care
units. Clin Perform Qual Health Care.
1998;6(4):172-8. PMID 10351284.
14.
Jain M, Miller L, Belt D, et al. Decline in

ICU adverse events, nosocomial infections
and cost through a quality improvement
initiative focusing on teamwork and culture
change. Qual Saf Health Care.
2006;15(4):235-9. PMID 16885246.
6.
Elpern EH, Killeen K, Ketchem A, et al.

and associated urinary tract infections. Am J
Crit Care. 2009;18(6):535-41; quiz 42.
PMID 19880955.
15.
7.
Fakih MG, Dueweke C, Meisner S, et al.

Effect of nurse-led multidisciplinary rounds
on reducing the unnecessary use of urinary
catheterization in hospitalized patients.
2008;29(9):815-9. PMID 18700831.

21596671.
16.
Loeb M, Hunt D, OHalloran K, et al. Stop

orders to reduce inappropriate urinary
catheterization in hospitalized patients: a
randomized controlled trial. J Gen Intern
Med. 2008;23(6):816-20. PMID 18421507.
17.
Murphy D, Francis K, Litzenberger M, et al.

Reducing urinary tract infection: a nurseinitiated program. Pa Nurse. 2007;62(4):20.
PMID 18286841.
8.
Fakih MG, Pena ME, Shemes S, et al. Effect

of establishing guidelines on appropriate
urinary catheter placement. Acad Emerg
Med. 2010;17(3):337-40. PMID 20370769.
9.
Fakih MG, Watson SR, Greene MT, et al.

Reducing inappropriate urinary catheter use:
a statewide effort. Arch Intern Med.
2012;172(3):255-60. PMID 22231611.
D-19
18.
Patrizzi K, Fasnacht A, Manno M. A

collaborative, nurse-driven initiative to
reduce hospital-acquired urinary tract
infections. J Emerg Nurs. 2009;35(6):536-9.
PMID 19914479.
19.
Reilly L, Sullivan P, Ninni S, et al.

Reducing foley catheter device days in an
intensive care unit: using the evidence to
change practice. AACN Adv Crit Care.
2006;17(3):272-83. PMID 16931923.
20.
21.
22.
Robinson S, Allen L, Barnes MR, et al.

Development of an evidence-based protocol
for reduction of indwelling urinary catheter
usage. Medsurg Nurs. 2007;16(3):157-61.
PMID 17849921.
Rothfeld AF, Stickley A. A program to limit
urinary catheter use at an acute care hospital.
Am J Infect Control. 2010;38(7):568-71.
PMID 20381918.
Schultz P, Aljawawdeh A, Hopp T. EB105:

with a nurse-driven protocol. Crit Care
Nurse. 2011;31(2):e42.
24.
Seguin P, Laviolle B, Isslame S, et al.

Effectiveness of simple daily sensitization of
physicians to the duration of central venous
and urinary tract catheterization. Intensive
Care Med. 2010;36(7):1202-6. PMID
20237761.
Stephan F, Sax H, Wachsmuth M, et al.

Reduction of urinary tract infection and
antibiotic use after surgery: a controlled,
prospective, before-after intervention study.
16652311.
26.
Topal J, Conklin S, Camp K, et al.

Prevention of nosocomial catheterassociated urinary tract infections through
computerized feedback to physicians and a
nurse-directed protocol. Am J Med Qual.
2005;20(3):121-6. PMID 15951517.
27.
van den Broek PJ, Wille JC, van Benthem

BHB, et al. Urethral catheters: can we
reduce use? BMC Urol. 2011;11:10. PMID
21605403.
28. Voss AB. Incidence and duration of urinary

catheters in hospitalized older adults: before
and after implementing a geriatric protocol.
J Gerontol Nurs. 2009;35(6):35-41. PMID
19537293.
Saint S, Kaufman SR, Thompson M, et al. A

reminder reduces urinary catheterization in
hospitalized patients. Jt Comm J Qual
Patient Saf. 2005;31(8):455-62. PMID
16156193.
23.
25.
29.
Weitzel T. To cath or not to cath? Nurs.

2008;38(2):20-1. PMID 18223406.
30.
Wenger JE. Cultivating quality: reducing

rates of catheter-associated urinary tract
infection. Am J Nurs. 2010;110(8):40-5.
PMID 20671495.
Evidence Tables for Chapter 10. Prevention of Central LineAssociated Bloodstream Infections: Brief Update Review
Evidence Tables for Chapter 11. Ventilator-Associated

Pneumonia: Brief Update Review
Evidence Tables for Chapter 12. Interventions to Allow the

Reuse of Single-Use Devices: Brief Review (NEW)
D-20
Evidence Tables for Chapter 13. Preoperative Checklists and Anesthesia Checklists
Table 1, Chapter 13. Studies of the World Health Organization surgical safety checklist at other locations
Author/Year
Description of PSP
Study
Design
Theory or Logic Model
Description of Organization
Safety Context
2008 WHO surgical

checklist, unmodified
Before and
after study,
comparing
pre-training
period to
post-training
The underlying philosophy

of the checklist is that a true
team approach with good
communication between
operating room team
members is safer and more
efficient than a hierarchical
system that relies on
individuals
A U.K. hospital, orthopedic

operations. 28% of operations
were urgent, and 77% involved
general anesthesia
Pre-training period Feb-May 2009

(480 operations). During this period: Correct
checklist use 8%, and 47% thought it
improved team communication Pre-training
staff perceptions: 55% thought it caused an
unnecessary time delay, 28% thought it
improves patient safety, 47% thought it
improves team communication and
teamwork, 64% would want the checklist
used if they were having an operation
Helmio 2011
2008 WHO surgical

checklist. No specialtyrelated changes, but
some minor changes.
Checklist included in
publication;
modifications did not
exclude any items
Before and
after study
The idea of the checklist is

to be an add-on security tool
for the defined safety
standard
Finland, otorhinolaryngology-head
and neck surgery ORs. 747
operations in the two month study
periods combined. All subgroups
of otorhinolaryngology-head and
neck surgery were included.
One-month pre-implementation period in

May 2009 (304 operations). 17% were urgent
operations. 24% were on children. 16% were
local anesthesia. Before implementation:
Knowledge of OR-teams names and roles
ranged from 61 % to 92%. Discussing risks
was 24%. Postop instructions recorded 74%84%. Successful communication 79%-93%.
Conley 2011
2008 WHO surgical

checklist, unmodified
Case series
None explicitly stated.
Five Washington state hospitals.

Two hospitals had <10 ORs,
one had 10-20, and two had >20.
Two urban, two suburban, and
one rural.
Nothing reported about pre-existing safety

culture. The Vice President for Patient Safety
at the Washington State Hospital Association
provided significant assistance. Checklist
introduction Dec 2008 to Jan 2009.
Interviews conducted Sept - Dec 2009. One
of the five hospitals had a recent wrong-site
incision that motivated surgical staff and
opened peoples eyes to the need for
ongoing patient safety efforts
Without a doubt, the

checklist works best when
all staff members are
engaged
Large two-hospital Trust in the

U.K. with 10,000 staff and
850,000 patients annually.
Nothing about pre-existing safety culture.

To prepare for the checklist, they set up a
Patient Safety Working Group
Sewell 2011
Bell 2010
4,5
2008 WHO checklist

Case series
adapted different for
different surgical
specialties. Checklist not
included in publication.
D-21
Author/Year
Sparkes 2010
Royal Bolton
7
2010
Vats 2010
Kearns 2011
Description of PSP
Study
Design
Safety Context
2008 WHO checklist

locally adapted.
publication;
exclude any items
Case series
Discussed various ways a

checklist could enhance
safety, including teamwork
and effective communication
Teaching hospital in the U.K. with

29 ORs in five locations
performing specialized complex
surgery
NR
2008 WHO checklist,

unmodified. Local
adaptation of it was
considered but
ultimately not done.
Case series
Improve patient safety by

enhancing teamwork and
communication
Trust in the U.K. with eight ORs
Prior to the checklist, the trust already had a

core group of patient safety experts
assembled; this group met to discuss how to
introduce the checklist. They examined the
previous years 41 safety incidents and all
were found to be avoidable had the checklist
been in use
2008 WHO surgical

checklist adapted for
England and Wales.
publication;
exclude any items
Case series
the checklist ensures that U.K. academic hospital

critical tasks are carried out
and that the team is
adequately prepared for the
operation
Nothing reported about pre-existing safety

culture. Piloted March-Sept 2008 at a London
hospital in 58% of operations (424/729)
among the two ORs selected (one for
trauma/orthopedics OR, the other for
GI/GYN).
Checklists may be used to U.K. study in obstetrics ORs.

improve patient safety by
Tertiary referral obstetric center
ensuring that all elements of with ~6,400 deliveries per year.
a practice are instituted for
each new clinical event.
Before introducing the checklist, they

measured staff attitudes, preserving
respondent anonymity: 30% felt familiar
with others in the OR 81% felt
communication could improve 85% felt that in
elective cases the checklist would be useful
53% felt that in emergency cases the
checklist would be inconvenient
WHO surgical checklist, Before and

after study
version NR. Some
obstetric-specific checks
had been added, but the
list of revisions was not
reported. Checklist not
D-22
Author/Year
Norton 2010
10
Description of PSP
Study
Design
2008 WHO checklist

Case series
modified for pediatric
operations and also to
meet the 2009 Joint
Commission Universal
Protocol. Checklist
Removed the following
three items from the
WHO checklist: pulse
oximetry, difficult airway,
anticipated blood loss
Checklist can help to reduce Childrens hospital in the US

breakdowns in
performing numerous types of
pediatric surgery
communication, ineffective
teamwork, and lack of
compliance with process
measures
Safety Context
At this hospital they had been building a
quality infrastructure for five years prior, and
had already implemented the Universal
Protocol.
Note: NR=Not reported; Int=Intervention; OR=Operating room; GI=Gastrointestinal; GYN=Gynecology
Table 2, Chapter 13. Implementation findings in studies of the World Health Organization surgical safety checklist at other locations
Author/Year
1
Sewell 2011
Training
Study Phases and

Checklist Fidelity
Reasons for Success or

Failure
Opinions, Knowledge and

Behavior
Health Outcomes
Checklist forms placed in

ORs, compulsory training
video detailing correct and
incorrect uses of the
checklist, emphasis placed
on all team members being
responsible. Active
discouragement of a simple
tickbox approach. Checklist
training was not associated
with reductions in any
complications or mortality
Training phase first

(unreported duration). Posttraining period June-Oct
2009 (485 operations).
Correct checklist use 97%.
2 minutes. 20% thought it
caused an unnecessary
time delay.
The initial implementation

of the checklist was met
with resistance by some
operating room team
members as there was a
belief that many of the
points were already in
practice.
77% thought it improved

team communication, 68%
thought it improves patient
safety, 80% would want the
checklist used if they were
having an operation
Early complications 8.5%

before checklist training and
7.6% after. Mortality 1.9%
1.6% after. Lower
respiratory tract infections
2.1% before checklist
training and 2.5% after.
Surgical site infection 4.4%
3.5% after. Unplanned
return to OR 1.0% before
checklist training and 1.0%
after.
D-23
Author/Year
Helmio 2011
Training
Study Phases and

Checklist Fidelity
Training involved a
One-month implementation
presentation from an
period in Sept 2009
(443 operations).
outside expert and three
45 minute lectures. Specific
guidelines were in the OR,
and short instructions on the
back of the checklist.

Failure

Behavior
Use of the checklist

improved verification of
patient identity, but this was
still inadequate. Our study
confirms that the surgical
checklist fits well into
otolaryngology. We
recommend the use of this
checklist in all operations
overall, the operating

NR
room personnel were
supportive.
Anesthesiologists
knowledge about patients
had improved as compared
to the pre-implementation
period. Preoperative check
of anesthesia equipment
increased from 71% to 84%.
After implementation, staff
were more likely to
accurately report patient
identity, procedure, and
operative side. After
implementation, there was
improvement in: Knowledge
of OR-teams names and
roles ranged from 81 % to
94%. Discussing risks was
38%. Postop instructions
recorded 86%. Successful
communication 87%-96%.
D-24
Health Outcomes
Author/Year
Conley 2011
Bell 2010
4,5
Training
Study Phases and

Checklist Fidelity

Failure

Behavior
Health Outcomes
NR
Duration of rollout:
<2 months at three
hospitals, >6 months at two
hospitals.
The key is whether the local

champion can persuasively
explain why and adaptively
show how to use the
checklist. Implementation
was incomplete at three
hospitals: One cancelled
attempts to implement the
checklist due to fear of
insurmountable resistance
and poor interdisciplinary
communication Another
cancelled attempts because
they were unable to move
beyond pilot testing. The
third had less effective
implementation because of
a laissez-faire leadership
style; no training; staff
understood neither why nor
how the checklist could be
implemented
Interviews conducted, but

no quantitative summary of
opinions provided. Three
hospitals were discussed in
detail.
NR
Training provided to prevent Piloted the checklist at one

teething problems. Instead of the two hospitals first.
of requiring paperwork, they
used in each OR an A3
board (a drawing board
about 14x20 inches) that
was color-coded to aid
completion. Publicity
campaign in both hospitals.
To implement the checklist

effectively, it was essential
to engage all staff to ensure
the theatre team worked
together. Working with
individuals to identify any
gaps or issues with
implementation. Currently it
is being used as standard
throughout theatres
Communication and staff

NR
morale have definitely
improved since the checklist
was implemented.
D-25
Author/Year
Sparkes 2010
Royal Bolton 2010
Training
Study Phases and

Checklist Fidelity

Failure

Behavior
Extensive educational
support and training
3 month pilot, during which

changes to the checklist
were made. After the pilot,
and training, the checklist
was introduced to all
29 ORs in Nov 2009.
Even though people agreed

with the checklist in theory,
it was difficult to change
attitudes and behaviors,
particularly the senior team.
The checklist was required
to be signed by team
members, and This had led
to the fear that legal
colleagues will apportion
blame to those who have
signed the checklist when
complications occur.
Before checklist
NR
introduction: Although all
found the checklist to be
useful, many senior
clinicians felt that such
communication already took
place informally, and that
more paperwork would not
add to safety. Audit of 250
cases in Feb 2010 found
that team briefings occurred
in 77% and time outs in
86%.
Drop-in educational
sessions which involve
120 participants
May and June of 2009 were

spent getting the word out
about plans to start using
the checklist. Piloted first for
one month in two of the
Trusts hospitals in 62
operations. Sept 2009 was
the trust-wide launch of the
checklist. Every Trust is
different but implementing
the checklist across the
trust rather than a
prolonged pilot period.
Within the first week 33% of
operations employed the
checklist. By one month it
was at 72%. Currently all
eight ORs use it.
The importance of
communicating with and
involving people beyond this
core group was recognised
straight away. Essentially
it is all about changing the
culture, which can be a long
process, but its well worth
it.
The feedback we received

from staff was very positive.
Most people were keen to
introduce the checklist as
quickly as possible.
D-26
Health Outcomes
One-month pilot identified

nine potential incidents that
were avoided as a result of
the checklist.
Author/Year
Vats 2010
Kearns 2011
Training
Study Phases and

Checklist Fidelity

Failure

Behavior
Health Outcomes
Limited time given to

training.
Checklist accelerated with

use. Large variability in how
the checklist was used:
sometimes incompletely,
hurried, dismissive replies,
and without some key
participants. Compliance
was initially good, then fell
when the research team
was absent, and so the
team had to re-enter ORs to
encourage greater use.
Compliance ranged from
42% to 80% in the
six month period.
Need a local champion as

well as local organizational
leadership. Importance of
being able to modify to fit
local needs, for example
there was no need to check
pulse oximetry because it is
already used always.
Anesthetists and nurses

were largely supportive.
Some surgeons were not
very enthusiastic. Awkward
self-introductions, takes
time to achieve comfort,
Steep interpersonal
hierarchy, ID the patient
BEFORE draping, not after.
Complaints about
duplication; perhaps a
revised checklist could have
less duplication
At our hospital, we found

no significant change in
overall morbidity or
mortality, which were
already very low, after the
introduction of the checklist.
However, there was a
noticeable improvement in
safety processes such as
timely use of prophylactic
antibiotics, which rose from
57% to 77% of operations
after the checklist was
introduced.
Training, humorous posters

provided, and all staff
empowered to remind the
team to perform the
checklist if it was forgotten.
Compliance with the

preoperative part of the
checklist was 61% after
three months and 80% after
one year. Compliance with
the postoperative part of the
checklist was 68% after
three months and 85% after
one year.
Authors cited four

contributors to success:
allocation of responsibilities,
local champion, sense of
ownership by team
members, and ongoing staff
consultation.
Staff attitudes three months NR

after checklist introduction:
50% now felt familiar with
others in the OR. 70% felt
communication had
improved. 80% felt that in
elective cases the checklist
was useful. 30% felt that in
emergency cases the
checklist was inconvenient.
Fifty-eight patients were
asked whether they noticed
the operating team
performing a series of
checks before the
operation, and 75% said
they did, and another 19%
remembered it after being
prompted. Of the combined
94%, they all disagreed with
the idea that the checks
would make them worried,
and 93% said they were
reassuring.
D-27
Author/Year
Norton 2010
10
Training
Study Phases and

Checklist Fidelity

Failure

Behavior
Health Outcomes
3x5 foot posters in each

OR. Launch involved formal
letter to staff, electronic
training application, multiple
in-service training sessions,
and mention in hospital
newsletter
December 2008 pilot test in

six pediatric surgical
services (general, neuro,
orthopedic, otolaryngology,
plastic surgery, and
urology). Feb 2009 pilot test
on the revised procedures,
and more minor edits were
made. Go-live date
April 1, 2009 in all of the
hospitals ORs. Surgical
chiefs were local
champions, and one nurse
champion was paired with
each surgeon champion.
They divided the
responsibility for leading the
Time Out phase among all
team members, and
identified key speaking
points. Compliance at ORs
improved over time during
this period from July 2009 to
Feb 2010.
Use of the Pediatric

Surgical Safety Checklist
encourages multidisciplinary
teamwork and has brought
increased communication to
our ORs and in other
areas.
Dec 2008 pilot test of

30 procedures had 80-90%
compliance, with
overwhelmingly positive
feedback. Team members
have expressed satisfaction
with the flow and content of
the checklist.
Checklist caught one near

miss during sign in (site not
marked), several near
missed during time out,
(antibiotics not given,
problems with consent
forms, site marking not
visible after draping,
missing equipment), and
sign out (one team realized
a patient needed straight
catheterization, and
reviewing procedure name
helped nurse
documentation, one
specimen was incorrectly
labeled).
Notes: NR=Not reported; Int=Intervention; OR=Operating room; GI=Gastrointestinal; GYN=Gynecology
D-28
Table 3, Chapter 13. Studies of wrong-site-surgery checklists implementing the universal protocol
Author/Year
Description of PSP
Study Design
Garnerin et al.
11
2008
Verification protocol
for checking patient
identity and the site of
surgery
Case series
Theory or Logic
Model
the prevention
of wrong patients
and wrong site
surgery, not to
mention
accountability,
demanded an
intervention aimed
at improving the
way both patient
identity and site of
surgery checks
were performed,
while acquiring the
ability to identify
and correct
deficiencies
D-29
Description of
Organization
Swiss
anaesthesiology
service located
within a 1200 bed
university hospital
Safety Context
Implementation Details
Prior to introduction
of the checklist, all
patients were
required to wear ID
bracelets, and the
operative site had to
be signed by the
surgeon.
Anesthesiologists
were made aware
that they were being
monitored.
Verification protocol developed by an

interdisciplinary team. It required patients
to state their identity, comparing the
statement to the ID bracelet, OR
schedule, and medical record. Similar
types of checks for correct site of surgery.
Nine consecutive months of data were
obtained (October 2003 to June 2004),
and later three subsequent months
(October 2004, March 2005, and October
2005).
Compared to the first three months of
implementation, the next three months
saw better compliance in checking patient
identify (63% up to 81%), complete
compliance with identity checks (10% up
to 38%), proportion of surgical site checks
performed (77% up to 93%), and
complete compliance with surgical site
checks (32% up to 52%). Compliance
was stable in subsequent periods.
Authors attributed the improvements to
increased use of wristbands upon
admission into the OR, the switch from to
using an open-ended questioning format,
and the use of three different sources for
verification.
Barriers included 1) surgeons saying they
already knew that patients or the surgical
site was obvious, and 2) the failure to
develop the protocol with the input of ALL
surgical services
Author/Year
Description of PSP
Study Design
Nilsson et al.
12
2010
Preoperative timeout checklist
Questionnaire
after
implementation
Theory or Logic
Model
None explicitly
stated
D-30
Description of
Organization
Two Swedish
hospitals, bed
sizes not reported
Safety Context
In the autumn of
2007, there were
two incidents of
wrong-side surgery
at these hospitals,
and a root-causes
analysis suggested
that a time-out
procedure might
help. The checklist
was pre-approved
by the heads of the
operating and
anesthesia
departments.
Implementation began in December

2007. Checklist was a shared
responsibility of the OR team. One year
later, a questionnaire was sent to all 704
surgeons, anesthesiologists, operation
nurses, anesthetic nurses, and nurse
assistants, soliciting their opinions about
the new time-out checklist.
Of the 331 responders, 93% felt that the
checklist contributes to increased patient
safety (either without a doubt, or
probably). When asked about eight
specific components of the time-out
checklists, the percentage of respondents
who felt the component was very
important varied widely, from a low of
14% for the introduction of team
members to highs of over 80% for patient
identity, correct procedure, and correct
side. Regarding the sign-out, 91% felt
that the item involving the count of
surgical instruments and sponges was
very important.
Author/Year
Description of PSP
Study Design
Owers et al.
13
2010
Correct site surgery

checklist incorporate
into an existing
preoperative checklist
Case series
Anonymous
11
2007
Checklist to implement Case series

the Universal Protocol,
tailored to this
hospitals preferences
and procedures
Theory or Logic
Model
None explicitly
stated
Description of
Organization
English childrens
hospital, bed size
not reported
Stated that the

checklist provides
cues for staff when
preparing for a
procedure.
Hospital in North
Carolina, bed size
not reported
Notes: NS=Not stated; Int=Intervention
D-31
Safety Context
A preoperative
checklist already
existed at this
facility; they added a
correct site surgery
component
Five people were required to sign the

documentation: marking surgeon,
operating surgeon, ward nurse, scrub
nurse, and anesthetist. Two audit cycles:
once in 2006 (sooner after
implementation) and once in 2008 (two
years later).
Comparing 2008 to 2006, correct
completion of the eight items was not at
all improved for four items (ward nurse
signed, operating surgeon signed, scrub
nurse, signed, and operating department
practitioner signed) but was improved for
the other four (mark site documented, no
mark required documented, entries
legible, and marking surgeon signed).
The lack of documentation, of course,
may not reflect that the new guidance and
processes are not being followed, but
rather that the documentation is regarded
as a low priority part of the process.
Before this checklist, Original checklist in 2005, minor revisions
they were using a
for 2006. Demonstrated the checklist
cumbersome form during educational staff meetings, and
to document their
new staff were given a primer. Staff gave
compliance with the positive comments that they no longer
Universal Protocol.
had to remember everything. The
completed checklist is kept as part of the
medical record.
Table 4, Chapter 13. Studies of anesthesia equipment checklist implementation

Author/Year
Study
Design
Description of
Organization
Study Phases and

Checklist Fidelity
Reasons for Success

or Failure
Opinions, Knowledge
and Behavior
Health Outcomes
Thomassen et al.
14
2010
Case Study
Anaesthesia and
intensive care
department of a
1,100-bed tertiary
teaching hospital
Developed 26-item
checklist after review of
adverse events, PubMed
review of literature, and
expert panel discussions.
Modified Delphi
technique used.
Checklist used on
502 patients.
Emphasized avoiding
checklist fatigue.
Process was supervised
by participating senior
clinician; researchers
were also present.
85 checklists identified
one or more missing
items (17%).
There was a low

compliance (61%) during
the testing period; a few
persons in leading
positions discouraged
use of checklists.
Median checklist
completion time was
88.5 seconds; did not
substantially increase
pre-induction time.
Thomassen et al.
15
2010
Case Study
Anaesthesia and
intensive care
department of a
1,100-bed tertiary
teaching hospital
Follow up study from

15
Thomassen et al. 2010
Checklist improved
confidence in unfamiliar
contexts. It revealed
insufficient equipment
standardization.
Checklists could divert

attention away from the
patient. Senior
consultants were both
skeptical and supportive.
N/A
Focus group interviews

were conducted after
previous study was
completed.
Notes: NS=not stated; Int=Intervention
D-32
References
1.
Sewell M, Adebibe M, Jayakumar P, et al.

Use of the WHO surgical safety checklist in
trauma and orthopaedic patients. Int Orthop
2011 Jun;35(6):897-901. PMID: 20730425
2.
Helmio P, Blomgren K, Takala A, et al.

Towards better patient safety: WHO
Surgical Safety Checklist in
otorhinolaryngology. Clin Otolaryngol 2011
Jun;36(3):242-7. PMID: 21481197
3.
4.
5.
6.
7.
8.
9.
10.
Conley DM, Singer SJ, Edmondson L, et al.

Effective surgical safety checklist
implementation. J Am Coll Surg 2011
May;212(5):873-9. PMID: 21398154
Bell R, Pontin L. How implementing the
surgical safety checklist improved staff
teamwork in theatre. Nurs Times 2010 Mar
30-Apr 5;106(12):12. PMID: 20432772
Bell R. Implementing the surgical safety
checklist at Guys and St Thomas.
[internet]. London (UK): Patient Safety
First. Accessed October 6, 2011. Available:
www.patientsafetyfirst.nhs.uk/Content.aspx?
path=/interventions/CasestudyGuys.
Sparkes D, Rylah B. The World Health
Organization Surgical Safety Checklist. Br J
Hosp Med (Lond) 2010 May;71(5):276-80.
PMID: 20448496
Anonymous. Communicating for success:
Royal Bolton Hospital NHS Foundation
Trust introduces the Surgical Safety
Checklist. J Perioper Pract 2010
Mar;20(3):85-6. PMID: 20642232
Vats A, Vincent CA, Nagpal K, et al.
Practical challenges of introducing WHO
surgical checklist: UK pilot experience.
BMJ 2010 Jan 16;340(7738):133-5. PMID:
20071413
Kearns RJ, Uppal V, Bonner J, et al. The
introduction of a surgical safety checklist in
a tertiary referral obstetric centre. BMJ Qual
Saf 2011 Sep;20(9):818-22. PMID:
21693466
Norton EK, Rangel SJ. Implementing a
pediatric surgical safety checklist in the OR
and beyond. AORN J 2010 Jul;92(1):61-71.
PMID: 20619773
D-33
11.
Garnerin P, Ares M, Huchet A, et al.

Health Care 2008 Dec;17(6):454-8. PMID:
19064662
12.
Nilsson L, Lindberget O, Gupta A, et al.

Implementing a pre-operative checklist to
increase patient safety: a 1-year follow-up of
personnel attitudes. Acta Anaesthesiol
Scand 2010 Feb;54(2):176-82. PMID:
19719811
13.

surgeryAre we up to standard? Cent Eur J
Med 2010 Aug;5(4):407-10.
14.
Thomassen O, Brattebo G, Softeland E, et

al. The effect of a simple checklist on
frequent pre-induction deficiencies. Acta
Anaesthesiol Scand 2010 Nov;54(10):117984. PMID: 21069898
15.
Thomassen O, Brattebo G, Heltne JK, et al.

Checklists in the operating room: help or
hurdle? A qualitative study on health
workers experiences. BMC Health Serv Res
2010;10:342. PMID: 21171967
16.
Surgical safety checklist. Geneva: World

Health Organization (WHO); 2009 Jan. 1 p.
Available at:
http://whqlibdoc.who.int/publications/2009/
9789241598590_eng_Checklist.pdf.
17.

guidelines for safe surgery. Geneva: World
Health Organization (WHO); 2008.
18.
Haynes AB, Weiser TG, Berry WR, et al. A

J Med 2009;360(5):491-9. Available:
Msa0810119. PMID: 19144931.
19.
de Vries EN, Prins HA, Crolla RM, et al.

Effect of a comprehensive surgical safety
system on patient outcomes. N Engl J Med
2010 Nov 11;363(20):1928-37. PMID:
21067384
20.
21.
22.
de Vries EN, Hollmann MW, Smorenburg

SM, et al. Development and validation of
the SURgical PAtient Safety System
(SURPASS) checklist. Qual Saf Health Care
2009 Apr;18(2):121-6. PMID: 19342526
de Vries EN, Dijkstra L, Smorenburg SM, et
al. The SURgical PAtient Safety System
(SURPASS) checklist optimizes timing of
antibiotic prophylaxis. Patient Saf Surg 2010
Apr 13;4(1):6. PMID: 20388204
AM, et al. Prevention of surgical
safety checklist. Ann Surg 2011
Mar;253(3):624-8. PMID: 21209590
23.
Greenberg CC, Regenbogen SE, Studdert

DM, et al. Patterns of communication
breakdowns resulting in injury to surgical
patients. J Am Coll Surg 2007
Apr;204(4):533-40. PMID: 173822111
24.
Griffen FD, Stephens LS, Alexander JB, et

al. The American College of Surgeons
closed claims study: new insights for
improving care. J Am Coll Surg 2007
Apr;204(4):561-9. PMID: 17382214
25.
Patient safety tip of the week: SURPASS:

the mother of all checklists. Lewiston (NY):
The Truax Group. Accessed November 30,
2010. Available at:
www.patientsafetysolutions.com/docs/Nove
mber_30_2010_SURPASS_The_Mother_of
_All_Checklists.htm.
26.
Windsor JA, Petrov MS. Patient safety in

medicine: are surgeons ready for checklists.
ANZ J Surg 2010 Jan;80(1-2):3-5. PMID:
20575869
27.
28.
Facts about the universal protocol.

Commission; 2004. Accessed January 9,
2008. Available:
www.jointcommission.org/assets/1/18/Univ
ersal%20Protocol%201%204%20111.PDF.
Mulloy DF, Hughes RG. Chapter 36.
Wrong-site surgery: a preventable medical
error. In: Patient safety and quality: an
evidence-based handbook for nurses. AHRQ
Publication No. 08-0043. Rockville (MD):
(AHRQ). Accessed April 2008. Available:
www.ahrq.gov/qual/nurseshdbk/docs/Mullo
yD_WSS.pdf.
D-34
29.
Using surgical checklists to comply with the

universal protocol. Jt Comm Perspect
Patient Saf 2011 Jul;11(7):9-11.
30.
Joint Commission field tests changes to

2009 safety goals, Universal Protocol. OR
Manager 2008 Mar;24(3):32.
31.
Watson DS. Implementing the universal

protocol. AORN J 2009 Aug;90(2):283-7.
PMID: 19664421
32.
Catalano K. The unveiling of the Joint

Commissions National Patient Safety Goals
for 2010. Plast Surg Nurs 2010 JanMar;30(1):44-9. PMID: 20220501
33.
Follow-up Survey on Universal Protocol.

Oakbrook Terrace (IL): The Joint
Commission; December 14, 2010. Accessed
October 18, 2011. Available:
www.jointcommission.org/followup_survey_universal_protocol.
34.
National patient safety goals effective July

1, 2011. Office-based surgery accreditation
program. Oakbrook Terrace (IL): The Joint
Commission; 2011. Also available:
www.jointcommission.org/assets/1/6/NPSG
_EPs_Scoring_OBS_20110707.pdf.
35.
Possible integration of actions to satisfy

Joint Commission universal protocol and
WHO surgical safety checklist. Harrisburg
(PA): Pennsylvania Patient Safety
Authority; 2011 Accessed November 21,
2011. Available: patientsafetyauthority.org/
EducationalTools/PatientSafetyTools/PWSS
/Documents/crosswalk.pdf.
36.
Merry AF. Role of anesthesiologists in

WHO safe surgery programs. Int
Anesthesiol Clin 2010 Spring;48(2):137-50.
PMID: 20386233
37.
International standards for a safe practice of

anaesthesia 2010. London (UK): World
Federation of Societies of
Anaesthesiologists; 2010 Accessed
November 11, 2011. Available:
www.anaesthesiologists.org/guidelines/pract
ice/2008-international-standards-for-a-safepractice-of-anaesthesia.
38.

(IL): American Society of Anesthesiologists;
2008.
39.
Wicker P, Smith B. Checking anaesthetic

machine. J Perioper Pract 2008
Aug;18(8):354-9. PMID: 18751494
Evidence Tables for Chapter 14. Use of Report Cards and

Outcome Measurements To Improve the Safety of Surgical
Care: American College of Surgeons National Quality
Improvement Program (NEW)
This review had no additional evidence tables. There is one evidence table in the text.
Evidence Tables for Chapter 15. Prevention of Surgical Items

Being Left Inside A Patient: Brief Update Review
This review had no additional evidence tables.
Evidence Tables for Chapter 16. Operating Room Integration

and Display Systems: Brief Review (NEW)
Evidence Tables for Chapter 17. Use of Beta Blockers To

Prevent Perioperative Cardiac Events: Brief Update Review
Evidence Tables for Chapter 18. Use of Real-Time Ultrasound

Guidance During Central Line Insertion: Brief Update Review
D-35
Evidence Tables for Chapter 19. Preventing In-Facility Falls

This topic modifies an evidence table from an existing systematic review, and therefore some of the columns and entries are
different than our normal format.
Table 1, Chapter 19. Evidence table adapted from Oliver and colleagues
Assessment/
Intervention
Performed
By
Research
Staff
Discipline
involved in
intervention
Nursing
Ward Staff
Nursing
References
1
Ang et al, 2011 *
Study Design
RCT
Setting
8 medical wards in an acute
care hospital in Singapore
Barker et al,
2
2009
Before-and-after
study
Small acute hospital in Australia
Barry et al,
3
2001
Before-and-after
study
Small long-stay and

rehabilitation hospital in Ireland
All patients admitted to 95 beds for

1 year preintervention and 2 years
postintervention
Yes/Local
Ward Staff
Multi
Before-and-after
study
An acute hospital in Australia

(including pediatric wards)
All patients admitted to 500 beds

for 1 year preintervention and
second year postintervention (no
data provided for first year of
intervention)
No/No
Ward Staff
Nursing
Cluster RCT
24 acute and rehabilitation

elderly care wards in 12
Australian hospitals
3999 patients admitted during the

3-month study period on each ward
Yes/No
Research
Staff/Ward
Staff
Nursing and
Physiotherapy
Brandis, 1999
Cumming et al,
5
2008
Participants
1822 newly admitted patients who
were age 21 or older, and scored 5
or above on fall risk model were
randomized.
271,095 patients admitted over 3
years before, and 6 years after
intervention
Individualized/
Use of Risk
Score
Yes/Local
* New studies added from update search

1
While based on STRATIFY, extensive changes were made.
D-36
Yes/Local
Individualized/
Use of Risk
Score
Yes/Local
Assessment/
Intervention
Performed
By
Research
Staff/ Ward
Staff
Discipline
involved in
intervention
Multi
References
Dykes et al,
6
2010 *
Study Design
Cluster RCT
Setting
8 medical units in 4 urban
United States hospitals
Participants
All patients admitted or transferred
to units over 6 month study period
Fonda et al,
7
2006
Before-and-after
study
Four elderly acute and

rehabilitation wards in an
Australian acute hospital
All admitted patients (1905 before,

2056 after) over 1 year before, 2
years after
Yes/Local
Ward Staff
Multi
GrenierSennelier et al,
8
2002
Before-and-after
study
A 400-bed rehabilitation hospital

in France
All admitted patients over 2 years

before and 2 years after (ca 800
admissions per year)
Yes/Local
Ward Staff
Nursing
Haines et al,
9
2004
RCT
Three subacute wards within an

Australian rehabilitation and
elderly care hospital
626 patients consenting to

randomization drawn from 1040
consecutive admissions
Yes/No
Ward Staff/
Research
Staff
Physiotherpay
and
Occupational
Therapy
Healey et al,
10
2004
Cluster RCT
Four acute and 4 rehabilitation

wards in one acute and 2
rehabilitation hospitals in the UK
All admitted patients over 1 year

(3386 patients)
Yes/No
Ward Staff
Multi
Koh et al, 2009
Cluster RCT
Two acute hospitals in

Singapore
All admissions during 1 year before

and 6 months after
Yes/Local
Ward Staff
Nursing
Krauss et al,
12
2008
Before-and-after
study with
contemporaneous
cohort
General medical wards in an

acute academic hospital
All admissions during 9 months

before and 9months after period (N
not given)
Yes/No
Ward Staff
Nursing
11
D-37
References
Mitchell et al,
13
1996 *
Study Design
Before-and-after
Oliver et al,
14
2002
Before-and-after
study measure
Schwendimann
15
et al, 2006
Individualized/
Use of Risk
Score
Yes/Local
Assessment/
Intervention
Performed
By
Ward Staff
Discipline
involved in
intervention
Nursing
Setting
The intervention took place in a
32 bed medical ward serving
both acute and subacute
patients with high acuity needs,
which was compared to fall
rates in the entire 225 bed
acute care teaching hospital in
Australia prior to the pilot.
An elderly medical unit within an
acute hospital in England
Participants
All admissions in the hospital for 6
months prior to the pilot compared
to 6 months in the pilot ward after
implementation.
3200 patients admitted annually;

data collected for 1 year
preintervention and 1 year
postintervention
Yes/STRATIFY
Ward Staff
Multi
Before-and-and
after study
Internal medicine, geriatric and

surgical wards in a 300-bed
Swiss acute hospital
All admissions (34,972) over an 18month before and 42-month after

period
Yes/Local
Ward Staff
Multi
Stenvall et al,
16
2007
RCT
Orthogeriatric ward
(intervention) and orthopedic
ward and geriatric ward
(control) in a Swedish acute
hospital
199 consecutively admitted

patients with femoral neck fracture
2
consenting to randomization and
without complex needs
Yes/No
Research and
ward staff
Multi
Uden et al,
17
1999
Before-andafterstudy
A geriatric department in an
acute hospital in Sweden
47 randomly selected patients from

the year before intervention, all 332
admitted patients in the intervention
year
Yes/No
Ward Staff
Nursing
There were apparently no ward capacity issues as there is no mention of any patients not being admitted to the ward to which they were randomized.
D-38
References
Van der Helm et
318
al, 2006
Study Design
Before-and-after
study
Vassallo et al,
19
2004
Cohort Study
Von RentelnKruse and

20
Krause, 2007
Before-and-after
study
Williams et al,
21
2007 *
Before-and-after
study
Individualized/
Use of Risk
Score
No/Local
Assessment/
Intervention
Performed
By
Ward Staff
Discipline
involved in
intervention
Nursing
Setting
One internal medicine ward and
one neurology ward within an
acute hospital in the
Netherlands
Participants
All admitted patients (2670) during
a 6-month before and 18-month
after period
Three rehabilitation wards

within a UK rehabilitation
hospital
825 patients (the first 275 patients

to be admitted to each of the 2
control and 1 intervention wards)
Yes/Downton
Ward Staff
Multi
Elderly acute and rehabilitation

wards in an acute hospital in
Germany
4272 patients admitted in a 235

months before period, 2982
admitted in a 16-month after
period
Yes/STRATIFY
Ward Staff
Multi
Three medical wards (72 beds

1,357 patients admitted during the
Yes/Local
Ward Staff
Multi
total) and a 17 bed geriatric
6 month study period were
evaluation management unit in
compared to aggregate hospital fall
a 755 bed metropolitan tertiary
rates over the same period a year
care teaching hospital in
earlier.
Australia
Reprinted from Clin Geriatr Med. 26(4), Oliver D, Healey F, Haines TP., Preventing falls and fall-related injuries in hospitals, 645-92, 2009 with permission from Elsevier.
Note that the investigators describe the before period as a pilot study, but actually appear to be describing the falls rate and practice before the intervention,
that is, a baseline rather than the piloting of the intervention.
4
Although the investigators refer to the study as quasi-randomized and Oliver et al (2007)1 refer to it as a cluster RCT, it appears the intervention ward was
selected (not randomized) on the basis of being the ward where the researchers worked, and the quasi-randomization relates only to the fact that patients would
be allocated from a waiting list to whichever ward was the first to have an empty bed. The study also refers to matching patients, but this appears to be comparison
of the cohorts for differences rather than matching at individual patient level.
5
A separate publication (von Renteln-Kruse and Krause, 2004) describes a review of reported falls from January 2000 to December 2002 when 5946 patients
were admitted of whom 1015 were fallers and who had 1596 falls. This suggests that the proportion of fallers had been reducing substantially year-on-year
even before the intervention was introduced (ie, 17% [1015/5946] of patients fell before intervention in 20002002, 14% [611/4272] of patients fell before intervention
in 20032004, and 11% [330/2982] of patients fell after intervention in 2005-early 2006).
D-39
Table 2, Chapter 19. Stenvall et al. 2007, Main content of the postoperative program and differences between the two groups:
20
teamwork
Teamwork
Intervention group
Team included registered nurses (RN), licensed
practical nurses (LPN), physiotherapists (PT),
occupational therapists (OT), dietician, and
geriatricians.
Close cooperation between orthopedic surgeons and
geriatricians in the medical care of the patients
Control group
No corresponding teamwork at the
orthopedic unit.
D-40
Table 3, Chapter 19. Implementation studies evidence table

Author/
Year
Browne et
22
al., 2004
Description of Fall
program
A new tool, the
ADAPT Fall
Assessment Tool,
was developed,
piloted, and
implemented as a
redesign for the
existing fall
prevention
program. The tool
automatically
calculates a fall risk
score from nurse
shift assessments
and produces a
score and
categorical
recommendation.
The 4 categories
were disorientation,
activity,
postmedication,
and toileting
precautions, and
each had a
corresponding
protocol and
suggested
interventions of
care.
Study Design
Theory or Logic
Model?
Descriptive with Redesign process
summative
looked for current
evaluation
recommendations in
the literature for fall
N=6402 inpatient risk factors. This
and observation included 4 authors.
records reviewed The tool of one of
from all adult
these authors,
medical-surgical Hendrich, was used
to validate the ADAPT
units, all
tool.
intensive care,
rehabilitation,
skilled nursing,
and psychiatric
units.
Description of
Organization
The Methodist
Healthcare System
(MHS) of San Antonio
used the Meditech
Clinical Documentation
Module for electronic
health records. This
system includes 7
inpatient facilities that
deliver full pediatric,
adult, rehabilitation,
maternal-child, and
psychiatric services.
Implementation Themes focus on

Additional
association with effectiveness
themes
Context: 7 years of effort had failed to
produce appreciable decreases in falls
on injury. A fall committee identified
reasons that might undermine fall
prevention efforts.
Missed partial or incorrect

documentation of fall events
(missed opportunities)
Overidentification of fall risk

patients with 60% of case plans
listing a fall risk problem.
Committee Goals: To develop: a
computerized documentation to promote
reassessment; an evidence-based risk
assessment tool; a tailored intervention
program; and a system for integrating
information into documentation and
communication.
Considerations:
Complete, consistent, and accurate fall
risk reassessment by nursing staff was
essential to success of the project.
Pilot: Once the tool was developed, it
was piloted and validated. The results
were presented to the MHS Falls
Committee, who gave permission for
automated implementation system-wide.
Education: fall and restraint fairs at the
time of implementation served to educate
nurses about the redesigned program.
Iterative change:
Nurse dissatisfaction with fall

risk appearing in a list of acute
care problems led to ongoing
evaluation of where best to
D-41
Comments
Fall assessment
documentation
compliance on
admission and
daily increased to
100% for all units
in all hospitals.
Fall rates
decreased from
3.41 to 3.21 per
1000 adjusted
patient days.
Injuries per 100
falls decreased
from 1.44 to 0.95.
Author/
Year
Description of Fall Study Design

program
Capan et al., A new fall risk

23
assessment tool
2007
was developed to
evaluate 7 risk
factors every 12
hours for all
patients. All
significant risk
patients received a
wrist band, door
sign, written guide,
hip protectors, and
orthostatic
hypotension
assessment.
Specific risk
factors(unsteady
gait, disorientation,
toileting issues,
medication issues),
have additional
tailored
interventions.
Time series
design
No sample size
given
Theory or Logic
Model?
a literature search
looking at best
practices and
reviewing existing fall
risk assessment
tools
Description of
Organization

Additional
themes
place the fall risk items.
Certain units with specific

issues (short-stay areas with
less routine review, specialty
units with at-risk and actual fall
discrepancies) have resulted in
special work teams tasked with
customizing fall prevention
program to their unique
challenges.
One size fits all where all

patients get the same set of
interventions
Franklin Square
There was an external pressure to
Hospital Center , a 357 improve, since this hospital had higher
bed acute care hospital fall rate than benchmark from the
in Baltimore, MD, is
Maryland Hospital Assoc. Quality
part of the MedStar
Indicator Project (MHA QI). The existing
Health System, which fall risk tool was not identifying high risk
is a community-based patients.
network of 7 hospitals
in the BaltimoreA root cause analysis of one years data
Washington Area.
found that, as opposed to the prior
assumption that most fallers were
confused, 70% of fallers were not
confused. This meant the hospitals
existing falls risk assessment tool was
not identifying the majority of the patients
who fell.
The intervention and implementation
were guided by a multidisciplinary team.
A pilot test was performed in a unit with
high fall rate and readiness for change
indication. Staff was involved in choosing
equipment. An internal financial incentive
was used; a contest for gift cards was
introduced for the first 25 staff
documenting a prevented fall. Unit
champions were considered key to the
acceptance and needed to be passionate
mentors; when nurses were only partially
Additionally, care
coordination rounds
had an
interdisciplinary
team meet to
D-42
Comments
In pilot, the fall rate

declined from 1.17
to 0.45 per 100
patient days over a
year.
In full
implementation,
the rate dropped
from 0.45 to 0.32
per 100 patient
days, below the
benchmark target
of 0.35. Severity of
injury has also
declined, and
declines have
continued, with the
fall rate cut in half
over two years.
Author/
Year

program
discuss total plan of
care for each
patient on the unit.
Theory or Logic
Model?
Description of
Organization
Dempsey,
24
2004
A new injury risk

assessment form
was used to match
individual risk
factors to
interventions,
educational
materials, and
illuminated graphics
at patients
bedside.
Pre-post study of
implementation in
1995-1996,
follow up
assessment in
2001.
The falls intervention

was devised using
the literature and the
collective experience
of the clinicians.
No other details are
given.
A regional teaching
hospital
In Australia.
Gutierrez,
25
2008
A specific specialty
adult focused
environment
(SAFE) unit as a
part of the definitive
observation unit.
The SAFE unit had
3 rooms with 2
beds each, staffed
by 2 RNs and 1
technical partner.
Fall protocol order
sets, post fall order
Time series
design
Total number of
patients not
responding
A literature review
was performed to
identify potentially
promising
interventions. Values
of physicians, and
nursing staff were
solicited to assess the
potential intervention
components.
Scripps
Mercy Hospital in San
Diego California
No other information
provided
D-43

Additional
themes
using the tool, champions analysis of fall
incidents helped build value and falls rate
declined.
Expansion: Initially planned for one unit
at a time, increasing fall rates led to
immediate hospital-wide implementation.
Extensive education efforts included inservicing and scheduled classes, with
95% of staff completing education prior
to implementation.
Compliance with the intervention was
monitored. Compliance was 88% and
was on a downward trend at the 2001
assessment (no data given).
No change in staffing, but occupancy
rates rose over time and could be related
to decline of effectiveness.
No significant differences in case mix.
1.
Assess values of staff and available

resources.
2. Identify clinical champions
3. Develop an Elevator Speech to
motivate nurses.
Our project goal is to improve the
patient care quality by preventing
inpatient falls.
Our patient population is more
educated about healthcare quality
and is seeking the highest-quality
care available.
Nurses play a primary role in
A possible reason
for the increase in
falls was
increased
reporting and not
an increase in
falls.
Comments
After an initial
reduction in falls, in
1995-1996,
beginning in 1998
falls reporting
began to increase
until they exceeded
pre 1995 levels.
The researcher
concluded that
falling compliance
associated with
increased
occupancy was
partly causative for
the decline in
effectiveness of the
program.
This project found
a lower rate of falls
(p-37/1000 patient
days) 3-6 months
after the
intervention
compared to the 9
months prior (3.0,
4.18, and 4.87
falls/1000 patient
days).
Author/
Year

program
sets, quiet zones,
use of recliners in
the hallways, low
beds with internal
alarms, keeping
doors open and
curtains back, and
nurses use of
portable computers
for documentation
within sight of
patients.
Theory or Logic
Model?
Description of
Organization
D-44

Additional
themes
preventing falls.
We want to be able to advise to the
public that we have the highestquality nursing care available in
California; to this end, we must
reduce patient falls.
Historically, the DOU floor has
exceeded minimum acceptable fall
occurrence standards as
benchmarked by CalNOC.
If provided enough resources and
staff and nursing is practiced
according to evidence, we can likely
minimize falls and the related
negative outcomes.
We wish to prove that we can
reduce our fall rates by eliminating
practice barriers in our existing
nursing-centered multidisciplinary
fall prevention plan.
Our project goal is to identify and
eliminate practice barriers within our
existing evidence-based fall
prevention protocol, improve its
effectiveness, and thereby reduce
falls and improve our quality of
patient care.
We think that we can reduce our fall
rates dramatically by being more
vigilant about a good fall prevention
plan; for instance, toileting our highrisk patients per protocol.
I know it sounds simple, but these
strategies have been used in other
hospitals and they are known to
work.
4. One champion for day and one for
night shift came to ensure
compliance with protocol for a total
of 192 hours.
5. Champions=Change, the belief
that champions change not only the
Comments
Author/
Year
Kolin et al.,
26
2010

program
Theory or Logic
Model?
Description of
Organization

Additional
themes
practice but also the culture.
6. The process is slow.
7. Leadership support, staff
involvement, time, money, and
energy are needed.
8. A no blame culture for fall
reporting.
Fall risk
Time series; data A fall literature review University of Pittsburgh UPMC leadership formed a system-wide
assessments were presented on fall was conducted on
team, including expert members and the
Medical Center
completed on
rates and injury
multiple databases
(UPMC) has 19 acute paper authors, to prioritize falls, identify
admission, at least rates for a year
best practices, compare UPMC strengths
(CINAHL, Medline,
care facilities in
every 24 hours, and preceding
Cochrane) and
Western Pennsylvania. and weaknesses, and determine a model
after certain trigger implementation, categorized into
for implementation. The team had regular
ongoing meetings beyond the duration of
events. At-risk
as well as for the levels of evidence.
the project.
patients receive
intervention year Evidence was then
visual identification in the UPMC
synthesized to
Data on the overall fall and injury rates
(arm band, door
system overall,
determine
were collected and compared to
sign, etc.)
as well as for one components for
benchmarks, which was then presented
specific units.
inclusion in a
A new tool was
to leadership.
multifactorial
developed and
intervention.
implemented.
A survey was then taken at each facility,
Depending on the
which revealed variability of compliance
number of
with risk reassessment, type of post-fall
questions on which
follow-up for, and patient assessment
a patient screened
form.
positive, levels of
Team members participated in falls
interventions were
education, and then held a rapid
applied.
improvement event, where experts were
divided amongst groups to address 5
Lightning Rounds,
specific issues: assessment and
which focused on a
reassessment, prevention equipment and
vital few patients,
interventions, hospital environment, staff
were implemented
and patient/family education, and posthourly.
fall follow-up. The first group tested and
compared different tools in a
A standard post-fall
convenience sample before developing
form was adopted.
their own assessment tool.
New patient
educational
A comprehensive education for the
materials were
nursing staff was provided before the
developed, and
implementation of the new tool.
D-45
Comments
Author/
Year
McCollam,
27
1995

program
environmental
modification
recommendations
were proposed.
The Morse Fall

Scale (MFS) was
adopted.
Nursing staff were
trained using a
video and
instructions for
scoring the scale.
Their
understanding was
then checked using
an evaluation.
Theory or Logic
Model?
Description of
Organization
Descriptive
quantitative
a careful review of
Veterans Affairs
research-based falls Medical Center,
literaturefound only Portland, Oregon
Data provided on one falls assessment
fall rates,
instrument that met
compliance and [our criteria]. The
tool reliability
identified scale is the
MFS.

Not all facilities use electronic medical
records, so roll-out was staggered
between those with and without EMRs.
Those with EMRs began working to
connect the records to the event
reporting system.
Research in Practice Committee led
effort.
The MFS was pilot for 3 months on the
hospitals 40-bed Cardiology General
Medicine Unit to determine if: 1. Patients
were accurately identified as at risk; 2.
Nurses could use it reliably; 3. MFS was
practical for routine clinical use. At the
end of data collection, a staff nurse
survey was used to evaluate aims 2 and
3.
Before this, near falls had not been part
of the reporting.
Patients scoring 45
or above received
nursing
implementations.
Problems identified during the pilot

included inconsistent and incomplete
reassessment, identification of secondary
diagnoses, and score consistencies
between shifts. Cut-off score was
adjusted from 45 to 55.
Full implementation included approval
from Nursing Administration, inclusion of
the scale in admission forms, and staff
education.
Although instrument completion
compliance ranged from 75 to 85%, care
plans or interventions for fall prevention
were only in the 50-58% range. This
could be due to a lack of knowledge or
skepticism about the program.
Strategies needed to maintain MFS use,
D-46
Additional
themes
Comments
In the year after

MFS
implementation,
reported falls had
risen 24%, and
serious injuries had
decreased 175%.
Author/
Year
Neily, 2005
OConnell,
29
2001
28

program
Theory or Logic
Model?
Description of
Organization
Collaborative
breakthrough series
(BTS); the
intervention
includes signs to
identify high risk
patients, toileting
interventions, use
of hip pads,
environmental
rounds, staff
education, and
post-fall
assessment.
Pre-post study
with summary
evaluation
exploring the
influence of
context on
effectiveness.
Number of
patients not
reported.
The intervention
implementation was
based on the
collaborative
breakthrough series.
32 Veterans Affairs
facilities (a mix of
acute and long term
care facilities). State
veterans homes and
one private long term
care facility.
Assess fall risk

using standardized
scale; patients at
high risk of falling
identified with
stickers and
wristbands;
standard fall
prevention
measures could be
implemented for
this group of
patients
Pre-post test with Literature review to

summative
assess potentially
evaluation
effective
interventions. No
Study sample N- additional
1065 patients, 2 specification.
wards in an acute
care hospital.
Acute care hospital in

Australia; no details
except mean length of
stay = 34 days
No other patient
data provided
D-47

strengthen interventions for at-risk
patients, and assigning responsibilities
for follow-up program monitoring and
evaluation.
In 4 sites where the intervention was
spread, leadership support was cited as
one of the strongest factors for continued
change. Root cause analysis and a
multidisciplinary approach were also
cited as important risk factors.
Leadership support, teamwork skills
correlated with one-year high team
performance.
At the one year follow up, high
performing sites, compared to low
performing sites, reported higher
agreement with questions about the
presence of useful information systems,
the sites gained and exchanged overall
value, teamwork skills, and leadership
support.
The authors themselves identified these
themes.
Confounding Contextual Issues: Hiring
freeze during the middle of the study
period meant staff vacancies could not
be filled.
Concurrent implementation of a program
to increase physical activity led to
feelings by staff of being overwhelmed by
the requirements of two projects, and lost
motivation. Implied was the notion that a
project driven by middle management
would receive less support.
Concurrent implementation of another
falls prevention program by the
occupational therapy department may
also have contributed to confusion.
Difficulties with the fall prevention
program:
The risk assessment instrument
identified about 75% of patients as high
Additional
themes
Comments
The primary
effectiveness
assessment of the
intervention was a
decrease in major
injury rate of 62%.
Methodological
barriers: Initial
attempt to design
evaluation as
RCT, then
controlled beforeand-after, left
evaluation team
discouraged that
pre-post study
was the only
feasible design.
In this study, no
statistically
significant benefit
of the program was
observed.
Author/
Year

program
Rauch et al., The Schmid Risk

30
2009
Assessment tool
was used to identify
at-risk patients.
Depending on
specific risk factors,
multiple
interventions were
specified, including
a general
intervention and
interventions
tailored for specific
risk factors like
medications or
altered mobility.
Time series, data

provided about
fall rates and
compliance
Theory or Logic
Model?
Description of
Organization
Ishikawa case and

effect chart and root
cause analysis
process
University Medical
Center at Princeton
Plan-Do-Study-Act
(PDSA) performance
improvement model
was used throughout
the implementation
process.

Additional
themes
risk. Consequently most patients had
stickers and bracelets. But 70% of falls
occurred in patients who were not
classified as high risk. This may have
undermined staff confidence in the
intervention and that the program lost
some of its significance. No or limited
ability to audit whether standard fall
prevention measures were being done.
Some staff said they were already doing
everything to prevent falls and this new
program did not add anything new to this.
The project began with a current practice
evaluation based on the Ishikawa
methods, and uncovered improvement
opportunities including communication,
care-planning and assessment,
equipment, education, process and
staffing.
Leadership hired the Hill-Rom Clinical
Excellence team as an outside
consultant with experience and expertise.
All levels of leadership were engaged
and accepted ownership of the process.
it is imperative to obtain frontline staff
input and feedback to ensure that
successful change management occurs
in the clinical arena.
Visual identifiers
were used in the
general
intervention, with a
daily list of at-risk
patients, arm
bands, and door
signs.
Policy was reviewed and rewritten to

include specific intervention components,
including a valid assessment tool,
assessment frequency, etc.
A multidisciplinary fall team including
managers and frontline staff identified the
Schmid Risk Assessment Tool for use in
the intervention.
A postfall protocol
was developed and
introduced.
The tool was first piloted in a unit with

high fall risk and willing staff. Originally
D-48
Comments
The rate of falls

with injury in the
pilot unit
decreased from
43% to 14% over
the year.
Staff compliance is
steadily improving.
Author/
Year

program
Theory or Logic
Model?
Description of
Organization

planned for 30 days, the pilot was
extended another 30 days to incorporate
changes and solidify the process before
full roll-out. Significant changes were
made, including activity distribution
between shifts, additional staffing, and
ongoing education and communication.
Additional
themes
Comments
Nurses found
filling out falls
incident report
forms
troublesome.
In this study, no
statistically
significant benefit
of the program was
observed.
Weekly teleconferences between

consultants and key hospital members,
as well as monthly fall team meetings
support the ongoing status of the
implementation.
After 8 weeks of fine tuning, there was an
incremental roll out in the rest of the
hospital.
Routine monitoring of staff compliance
and understanding was measured using
the GAP analysis tool.
SeminGoossens,
31
2003
A guideline
developed by an
internal project
team of 11, with 4
nurses from each
ward, that focused
on identifying
patients, at
increased risk on
the basis of 3 main
risk factors and
then for patients at
increased risk
doing one or more
of: moving bed to
lowest position;
raising side rails;
noting the
increased risk in
Longitudinal time
series study
sample N=2670
patients. No
other patient data
provided.
The intervention used

Grols 5 step
implementation
model:1) develop and
change protocol; 2)
identify obstacles to
change; 3) link
intervention to
obstacles; 4) develop
and plan; and 5)
evaluate the process
The implementation
was a bottom up
approach with input
from ward nurses at
every step of the way
and attention paid to
attractiveness of the
educational materials
Acute care hospitals in

the Netherlands, 2
voluntary cooperating
wards. A 32 bed
neurology ward with 33
nurses and 850
admissions/ year. A 32
bed internal medicine
ward with 34 nurses
and 1500
admissions/year.
Overall, the hospital
has 1000 beds and is a
teaching hospital. The
motive for the
intervention was the
high number of falls
reported to the Incident
Reporting Committee.
D-49
The program was well received by the

staff.
The investigators judged nurses may
have been resistant to the idea that falls
could be predicted and prevented. On
the neurology ward, nurses stated it was
simply impossible to prevent falls.
Falling was considered to be an
inevitable part of aging. These feelings
of helplessness did not change during
the intervention. The authors speculate
that an important aspect of success is
changing the attitude of nurses. The
authors conclude with three things they
would do differently:
Get more buy-in from floor nurses

and not just the head nurse
Assess the prior experience with

implementation of practice
guidelines
Author/
Year
Weinberg et
32
al., 2011

program
the nursing file;
informing patients
and relatives about
the measures;
putting the call bell
within reach; and a
restraining waist
belt could be used.
Additionally, efforts
were made to
reduce
environmental
hazards.
Fall prevention
initiative (FPI)
included:
1. Monthly Fall
reviews were
attended by unit
managers, staff
involved in patient
care, and the FPI
co-chairs.
2. Patient care staff
and managers were
made more
accountable for
breaches, and a fall
index report by unit,
as well as daily
rounds, were
instituted.
3. Policy changes
included:
Formalized use of
bed alarms;
improved fall
documentation;
medication
restrictions; fall risk
Time series
design.
All beds were
included in
analysis, 714
beds in hospital.
Theory or Logic
Model?
and feedback on fall
rates. An
organizational plan to
use stickers to identify
high risk patients was
abandoned because
nurses judged them
stigmatizing with little
evidence of
effectiveness.
Description of
Organization
A previous fall
prevention program
failed. The belief was
that the
implementation was
simplistic.

Get more organizational buy-in in

order to create and environment in
which it is easier to implement
change. Attempt to involve medical
chiefs and nurse managers.
The FPI was related

to adaptive and
business
management models
used in industries
that cannot permit
failures. These
models institutionalize
continuous quality
improvement and
evidence-based
strategies for
implementing cultural
change through
modification of
system failures,
leadership support,
communication, clear
goals for each
member, lateral
accountability and
cooperation, and
correction of system
failures.
Staten Island
University Hospital has
two campuses and 714
beds. Services include
medical/surgical.
pediatric, maternity,
behavioral sciences
and physical
rehabilitation.
Adaptive challenges, including poor

institutional prioritization and poor
compliance with existing protocols, were
identified. Prior to the FPI, reactions to
falls rates included policy and procedure
changes that failed to reduce incidence.
Two events provided motivation for the
intervention: the highest recorded fall
rate at the hospital and the introduction
of fall prevention as a National Patient
Safety Goal.
Hospital leadership initiated the effort
and prioritized falls, forming a
multidisciplinary hospital falls committee
to review fall-related policy breaches.
Committee attendance was mandated.
FPI provided forum for staff to define
and solve problems, encouraged
collaborations between units, and the
sharing of best practices.
FPI co-chairs evaluated cultural factors,
and found that although existing
protocols followed best practices, low
prioritization of falls, superficial fall
analysis, and lack of accountability all
The normal accident

theory and highreliability theory,
which emphasize
D-50
Additional
themes
Comments
After four years,

yearly inpatient fall
rates decreased by
63.9% (p<.0001).
Documentation of
injury level
increased and
minor and
moderate fallrelated injuries
decreased, all
statistically
significant.
Author/
Year

program
and postfall
assessments.
Theory or Logic
Model?
documentation and
the role of a just
culture, were also
utilized.
Description of
Organization

decreased protocol success.
Initial reviews revealed partial or
superficial compliance, highlighting
compliance as a main issue for effective
prevention.
Most protocols and policies stayed from
before FPI, the biggest change was
culture. as the initiative processed, the
culture of the hospital appeared to
change to one in which, rather than being
burdensome, fall prevention engendered
pride and enthusiasm
D-51
Additional
themes
Comments
References
1.
2.
3.
4.
5.
6.
Ang E, Mordiffi SZ, Wong HB. Evaluating

the use of a targeted multiple intervention
strategy in reducing patient falls in an acute
care hospital: a randomized controlled trial.
J Adv Nurs. 2011 Sep;67(9):1984-92. PMID
21507049.
Barker A, Kamar J, Morton A, et al.
Bridging the gap between research and
practice: review of a targeted hospital
inpatient fall prevention programme. Qual
Saf Health Care. 2009 Dec;18(6):467-72.
PMID 19955459.
Barry E, Laffoy M, Matthews E, et al.
Preventing accidental falls among older
people in long stay units. Ir Med J. 2001
Jun;94(6):172, 4-6. PMID 11495234.
Brandis S. A collaborative occupational
therapy and nursing approach to falls
prevention in hospital inpatients. J Qual Clin
Pract. 1999 Dec;19(4):215-20. PMID
10619149.
Cumming RG, Sherrington C, Lord SR, et
al. Cluster randomised trial of a targeted
multifactorial intervention to prevent falls
among older people in hospital. BMJ. 2008
Apr 5;336(7647):758-60. PMID 18332052.
Dykes PC, Carroll DL, Hurley A, et al. Fall
prevention in acute care hospitals: a
randomized trial. JAMA. Nov
3;304(17):1912-8. PMID 21045097.
7.
Fonda D, Cook J, Sandler V, et al. Sustained

reduction in serious fall-related injuries in
older people in hospital. Med J Aust. 2006
Apr 17;184(8):379-82. PMID 16618235.
8.
Grenier-Sennelier C, Lombard I, JenyLoeper C, et al. Designing adverse event

prevention programs using quality
management methods: the case of falls in
hospital. Int J Qual Health Care. 2002
Oct;14(5):419-26. PMID 12389808.
9.
Haines TP, Bennell KL, Osborne RH, et al.

Effectiveness of targeted falls prevention
programme in subacute hospital setting:
randomised controlled trial. BMJ. 2004 Mar
20;328(7441):676. PMID 15031238.
D-52
10.
Healey F, Monro A, Cockram A, et al.

Using targeted risk factor reduction to
prevent falls in older in-patients: a
randomised controlled trial. Age Ageing.
2004 Jul;33(4):390-5. PMID 15151914.
11.
Koh SL, Hafizah N, Lee JY, et al. Impact of

a fall prevention programme in acute
hospital settings in Singapore. Singapore
Med J. 2009 Apr;50(4):425-32. PMID
19421690.
12.
Krauss MJ, Tutlam N, Costantinou E, et al.

Hosp Epidemiol. 2008 Jun;29(6):539-45.
PMID 18476777.
13.
Mitchell A, Jones N. Striving to prevent

falls in an acute care setting--action to
enhance quality. J Clin Nurs. 1996
Jul;5(4):213-20. PMID 8718053.
14.
Oliver D, Martin F, Seed P. Preventing

patient falls. Age Ageing. 2002
Jan;31(1):75-6. PMID 11850313.
15.
Schwendimann R, Buhler H, De Geest S, et

al. Falls and consequent injuries in
hospitalized patients: effects of an
interdisciplinary falls prevention program.
BMC Health Serv Res. 2006;6:69. PMID
16759386.
16.
Stenvall M, Olofsson B, Lundstrom M, et al.

A multidisciplinary, multifactorial
intervention program reduces postoperative
falls and injuries after femoral neck fracture.
Osteoporos Int. 2007 Feb;18(2):167-75.
PMID 17061151.
17.
Uden G, Ehnfors M, Sjostrom K. Use of

initial risk assessment and recording as the
main nursing intervention in identifying risk
of falls. J Adv Nurs. 1999 Jan;29(1):145-52.
PMID 10064293.
18.
van der Helm J, Goossens A, Bossuyt P.

When implementation fails: the case of a
nursing guideline for fall prevention. Jt
Mar;32(3):152-60. PMID 16617946.
19.
20.
21.
Vassallo M, Vignaraja R, Sharma JC, et al.

The effect of changing practice on fall
prevention in a rehabilitative hospital: the
Hospital Injury Prevention Study. J Am
Geriatr Soc. 2004 Mar;52(3):335-9. PMID
14962145.
von Renteln-Kruse W, Krause T. Incidence
of in-hospital falls in geriatric patients
before and after the introduction of an
interdisciplinary team-based fall-prevention
intervention. J Am Geriatr Soc. 2007
Dec;55(12):2068-74. PMID 17971140.
Williams TA, King G, Hill AM, et al.
Evaluation of a falls prevention programme
in an acute tertiary care hospital. J Clin
Nurs. 2007 Feb;16(2):316-24. PMID
17239067.
22.
Browne JA, Covington BG, Davila Y. Using

information technology to assist in redesign
of a fall prevention program. J Nurs Care
Qual. 2004;19(3):218-25.
23.
Capan K, Lynch B. A hospital fall

assessment and intervention project. JCOM.
2007;14(3):155-60.
24.
Dempsey J. Falls prevention revisited: a call

for a new approach. J Clin Nurs. 2004
May;13(4):479-85. PMID 15086634.
25.
Gutierrez F, Smith K. Reducing falls in a

Definitive Observation Unit: an evidencebased practice institute consortium project.
Crit Care Nurs Q. 2008 Apr-Jun;31(2):12739. PMID 18360143.
D-53
26.
Kolin MM, Minnier T, Hale KM, et al. Fall

initiatives: redesigning best practice. J Nurs
Adm. Sep;40(9):384-91. PMID 20798621.
27.
McCollam ME. Evaluation and

implementation of a research-based falls
assessment innovation. Nurs Clin North Am.
1995 Sep;30(3):507-14. PMID 7567575.
28.
Neily J, Howard K, Quigley P, et al. Oneyear follow-up after a collaborative

breakthrough series on reducing falls and
fall-related injuries. Jt Comm J Qual Patient
Saf. 2005 May;31(5):275-85. PMID
15960018.
29.
OConnell B, Myers H. A failed fall

prevention study in an acute care setting:
lessons from the swamp. Int J Nurs Pract.
2001 Apr;7(2):126-30. PMID 11811315.
30.
Rauch K, Balascio J, Gilbert P. Excellence

in action: developing and implementing a
fall prevention program. J Healthc Qual.
2009 Jan-Feb;31(1):36-42. PMID 19343900.
31.
Semin-Goossens A, van der Helm JM,

Bossuyt PM. A failed model-based attempt
to implement an evidence-based nursing
guideline for fall prevention. J Nurs Care
Qual. 2003 Jul-Sep;18(3):217-25. PMID
12856906.
32.
Weinberg J, Proske D, Szerszen A, et al. An

inpatient fall prevention initiative in a
tertiary care hospital. Jt Comm J Qual
Patient Saf. Jul;37(7):317-25. PMID
21819030.
Evidence Tables for Chapter 20. Preventing In-Facility Delirium

Table 1, Chapter 20. Risk factors for delirium
Author/Year/
Country
Study Design
Patient
Population
Slor et al.
3
2011
The
Netherlands
Secondary
analysis of
RCT
526 patients
Adults aged
70 years or older
undergoing
acute or elective
hip surgery,
without delirium
at admission
(or profound
dementia
precluding
communication)
Burkhart et al.
4
2010
Switzerland
Cohort study
(post-hoc
analysis of
RCT)
113 patients
Adults aged
65 years or older
undergoing
cardiac surgery
with cardiopulmonary
bypass (CPB);
patients with
Mini-Mental
State Exam
(MMSE) score
<15/30 were
excluded
Description
of
Organization
Academic
hospital
(915 beds)
Academic
hospital
Diagnosis of
Delirium
Diagnostic
and Statistical
Manual of
Mental
Disorders,
Fourth Edition
(DSM-IV)
criteria and
Confusion
Assessment
Measure
(CAM)
CAM
D-54
Type of Analysis
and factors adjusted
for
Univariate analyses
followed by
multivariable logistic
regression; factors
controlled for include
age, APACHE II
score, MMSE score,
Snellen test score,
benzodiazepines,
anticholinergics,
opioids, type of
anesthesia
Univariate and
multivariable logistic
regression with
stepwise backward
elimination; factors
adjusted for include
C-reactive protein
(CRP), intraoperative
fentanyl, and duration
of mechanical
ventilation
Risk Factors
Modifiable risk
factors
No significant risk factors

for delirium were
identified.
None
Multivariable logistic
regression analyses:
Maximum value of
C-reactive protein
measured post-op:
OR: 1.1
(95% CI: 1.01-1.16)
P = 0.02
Fentanyl intraoperatively:
OR: 4.9
(95% CI: 1.72-13.8)
P = 0.003
Duration of mechanical
ventilation:
OR: 1.1
(95% CI: 1.04-1.21)
P = 0.004
Fentanyl amount,
duration of
mechanical
ventilation
Overall
risk of
bias
High
High
Author/Year/
Country
Study Design
Patient
Population
Hudetz et al.
5
2010
USA
Prospective
cohort study
40 patients
Kazmierski et
6
al. 2010
Poland
Prospective
cohort study
563 patients
Adult males
aged 55 years or
older scheduled
for elective
CABG and/or
valve
replacement/
repair
procedures with
CPB. Patients
with prior
documented
cognitive deficits
or vascular
dementia were
excluded.
Adult patients
admitted for
cardiac surgery
with
cardiopulmonary
bypass;
patients with
preop dementia
were excluded.
Description
of
Organization
Veterans
Affairs (VA)
medical center
Diagnosis of
Delirium
Academic
hospital
DSM-IV
criteria
Intensive
Care Delirium
Screening
Checklist
(ICDSC)
D-55
Type of Analysis
for
Univariate and
multiple logistic
regression; factors
psychosocial
variables
(dispositional
optimism, perceived
social support,
perceived stress level,
and depression)
Risk Factors
Modifiable risk
factors
Incidence of post-op
delirium within 5 days
of surgery was reduced
by:
dispositional optimism:
OR: 0.57
(95% CI: 0.35-0.92)
p<0.02
None
Univariate analyses
followed by
multivariate backward
stepwise logistic
regression; factors
age, MMSE score,
major depression,
anemia,
atrial fibrillation (AF),
intubation time, and
pO2 level.
Risk factors for

delirium:
Age 65 years:
OR: 4.23
(95% CI: 2.24-7.96)
MMSE <25:
OR: 6.14
(95% CI: 3.31-11.39)
Intubation >24 hr:
OR 5.29
(95% CI: 2.14-13.06)
pO2 <60 mmHg:
OR: 3.24
(95% CI: 1.77-5.94)
Major depression:
OR: 4.69
(95% CI 1.84-11.93)
Anemia:
OR: 4.77
(95% CI: 1.35-16.82)
AF:
OR: 3.67
(95% CI: 1.40-9.60)
Cognitive
impairment,
depression,
anemia, and AF
could be treated
prior to surgery
Overall
risk of
bias
Moderate
Moderate
Author/Year/
Country
Study Design
Patient
Population
Koebrugge et
7
al. 2010
The
Netherlands
Retrospective
cohort study
107 patients
Patients aged
65 years or older
undergoing
aortoiliac
surgery; patients
with Alzheimers
disease or
dementia were
excluded.
Retrospective
cohort study
26,057,988
hospitalizations
Hospitalizations
recorded in the
National
Inpatient Sample
(NIS) for
DRG categories
pneumonia,
orthopedic
surgery of the
lower extremity,
congestive heart
failure, and
urinary tract/
kidney infections
Lin et al. 2010

USA
Description
of
Organization
Suburban
teaching
hospital
Diagnosis of
Delirium
NIS database
from
1998-2005
ICD-9 codes
for delirium
with
dementia,
drug-induced
delirium, and
nondementia,
non-drug
(NDND)
delirium
DSM-IV
criteria
D-56
Type of Analysis
for
Univariate and
multivariate step
forward logistic
regression; factors
age and urgency of
surgery (emergency
vs. elective)
Multivariate stepwise
forward logistic
regression; factors
age, gender,
logarithm base e,
length of stay, payor,
DRG, cerebrovascular
disease, dementia,
adverse drug effect,
sodium imbalance,
volume depletion,
anemia, atrial
fibrillation, respiratory
intervention, and
diabetes mellitus
Risk Factors
Modifiable risk
factors
Post-op delirium:
Age 70 years:
OR: 7.7
(95% CI: 1.9-30.4)
P<0.01
Emergency (vs. elective)
surgery:
OR: 5.3
(95% CI: 1.3-21.2)
P<0.01
Dementia-associated
delirium:
Age, logarithm base e,
length of stay,
cerebrovascular disease,
dementia, adverse drug
effect, sodium
imbalance, volume
depletion, atrial
fibrillation were all
significant risk factors for
delirium.
Female gender,
Medicaid as payor,
congestive heart failure
DRG, pneumonia DRG,
anemia, and diabetes
were associated with
significantly lower risk of
delirium.
Drug-induced delirium:
length of stay,
orthopedic DRG,
dementia, adverse drug
effect, were all significant
risk factors for delirium.
Female gender,
Medicaid as payor,
None
Sodium
imbalance,
volume depletion,
atrial fibrillation,
and anemia
Overall
risk of
bias
High
High
Author/Year/
Country
Lin et al. 2010

USA
Study Design
Retrospective
cohort study
1,968,527
hospitalizations
Patient
Population
Acute care
hospitalizations
(for pneumonia,
lower extremity
orthopedic
surgery,
congestive heart
failure [CHF],
and kidney/
urinary tract
infection [UTI]) of
patients aged
18 years or older
in New York
Description
of
Organization
De-identified
inpatient data
obtained from
the New York
State Dept of
Health
Statewide
Planning for
Research
Cooperative
System
(SPARCS)
database
Diagnosis of
Delirium
ICD-9 codes
used to
identify
delirium
cases;
original
diagnostic
criteria not
reported
D-57
Type of Analysis
for
Forward stepwise
logistic regression;
factors adjusted for
include comorbidities,
DRG categories,
adverse drug effects
(ADEs), dementia,
mechanical
ventilation/ ventilator
assistance, gender,
age (in decade),
year of discharge,
Caucasian ethnicity,
Medicaid reimburse-
Risk Factors
DRG, pneumonia DRG,

sodium imbalance,
delirium.
Non-dementia, non-drug
delirium:
length of stay,
adverse drug effect,
sodium imbalance,
volume depletion,
atrial fibrillation, and
respiratory intervention
were all significant risk
factors for delirium.
Female gender,
Medicaid as payor,
orthopedic DRG,
DRG, pneumonia DRG,
delirium.
Any delirium after
admission:
Decade of age:
OR: 1.53
(95% CI: 1.49-1.58)
Female:
OR: 0.70
(95% CI: 0.66-0.75)
Caucasian:
OR: 1.45
(95% CI: 1.29-1.62)
Elective admission:
OR: 0.87
(95% CI: 0.80-0.94)
Modifiable risk
factors
Overall
risk of
bias
None
High
Author/Year/
Country
Study Design
Patient
Population
Description
of
Organization
Diagnosis of
Delirium
State
(1998-2007).
Type of Analysis
for
ment, and elective
admission status
Risk Factors
Medicaid:
OR: 0.74
(95% CI: 0.66-0.82)
CHF DRG:
OR: 0.76
(95% CI: 0.64-0.89)
Lower extremity
orthopedic surgery
DRGs:
OR: 7.36
(95% CI: 6.38-8.50)
Any ADE:
OR: 22.19
(95% CI: 20.72-23.76)
Dementia:
OR: 1.26
(95% CI: 1.12-1.41)
Respiratory intervention:
OR: 1.96
(95% CI: 1.62-2.36)
Cerebrovascular
disease:
OR: 1.18
(95% CI: 1.01-1.39)
Atrial fibrillation:
OR: 1.24
(95% CI: 1.15-1.34)
Diabetes mellitus:
OR: 1.14
(95% CI: 1.06-1.23)
Volume depletion:
OR: 1.41
(95% CI: 1.28-1.57)
Anemia:
OR: 1.15
(95% CI: 1.05-1.25)
Hyponatremia:
OR: 1.42
(95% CI: 1.25-1.60)
D-58
Modifiable risk
factors
Overall
risk of
bias
Author/Year/
Country
Study Design
Patient
Population
Radtke et al.
10
2010
Germany
Cohort study
910 patients
Patients received
elective general
anesthesia and
were observed in
recovery room
and hospital
ward on first
postoperative
day
Prospective
cohort study
44 patients
Patients aged
65 or older who
spoke and
understood
English; Patients
with prevalent
delirium or
moderate to
severe cognitive
dysfunction were
excluded.
Rigney 2010
USA
11
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
Academic
hospital
(365 beds)
CAM
Nursing
delirium
screening
scale
(Nu-DESC)
Type of Analysis
for
Univariate and
multivariate logistic
regression with
delirium as the
response. Regression
analyses were
supplemented with a
feature selection
process using
backward elimination.
Factors adjusted for
include age, gender,
duration of surgery,
site, intraop opioids,
anesthetic, preop
fasting (solids and
fluids)
Univariate and
bivariate analyses
followed by logistic
regression; factors
total allostatic load
(AL) scores,
primary mediators
score, secondary
outcomes score, and
individual AL
parameters
D-59
Risk Factors
Modifiable risk
factors
Multiple logistic
regression analyses:
Longer preoperative fluid
fasting time (>6 hr) was
the only significant risk
factor for delirium in both
the recovery room
(OR: 2.69,
95% CI: 1.38-5.24) and
the ward (OR: 10.57,
95% CI: 1.42-78.62).
Older age (OR: 1.02,
95% CI: 1.01-1.03) and
surgical site
(intraabdominal or
intrathoracic vs.
other sites) (OR: 1.83,
95% CI: 1.09-3.07) were
significant risk factors in
the recovery room.
Intraoperative opioid
choice (fentanyl vs.
remifentanil) was a
significant risk factor in
the ward (OR: 2.27,
95% CI: 1.01-5.06).
Primary mediators score
was the only significant
factor predicting delirium.
Preoperative fluid
fasting time,
choice of
intraoperative
opioid
None
Overall
risk of
bias
Moderate
Moderate
Author/Year/
Country
Study Design
Patient
Population
Sieber et al.
12
2010
USA
Double-blind
randomized
controlled trial
(RCT)
114 patients
Patients aged
65 or older
undergoing hip
fracture repair
under spinal
anesthesia with
propofol
sedation;
patients with
mental
barriers that
would preclude
data collection
were excluded.
Bo et al.
13
2009
Italy
Prospective
cohort study
252 patients
Patients aged
70 years
admitted from
emergency dept
(ED) to an acute
geriatric ward
(AGW) or an
acute general
medical ward
(AGMW);
patients with
delirium during
ED stay or at
ward entry were
excluded.
Description
of
Organization
Academic
medical center
Diagnosis of
Delirium
Academic
hospital
CAM
CAM
Type of Analysis
for
Univariate and
multivariate
regression; factors
deep sedation,
dementia, units of
packed erythrocytes
transfused, and
admission to the ICU
Univariate analyses,
then multivariate
forward stepwise
modeling of variables
associated with
incident delirium;
include APACHE II
score, SPMSQ score,
stressful events,
AGW hospitalization
(vs. AGMW
hospitalization)
D-60
Risk Factors
Modifiable risk
factors
Multivariate regression
significant risk factors:
Deep sedation:
OR: 2.69
(95% CI: 1.04-6.93)
p = 0.04
preoperative dementia:
OR: 3.97
(95% CI: 1.54-10.2)
p = 0.004),
units of packed
erythrocytes transfused:
OR: 1.62
(95% CI: 1.10-2.38)
p = 0.01), and
admission to the ICU:
OR: 3.69
(95% CI: 1.17-11.7)
p = 0.02).
Risk of incident
delirium:
APACHE II:
RR: 1.30
(95% CI: 1.11-1.51)
P = 0.001
SPMSQ:
RR: 2.06
(95% CI: 1.62-2.64)
P<0.001
Stressful events:
RR: 3.36
(95% CI: 2.86-5.44)
P = 0.001
AGW hospitalization:
RR: 0.04
(95% CI: 0.01-0.21)
P<0.001
Sedation
More patients
can be admitted
to AGW vs.
AGMW, some
stressful events
might be reduced
Overall
risk of
bias
Moderate
Moderate
Author/Year/
Country
Study Design
Patient
Population
Greene et al.
14
2009
USA
Prospective
cohort study
100 patients
Patients aged
50 years or older
admitted for
major elective
noncardiac
surgery with
at least a 2-day
postop stay
Hattori et al.
15
2009
Japan
Prospective
cohort study
160 patients
Patients aged
75 years
admitted for
abdominal
surgery,
vascular surgery,
or orthopedic
surgery (all
non-emergency);
patients with
severe dementia
were excluded.
Description
of
Organization
Academic
medical center
Diagnosis of
Delirium
4 hospitals
(1 academic),
bed size
ranged from
300 to 887
NEECHAM
Confusion
Scale
CAM
D-61
Type of Analysis
for
Bivariate analyses
then multivariate
analysis: factors
Geriatric Depression
Score-Short Form,
Trails B time,
Digit Symbol Test,
and Symbol Search
Test
Univariate and
multivariate analyses;
include age, gender,
department,
anesthesia, MMSE,
and preop NEECHAM
score
Risk Factors
Modifiable risk
factors
Geriatric Depression
Score-Short Form:
OR per unit: 1.53
(95% CI: 1.22-2.05)
P = 0.0001);
Trails B time:
OR: 1.02
(95% CI: 1.01-1.04)
Depression
Overall
risk of
bias
Moderate
Risk of postop
delirium:
Age >80 years:
OR: 3.14
(95% CI: 1.35-7.26)
Male:
OR: 2.86
(95% CI: 1.09-7.47)
Preop MMSE <25:
OR: 3.96
(95% CI: 1.52-10.39)
Preop NEECHAM <27:
OR: 5.33
(95% CI: 1.84-15.31)
None
Moderate
Author/Year/
Country
Study Design
Patient
Population
Katznelson et
16
al. 2009
Canada
Prospective
cohort study
1,059 patients
Patients
undergoing
cardiac surgery
with CPB
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
CAM-ICU
D-62
Type of Analysis
for
Univariate analysis
then multivariate
logistic regression
with backward and
stepwise selection;
include older age,
gender,
preop depression,
preop renal
dysfunction,
hypertension,
peripheral vascular
disease, New York
Heart Association
(NYHA) class >2,
preop anemia,
diabetes,
preop history of
cerebrovascular
accident/TIA,
prolonged CPB,
intraop anemia and
hyperglycemia,
complex cardiac
surgery, perioperative
intraortic balloon
pump support, and
massive blood
transfusion
Risk Factors
Modifiable risk
factors
Risk of postop
delirium:
Red blood cell
transfusion (>5 units):
OR: 3.29
(95% CI: 2.09-5.19)
Perioperative intraaortic
balloon pump support:
OR: 3.84
(95% CI: 1.72-8.56)
Preop depression:
OR: 3.06
(95% CI: 1.36-6.90)
Preop creatinine
>150 mM:
OR: 2.96
(95% CI: 1.9-4.63)
Age 60 years:
OR: 2.47
(95% CI: 1.43-4.23)
Combined CABG and
valvular surgery:
OR: 1.86
(95% CI: 1.16-2.98)
Preop administration of
statins:
OR: 0.54
(95% CI: 0.35-0.84)
Preop
administration of
statins, preop
depression,
preop creatinine
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
Maldonado et
17
al. 2009
USA
RCT
118 patients
were
randomized to
three different
sedatives
Patients aged
18-90 years
admitted to the
ICU following
elective cardiac
surgery. Patients
with prior
diagnosis of
dementia were
excluded.
Pisani et al.
18
2009
USA
Prospective
cohort study
304 patients
Patients aged
60 years
admitted to ICU
Description
of
Organization
Academic
medical center
and a VA
medical center
Diagnosis of
Delirium
Academic
hospital
(900 beds,
with
14-bed ICU)
CAM-ICU
DSM-IV
criteria
applied by a
neuropsychiatrist
Type of Analysis
for
Univariate followed by
multiple logistic
regression; factors
age, gender,
ASA class, baseline
MMSE score,
Midazolam (vs.
Dexmedetomidine),
and Propofol (vs.
Dexmedetomidine)
Bivariate analyses,
then multivariate
forward selection
regression of
variables associated
with delirium
(P<0.20); factors
benzodiazepine or
opioid use,
Haloperidol use,
steroid use,
ADL impairment,
history of depression,
dementia,
ICU diagnosis of
respiratory disease,
APACHE II score
(minus Glasgow
Coma Scale), Alanine
aminotransferase
level, intubated during
ICU stay, restraint use
during ICU stay
D-63
Risk Factors
Modifiable risk
factors
Post-op sedation:
Midazolam vs.
Dexmedetomidine:
OR: 28.6
(95% CI: 3.7-262.5)
p = 0.01
propofol vs.
dexmedetomidine:
OR: 29.6
(95% CI: 4.8-280.6)
p = 0.01
Age (increasing
10 years):
OR: 1.3
(95% CI: 1.1-1.5)
p = 0.01
Benzodiazepine or
opioid use:
Rate Ratio: 1.64
(95% CI: 1.27-2.10)
Dementia:
Rate Ratio: 1.19
(95% CI: 1.07-1.33)
Haloperidol:
Rate Ratio: 1.35
(95% CI: 1.21-1.50)
APACHE II score:
Rate Ratio:1.01
(95% CI: 1.00-1.02)
Other models showed:
Benzodiazepines or
opioids are a significant
risk for delirium when
dementia is absent, but
not when it is present.
Haloperidol is a
significant risk for
delirium when dementia
is absent, but not when it
is present.
Post-op sedation
Overall
risk of
bias
High
Medication use
Moderate
Author/Year/
Country
Study Design
Patient
Population
Rudolph et al.
19
2009
USA
Prospective
cohort study
122 patients
(derivation
set),
109 patients
(validation set)
Patients aged
60 years who
underwent
cardiac surgery
under general
anesthesia;
patients with
delirium prior to
surgery were
excluded.
Description
of
Organization
Two academic
medical
centers and a
VA hospital
Diagnosis of
Delirium
CAM
D-64
Type of Analysis
for
Multivariate modeling
with bootstrap
resampling was used
to develop a
prediction rule.
Risk Factors
Modifiable risk
factors
Mini mental state

examination (MMSE)
23, prior stroke/TIA,
abnormal albumin, and
geriatric depression
scale >4 were included
in the prediction rule.
Both cohorts showed
increasing risk of
delirium with increasing
risk score (C-statistic =
0.74).
Depression,
cognitive
impairment,
abnormal
albumin
Overall
risk of
bias
Moderate
Author/Year/
Country
Study Design
Patient
Population
Smith et al.
20
2009
USA
Retrospective
cohort study
998 patients
Adults aged
18 years
undergoing
non-cardiac
surgery, with a
minimum of
2 days inpatient
stay. Patients
with history of
dementia or
MMSE score 23
were excluded.
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
Retrospective
chart review
and/or CAM
D-65
Type of Analysis
for
General linear
modeling and logistic
regression with all
covariates entered
simultaneously.
Factors adjusted for
include age, years of
education, Charlson
comorbidity scale,
alcohol consumption
(drinks per week),
pain, and depressive
symptoms
Risk Factors
Modifiable risk
factors
After adjustment for

covariates, older age:
OR: 1.85
(95% CI: 1.11-3.09)
P = 0.019
greater medical
comorbidities:
OR: 1.38
(95% CI: 1.02-1.86)
P = 0.036)
higher levels of
depressive symptoms:
OR: 1.37
(95% CI: 1.00-1.88)
P = 0.049 and
poorer executive
function: OR: 1.23
(95% CI: 1.06-1.43)
P = 0.007
continued to predict
postoperative delirium.
In a post-hoc multivariate
analysis, Stroop task
was the only index of
executive function that
predicted postoperative
delirium:
OR: 1.56
(95% CI: 1.14-2.14)
P = 0.006
Depressive
symptoms are
modifiable with
treatment
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
Van Rompaey
21
et al. 2009
Belgium
Prospective
cohort study
523 patients
Patients aged
18 years were
in the ICU for
at least 24 hours.
Description
of
Organization
One academic
hospital,
one private
hospital, and
two community
hospitals
Diagnosis of
Delirium
Neelon and
Champagne
Confusion
Scale
D-66
Type of Analysis
for
Univariate logistic
regression followed
by multivariate
forward conditional
regression analysis;
include daily alcohol
use, cognitive
impairment,
admission for internal
medicine,
psychoactive
medication,
endotracheal tube or
tracheostomy,
more than 3
perfusions, isolation,
no visible daylight,
and no visit
Risk Factors
Modifiable risk
factors
Daily use of more than

3 units of alcohol:
OR: 3.23
(95% CI: 1.30-7.98)
predisposing cognitive
impairment:
OR: 2.41
(95% CI: 1.21-4.79)
admission for internal
medicine:
OR: 4.01
(95% CI: 1.46-11.01)
psychoactive medication:
OR: 3.34
95% CI: 1.50-11.23
endotracheal tube or
tracheostomy:
OR: 8.07
(95% CI: 1.18-55.06)
more than 3 perfusions:
OR: 2.74
(95% CI: 1.07-7.05)
isolation:
OR: 2.89
(95% CI: 1.00-8.36)
no visible daylight:
OR: 2.39
(95% CI: 1.28-4.45)
and no visit:
OR: 3.73
(95% CI: 1.75-7.93)
Alcohol intake,
psychoactive
medication dose,
isolation,
daylight, allowing
visitors
Overall
risk of
bias
Moderate
Author/Year/
Country
Study Design
Patient
Population
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
Vidan et al.
22
2009
Spain
Controlled
clinical trial
542 patients
Patients aged
70 years
admitted to the
geriatric acute
care unit and two
internal medicine
wards. Patients
had to be free of
delirium yet have
risk factors for
delirium at time
of admission.
Voyer et al.
23
2009
Canada
Crosssectional study
155 patients
Patients aged
65 years with a
prior diagnosis of
dementia
Three longterm care

(LTC) facilities
and one LTC
unit of a large
regional
hospital
CAM
Bivariate analyses
then multivariate
regression; factors
age, severity of
dementia, and
risk factor scores
Koster et al.
24
2008
The
Netherlands
Prospective
cohort study
112 patients
Patients aged
45 years who
underwent
elective cardiac
surgery (with or
without CPB).
Patients with
preop delirium
were excluded.
Hospital
DSM-IV
criteria
Univariate and
multivariate analysis;
include age, type of
operation, anxiety
score, disturbed
sodium/potassium,
diabetes mellitus,
use of CPB, and
EuroSCORE
CAM
D-67
Type of Analysis
for
Logistic regression
with adjustment for
confounders;
these included age
(per decade),
dementia, baseline
ADL independence,
in-hospital stay
(per day),
intervention group
Risk Factors
Modifiable risk
factors
Dementia:
OR: 2.14
(95% CI: 1.15-3.99)
P = 0.02
Baseline ADL
independence:
OR: 0.78
(95% CI: 0.69-0.89)
P = 0.001
In-hospital stay (per
day): OR: 1.02
(95% CI: 1.00-1.05)
P = 0.05
Intervention group:
OR: 0.43
(95% CI: 0.24-0.77)
P = 0.005
Severity of dementia:
OR: 1.04
(95% CI: 1.02-1.06)
risk factor scores:
OR: 1.67
(95% CI: 1.11-2.51)
Risk factor scores based
on number of
predisposing factors for
each patient.
EuroSCORE:
OR: 1.12
(95% CI: 1.05-1.19)
P = 0.001
Electrolytes disturbance:
OR: 3.29
(95% CI: 1.16-9.34)
P = 0.025
Intervention
Overall
risk of
bias
Moderate
Dehydration,
fever, number of
medications,
depression were
modifiable factors
associated with
higher risk scores
Moderate
Electrolytes
disturbance
High
Author/Year/
Country
Study Design
Patient
Population
Lin et al.
25
2008
Taiwan
Prospective
cohort study
151 patients
Mechanicallyventilated adult
patients admitted
to ICU; delirium
assessed for first
5 days; history of
dementia was an
exclusion
criterion
Oh et al.
26
2008
Korea
Retrospective
cohort study
224
All patients aged

70 years who
had undergone
neurosurgery
during a 2-year
period
Description
of
Organization
Academic
medical center
Diagnosis of
Delirium
Academic
medical center
MMSE and
CAM
CAM-ICU
D-68
Type of Analysis
for
then multivariate
stepwise regression
using selected
variables (P<0.1);
include diabetes
mellitus, sepsis, and
hypoalbuminemia
Univariate analyses
followed by
multivariate
regression of
significant factors;
include prior
dementia/ delirium,
abnormal preop
serum glucose,
diabetes, local or
regional anesthesia,
duration of surgery,
recovery room stay,
VAS score (>6.8), and
analgesics usage
Risk Factors
Modifiable risk
factors
Sepsis:
OR: 3.65
(95% CI: 1.03-12.90)
Hypoalbuminemia:
OR: 5.94
(95% CI: 1.23-28.77)
Note: Medications were
not associated
with delirium in
univariate
analyses.
Multivariate model risk
factors:
Previous
dementia/delirium:
OR: 630.4
(95% CI: 289.2-852.4)
P<0.0001
Pre-existent diabetes:
OR: 1.47
(95% CI: 1.17-2.45)
P = 0.012
Local or regional
anesthesia:
OR: 2.21
(95% CI: 1.34-3.47)
P<0.001
VAS score (>6.8):
OR: 1.99
(95% CI: 1.45-4.16)
P<0.001
Analgesics usage:
OR: 1.38
(95% CI: 1.06-2.14)
P = 0.038
None
Preop serum
glucose, type of
anesthesia,
analgesics usage
Overall
risk of
bias
Moderate
High
Author/Year/
Country
Study Design
Patient
Population
Redelmeier et
27
al. 2008
Canada
Retrospective
cohort study
284,158
patients
All patients aged

65 years who
underwent
elective surgery
Description
of
Organization
Database
representing
all Ontario
hospitals
Diagnosis of
Delirium
ICD codes
used to
identify cases
Type of Analysis
for
Multivariable logistic
regression; factors
age, sex,
neuropsychiatric drug,
type of surgery,
duration of surgery
Risk Factors
Modifiable risk
factors
Age (per year increase):

OR: 1.09
(95% CI: 1.09-1.10)
Sex (male vs. female):
OR: 1.71
(95% CI: 1.59-1.86)
Cholinesterase inhibitor:
OR: 3.99
(95% CI: 2.26-7.05)
Antipsychotic:
OR: 1.57
(95% CI: 1.26-1.95)
Antidepressant:
OR: 2.01
(95% CI: 1.75-2.25)
Benzodiazepine:
OR: 1.40
(95% CI: 1.28-1.53)
Thoracic surgery:
OR: 1.54
(95% CI: 1.29-1.84)
Neurosurgery:
OR: 1.22
(95% CI: 1.00-1.49)
Vascular surgery:
OR: 1.20
(95% CI: 1.06-1.36)
Neuropsychiatric
drug use
Musculoskeletal surgery:
OR: 1.19
(95% CI: 1.08-1.31)
Lower urologic and
gynecologic:
OR: 0.55
(95% CI: 0.48-0.62)
Breast and skin surgery:
OR: 0.46
(95% CI: 0.36-0.59)
External head and neck
surgery:
OR: 0.39
D-69
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
Description
of
Organization
Diagnosis of
Delirium
Type of Analysis
for
Inouye et al.
28
2007
USA
Prospective
cohort study
491 patients
(development
cohort)
469 patients
(validation
cohort)
Patients
aged 70 years
admitted to
6 general
medicine units
at an academic
hospital
Academic
medical center
CAM
Bivariable analyses
then multivariate
model; factors
dementia, vision
impairment, activities
of daily living
impairment, Charlson
score, and restraint
use during delirium
Ely et al.
29
2007
USA
Prospective
cohort study
53 patients
Patients aged
18 years
admitted to the
ICU for >24 hrs
Community
teaching
hospital
(541 beds)
CAM-ICU
(intensive
care unit)
Ordinal logistic
regression
(dependent variable
was delirium days);
include APOE4, age,
APACHE II score,
coma days,
sepsis/ARDS/
pneumonia, and
Lorazepam total dose
D-70
Risk Factors
(95% CI: 0.30-0.50)

Opthalmologic surgery:
OR: 0.08
(95% CI: 0.05-0.13)
Duration of surgery (per
30 min increase):
OR: 1.20
(95% CI: 1.19-1.21)
Dementia:
OR: 2.3
(95% CI: 1.4-3.7)
vision impairment:
OR: 2.1
(95% CI: 1.3-3.2)
activities of daily living
impairment:
OR: 1.7
(95% CI: 1.2-3.0)
Charlson score 4:
OR: 1.7
(95% CI: 1.1-2.6)
restraint use during
delirium:
OR: 3.2
(95% CI: 1.9-5.2)
APOE4:
OR: 7.32
(95% CI: 1.82-29.5)
P = 0.005
Coma days, quintiles:
OR: 1.32
(95% CI: 1.08-1.60)
P = 0.006
Modifiable risk
factors
Overall
risk of
bias
Restraint use,
vision
impairment,
functional
impairment
Moderate
None
High
Author/Year/
Country
Study Design
Patient
Population
Leung et al.
30
2007
USA
Prospective
cohort study
203 patients
Patients aged
65 years
scheduled for
major noncardiac
surgery requiring
anesthesia
Ouimet et al.
31
2007
Canada
Prospective
cohort study
203 patients
Patients age
18 years
admitted for
more than 24 hr
to an ICU
Description
of
Organization
Academic
medical center
Diagnosis of
Delirium
Academic
hospital
Intensive care
delirium
screening
checklist
(ICDSC)
CAM
D-71
Type of Analysis
for
Univariate analysis
then multivariate
logistic regression
with the most
promising factors
(APOE, age, history
of CNS disorders,
education, pain,
ADLs, alcohol intake,
cognitive status,
GDS score)
Univariate then
multivariate stepwise
logistic regression on
selected variables;
included age,
hypertension,
tobacco consumption,
alcohol consumption,
APACHE II score,
epidural catheter use,
opiate dose,
benzodiazepine dose,
propofol dose,
indomethacin dose,
coma, anxiety, and
pain
Risk Factors
Modifiable risk
factors
Risk factors for

delirium:
APOE (with e4 vs.
without e4):
OR: 3.64
(95% CI: 1.51-8.77)
Age:
OR: 1.08
(95% CI: 1.00-1.16)
History of CNS disorders
(yes vs. no):
OR: 3.42
(95% CI: 1.44-8.09)
Hypertension:
OR: 1.88
(95% CI: 1.3-2.6)
Alcoholism:
OR: 2.03
(95% CI: 1.26-3.25)
APACHE II score:
OR: 1.05
(95% CI: 1.03-1.07)
Coma:
OR: 3.71
(95% CI: 2.32-5.9)
Anxiety:
OR: 1.8
(95% CI: 1.04-3.37)
None
These factors are

difficult to modify
in the short term
in an ICU
environment.
Overall
risk of
bias
High
Moderate
Author/Year/
Country
Study Design
Patient
Population
Pisani et al.
32
2007
USA
Prospective
cohort study
304 patients
Patients
60 years old
admitted to ICU
for at least
24 hrs
Rudolph et al.
33
2007
USA
Prospective
cohort study
1,218 patients
Patients aged
60 years
undergoing
noncardiac
surgery. Patients
with dementia
were excluded.
Description
of
Organization
Academic
hospital
(900 beds,
with
14-bed ICU)
Diagnosis of
Delirium
13 hospitals in
8 countries
(Denmark,
France,
Germany, the
UK, Greece,
the
Netherlands,
Spain, and the
USA)
DSM-III
criteria
CAM-ICU
D-72
Type of Analysis
for
Univariate analysis
then multivariate
modeling; factors
alcohol, Medicaid
status, race, history of
depression,
medication use,
dementia, APACHE II
score, admitting
diagnosis, admitting
laboratory variables,
and admitting
physiologic variables
Bivariate analyses,
then stepwise
backward and forward
proportional hazard
regression models
using the most
promising variables;
included age,
gender (male),
cognitive
performance,
tobacco exposure,
diabetes, prior
myocardial infarction
(MI), and vascular
surgery
Risk Factors
Modifiable risk
factors
Dementia by
IQCODE >3.3:
OR: 6.3
(95% CI: 2.9-13.8)
Benzodiazepines
before ICU admission:
OR: 3.4
(95% CI: 1.6-7.0)
Creatinine >2 mg/dL:
OR: 2.1
(95% CI: 1.1-4.0)
Arterial pH <7.35:
OR: 2.1
(95% CI: 1.1-3.9)
Vascular risk factors
(tobacco exposure and
vascular surgery):
Rate Ratio: 3.2
(95% CI: 2.1-4.9)
Mildly impaired cognitive
performance:
Rate Ratio: 2.2
(95% CI: 1.4-2.7)
Age (per year):
Rate Ratio: 1.1
(95% CI: 1.0-1.1)
Benzodiazepine
use,
creatinine level,
and arterial pH
are modifiable
Cognitive deficit
might be
treatable prior to
surgery
Overall
risk of
bias
Moderate
Moderate
Author/Year/
Country
Study Design
Patient
Population
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
VelizReissmuller et
34
al. 2007
Sweden
Prospective
cohort study
107 patients
Patients aged
60 years
scheduled for
CABG,
valve surgery or
combined
procedures;
none had
dementia
Beaussier et
35
al. 2006
France
Double-blind
RCT
59 patients
Patients aged
>70 years
undergoing
surgical
resection of
cancer of the left
colon or rectum;
patients with
preoperative
mental
dysfunction were
excluded.
Academic
hospital
CAM
Furlaneto and
Garcez-Leme
36
2006
Brazil
Prospective
cohort study
103 patients
Patients aged
65 years
admitted to the
geriatric
orthopedic ward
for hip fracture
(almost all
underwent
surgery)
Academic
medical center
CAM
CAM
D-73
Type of Analysis
for
Univariate analysis
then logistic
regression of
significant variables;
include age,
alcohol consumption,
memory complaints,
CABG-valve vs.
CABG, valve vs.
CABG, MMSE score
Comparison of
randomized group
outcomes,
no adjustment for
other factors. General
anesthesia for colon
resection; pre-op
intrathecal morphine
(0.3 mg) + postop
patient-controlled
(PCA) intravenous
morphine vs.
PCA alone
Univariate regression
prior to logistic
regression modeling;
include mental
assessment factors
(MMSE,
clock drawing,
blessed), ADL and
length of hospital stay
Risk Factors
Modifiable risk
factors
Overall
risk of
bias
High
Memory complaints:
OR: 3.37
(95% CI: 1.0-11.5)
Valve vs. CABG:
OR: 3.90
(95% CI: 1.0-15.8)
MMSE score
(28 preop):
OR: 11.3
(95% CI: 2.7-47.7)
Cognitive deficit
may be treatable
prior to surgery
No significant difference
in delirium incidence was
found between the two
groups.
None, since
neither
intervention
showed a
difference
Moderate
Cognitive deficit:
OR: 3.04
(95% CI: 1.24-7.41)
Cognitive deficit
may be treatable
Moderate
Author/Year/
Country
Study Design
Patient
Population
Goldenberg et
37
al. 2006
USA
Prospective
cohort study
77 patients
Patients aged
>65 years
admitted for hip
surgery; patients
with existing
delirium were
excluded
Kazmierski et
38
al. 2006
Poland
Prospective
cohort study
260 patients
All patients
received cardiac
surgery; patients
with preop
delirium or
dementia were
excluded
Description
of
Organization
Community
teaching
hospital
Diagnosis of
Delirium
Academic
hospital
DSM-IV
criteria
CAM
D-74
Type of Analysis
for
Univariate logistic
analysis identified
12 factors as
predictors; these
were included in a
regression analysis
(age, morbidity index,
Hct, Alb, MMSE
score, set test score,
ADL score, dementia,
skilled nursing facility
(SNF) residence,
multiple medications,
CNS medications and
abnormal laboratory
values)
then significant
variables added to
multivariate
regression model
(backward stepwise
procedure); factors
MMSE score, AF,
peripheral vascular
disease, major
depression,
cerebrovascular
disease, and age
Risk Factors
Modifiable risk
factors
Risk factors for

delirium:
Multiple medications:
OR: 33.6
(95% CI: 1.9-591.6)
Set test score <20:
OR: 13.1
(95% CI: 2.1-82.7)
MMSE score <24:
OR: 6.9
(95% CI: 1.2-39.5)
Albumin <3.5 g/dl:
OR: 6.1
(95% CI: 1.2-39.5)
Multiple
medications and
cognitive
impairment, but
there may not be
time before
surgery to modify
these factors
Risk factors for

delirium:
MMSE 24:
OR: 10.2
(95% CI: 3.7-28.6)
AF:
OR: 7.2
(95% CI: 2.3-22.7)
Peripheral vascular
disease:
OR: 6.4
(95% CI: 1.9-21.6)
Major depression:
OR: 6.3
(95% CI: 1.4-29.7)
Age 65 years:
OR: 4.0
(95% CI: 1.5-10.4)
Depression,
cognitive
impairment,
AF can be
treated prior to
surgery
Overall
risk of
bias
Moderate
High
Author/Year/
Country
Study Design
Patient
Population
Leung et al.
39
2006
USA
Blind RCT
228 patients
Patients aged
65 years
undergoing
non-cardiac
surgery requiring
general
anesthesia,
expected to
remain in the
hospital 48 hr
Pandharipande
40
et al. 2006
USA
Prospective
cohort study
198 patients
All adult
mechanicallyventilated
patients admitted
to ICU; patients
with preop
neurological
diseases that
would confound
delirium
diagnosis were
excluded.
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
Academic
medical center
CAM-ICU and
Richmond
Agitation
Sedation
Scale (RASS)
CAM
D-75
Type of Analysis
for
Bivariate analyses
then multivariate
logistic regression
analysis with
variables associated
with delirium
(P0.20); factors
age, anesthetic type
(N2O vs. oxygen),
dependence on
performing 1 IADL,
Postop analgesia
(PCA vs.
oral opioids),
benzodiazepine use
on POD 1 or POD 2
Multivariable analysis
of sedative and
analgesic medications
as risk factors for
delirium in a Markov
model; factors
age, gender, visual
and hearing deficits,
dementia, depression,
severity of illness,
sepsis, neurologic
disease, hematocrit,
daily serum glucose
level, lorazepam,
midazolam, fentanyl,
morphine, and
propofol
Risk Factors
Modifiable risk
factors
Age:
OR: 1.07
(95% CI: 1.02-1.26)
Dependence on
performing 1 IADL:
OR: 1.54
(95% CI: 1.01-2.35)
Postop analgesia
(PCA vs. oral opioids:
OR: 3.75
(95% CI: 1.27-11.01)
Benzodiazepine use on
POD 1 or POD 2:
OR: 2.29
(95% CI: 1.21-4.36)
Postop
analgesia,
benzodiazepine
use
Risk factors for

delirium:
Lorazepam:
OR: 1.2
(95% CI: 1.1-1.4)
P = 0.003
No other sedative or
analgesic showed a
statistically significant
risk for delirium.
Use of lorazepam
(alternative
medications can
be substituted)
Overall
risk of
bias
Moderate
Moderate
Author/Year/
Country
Study Design
Patient
Population
Ranhoff et al.
41
2006
Italy
Prospective
cohort study
401 patients
Patients
60 years of age
admitted to a
sub-intensive
care unit for
elderly patients
(SICU)
Sheng et al.
42
2006
Australia
Prospective
cohort study
156 patients
Stroke patients
aged 65 years
recruited over
1 year
Description
of
Organization
General
hospital
Diagnosis of
Delirium
Academic
teaching
hospital
(450 beds)
DSM-IV
criteria
CAM
Type of Analysis
for
Bivariate analysis
then multiple logistic
regression of
variables with p<0.05
in bivariate analysis;
include heavy alcohol
use, fitted bladder
catheter, number of
drugs, visual
problems, Acute
Physiology Score
(APS), Age,
S-albumin, dementia
Binary logistic
regression then
multiple logistic
regression analyses
using significant
variables;
include age,
dementia prestroke,
hemorrhagic stroke,
metabolic factor,
able to lift both arms,
Glasgow coma scale
score <15, neglect,
dysphasia,
vision field loss,
urinary tract infection,
urinary incontinence,
fecal incontinence,
systolic blood
pressure, diastolic
blood pressure, and
one or more
metabolic factors
D-76
Risk Factors
Modifiable risk
factors
Heavy alcohol use:

OR: 6.1
(95% CI: 1.8-19.6)
Fitted bladder catheter:
OR: 2.7
(95% CI: 1.4-4.9)
Max no. of drugs (7+):
OR: 1.9
(95% CI: 1.1-3.2)
Disabled:
OR: 2.5
(95% CI: 1.3-4.7)
Probably demented:
OR: 11.5
(95% CI: 6.1-20.1)
Age:
OR: 1.1
(95% CI: 1.0-1.2)
Dementia prestroke:
OR: 5.7
(95% CI: 1.3-24.9)
Hemorrhagic stroke:
OR: 3.7
(95% CI: 1.2-11.6)
Metabolic factor:
OR: 6.1
(95% CI: 1.9-20.2)
Able to lift both arms:
OR: 0.3
(95% CI: 0.1-0.9)
Glasgow coma scale
score:
OR: 10
(95% CI: 3.7-26.7)
Use of bladder
catheters and no.
of drugs
None
Overall
risk of
bias
Moderate
Moderate
Author/Year/
Country
Study Design
Patient
Population
Korevaar et al.
43
2005
The
Netherlands
Prospective
cohort study
126 patients
All patients
>65 years and
acutely admitted
Shulman et al.
44
2005
Canada
Retrospective
cohort study
10,230
patients
All patients
>65 years who
were newly
dispensed 1 of 3
drugs: lithium,
valproate, or
benztropine
Description
of
Organization
Academic
medical center
Diagnosis of
Delirium
4 administrative databases
covering all
hospitals in
Ontario
Not reported
CAM
D-77
Type of Analysis
for
Univariate and
regression analysis;
include cognitive
impairment,
Katz ADL, Urea, and
leucocytes
Cox proportional
hazards regression,
adjusted for lithium,
valproate,
benztropine, age,
sex, comorbidity,
visual impairment,
and hearing
impairment
Risk Factors
Modifiable risk
factors
Risk factors for

delirium:
Cognitive impairment:
adjusted hazard ratio:
9.48
(95% CI: 2.27-39.54)
Katz ADL 5-6:
8.14
(95% CI: 1.08-61.31)
Katz ADL 7:
14.13
(95% CI: 2.26-88.24)
Urea:
1.10
(95% CI: 1.02-1.18)
9
Leucocytes(10 /L):
0.87
(95% CI: 0.79-0.97)
Benztropine (vs. lithium):
hazard ratio: 1.88
(95% CI: 1.35-2.62)
Cognitive
impairment
Benztropine use
Overall
risk of
bias
Moderate
High
Author/Year/
Country
Study Design
Patient
Population
Yildizeli et al.
45
2005
Turkey
Retrospective
cohort study
432 patients
Patients aged
18 years
admitted for
major elective or
urgent thoracic
surgery
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
DSM-IV
criteria
D-78
Type of Analysis
for
then multivariate
stepwise logistic
regression; factors
age, gender,
chronic disease,
alcohol abuse,
psychiatric problems,
diabetes,
cerebrovascular
disease,
chemotherapy usage,
operation due to
malignancy,
urgent operation,
respiratory
insufficiency,
markedly abnormal
serum chemistry
values, operation
time, length of
hospital stays,
length of intensive
care unit stays,
sleep deprivation,
hypertension,
infection, blood
transfusion, use of
various drugs,
immobilization
Risk Factors
Modifiable risk
factors
Markedly abnormal
serum chemistry values:
OR: 3.01
p = 0.038
Sleep deprivation:
OR: 5.64
p = 0.05
Age:
OR: 1.04
p = 0.03
Operation time:
OR: 1.29
p = 0.04
Sleep
deprivation,
abnormal serum
chemistry
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
Bucerius et al.
46
2004
Germany
Retrospective
cohort study
16,184
patients
All patients
receiving cardiac
surgery
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
Physician
diagnosis
based on
American
Psychiatric
Association
(APA)
guidelines
Type of Analysis
for
then significant
variables added to
multivariate
regression model
(backward stepwise
procedure); factors
age, beating-heart
surgery, atrial
fibrillation,
cerebrovascular
disease, diabetes,
peripheral vascular
disease, LVEF,
preop cardiogenic
shock,
urgent operation,
operating time,
intraop hemofiltration,
and RBC transfusion
Risk Factors
Modifiable risk
factors
Risk factors for

delirium:
Cerebrovascular
disease: OR: 2.15
(95% CI: 1.69-2.72)
Atrial fibrillation:
OR: 1.36
(95% CI: 1.14-1.62)
Diabetes:
OR: 1.31
(95% CI: 1.16-1.49)
Peripheral vascular
disease:
OR: 1.34
(95% CI: 1.17-1.53)
LVEF 30%:
1.30
(95% CI: 1.09-1.49)
Preop cardiogenic shock:
OR: 1.23
(95% CI: 1.05-1.45)
Urgent operation:
OR: 1.17
(95% CI: 1.02-1.34)
Operating time 3 hr:
OR: 1.26
(95% CI: 1.01-1.45)
Type of surgery
(if patient is
candidate for
beating-heart
surgery); AF can
be treated prior
to surgery
Intraop hemofiltration:
OR: 1.26
(95% CI: 1.06-1.49)
RBC transfusion
2000 ml:
OR: 3.15
(95% CI: 2.71-3.65)
Lower risk of delirium:
Beating-heart surgery:
OR: 0.47
(95% CI: 0.32-0.69)
YOUNGER AGE:
Age <50 years:
D-79
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
Description
of
Organization
Diagnosis of
Delirium
Type of Analysis
for
Caeiro et al.
47
2004
Portugal
Prospective
cohort study
218 patients
Consecutive
acute stroke
patients admitted
to stroke unit
Academic
hospital with
12-bed stroke
unit
Delirium
Rating Scale
(DRS) score
10 and
fulfilled
DSM-IV-TR
criteria
Univariate and
multivariate analysis
with stepwise logistic
regression
Santos et al.
48
2004
Brazil
Prospective
cohort study
220 patients
Patients aged
60 years
admitted for
nonemergency
CABG; patients
with severe
cognitive deficits
were excluded.
Academic
tertiary referral
hospital
DSM-IV
criteria
3 multivariate
analyses:
(1) preop variables;
(2) preop and intraop
variables;
(3) preop, intraop, and
postop variables);
include age,
blood urea,
cardiothoracic index,
hypertension,
smoking, blood
replacement, AF,
pneumonia,
blood balance
nd
2 postop day
D-80
Risk Factors
OR: 0.22
(95% CI: 0.15-0.31)
Age 50 and <60 years:
OR: 0.34
(95% CI: 0.27-0.43)
Age 60 and <70 years:
OR: 0.6
(95% CI: 0.52-0.68)
Non-neuroleptics
anticholinergics (ACH)
during hospitalization,
medical complications,
ACH taken before stroke,
and intracerebral
hemorrhage (ICH) all
remained in the final
regression model.
Age:
OR: 1.1
(95% CI: 1.01-1.19)
Blood urea:
OR: 1.03
(95% CI: 1.01-1.05)
Cardiothoracic index:
OR: 3.38
(95% CI: 1.39-8.25)
Hypertension:
OR: 3.55
(95% CI: 1.25-10.14)
Smoking:
OR: 4.19
(95% CI: 1.35-13.05)
AF:
OR: 2.62
(95% CI: 1.05-6.58)
U:
OR: 6.36
(95% CI: 1.24-32.71)
Modifiable risk
factors
Overall
risk of
bias
Use of ACH
medications
High
Blood urea,
hypertension and
AF are potentially
modifiable
prior to
nonemergency
surgery
Moderate
Author/Year/
Country
Study Design
Patient
Population
Bohner et al.
49
2003
Germany
Prospective
cohort study
153 patients
Patients
undergoing
elective arterial
surgery with an
expected time of
90 minutes
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
DSM-IV
criteria plus
DRS score
12 points
Type of Analysis
for
Univariate then
stepwise multivariate
analysis, which
adjusted for age,
depression,
major amputation,
supraortic occlusive
disease, body length,
cognitive impairment
(MMSE), colloid
infusion, minimal
potassium level,
hypercholesterinemia
Risk Factors
Modifiable risk
factors
Risk factors for

delirium:
No history of supraortic
occlusive disease:
OR: 6.73
P = 0.001
History of major
amputation:
OR: 24.4
P = 0.001
No history of
hypercholesterinemia:
OR: 5.51
P = 0.001
Age >64 years:
OR: 3.03
P = 0.018
Body length <170 cm:
OR: 3.95
P = 0.004
MMSE <25 points:
OR: 28.0
P = 0.001
Intraop colloid infusion
>800 ml:
OR: 2.62
P = 0.035
Intraop colloid
infusion,
intraop minimal
potassium;
cognitive
impairment can
be treated prior
to surgery
Intraop minimal
potassium <3.4 mmol/L:
OR: 3.18
P = 0.021
D-81
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
Centorrino et
50
al. 2003
USA
Retrospective
cohort study
139 patients
Consecutive
adult
hospitalized
patients given
clozapine
Description
of
Organization
Academic
hospital
Diagnosis of
Delirium
Investigator
consensus
based on
signs and
symptoms in
medical chart,
and rated by
consensus on
a 3-point
severity scale
(mild,
moderate,
severe)
D-82
Type of Analysis
for
Bivariate analysis
followed by
regression of factors
with associations with
delirium (p0.10);
include anticholinergic
meds, clinical
responder, age,
hospitalized 20 days,
antipsychotic meds,
CNS agent,
anticonvulsants,
any mood stabilizer,
clozapine dose
>250 mg/day,
tricyclic antidepressan
ts, benzodiazepines,
serotonin reuptake
inhibitors, women,
lithium, any
antidepressant
Risk Factors
Modifiable risk
factors
Any centrally active

anticholinergic:
X2 = 9.69
p = 0.002
Age 39 years:
X2 = 5.69
p = 0.017
Anticholinergic
exposure
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
Morrison et al.
51
2003
USA
Prospective
cohort study
541 patients
Patients
admitted for
hip fracture
without evidence
of delirium
Zakriya et al.
52
2002
USA
Prospective
cohort study
168 patients
Patients
admitted for
hip fracture
service
(age 50-98);
patients with preexisting delirium
or dementia
were excluded.
Description
of
Organization
4 metropolitan
hospitals
Diagnosis of
Delirium
Academic
hospital
CAM
CAM
Type of Analysis
for
Univariate analyses
then multivariate
logistic regression;
included age, gender,
residence,
cognitive impairment,
FIM score,
RAND score,
abnormal BP,
abnormal heart
rhythm, chest pain,
heart failure,
medical complication,
morphine, meperidine
Univariate analyses
then multiple logistic
regression of
variables with P0.1
from univariate;
include normal white
blood cell count,
abnormal serum
sodium, ASA class,
history of congestive
heart failure,
history of AF,
history of peripheral
vascular disease
D-83
Risk Factors
Modifiable risk
factors
Risk factors for

delirium:
Cognitive impairment:
OR: 3.6
(95% CI: 1.8-7.2)
Abnormal BP:
OR: 2.3
(95% CI: 1.2-4.7)
Heart failure:
OR: 2.9
(95% CI: 1.6-5.3)
Parenteral morphine
sulfate equivalents/d
<10 mg:
OR: 5.4
(95% CI: 2.4-12.3)
Received meperidine:
OR: 2.4
(95% CI: 1.3-4.5)
Normal white blood cell
count:
OR: 2.2
(95% CI: 1.2-4.1)
Abnormal serum sodium:
OR: 2.4
(95% CI: 1.1-5.3)
ASA class >II:
OR: 11.3
(95% CI: 2.6-49.2)
Morphine dose,
meperidine use;
cognitive
impairment can
be treated prior
to surgery
Abnormal serum
sodium and white
blood cell count
Overall
risk of
bias
Moderate
High
Author/Year/
Country
Study Design
Patient
Population
Agostini et al.
53
2001
USA
Prospective
cohort study
426 patients
Andersson et
54
al. 2001
Sweden
Prospective
cohort study
457 patients
Patients aged
70 years with
no baseline
delirium admitted
to general
medical service
(non-ICU);
profound
dementia
precluding verbal
communication
was an exclusion
criterion.
Patients aged
65 years
referred for
orthopedic
surgery (hip
fracture or
elective
coxarthros or
gonarthros
surgery)
Description
of
Organization
Academic
hospital
(900 beds)
Diagnosis of
Delirium
Hospital
Modified
Organic Brain
Syndrome
(OBS) Score;
also
considered
DSM-IV
criteria
CAM
D-84
Type of Analysis
for
Logistic regression
model adjusted for
baseline delirium risk,
gender, and age
Risk Factors
Modifiable risk
factors
Diphenhydramine:
OR: 2.3
(95% CI: 1.4-3.6)
Diphenhydramine
Multiple regression,
stepwise model;
included gender, age,
vision, hearing,
reason for hospital
admission, number of
other diseases,
postop complications,
bladder catheter,
preop medical
treatment, anesthesia
time and method,
loss of blood during
surgery, time from
admission to surgery,
surgery time,
time of admission,
marital status,
cohabitation,
type of housing
Risk of developing
delirium:
Four or more physical
diseases:
Exp (B): 15.94
(95% CI: 4.60-55.31)
Reason for admission:
Exp (B): 4.74
(95% CI: 1.76-12.80)
Impaired vision:
Exp (B): 4.52
(95% CI: 2.27-8.98)
Preop medical treatment:
Exp (B): 2.66
(95% CI: 1.26-5.62)
Anesthesia time:
Exp (B): 1.82
(95% CI: 1.31-2.53)
OBS-score on
admission:
Exp (B): 1.28
(95% CI: 1.06-1.54)
Age:
Exp (B): 1.10
(95% CI: 1.04-1.15)
Impaired vision,
anesthesia time,
possibly preop
medical
treatment;
cognitive
impairment can
be treated prior
to surgery
Overall
risk of
bias
Moderate
Moderate
Author/Year/
Country
Study Design
Patient
Population
Dubois et al.
55
2001
Canada
Prospective
cohort study
418 patients
Consecutive
patients aged
18 years
admitted for
>24 hrs to the
ICU
Description
of
Organization
Academic
hospital with
16-bed
medical and
surgical ICU
Diagnosis of
Delirium
Intensive care
delirium
screening
checklist
D-85
Type of Analysis
for
Univariate analyses
then multivariate
analysis using the 5
best factors
(morphine, use of
epidural, smoking
history, bilirubin level,
hypertension)
Univariate nonsignificant factors:
COPD, alcohol abuse,
sodium level, glucose
level, lorazepam,
rooms without
windows, rooms with
windows
Risk Factors
Modifiable risk
factors
Risk of developing
delirium:
Hypertension:
OR: 2.6
(95% CI: 1.14-5.72)
Bilirubin level (% days
abnormal):
OR: 1.2
(95% CI: 1.03-1.40)
Use of Epidural:
OR: 3.5
(95% CI: 1.20-10.39)
Morphine (mean daily
dose):
0.01-7.1 mg:
OR: 7.8 (1.76-34.4)
7.2-18.6 mg:
OR: 9.2 (2.17-39.0)
18.7-331.6 mg:
OR: 6.0 (1.41-25.4)
Hypertension,
bilirubin level,
use of epidural,
morphine dose
Overall
risk of
bias
High
Author/Year/
Country
Study Design
Patient
Population
McCusker et
56
al. 2001
Canada
Retrospective
cohort study
444 patients
(326 with
delirium,
118 without)
Patients
65 years
admitted from
ED to
medical services;
59.5% had
dementia.
Christe et al.
57
2000
Switzerland
Double-blind
RCT
65 patients
Consecutive
geriatric
inpatients
requiring upper
gastrointestinal
endoscopy
Description
of
Organization
Primary acute
care general
hospital
Diagnosis of
Delirium
Academic
geriatric
hospital
(304 beds)
MMSE
decrease of
3 points or
more
CAM
D-86
Type of Analysis
for
Multivariable analyses
of variance;
include age,
delirium index score,
comorbidity, length of
follow-up, dementia,
study group,
prevalent delirium,
visual or hearing
impairment,
number of room
changes,
hospital unit,
in isolation,
stimulation level,
not in the same room,
in a single room,
physical restraint,
medical restraint,
surroundings not welllit, surroundings
noisy/quiet, radio/TV
on, clock/watch
absent,
calendar absent,
no personal
possessions,
not wearing glasses,
not using hearing
aids, family absent
Univariate analyses
then multivariate
stepwise forward and
backward logistic
regressions; factors
adjusted for were not
stated
Risk Factors
Modifiable risk
factors
Final model for

prediction of delirium
severity:
Delirium index score:
Beta: 0.54 0.03,
(P<0.01)
Dementia:
Beta: 1.09 0.28,
(P<0.01)
Number of room
changes:
Beta: 0.40 0.16,
(P = 0.01)
ICU vs. medical:
Beta: 4.62 0.60,
(P<0.01)
Physical restraint:
Beta: 1.21 0.17,
(P<0.01)
Medical restraint:
Beta: 0.42 0.19,
(P = 0.02)
Not wearing glasses:
Beta: 0.81 0.19
(P<0.01)
Room changes,
physical and
medical restraint,
glasses
Basal MMSE <21:

OR: 6.4
(95% CI: 1.1-37.3)
None
Overall
risk of
bias
Moderate
High
Table 2, Chapter 20. Delirium preventionmulti-component interventions

Author/Year Description of Study Design
PSP
and Patients
Inpatient hospital care
Allen et al.
System-wide
58
2011
quality
USA
improvement
(QI) project to
prevent
delirium in
hospitalized
patients
Prospective
controlled
before-after
(CBA) study
199 patients
Theory or
Logic Model
Not reported
Contexts
Implementation
Details
Outcomes: Benefits
External:
None mentioned
Organizational
Characteristics:
6 community
hospitals (part of
Summa Health
System), over 2,000
licensed beds. Acute
Care for Elders
(ACE) unit had prior
experience using
delirium prevention
guidelines.
Teamwork:
Multidisciplinary
delirium workgroup
with physicians and
ACE nurses, director
of hospital quality.
Nurse quality
management and
leadership, clinical
informatics nurses,
geriatric pharmacy,
and geriatric
medicine fellows.
Leadership:
3 of the authors led
the pilot in the ACE
unit.
Culture:
Statement that
Summa Health
System maintains a
strong commitment
to patient safety and
quality
Implementation tools:
First obtained
stakeholder
agreement, then
multidisciplinary
workgroup devised
strategy and carried
out the pilot project.
It involved education
of ACE unit staff on
delirium screening,
prevention and
treatment protocols
that were then
implemented.
Delirium incidence
decreased from 8.8%
in pre-implementation
group to 7.2% in
implementation group
(not statistically
significant).
Mean length of stay
decreased from
7.6 days to 4 days
(difference 3.6 days,
95% CI: 0.66 to
6.49 days).
D-87
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
No harms
Not reported High
reported for
intervention.
Deaths,
ICI transfers, and
30-day
readmissions
all decreased in
intervention
group.

PSP
and Patients
Black et al.
59
2011
Northern
Ireland
Nursefacilitated
family
participation
Chen et al.
60
2011
Taiwan
Modified
Hospital Elder
Life Program
(HELP);
modified to
include
3 shared risk
Theory or
Logic Model
Contexts
Implementation
Details
Staff education and

training, use of audit
and feedback
Prospective
Neumans
External:
CBA study
system model None mentioned
170 patients for nursing
Organizational
Characteristics:
aged
interventions
Inner city public
18 years
hospital with
admitted to a
7-bedded general
general
ICU
intensive care
Teamwork:
unit (ICU)
Researchers, nurses,
and family members
work together.
Leadership:
Researchers
(Director of School of
Nursing and
Emeritus Professor
of Nursing)
Culture:
Not reported
Researcher or nurse
provides family
members with an
information booklet
describing how to
prevent delirium; they
also provide verbal
and printed
introduction to the
study and booklet.
CBA study
Prior evidence External:
(historical
suggests the
None mentioned
Organizational
control)
HELP model
189 patients can prevent and Characteristics:
reduce older
aged
Urban medical
patients posthospital (2,200 beds,
65 years
admitted to a surgical
36-bed
Outcomes: Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Nurses gave family

members the
information booklet
at admission to the
unit; researcher
provided
explanation of the
study and booklet
on Day 1; from Day
2 to transfer to ward,
nurses facilitated
family access to
patient, and families
implemented the
booklets advice.
Incident delirium:
Not reported
Intervention: 25/87
(29%)
Control: 64/83 (77%)
OR = 0.12
(95% CI: 0.06-0.24)
P<0.0001
Authors also state
there were no
significant differences
in mean scores
between groups.
Not reported High
The trained HELP

nurse helped
(sometimes with
family members)
mobilize patients
and simultaneously
engaged them in
Delirium at
discharge:
HELP:
0/102 (0%)
Control:
12/77 (15.6%)
OR = 0.03
Not reported High
D-88
Not reported

PSP
and Patients
Theory or
Logic Model
Contexts
factors
(functional,
nutritional, and
cognitive
status)
targeted by
3 modified
HELP
protocols (early
mobilization,
nutritional
assistance,
and therapeutic
cognitive
activities
gastrointestinal ward
for elective
surgery, with
expected
length of stay
>6 days
functional
decline.
The authors
earlier work
suggests that 3
key elements
are the most
relevant for
surgical patients
and those were
used in this
study.
Hospital Elder
Life Program
(HELP)
CBA study
(historical
control)
Thousands of
patients (aged
70 years)
from 2002 to
2008
HELP provides
skilled interdisciplinary staff
and trained
volunteers to
conduct
intervention
protocols
targeted toward
6 delirium
risk factors:
orientation,
therapeutic
activities, early
mobilization,
vision and
hearing
protocols,
oral volume
repletion, and
sleep
enhancement; it
has been
gastrointestinal ward)
Leadership:
Researchers
designed program
and led the study
Teamwork:
Not reported
Culture:
Not reported
A full-time trained
HELP nurse, blinded
to the study
hypothesis and
not an outcomes
assessor,
implemented the
program.
External:
None mentioned
Organizational
Characteristics:
Community teaching
hospital (500 beds)
Leadership:
The project director
was primarily
responsible for
implementation
Teamwork:
Interdisciplinary staff
and trained
volunteers
Culture:
Not reported
Volunteers were
trained by staff; some
received additional
training by speech
therapists and
Rubin et al.
61
2011
USA
Implementation
Details
Outcomes: Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
cognitive activities (95% CI: 0.001-0.44)

(such as discussing P<0.001
things that
interested the
patient); the nurse
also provided
nutritional
assistance
(oral care,
assisted feeding
if necessary).
HELP was first

implemented in one
40-bed medical unit
in 2002; by 2008
it had spread to
6 units with a total of
184 beds; The
project director
initially worked with
hospital leadership
to determine metrics
for measuring
success; initial
success in the
proposed metrics
was demonstrated,
so the hospital
agreed to continue
funding and allowed
expansion to
additional units;
before starting in a
new unit, the project
D-89
Delirium rate:
Pre-HELP (2001):
41%
HELP (2002): 26%
HELP (2005): 16%
HELP (2008): 18%
Nurse satisfaction:
Nurses and
nurses aides reported
benefit and
satisfaction with HELP
and agreed with a
questionnaire item that
their job was more
satisfying due to
HELP.
Not reported
Not reported High

PSP
and Patients
Theory or
Logic Model
shown to be
effective for
delirium
prevention.
Inouye et al.
62
63
2003 , 1999
USA
Contexts
Implementation
Details
director solicited
input from each
nursing unit director;
the project director
worked with the
Chief Nursing
Director to identify
subsequent units to
target; as patient
volume increased,
paid HELP staff and
volunteers were
added; one Elder
Life Specialist
became the
lead volunteer
coordinator; weekly
meetings of staff
were held to
maintain quality and
document
modifications to the
original HELP
protocols
Delirium has
External:
With oversight by a
HELP for
Prospective
None mentioned
geriatric nurse
prevention of matched CBA been
associated with Organizational
specialist and
delirium in
study
Characteristics:
geriatrician,
elderly patients 852 patients several risk
at least
Urban teaching
factors;
the Elder Life
70 years old the HELP
hospital (900 beds) specialists
Teamwork:
admitted to
targets 6 of
implemented
Interdisciplinary team 6 interventions:
general
these risk
including a geriatric orientation,
medicine floor factors
nurse-specialist,
(later study
(cognitive
therapeutic
two Elder Life
included
impairment,
activities, mobility,
specialists, a certified sleep, hearing or
422 patients sleep
from the HELP deprivation,
therapeuticvision, and
arm of the
recreation specialist, volume repletion (for
immobility,
dehydration); they
study)
a physical therapy
visual
were assisted by
consultant, a
impairment,
hearing
geriatrician, and
trained volunteers;
Outcomes: Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
physical therapists
D-90
In the earlier
Not reported
publication, incident
delirium was
significantly lower in
the intervention group
vs. the usual care
group (9.9% vs. 15%,
OR: 0.60
(95% CI: 0.390.92); P
= 0.02
High patient Moderate

adherence to
individual
interventions
significantly
reduced
incident
delirium
rates.
Adherence
(each 1 point
increase):
OR: 0.69
(95% CI:
0.56-0.87)
P = 0.001

PSP
and Patients
Theory or
Logic Model
Contexts
Implementation
Details
impairment, and trained volunteers.

dehydration)
Leadership:
Not reported
Culture:
Not reported
All staff and
volunteers underwent
quarterly
standardization to
ensure consistent
application of all
intervention protocols
Bjorkelund et Multifactorial
Prospective
Authors cite
External:
64
al. 2010
intervention
CBA study
prior
None mentioned
Sweden
including pre- 263 patients multifactorial
Organizational
Characteristics:
hospital and
aged
intervention
Academic hospital,
perioperative 65 years with studies; they
also pre-hospital
treatment and hip fracture
added preambulance care
care of patients
hospital
Leadership:
with hip
component
Researchers were in
fracture.
because prior
charge
Components
studies have
include
identified preop Teamwork:
supplemental
risk factors for Nurses and
orthopedic surgeons
oxygen
delirium
were part of the team
3-4l/min,
Culture:
IV fluid
Not reported
supplementation and extra
nutrition,
Two of the authors
increased
were in charge of
monitoring of
implementation
vital
physiological
parameters,
adequate pain
relief, avoid
delay in
transfer
logistics, daily
Outcomes: Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
all patients were

assigned
orientation,
therapeutic
activities, and
mobility; other
protocols were
targeted to a
subgroup of patients
with the identified
risk factor.
Patients underwent
pre-hospital care,
nurse assessment
immediately
after admission,
orthopedic surgeon
assessment 30 min
before referral to Xray department,
then transfer to
orthopedic ward,
then surgery for hip
fracture with general
or spinal anesthesia.
Delirium was
assessed by
researchers within 4
hrs of admission
and 8 hr after the
end of anesthesia.
D-91
Post-op delirium:
Intervention:
28/131 (21.4%)
Control:
44/132 (33.3%)
OR = 0.54 (0.31-0.95)
P = 0.03
Any
complications:
Intervention:
66/131 (50.4%)
Control:
70/132 (53.0%)
P = 0.67
Not reported High

PSP
and Patients
Needham et
65
al. 2010
USA
delirium
screening
using Organic
Brain
Syndrome
(OBS) scale,
avoid polypharmacy, and
perioperative/
anesthetic
period protocol
Structured
quality
improvement
(QI) model with
components
including:
understanding
the problem
within the
larger
healthcare
system,
creating a
multidisciplinary
improvement
team, enlisting
all
stakeholders to
identify barriers
to change and
appropriate
solutions, and
creating a
change in
practice
through
engagement,
education,
execution, and
CBA study
(historical
control)
57 patients
with acute
respiratory
failure
Theory or
Logic Model
Contexts
Implementation
Details
Outcomes: Benefits
The QI model
was based on a
4 Es model
(engage,
educate,
execute, and
evaluate).
Previous
studies have
shown that
early physical
medicine and
rehabilitation
(PM&R) in the
ICU provides
benefits for
critically ill
patients, and
the QI model
applied this
evidence to
patients in the
medical ICU
(MICU).
External:
None mentioned
Organizational
Characteristics:
Academic hospital
with 16-bed MICU
Leadership:
The lead author was
the project leader.
Researchers were in
charge
Teamwork:
A multidisciplinary QI
team with
representatives from
each relevant
clinician group in the
MICU and PM&R
Culture:
Not reported
Education and
training of nurses,
physical therapists,
occupational
therapists, and
respiratory therapists
to obtain specific
skills related to rehab
of mechanically
Standardized MICU
admission modified
to change default
activity from
bed rest to
as tolerated;
change in sedation
practice from
continuous
intravenous
infusions to
as needed bolus
doses; establishing
guidelines for PT
and OT
consultation;
developing safetyrelated guidelines
for PM&R-related
consultation;
including a full-time
PT and OT and a
part-time rehab
assistant; consulting
a physiatrist; and
increasing
consultations to
neurologists for
MICU patients with
severe or prolonged
Incident delirium:
QI period:
125/482 (28%) MICU
patient days
Pre-QI period: 107/312
(36%) MICU patient
days
P = 0.003
D-92
Outcomes:
Harms
Unexpected
events:
QI period:
4 cases of rectal
or feeding tube
removal, without
any significant
complications
Pre-QI period:
No unexpected
events
P>0.99
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Not reported High

PSP
and Patients
Vidan et al.
22
2009
Spain
Harari et al.
66
2007
U.K.
evaluation
Education
measures and
specific actions
in 7 risk areas
(orientation,
sensory
impairment,
sleep,
mobilization,
hydration,
nutrition, drug
use), with daily
monitoring of
adherence
Proactive care
of older people
undergoing
surgery
(POPS); multidisciplinary
preoperative
comprehensive
geriatric
assessment
(CGA) service
with postoperative
Controlled
clinical trial
542 patients
aged
70 years
admitted to a
geriatric acute
care unit and
two internal
medicine
wards
CBA study
(historical
control)
108 patients
aged
65 years
undergoing
elective
surgery
Theory or
Logic Model
Authors discuss
the HELP
program
as inspiration,
but the new
protocol was
designed to be
implemented in
daily practice
without extra
staff (unlike
HELP).
The authors
hypothesized
that
preoperative
CGA
incorporating
prediction of
adverse
outcomes
combined with
targeted
interventions,
would reduce
Contexts
Implementation
Details
ventilated patients.
External:
None mentioned
Organizational
Characteristics:
Academic hospital
Leadership:
A specialist geriatric
nurse coordinated
the intervention and
monitored
adherence.
Teamwork:
team including
geriatricians,
residents and nurses
who worked in the
geriatric ward.
muscle weakness.
Intervention
implemented within
first 24 hours of
admission to
geriatric ward by
geriatricians,
residents, and
nurses. A specialist
geriatric nurse
coordinated the
intervention and
monitored
adherence.
Culture:
Not reported
Educational program
aimed at changing
the patient care
approach of geriatric
ward staff.
External:
None mentioned
Organizational
Characteristics:
Urban teaching
hospital
Leadership:
Not reported
Teamwork:
team including a
consultant
geriatrician,
The multidisciplinary
QI team
implemented POPS.
Most patients
received pre-op
home visits from
occupational
therapist and
physiotherapy.
Social worker
provided inputs if
needed. Patients
were educated in
D-93
Outcomes: Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
New delirium
episodes:
Intervention:
20/170 (11.7%)
Usual care:
69/372 (18.5%)
OR = 0.59
(95% CI: 0.34-1.00)
P = 0.05
Not reported
Not reported High
Post-op delirium:
POPS:
3/54 (5.6%)
Pre-POPS:
10/54 (18.5%)
OR = 0.26 (0.07-1.00)
P = 0.036
Only reported
complications
were related to
surgery,
not POPS
Not reported High

PSP
and Patients
follow-through
Lundstrom et Postoperative
67
al. 2007
multifactorial
Sweden
intervention to
reduce delirium
and improve
outcomes in
patients with
femoral neck
fractures
Theory or
Logic Model
Contexts
post-operative
complications
and hence
length of stay
(LOS) in
older people
undergoing
elective
surgery.
This strategy
did not target
delirium alone,
but any factor
that might
contribute to
complications or
longer LOS.
nurse specialist in
older people,
occupational
therapist,
physiotherapist and
social worker
Culture:
Not reported
Geriatrician and
nurse provided staff
education in post-op
early detection and
treatment of medical
complications, early
mobilization, pain
management, bowelbladder function,
nutrition and
discharge planning.
Randomized Not reported

controlled trial
(RCT)
199 patients
aged
70 years with
femoral neck
fractures
Implementation
Details
optimizing post-op
recovery. The
geriatrician and
nurse reviewed
patients in
surgical wards and
provided
staff education in
post-op early
detection and
treatment of medical
complications, early
mobilization, pain
management,
bowel-bladder
function, nutrition
and discharge
planning. Follow-up
therapy home visits
were provided to
those with functional
difficulties, and
outpatient clinical
review in those with
ongoing medical
problems.
External:
After education,
None mentioned
all team members
Organizational
(except dietician)
Characteristics:
assessed each
Academic hospital
patient, usually
with 24-bed geriatric within 24 hours after
unit (used only for
admission; team
intervention group)
planning of
Leadership:
individual rehab
Not reported
performed twice a
Teamwork:
week; assessment
A multidisciplinary
of patients with
team including RNs, delirium for
LPNs, registered
precipitating factors;
physiotherapists,
prevention and
D-94
Outcomes: Benefits
Outcomes:
Harms
Post-op delirium:
Not reported
Intervention:
56/102 (54.9%)
Control:
73/97 (75.3%)
OR = 0.40 (0.22-0.73)
p = 0.003
Days with post-op
delirium:
Intervention:
5.0 7.1 days
Control:
10.213.3 days,
p = 0.009
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Not reported High

PSP
and Patients
Lundstrom et Education
68
al. 2005
program and
Sweden
reorganization
of nursing and
medical care
Quasi-RCT
400 patients
aged
70 years
admitted to
two wards
(intervention
and usual
care)
Theory or
Logic Model
Not reported
Contexts
Implementation
Details
registered
occupational
therapists, a dietician
and geriatricians
Culture:
Not reported
All nursing and
medical staff
members attended a
4-day course in
caring, rehabilitation,
teamwork, and
medical knowledge.
treatment of
complications
(infection, anemia,
embolism);
assessment of
bowel/bladder
function; treatment
of sleep apnea;
prevention and
treatment of
decubitus ulcers;
prevention and
treatment of post-op
pain; ensure oxygen
saturation during
first post-op day;
measure blood
pressure for first 2
post-op days;
ensure adequate
nutrition;
mobilization within
first post-op day;
and secondary
prevention of falls
and fractures.
All nursing and
medical staff
members attended a
2-day course
focusing on
dementia and
delirium in geriatric
patients. Staff were
also trained in the
caregiver-patient
interaction. Nursing
care was
reorganized to
support
individualized care,
External:
None mentioned
Organizational
Characteristics:
Academic hospital
Leadership:
Two of the authors
were the leaders of
the program
Teamwork:
A multidisciplinary
team including all
staff in the
intervention ward
Culture:
D-95
Outcomes: Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Prevalent delirium
No harms related Not reported High
within 24 hrs of
to intervention
admission:
were reported
Intervention:
63/200 (31.5%)
Control:
62/200 (31%)
P = 0.91
Delirium on Day 7:
Intervention:
19/63 (30.2%)
Control:
37/62 (59.7%)
OR = 0.29 (0.14-0.61)
P = 0.001

PSP
and Patients
Tabet et al.
Educational
69
70
2005 ; 2006 package for
U.K.
medical and
nursing staff to
reduce
incidence of
delirium in
hospitalized
elderly
patients; a
control ward
did not receive
the educational
package and
performed
usual practice.
CBA study
(concurrent
control)
250 patients
aged
70 years
admitted to
two acute
admission
wards
Theory or
Logic Model
The authors cite

prior studies of
educational
programs
directed at staff
that have
influenced
nursing practice
in relation to
mental health
issues in elderly
people.
Contexts
Implementation
Details
Not reported
All nursing and
medical staff
members attended a
2-day course
focusing on dementia
and delirium in
geriatric patients.
Staff were also
trained in the
caregiver-patient
interaction.
External:
None mentioned
Organizational
Characteristics:
Two acute admission
wards in an inner-city
teaching hospital
Teamwork:
Geriatric psychiatrist
educated staff,
who altered their
practice based on the
education.
Leadership:
The lead investigator
(a geriatric
psychiatrist)
supervised the
project.
Culture:
Not reported
Education program
on delirium for staff
with follow-up
sessions delivered by
geriatric psychiatrist
and nursing staff

received guidance
once a month by a
supervisor
observing a nursing
action.
The educational
package, which
highlighted delirium
risk factors, was
delivered on site
and at various times
to ensure all staff
were involved.
There was an initial
1 hr formal
presentation, written
information on
guidelines for
delirium prevention
and management,
and regular followup meetings to
reinforce learning;
researchers did not
intervene in day-today management or
provide specific
advice pertaining to
specific patients.
D-96
Outcomes: Benefits
Point prevalence of
delirium:
Intervention ward:
12/122 (9.8%)
Usual care ward:
25/128 (19.5%)
OR: 0.45
(95% CI: 0.21-0.94,
P<0.05)
Outcomes:
Harms
Not reported
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
The
High
educational
package was
found to more
effectively
prevent
delirium in
men
(OR: 0.17,
95% CI: 0.050.65) than in
women (OR:
1.04, 95% CI:
0.38-2.81).

PSP
and Patients
Wong et al.
71
2005
Australia
Delirium
education for
hospital staff
plus
recommendations by
geriatric
registrar for up
to 10 possible
targeted
intervention
strategies to
prevent
delirium after
hip fracture
CBA study
(historical
control)
99 patients
aged
>50 years with
hip fracture
admitted to a
general
orthopedic unit
Theory or
Logic Model
Contexts
Implementation
Details
This strategy
had been
successfully
used at a U.S.
hospital in a
previouslypublished study;
targeted
recommendations include
regulation of
bladder and
bowel function,
early detection/
treatment of
major
complications,
correction of
fluid and
electrolyte
imbalance,
discontinuation
of unnecessary
medications,
provision of
oxygen, severe
pain treatment,
agitated
delirium
treatment, use
of appropriate
environmental
stimuli,
adequate
nutritional
intake, and
early
mobilization and
rehabilitation.
External:
None mentioned
Organizational
Characteristics:
Urban teaching
hospital (460 beds)
Teamwork:
Multidisciplinary
committee with
medical, nursing, and
allied health
members of the
orthopedic, geriatric,
and anesthetic depts.
Leadership:
The lead investigator
supervised the
project.
Culture:
Not reported
The leader educated
frontline staff
(interns, ward
nurses, and allied
health staff) on
delirium every
10 weeks
The lead
investigator
educated staff,
supervised data
collection and
assessed patients;
the project team met
fortnightly to
supervise the
program; the
intervention was
implemented over a
3-month period; the
major barrier was a
high turnover of
nursing staff that
was partly overcome
by the nurse
manager of the
orthopedic unit
ensuring that all
nursing staff
attended the
tutorials and
received education
about the use of the
CAM.
D-97
Outcomes: Benefits
Incident delirium:
Intervention:
9/71 (12.7%)
Pre-intervention:
10/28 (37.5%)
OR = 0.26
(95% CI: 0.09-0.74)
P = 0.012
Outcomes:
Harms
None reported
Influence of
Overall
Contexts on
Risk of
Outcomes
Bias
Not reported High

PSP
and Patients
Marcantonio
72
et al. 2001
USA
Proactive
geriatrics
consultation
with target
recommendations based
on a structured
protocol for
patients after
hip fracture
(target
recommendations same
as in Wong et
al. 2005)
Long-term care
Lapane et al. Pharmacist-led
73
2011
Geriatric Risk
USA
Assessment
MedGuide
(GRAM)
reports and
automated
monitoring
plans focusing
on medication
monitoring
phase to
prevent
potential
Theory or
Logic Model
Contexts
Implementation
Details
Single-blind
RCT
126 patients
aged
65 years
admitted
emergently for
surgical repair
of hip fracture
Not clearly
stated, other
than that
geriatrics
consultation is
easily
implementable
and that a
targeted,
proactive
strategy with
intervention on
defined
outcomes has
shown
effectiveness,
although it is
not clear
whether it has
shown prior
effectiveness in
delirium
prevention.
External:
None mentioned
Organizational
Characteristics:
Academic tertiary
medical center
Teamwork:
Geriatrician and
orthopedics team
worked together
Leadership:
Not reported
Culture:
Not reported
Not reported
A geriatrician
evaluated patients
preoperatively or
within 24 hours
postop, performed
daily visits for
duration of
hospitalization and
made targeted
recommendations.
The orthopedics
team (surgeons and
nurses)
implemented the
recommendations
(adherence rate:
77%).
The usual care
group received
management by the
orthopedics team,
including internal
medicine or geriatric
consults on a
reactive rather than
proactive basis.
Post-op delirium:
None reported
Consult:
20/62 (32%)
Usual care:
32/64 (50%)
P = 0.04
However, when
adjusted for baseline
imbalances the effect
size was no longer
statistically significant:
OR: 0.6
(95% CI: 0.3-1.3)
No significant
between-group
difference in days of
delirium per episode
Quasi-RCT:
3,202 patients
(2003)
3,321 patients
(2004)
25 nursing
homes were
randomized to
receive
intervention or
control
GRAM was
designed to
assist
healthcare
professionals
with expertise in
geriatric
pharmacotherapy in
problem
identification
when evaluating
complex
medication
External:
None mentioned
Organizational
Characteristics:
25 nursing homes
(each with 50 or
more geriatric beds).
All nursing homes
had stable contracts
with Omnicare and
had few short stay
residents
Teamwork:
Pharmacists shared
After training in
2003, GRAM
database for falls
and delirium was
integrated in
January 2004 into
the pharmacies
commercial
pharmacy software
system for the
intervention homes.
Reports were
generated on
medications that
Potential delirium
indicator:
In home 2003/04:
Adjusted hazard ratio:
0.93 (0.80-1.09)
New admits 2004:
Adjusted hazard ratio:
0.42 (0.35-0.52)
D-98
Outcomes: Benefits
Outcomes:
Harms
No significant
difference
between groups
for potential
adverse-event
related hospitalization, falls, or
death
Influence of
Overall
Contexts on
Risk of
Outcomes
Bias
Consultation Moderate
showed a
trend toward
being more
effective
among
patients
without
prefracture
dementia or
ADL
impairment,
but the
differences
were not
statistically
significant
Not reported High

PSP
and Patients
Theory or
Logic Model
Contexts
Implementation
Details
adverse drug
events (falls
and delirium) in
nursing homes
regimens of
older adults to
identify, resolve,
and prevent
medicationrelated
problems, aid in
evaluation of
medications as
a cause or
aggravating
factor
contributing to
an older adults
physical,
cognitive, or
functional
decline, and
facilitate
incorporation of
medication
monitoring
information into
the older adults
plan of care.
reports with facility

nurses.
Leadership:
Consultant
pharmacists
Culture:
Not reported
The ASCP
Foundation
developed and
delivered in-service
programs for nursing
staff and consultant
pharmacists. Two of
the authors were
instructors.
contribute to falls
and delirium, as well
as medication
monitoring care
plans and flow
records. Facility
nurses received
reports within 24 h
of admission for new
admissions;
consultant
pharmacists did onsite reviews of drug
regimens for each
resident once every
30 days.
D-99
Outcomes: Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Table 3, Chapter 20. Delirium preventionsingle interventions

Author/Year
Description
of PSP
Inpatient hospital care

Low dose
Al-Aama et al.
74
melatonin for
2010
Canada
patients with
hip fracture
Larsen et al.
75
2010
USA
Atypical
antipsychotic
Perioperative
olanzapine
(5 mg orally
before and
after surgery)
or placebo to
prevent
postop
delirium in
elderly
patients after
joint
replacement
surgery
Study
Design
and
patients
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Doubleblind RCT
145
patients
aged
65 years
admitted to
internal
medicine
service
The article
cites a theory
that delirium
may be related
to abnormal
tryptophan
metabolism,
which can be
regulated by
melatonin
supplementation
Olanzapine is
an
antipsychotic
with some
prior evidence
of efficacy for
delirium
treatment and
prevention.
Internal
medicine
service in a
tertiary care
center
Study medication
was administered (in
double-blind fashion)
daily between
1,800 and 2,400 h
depending upon
patient availability
and medication
administration
schedules for up to
14 days
Incidence of
delirium:
Melatonin:
2/56 (3.6%)
Placebo:
10/52 (19.2%)
RR = 0.19
(95% CI: 0.04-0.81)
P<0.02
2/61 patients
on melatonin
had side
effects of
nightmares or
hallucinations
Not
applicable
Moderate
Academic
hospital
Perioperative
olanzapine (5 mg
orally) or placebo was
administered before
and after surgery by
nurses not involved in
ongoing care of the
patients.
Incidence of
delirium:
Olanzapine:
28 (14.3%)
Placebo:
82 (40.2%)
RR = 0.36
(95% CI: 0.24-0.52)
P<0.0001
The difference was
also significant in
separate subgroups
(knee replacement,
hip replacement)
Severity of
delirium was
greater in the
olanzapine
group
(DRS-R-98
score:
16.44 vs.
14.5,
p = 0.02), and
lasted longer
(2.2 vs.
1.6 days,
p = 0.02).
Medical
complications
did not differ
significantly
between
groups.
Not
applicable
Moderate
Doubleblind RCT
400
patients
aged
65 years
undergoing
elective
knee or hip
replacement
surgery
D-100
Author/Year
Description
of PSP
Prakanrattana
and
Prapaitrakool
76
2007
Thailand
Atypical
antipsychotic
Risperidone
(1 mg) or
placebo taken
orally
(sublingually)
a single time
following
cardiac
surgery
Light propofol
sedation
during
hip repair
surgery
Sieber et al.
12
2010
USA
Study
Design
and
patients
Doubleblind RCT
126
patients
aged
>40 years
undergoing
elective
cardiac
surgery
Doubleblind RCT
114
patients
aged
65 years
undergoing hip
fracture
repair
Maldonado et
17
al. 2009
USA
Different
types of
post-op
sedation
after cardiac
surgery
RCT
118
patients
aged
18 years
undergoing
elective
cardiac
valve
surgery
Shehabi et al.
77
2009
Australia
Sedation
Dexmedetomidine vs.
Doubleblind RCT
306
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Risperidone is
an
antipsychotic
with some
previous
evidence of
efficacy for
treatment of
delirium
Academic
hospital
Risperidone (1 mg
orally) or placebo was
given by nurses when
patients began to
wake in the ICU
Post-op delirium:
Risperidone:
7/63 (11.1%)
Placebo:
20/63 (31.7%)
RR = 0.35
(95% CI: 0.16-0.77)
P = 0.009
None
reported
(post-op
complications
did not differ
significantly
between
groups)
Not
applicable
Low
The authors
hypothesized
that minimizing
sedation depth
during spinal
anesthesia for
hip fracture
repair in
elderly
patients could
decrease the
occurrence of
postop
delirium
The authors
hypothesized
that dexmedetomidine
may be
associated
with a lower
incidence of
delirium due to
its
pharmacologic
properties
Dexmedetomidine is a
selective and
Academic
medical center
Implemented by
anesthesiologists
during surgery.
Post-op delirium:
Light sedation:
11/57 (19%)
Deep sedation:
23/57 (40%)
RR = 0.48
(95% CI: 0.26-0.89)
P = 0.02
Complication
rates were
similar in
both groups.
Light
sedation:
26/57 (46%)
Deep
sedation:
30/57 (53%
p = 0.57
Not
applicable
Moderate
Academic
medical center
Implemented in the
ICU following cardiac
surgery. Patients
were randomized to
three different
sedatives.
Not reported
Not
applicable
High
Two tertiary
referral
academic
Study drug infusion

began at 3 ml/h
within 1 h of
Post-op delirium
(Intention-to-treat):
Dexmedetomidine:
4/40 (10%)
Propofol:
16/36 (44%)
Midazolam:
17/40 (44%)
p<0.001
Per protocol analysis
also significantly
different (p<0.001)
Incident Delirium:
Dexmedetomidine:
13/152 (8.6%)
Bradycardia
occurred
more often in
Not
applicable
Moderate
D-101
Author/Year
Description
of PSP
morphine,
effect on
prevalence of
delirium in
patients
at least
60 years old
after cardiac
surgery
Study
Design
and
patients
patients
aged
60 years
undergoing
cardiac
surgery
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
potent
2 adrenergic
receptor
agonist.
In theory,
its specificity
may provide
an advantage
for delirium
prevention
compared to
other
post-surgical
sedatives or
analgesics
hospitals
admission to the ICU;

dexmedetomidine
dose was
0.1-0.7 g/kg;
morphine dose was
10-70 g/kg; a
propofol infusion
and/or boluses were
given if deemed
necessary for rapid
control of
hypertensive
episodes or
unplanned
awakening; open
label morphine was
allowed in the
dexmed group to
achieve equivalent
analgesia, and
propofol was allowed
in the morphine arm
to maintain equivalent
sedation; drug
infusion was
continued until
removal of chest
drains when patient
was ready for
discharge from ICU,
or for up to 48 h of
mechanical
ventilation.
Morphine:
22/147 (15%)
Rate Ratio: 0.57
(95% CI: 0.26-1.1),
P = 0.09
Dex group
(16.5%) than
in the
Morphine
group (6.1%)
P = 0.006
Systolic
hypotension
occurred
more often in
Morphine
group
(38.1%)
compared to
Dex group
(23%)
P = 0.006
D-102
Duration of
delirium, median:
Dexmedetomidine:
2 days
Morphine: 5 days
(95% CI: 1.1-6.7)
P = 0.03
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Author/Year
Description
of PSP
Hudetz et al.
78
2009
USA
Anesthetic
(NMDA
receptor
antagonist)
Ketamine
during
anesthetic
induction in
older patients
undergoing
cardiac
surgery with
CPB.
Mouzopoulos
79
et al. 2009
Greece
Local
anesthetic
Fascia iliac
block
prophylaxis
(via
Bupivacaine)
for
hip fracture
patients
Study
Design
and
patients
RCT
58 patients
aged
55 years
undergoing
cardiac
surgery
with CPB.
RCT
207
patients
aged
70 years
admitted
for hip
fracture
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Citing prior
evidence that
ketamine may
have neuroprotective
effects, the
authors
hypothesized
that a single
dose of
ketamine
during
anesthetic
induction
would
attenuate
postop
delirium in
older patients
undergoing
cardiac
surgery with
CPB.
The authors
cite prior
studies
suggesting
that hip
fracture
patients are at
increased risk
of delirium due
to severe pain;
therefore, a
fascia iliac
block might
prevent
delirium by
preventing
severe pain.
Veterans
Affairs medical
center
Ketamine (0.5 mg/kg)

or placebo was
administered
intravenously during
anesthetic induction
for cardiac surgery.
Post-op delirium:
Ketamine:
1/29 (3.4%)
Placebo:
9/29 (31%)
RR = 0.11
(95% CI: 0.02-0.81)
P = 0.01
Not reported
Not
applicable
Moderate
Hospital (type
not reported)
(980 beds)
Bupivacaine was
injected on admission
(in blinded fashion)
and repeated daily
every 24 h until
delirium occurrence
or hip surgery; 24 hr
after surgery it was
re-administered and
repeated daily until
delirium occurrence
or discharge.
Incident delirium:
Prophylaxis:
10.8% (11/102)
Placebo:
23.8% (25/105)
OR = 0.45
(95% CI: 0.23-0.87)
No
complications
other than
3 local
hematomas
at injection
site which
resolved
spontaneously
Not
applicable
Moderate
D-103
Author/Year
Description
of PSP
Gamberini et
80
al. 2009
Switzerland
Acetylcholinesterase
inhibitor
Rivastigmine
administered
every 8 hrs
from night
before
surgery until
6th postop
day in a highrisk group of
elderly
patients
undergoing
elective
cardiac
surgery with
CPB
Study
Design
and
patients
Doubleblind RCT
120
patients
aged
65 years
undergoing
elective
cardiac
surgery
with CPB
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Based on prior
studies
suggesting
cholinesterase
inhibitors can
successfully
treat delirium,
the authors
hypothesized
that short-term
administration
of oral
rivastigmine
would reduce
the incidence
of postop
delirium in a
high-risk group
of elderly
patients
undergoing
elective
cardiac
surgery with
CPB
Academic
hospital
Rivastigmine
administered every
8 hrs as a colorless
odorless solution from
night before surgery
th
until 6 postop day
Incident delirium
as assessed by
CAM:
Rivastigmine:
18/56 (32%)
Placebo:
17/57 (30%)
RR = 1.12
(95% CI: 0.50-2.48)
P = 0.80
No significant
betweengroup
difference for
any adverse
events.
Not
applicable
Low
D-104
Author/Year
Description
of PSP
Liptzin et al.
81
2005
USA
Acetylcholinesterase
inhibitor
Donepezil
(given at
5 mg/day) or
placebo for
14 days
preop and
14 days
postop in
patients
undergoing
total joint
replacement
(knee or hip)
Study
Design
and
patients
Doubleblind RCT
80 patients
aged
50 years
undergoing knee
or joint
replacement
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Donepezil is a
cholinesterase
inhibitor
(disruption in
cholinergic
transmission is
thought to be
in causal
pathway of
delirium)
Academic
medical center
Each patient was

evaluated before
surgery then given
either Donepezil
(given at 5 mg/day) or
placebo for 14 days
preop and 14 days
postop
Post-op delirium:
Donepezil:
8/39 (20.5%)
Control:
7/41 (17.1%)
Rate Ratio = 1.2
(95% CI: 0.6-2.6)
P = 0.69
Not reported
Not
applicable
Moderate
D-105
Author/Year
Description
of PSP
McCaffrey et
82
al. 2006
USA
Music therapy
(musical
selection with
bedside CD
turned on
1-3 times/day
+ standard
postop care
from
anesthesia
awakening
time until
discharged)
for patients
undergoing
hip or knee
surgery
Study
Design
and
patients
RCT
(music
therapy +
usual care
vs.
usual care
alone)
124
patients
aged
65 years
undergoing
elective
hip or knee
surgery
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Prior studies
have shown
evidence that
music can
improve
cognition and
calm agitated
patients
Large tertiary
care center
Nurses blinded to
room designation
made room
assignments. Various
CDs were available in
the music therapy
rooms. Music was
played when patients
were awakening from
anesthesia. CD was
set to play for 1 hour
4 times daily. Also,
nurses were asked to
turn on the music
each time they
entered the room,
and family members
and patients were
instructed how to use
the CD player.
Research assistants
checked that CD
players were working
and that the music
and timing of play
suited patient
preferences.
Patients who
experienced acute
confusion:
Music therapy:
2/62 (3.2%)
Usual care:
36/62 (58.1%)
RR = 0.06
(95% CI: 0.01-0.22)
P<0.0001
None
reported
Not
applicable
High
D-106
Author/Year
Description
of PSP
McCaffrey and
83
Locsin 2004
USA
Music therapy
(musical
selection with
bedside CD
turned on
1-3 times/day
+ standard
postop care
from
anesthesia
awakening
time until
discharged)
for patients
undergoing
elective hip
and knee
surgery
Study
Design
and
patients
RCT
(music
therapy +
usual care
vs.
usual care
alone)
66 patients
aged
65 years
undergoing
elective
hip or knee
surgery
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Prior studies
have shown
evidence that
music can
improve
cognition and
calm agitated
patients
Large tertiary
care center
Nurses blinded to
room designation
made room
assignments. Various
CDs were available in
the music therapy
rooms. Music was
played when patients
were awakening from
anesthesia. CD was
set to play for 1 hour
3 times daily. Also,
nurses were asked to
turn on the music
each time they
entered the room,
and family members
and patients were
instructed how to use
the CD player.
Research assistants
checked that CD
players were working
and that the music
and timing of play
suited patient
preferences.
Significantly fewer
patients in the music
therapy group had
episodes of
confusion and
delirium (F = 19.568,
P = 0.001)
None
reported
Not
applicable
High
D-107
Author/Year
Description
of PSP
Kalisvaart et
84
al. 2005
The
Netherlands
Typical
antipsychotic
Haloperidol or
placebo
(0.5 mg
3 times daily)
was started
on admission
and
continued
until 3 days
postop to
prevent
delirium after
hip surgery
Aizawa et al.
85
2002
Japan
Benzodiazepines
Diazepam +
flunitrazepam
drip infusion
and pethidine
drip infusion
for first
3 days (day
of operation
and first 2
postop days)
in patients
undergoing
gastrointestinal
surgery
Study
Design
and
patients
Doubleblind RCT
430
patients
aged
70 years
undergoing hip
surgery
RCT
(deliriumfree
protocol
[DFP] vs.
non-DFP)
40 patients
aged
>70 years
undergoing
gastrointestinal
surgery
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
Haloperidol is
a dopamine
antagonist
which can
enhance
acetylcholine
release. Since
delirium is
highly
associated
with
cholinergic
deficiency,
the authors
hypothesized
that
haloperidol
may have an
indirect
beneficial
effect on
delirium.
Sleep-wake
cycle disorders
have been
reported to be
associated
with postop
delirium, so
medications
that target
sleep cycle
disorders
might prevent
delirium
Teaching
hospital
Haloperidol (0.5 mg
3 times daily) or
placebo was started
on admission and
continued until 3 days
after surgery.
Experienced geriatric
nurses and
geriatricians provided
proactive geriatric
consultation (based
on a structured
multimodal protocol)
to all patients.
Post-op delirium:
Haloperidol:
32/212 (15.1%)
Placebo:
36/218 (16.5%)
RR = 0.91 (95% CI
0.59-1.42)
P = 0.69
Duration of
delirium (days):
Haloperidol: 5.44.9
Placebo: 11.87.5
P<0.001
No drugrelated side
effects were
reported
Not
applicable
Moderate
A city hospital,
no other
details
provided
Diazepam (0.1 mg/kg

intramuscular) +
flunitrazepam
(0.04 mg/kg drip
infusion) and
pethidine (1 mg/kg
drip infusion) at
specific times during
first 3 days (day of
operation and first
2 postop days)
Incidence of postop delirium:

DFP:
1/20 (5%)
Non-DFP:
7/20 (35%)
P = 0.023
DFP was
reported to
cause
morning
lethargy in
8/20 patients
(40%). No
other side
effects were
reported.
Not
applicable
High
D-108
Author/Year
Description
of PSP
Long-term care
Mentes and
Hydration
86
Culp 2003
(individually
USA
calculated
fluid intake
goal) over an
8-week
period in
nursing home
residents
aged
65 years
Moretti et al.
87
2004
Italy
Rivastigmine
(3-6 mg/day)
for 2 years in
patients with
vascular
dementia
Study
Design
and
patients
Theory or
Logic Model
Description of
Organization
Implementation
Details
Outcomes:
Benefits
Outcomes:
Harms
Influence of
Contexts on
Outcomes
Overall
Risk of
Bias
QuasiRCT
(randomization by
coin toss
of different
participating
facilities)
49 participants aged
65 years
Prior studies
have shown
that chronic
underhydration may
lead to
delirium and
other adverse
events
2 Veterans
Administration
(VA),
2 community
nursing homes
Episodes of acute
confusion:
Treatment:
0/25 (0%)
Control:
2/24 (8.2%)
P = not significant
None
reported
Not clear,
but the
possibility
was raised
that control
group staff
might have
altered their
standard
hydration
practices
due to
awareness
of research
staff data
collection.
High
RCT
(Rivastigmine vs.
cardioaspirin)
246
patients
aged 6885 years
with
vascular
dementia
Delirium in
patients with
vascular
dementia
might be due
to lack of
acetylcholine
in the brain.
Rivastigmine
is an anticholinesterase
inhibitor
Academic
hospital
All RNs responsible

for coordinating
implementation at
their site received
intensive training on
intervention/ usual
care implementation.
The project director
made weekly visits to
each site to ensure
that the protocol was
being implemented.
RNs were responsible
for most
implementation
details with
assistance from NAs.
NAs were responsible
for providing fluids for
participants.
Rivastigmine
(3-6 mg/day) or
aspirin (100 mg/day)
for 2 years in patients
with vascular
dementia
Patients with
episodes of
delirium during
follow-up:
Rivastigmine:
46/115 (40%)
Cardioaspirin:
71/115 (62%)
RR = 0.65
(95% CI: 0.50-0.85)
P<0.001
Mean duration of
delirium:
Rivastigmine:
4 1.71 days
Cardioaspirin:
7.86 2.73 days
P<0.01
Not reported
Not
applicable
High
D-109
References
1.
National Clinical Guideline Centre.

Delirium: diagnosis, prevention and
management. London (UK): Royal College
of Physicians; 2010 Jul. 662 p. (Clinical
guideline; no. 103). Available:
www.nice.org.uk/nicemedia/live/13060/499
08/49908.pdf.
8.
Lin RY, Heacock LC, Fogel JF. Druginduced, dementia-associated and nondementia, non-drug delirium hospitalizations
in the United States, 1998-2005: an analysis
of the national inpatient sample. Drugs
Aging 2010 Jan 1;27(1):51-61.
PMID: 20030432
2.
Cochrane Effective Practice and

Organisation of Care (EPOC) Group.
Risk of bias. Replaces quality criteria in the
EPOC data collection checklist.
Ottawa (ON): Cochrane Collaboration; 2001
May 1. 3 p. Also available:
http://epoc.cochrane.org/epoc-authorresources.
9.
Lin RY, Heacock LC, Bhargave GA, et al.

Clinical associations of delirium in
hospitalized adult patients and the role of on
admission presentation. Int J Geriatr
Psychiatry 2010 Oct;25(10):1022-9.
PMID: 20661879
10.
Radtke FM, Franck M, MacGuill M, et al.

Duration of fluid fasting and choice of
analgesic are modifiable factors for early
postoperative delirium. Eur J Anaesthesiol
2010 May;27(5):411-6. PMID: 19887950
11.
Rigney T. Allostatic load and delirium in the

hospitalized older adult. Nurs Res 2010 SepOct;59(5):322-30. PMID: 20671582
12.
Sieber FE, Zakriya KJ, Gottschalk A, et al.

Sedation depth during spinal anesthesia and
the development of postoperative delirium
in elderly patients undergoing hip fracture
repair.[Erratum appears in Mayo Clin Proc.
2010 Apr;85(4):400 Note: Dosage error in
article text. Mayo Clin Proc 2010
Jan;85(1):18-26. PMID: 20042557
13.
Bo M, Martini B, Ruatta C, et al. Geriatric

ward hospitalization reduced incidence
delirium among older medical inpatients.
Am J Geriatr Psychiatry 2009
Sep;17(9):760-8. PMID: 19705520
14.
Greene NH, Attix DK, Weldon BC, et al.

Measures of executive function and
depression identify patients at risk for
postoperative delirium. Anesthesiology 2009
Apr;110(4):788-95. PMID: 19326494
15.
Hattori H, Kamiya J, Shimada H, et al.

Assessment of the risk of postoperative
delirium in elderly patients using E-PASS
and the NEECHAM Confusion Scale.
Int J Geriatr Psychiatry 2009
Nov;24(11):1304-10.
16.
Katznelson R, Djaiani GN, Borger MA, et

al. Preoperative use of statins is associated
with reduced early delirium rates after
cardiac surgery. Anesthesiology 2009
Jan;110(1):67-73. PMID: 19104172
3.
Slor CJ, De Jonghe JF, Vreeswijk R, et al.

Anesthesia and postoperative delirium in
older adults undergoing hip surgery.
J Am Geriatr Soc 2011 Jul;59(7):1313-9.
4.
Burkhart CS, Dell-Kuster S, Gamberini M,

et al. Modifiable and nonmodifiable risk
factors for postoperative delirium after
cardiac surgery with cardiopulmonary
bypass. J Cardiothorac Vasc Anesth 2010
Aug;24(4):555-9. PMID: 20227891
5.
6.
7.
Hudetz JA, Hoffmann RG, Patterson KM, et

al. Preoperative dispositional optimism
correlates with a reduced incidence of
postoperative delirium and recovery of
postoperative cognitive function in cardiac
surgical patients. J Cardiothorac Vasc
Anesth 2010 Aug;24(4):560-7.
PMID: 20346702
Kazmierski J, Kowman M, Banach M, et al.
Incidence and predictors of delirium after
cardiac surgery: results from the IPDACS
study. J Psychosom Res 2010
Aug;69(2):179-85. PMID: 20624517
Koebrugge B, van Wensen RJ, Bosscha K,
et al. Delirium after emergency/elective
open and endovascular aortoiliac surgery at
a surgical ward with a high-standard
delirium care protocol. Vascular 2010 SepOct;18(5):279-87. PMID: 20822723
D-110
17.
Maldonado JR, Wysong A,

van der Starre PJ, et al. Dexmedetomidine
and the reduction of postoperative delirium
after cardiac surgery. Psychosomatics 2009
May-Jun;50(3):206-17. PMID: 19567759
27.
Redelmeier DA, Thiruchelvam D,

Daneman N. Delirium after elective surgery
among elderly patients taking statins.
CMAJ 2008 Sep 23;179(7):645-52.
PMID: 18809895
18.
Pisani MA, Murphy TE, Araujo KL, et al.

Benzodiazepine and opioid use and the
duration of intensive care unit delirium in an
older population. Crit Care Med 2009
Jan;37(1):177-83. PMID: 19050611
28.
Inouye SK, Zhang Y, Jones RN, et al.

Risk factors for delirium at discharge:
development and validation of a predictive
model. Arch Intern Med 2007
Jul 9;167(13):1406-13. PMID: 17620535
19.
Rudolph JL, Jones RN, Levkoff SE, et al.

Derivation and validation of a preoperative
prediction rule for delirium after cardiac
surgery. Circulation 2009
Jan 20;119(2):229-36. PMID: 19118253
29.
Ely EW, Girard TD, Shintani AK, et al.

Apolipoprotein E4 polymorphism as a
genetic predisposition to delirium in
critically ill patients. Crit Care Med 2007
Jan;35(1):112-7. PMID: 17133176
20.
Smith PJ, Attix DK, Weldon BC, et al.

Executive function and depression as
independent risk factors for postoperative
delirium. Anesthesiology 2009
Apr;110(4):781-7. PMID: 19326492
30.
21.
Van Rompaey B, Elseviers MM,

Schuurmans MJ, et al. Risk factors for
delirium in intensive care patients:
a prospective cohort study. Crit Care
2009;13(3):R77. PMID: 19457226
Leung JM, Sands LP, Wang Y, et al.

Apolipoprotein E e4 allele increases the risk
of early postoperative delirium in older
patients undergoing noncardiac surgery.
Anesthesiology 2007 Sep;107(3):406-11.
PMID: 17721242
31.
Ouimet S, Kavanagh BP, Gottfried SB, et al.

Incidence, risk factors and consequences of
ICU delirium. Intensive Care Med 2007
Jan;33(1):66-73. PMID: 17102966
32.
Pisani MA, Murphy TE, Van Ness PH, et al.

Characteristics associated with delirium in
older patients in a medical intensive care
unit. Arch Intern Med 2007 Aug 1327;167(15):1629-34. PMID: 17698685
33.
Rudolph JL, Jones RN, Rasmussen LS, et al.

Independent vascular and cognitive risk
factors for postoperative delirium.
Am J Med 2007 Sep;120(9):807-13.
PMID: 17765051
34.
Veliz-Reissmuller G, Aguero Torres H,

van der Linden J, et al. Pre-operative mild
cognitive dysfunction predicts risk for postoperative delirium after elective cardiac
surgery. Aging Clin Exp Res 2007
Jun;19(3):172-7. PMID: 17607083
35.
Beaussier M, Weickmans H, Parc Y, et al.

Postoperative analgesia and recovery course
after major colorectal surgery in elderly
patients: a randomized comparison between
intrathecal morphine and intravenous PCA
morphine. Reg Anesth Pain Med 2006 NovDec;31(6):531-8. PMID: 17138196
22.
Vidan MT, Sanchez E, Alonso M, et al.

An intervention integrated into daily clinical
practice reduces the incidence of delirium
during hospitalization in elderly patients.
J Am Geriatr Soc 2009 Nov;57(11):2029-36.
PMID: 19754498
23.
Voyer P, Richard S, Doucet L, et al.

Predisposing factors associated with
delirium among demented long-term care
residents. Clin Nurs Res 2009
May;18(2):153-71. PMID: 19377042
24.
Koster S, Oosterveld FG, Hensens AG, et al.

Delirium after cardiac surgery and
predictive validity of a risk checklist.
Ann Thorac Surg 2008 Dec;86(6):1883-7.
25.
Lin SM, Huang CD, Liu CY, et al. Risk

factors for the development of early-onset
delirium and the subsequent clinical
outcome in mechanically ventilated patients.
J Crit Care 2008 Sep;23(3):372-9.
PMID: 18725043
26.
Oh YS, Kim DW, Chun HJ, et al. Incidence

and risk factors of acute postoperative
delirium in geriatric neurosurgical patients.
J Korean Neurosurg Soc 2008
Mar;43(3):143-8. PMID: 19096622
D-111
36.
Furlaneto ME, Garcez-Leme LE. Delirium

in elderly individuals with hip fracture:
causes, incidence, prevalence, and risk
factors. Clinics 2006 Feb;61(1):35-40.
PMID: 16532223
46.
Bucerius J, Gummert JF, Borger MA, et al.

Predictors of delirium after cardiac surgery
delirium: effect of beating-heart (off-pump)
surgery. J Thorac Cardiovasc Surg 2004
Jan;127(1):57-64. PMID: 14752413
37.
Goldenberg G. Predicting post-operative

delirium in elderly patients undergoing
surgery for hip fracture. Psychogeriatrics
2006;6(2):43-8.
47.
38.
Kazmierski J, Kowman M, Banach M, et al.

Preoperative predictors of delirium after
cardiac surgery: a preliminary study.
Gen Hosp Psychiatry 2006 NovDec;28(6):536-8. PMID: 17088170
Caeiro L, Ferro JM, Claro MI, et al.

Delirium in acute stroke: a preliminary study
of the role of anticholinergic medications.
Eur J Neurol 2004 Oct;11(10):699-704.
PMID: 15469455
48.
Santos FS, Velasco IT, Fraguas R Jr.

Risk factors for delirium in the elderly
after coronary artery bypass graft surgery.
Int Psychogeriatr 2004 Jun;16(2):175-93.
PMID: 15318763
49.
Bohner H, Hummel TC, Habel U, et al.

Predicting delirium after vascular surgery:
a model based on pre- and intraoperative
data. Ann Surg 2003 Jul;238(1):149-56.
PMID: 12832977
39.
Leung JM, Sands LP, Vaurio LE, et al.

Nitrous oxide does not change the incidence
of postoperative delirium or cognitive
decline in elderly surgical patients.
Br J Anaesth 2006 Jun;96(6):754-60.
PMID: 16670110
40.
Pandharipande P, Shintani A, Peterson J, et

al. Lorazepam is an independent risk factor
for transitioning to delirium in intensive care
unit patients. Anesthesiology 2006
Jan;104(1):21-26. PMID: 16394685
50.
Centorrino F, Albert MJ, Drago-Ferrante G,

et al. Delirium during clozapine treatment:
incidence and associated risk factors.
Pharmacopsychiatry 2003 Jul;36(4):156-60.
PMID: 12905102
41.
Ranhoff AH, Rozzini R, Sabatini T, et al.

Delirium in a sub-intensive care unit for the
elderly: occurrence and risk factors.
Aging Clin Exp Res 2006 Oct;18(5):440-5.
PMID: 17167309
51.
42.
Sheng AZ, Shen Q, Cordato D, et al.

Delirium within three days of stroke in a
cohort of elderly patients. J Am Geriatr Soc
2006 Aug;54(8):1192-8. PMID: 16913984
Morrison RS, Magaziner J, Gilbert M, et al.

Relationship between pain and opioid
analgesics on the development of delirium
following hip fracture. J Gerontol A Biol Sci
Med Sci 2003 Jan;58(1):76-81.
PMID: 12560416
52.
Zakriya KJ, Christmas C, Wenz JF Sr, et al.

Preoperative factors associated with
postoperative change in confusion
assessment method score in hip fracture
patients. Anesth Analg 2002
Jun;94(6):1628-32, table of content.
PMID: 12032042
53.
Agostini JV, Leo-Summers LS, Inouye SK.

Cognitive and other adverse effects of
diphenhydramine use in hospitalized older
patients. Arch Intern Med 2001
Sep 24;161(17):2091-7. PMID: 11570937
54.
Andersson EM, Gustafson L, Hallberg IR.

Acute confusional state in elderly
orthopaedic patients: factors of importance
for detection in nursing care. Int J Geriatr
Psychiatry 2001 Jan;16(1):7-17.
PMID: 11180480
43.
Korevaar JC, van Munster BC, de Rooij SE.

Risk factors for delirium in acutely admitted
elderly patients: a prospective cohort study.
BMC Geriatr 2005;5:6. PMID: 15826320
44.
Shulman KI, Sykora K, Gill S, et al.

Incidence of delirium in older adults newly
prescribed lithium or valproate: a
population-based cohort study. J Clin
Psychiatry 2005 Apr;66(4):424-7.
PMID: 15816783
45.
Yildizeli B, Ozyurtkan MO, Batirel HF, et

al. Factors associated with postoperative
delirium after thoracic surgery. Ann Thorac
Surg 2005 Mar;79(3):1004-9. PMID:
15734423
D-112
55.
Dubois MJ, Bergeron N, Dumont M, et al.

Delirium in an intensive care unit: a study of
risk factors. Intensive Care Med 2001
Aug;27(8):1297-304. PMID: 11511942
56.
McCusker J, Cole M, Abrahamowicz M, et

al. Environmental risk factors for delirium in
hospitalized older people. J Am Geriatr Soc
2001 Oct;49(10):1327-34. PMID: 11890491
57.
58.
65.
Needham DM, Korupolu R, Zanni JM, et al.

Early physical medicine and rehabilitation
for patients with acute respiratory failure:
a quality improvement project. Arch Phys
Med Rehabil 2010 Apr;91(4):536-42.
PMID: 20382284
66.
Christe C, Janssens JP, Armenian B, et al.

Midazolam sedation for upper
gastrointestinal endoscopy in older persons:
a randomized, double-blind, placebocontrolled study. J Am Geriatr Soc 2000
Nov;48(11):1398-403. PMID: 11083314
Harari D, Hopper A, Dhesi J, et al. Proactive

care of older people undergoing surgery
(POPS): designing, embedding, evaluating
and funding a comprehensive geriatric
assessment service for older elective
surgical patients. Age Ageing 2007
Mar;36(2):190-6. PMID: 17259638
67.
Allen KR, Fosnight SM, Wilford R, et al.

Implementation of a system-wide quality
improvement project to prevent delirium in
hospitalized patients. J Clin Outcomes
Manag 2011 Jun;18(6):253-8.
Lundstrom M, Olofsson B, Stenvall M, et al.

Postoperative delirium in old patients with
femoral neck fracture: a randomized
intervention study. Aging Clin Exp Res
2007 Jun;19(3):178-86. PMID: 17607084
68.
Lundstrom M, Edlund A, Karlsson S, et al.

A multifactorial intervention program
reduces the duration of delirium, length of
hospitalization, and mortality in delirious
patients. J Am Geriatr Soc 2005
Apr;53(4):622-8. PMID: 15817008
69.
Tabet N, Hudson S, Sweeney V, et al.

An educational intervention can prevent
delirium on acute medical wards.
Age Ageing 2005 Mar;34(2):152-6.
PMID: 15713859
70.
Tabet N, Stewart R, Hudson S, et al. Male

gender influences response to an educational
package for delirium prevention among
older people: a stratified analysis. Int J
Geriatr Psychiatry 2006 May;21(5):493-7.
PMID: 16676296
71.
Wong DM, Niam T, Bruce JJ, et al. Quality

project to prevent delirium after hip fracture.
Aust J Ageing 2005;24(3):174-7.
72.
Marcantonio ER, Flacker JM, Wright RJ, et

al. Reducing delirium after hip fracture:
a randomized trial. J Am Geriatr Soc 2001
May;49(5):516-22. PMID: 11380742
73.
Lapane KL, Hughes CM, Daiello LA, et al.

Effect of a pharmacist-led multicomponent
intervention focusing on the medication
monitoring phase to prevent potential
adverse drug events in nursing homes.
J Am Geriatr Soc 2011 Jul;59(7):1238-45.
PMID: 21649623
59.
Black P, Boore JR, Parahoo K. The effect of

nurse-facilitated family participation in the
psychological care of the critically ill
patient. J Adv Nurs 2011 May;67(5):1091101. PMID: 21214624
60.
Chen CC, Lin MT, Tien YW, et al. Modified

hospital elder life program: effects on
abdominal surgery patients. J Am Coll Surg
2011 Aug;213(2):245-52. PMID: 21641835
61.
Rubin FH, Neal K, Fenlon K, et al.

Sustainability and scalability of the hospital
elder life program at a community hospital.
J Am Geriatr Soc 2011 Feb;59(2):359-65.
PMID: 21314654
62.
Inouye SK, Bogardus ST Jr, Williams CS, et

al. The role of adherence on the
effectiveness of nonpharmacologic
interventions: evidence from the delirium
prevention trial. Arch Intern Med 2003
Apr 28;163(8):958-64. PMID: 12719206
63.
64.
Inouye SK, Bogardus ST Jr,

Charpentier PA, et al. A multicomponent
intervention to prevent delirium in
hospitalized older patients. N Engl J Med
1999 Mar 4;340(9):669-76.
PMID: 10053175
Bjorkelund KB, Hommel A, Thorngren KG,
et al. Reducing delirium in elderly patients
with hip fracture: a multi-factorial
intervention study. Acta Anaesthesiol Scand
2010 Jul;54(6):678-88. PMID: 20236093
D-113
74.
Al-Aama T, Brymer C, Gutmanis I, et al.

Melatonin decreases delirium in elderly
patients: a randomized, placebo-controlled
trial. Int J Geriatr Psychiatry 2011
Jul;26(7):687-94. PMID: 20845391
75.
Larsen KA, Kelly SE, Stern TA, et al.

Administration of olanzapine to prevent
postoperative delirium in elderly jointreplacement patients: a randomized,
controlled trial. Psychosomatics 2010 SepOct;51(5):409-18. PMID: 20833940
76.
Prakanrattana U, Prapaitrakool S.
Efficacy of risperidone for prevention of
postoperative delirium in cardiac surgery.
Anaesth Intensive Care 2007 Oct;35(5):7149. PMID: 17933157
77.
Shehabi Y, Grant P, Wolfenden H, et al.

Prevalence of delirium with
dexmedetomidine compared with morphine
based therapy after cardiac surgery: a
randomized controlled trial
(DEXmedetomidine COmpared to
Morphine-DEXCOM Study).
Anesthesiology 2009 Nov;111(5):1075-84.
PMID: 19786862
78.
79.
80.
81.
82.
Hudetz JA, Patterson KM, Iqbal Z, et al.

Ketamine attenuates delirium after cardiac
surgery with cardiopulmonary bypass.
J Cardiothorac Vasc Anesth 2009
Oct;23(5):651-7. PMID: 19231245
Mouzopoulos G, Vasiliadis G,
Lasanianos N, et al. Fascia iliaca block
prophylaxis for hip fracture patients at risk
for delirium: a randomized placebocontrolled study. J Orthop Traumatol
2009;10(3):127-33.
Gamberini M, Bolliger D, Lurati Buse GA,
et al. Rivastigmine for the prevention of
postoperative delirium in elderly patients
undergoing elective cardiac surgery-a randomized controlled trial. Crit Care Med
2009 May;37(5):1762-8. PMID: 19325490
Liptzin B, Laki A, Garb JL, et al. Donepezil
in the prevention and treatment of postsurgical delirium. Am J Geriatr Psychiatry
2005 Dec;13(12):1100-6. PMID: 16319303
on pain and acute confusion in older adults
undergoing hip and knee surgery. Holist
Nurs Pract 2006 Sep-Oct;20(5):218-24;
quiz 225-6. PMID: 16974175
D-114
83.

listening on acute confusion and delirium in
elders undergoing elective hip and knee
surgery. J Clin Nurs 2004;13(s2):91.
84.
Kalisvaart KJ, de Jonghe JF, Bogaards MJ,

et al. Haloperidol prophylaxis for elderly
hip-surgery patients at risk for delirium:
a randomized placebo-controlled study.
J Am Geriatr Soc 2005 Oct;53(10):1658-66.
PMID: 16181163
85.
Aizawa K, Kanai T, Saikawa Y, et al.

A novel approach to the prevention of
postoperative delirium in the elderly after
gastrointestinal surgery. Surg Today
2002;32(4):310-4. PMID: 12027195
86.
Mentes JC, Culp K. Reducing hydrationlinked events in nursing home residents.

Clin Nurs Res 2003 Aug;12(3):210-25;
discussion 226-8. PMID: 12918647
87.
Moretti R, Torre P, Antonello RM, et al.

Cholinesterase inhibition as a possible
therapy for delirium in vascular dementia:
a controlled, open 24-month study of 246
patients. Am J Alzheimers Dis Other Demen
2004 Nov-Dec;19(6):333-9.
PMID: 15633941
88.
Boorsma M, Frijters DH, Knol DL, et al.

Effects of multidisciplinary integrated care
on quality of care in residential care
facilities for elderly people: a cluster
randomized trial. CMAJ 2011
Aug 9;183(11):E724-32. PMID: 21708967
89.
Brausi M, Neri M, Bergamini L, et al.

Incidence of delirium in elderly patients
undergoing urologic surgery: the role of
comprehensive geriatric assessment (CGA)
in defining a high-risk population for
delirium. Anticancer Res 2011
May;31(5):1819-20.
90.
Campbell NL, Khan BA, Farber M, et al.

Improving delirium care in the intensive
care unit: the design of a pragmatic study.
Trials 2011;12:139. PMID: 21645330
91.
Keunen RW, Palmbergen W,

Van Sonderen A, et al. Improved perioperative neurological screening reduces
postoperative delirium after CABG surgery.
Cerebrovasc Dis 2011 May;31:24.
92.
93.
94.
95.
Sanchez Lorenzo I, Lopez Saracho I,

Ledo Rubio A, et al. Delirium in
hospitalized patients: a descriptive and
retrospective study in the clinical university
hospital of valladolid. Eur Psychiatry 2011
Mar;26.
Mouchoux C, Rippert P, Duclos A, et al.
Impact of a multifaceted program to prevent
postoperatrive delirium in the elderly: the
CONFUCIUS stepped wedge protocol.
BMC Geriatr 2011 May 18;11(1):25.
PMID: 21592324
Chaushu S, Becker A. Behaviour
management needs for the orthodontic
treatment of children with disabilities.
Eur J Orthod 2000 Apr;22(2):143-9.
PMID: 10822887
Ono H, Taguchi T, Kido Y, et al. The
usefulness of bright light therapy for patients
after oesophagectomy. Intensive Crit Care
Nurs 2011 Jun;27(3):158-66.
PMID: 21511473
96.
Siddiqi N. Preventing delirium in older

people. Nurs Times 2011 May 31Jun 6;107(21):22-3. PMID: 21800525
97.
Siddiqi N, Young J, House AO, et al.

Stop Delirium! A complex intervention to
prevent delirium in care homes: a mixedmethods feasibility study. Age Ageing 2011
Jan;40(1):90-8. PMID: 20861087
98.
Tognoni P, Simonato A, Robutti N, et al.

Preoperative risk factors for postoperative
delirium (POD) after urological surgery in
the elderly. Arch Gerontol Geriatr 2011
May-Jun;52(3):e166-9. PMID: 21084123
99.
Banerjee A, McGrane S, Pandharipande P.

Minimizing perioperative sedation to reduce
delirium. Future Neurol 2010 May;5(3):35761.
100.
Featherstone I, Hopton A, Siddiqi N.

An intervention to reduce delirium in care
homes. Nurs Older People 2010
May;22(4):16-21. PMID: 20503675
101.
Figueroa-Ramos MI, Arroyo-Novoa CM,

Puntillo KA. The effect of a sedation wakeup trial on the occurrence of delirium in
critically ill trauma patients. Intensive Care
Med 2010 Sep;36:S221.
D-115
102.
Flagg B, Cox L, McDowell S, et al. Nursing

identification of delirium. Clin Nurse Spec
2010 Sep-Oct;24(5):260-6. PMID:
20716979
103.
Gani H, Xhani M, Domi R, et al. Good pain

control reduces the risk of cognitive
complications to elderly after urology
procedures. Eur Urol Suppl 2010
Sep;9(6):608.
104.
Girard TD, Pandharipande PP, Carson SS, et

al. Feasibility, efficacy, and safety of
antipsychotics for intensive care unit
delirium: the MIND randomized, placebocontrolled trial. Crit Care Med 2010
Feb;38(2):428-37. PMID: 20095068
105.
Griffiths R, Rasmussen LS. Delirium in hip

fracture patients. Acta Anaesthesiol Scand
2010 Jul;54(6):661-2. PMID: 20618161
106.
Holroyd-Leduc JM, Abelseth GA,

Khandwala F, et al. A pragmatic study
exploring the prevention of delirium among
hospitalized older hip fracture patients:
applying evidence to routine clinical
practice using clinical decision support.
Implement Sci 2010;5:81. PMID: 20969770
107.
Mudge A, Denaro C, Maussen C. Reducing

delirious discharges: results of a quality
improvement program. Int J Med 2010
Mar;40(Suppl 1):39.
108.
Radtke FM, Franck M, Lorenz M, et al.

Remifentanil reduces the incidence of postoperative delirium. J Int Med Res
2010;38(4):1225-32.
109.
Rosenbloom-Brunton DA, Henneman EA,

Inouye SK. Feasibility of family
participation in a delirium prevention
program for hospitalized older adults.
J Gerontol Nurs 2010 Sep;36(9):22-33;
quiz 34-5. PMID: 20438016
110.
Sandhaus S, Zalon ML, Valenti D, et al.

A volunteer-based Hospital Elder Life
Program to reduce delirium. Health Care
Manag (Frederick) 2010 Apr-Jun;29(2):1506. PMID: 20436332
111.
Souda F, Sako H, Yatagai C, et al. Studying

factors helpful in predicting delirium in
acute stroke patients admitted to SCU.
Int J Stroke 2010 Oct;5:267.
112.
Stockl KM, Le L, Zhang S, et al. Clinical

and economic outcomes associated with
potentially inappropriate prescribing in the
elderly. Am J Manag Care 2010
Jan;16(1):e1-10. PMID: 20059286
113.
Strom T, Martinussen T, Toft P. A protocol

of no sedation for critically ill patients
receiving mechanical ventilation:
a randomised trial. Lancet 2010
Feb 6;375(9713):475-80. PMID: 20116842
114.
115.
116.
117.
118.
119.
120.
121.
122.
Pretto M, Spirig R, Milisen K, et al. Effects

of an interdisciplinary nurse-led Delirium
Prevention and Management Program
(DPMP) on nursing workload: a pilot study.
Int J Nurs Stud 2009 Jun;46(6):804-12.
PMID: 19249782
123.
Schweickert WD, Pohlman MC,

Pohlman AS, et al. Early physical and
occupational therapy in mechanically
ventilated, critically ill patients: a
randomised controlled trial. Lancet 2009
May 30;373(9678):1874-82.
PMID: 19446324
124.
Ushida T, Yokoyama T, Kishida Y, et al.

Incidence and risk factors of postoperative
delirium in cervical spine surgery. Spine
(Phila Pa 1976) 2009 Nov;34(23):2500-4.
PMID: 19927098
125.
Quetiapine + haloperidol = faster resolution

of ICU delirium. J Natl Med Assoc 2010
Jul;102(7):662.
Del Rosario E, Esteve N, Sernandez MJ, et

al. Does femoral nerve analgesia impact the
development of postoperative delirium in
the elderly? A retrospective investigation.
Acute Pain 2008 Jun;10(2):59-64.
126.
Benedict L, Hazelett S, Fleming E, et al.

Prevention, detection and intervention with
delirium in an acute care hospital:
a feasibility study. Int J Older People Nurs
2009 Sep;4(3):194-202. PMID: 20925776
Robinson S, Rich C, Weitzel T, et al.

Delirium prevention for cognitive, sensory,
and mobility impairments. Res Theory Nurs
Pract 2008;22(2):103-13. PMID: 18578220
127.
Schrader SL, Wellik KE,

Demaerschalk BM, et al. Adjunctive
haloperidol prophylaxis reduces
postoperative delirium severity and duration
in at-risk elderly patients. Neurologist 2008
Mar;14(2):134-7. PMID: 18332845
128.
Day J, Higgins I, Koch T. Delirium and

older people: what are the constraints to best
practice in acute care. Int J Older People
Nurs 2008 Sep;3(3):170-7.
PMID: 20925817
de Jonghe JF, Kalisvaart KJ, Dijkstra M, et

al. Early symptoms in the prodromal phase
of delirium: a prospective cohort study in
elderly patients undergoing hip surgery.
Am J Geriatr Psychiatry 2007
Feb;15(2):112-21. PMID: 17272731
129.
Fong JJ, Devlin JW. Can postoperative

delirium be prevented pharmacologically?
[Erratum appears in Crit Care Med.
2009 Jul;37(7):2331 Note: Devlin, John
[corrected to Devlin, John W]. Crit Care
Med 2009 May;37(5):1825-6.
PMID: 19373065
Sampson EL, Raven PR, Ndhlovu PN, et al.

A randomized, double-blind, placebocontrolled trial of donepezil hydrochloride
(Aricept) for reducing the incidence of
postoperative delirium after elective total
hip replacement. Int J Geriatr Psychiatry
2007 Apr;22(4):343-9. PMID: 17006875
130.
Pitkala KH, Laurila JV, Strandberg TE, et

al. Multicomponent geriatric intervention
for elderly inpatients with delirium: a
randomized, controlled trial. J Gerontol A
Biol Sci Med Sci 2006 Feb;61(2):176-81.
PMID: 16510862
Sultan SS. Assessment of role of

perioperative melatonin in prevention and
treatment of postoperative delirium after hip
arthroplasty under spinal anesthesia in the
elderly. Saudi J Anaesth 2010 Sep;4(3):16973. PMID: 21189854
Eijk MV, Roes K, Honing M, et al.
Rivastigmine does not decrease duration of
delirium and may increase mortality in
critically ill patients: a multicentre, doubleblind, randomized, placebo-controlled addon trial. Crit Care Med 2010 Dec;38:A10.
Day J, Higgins I, Koch T. The process of

practice redesign in delirium care for
hospitalised older people: a participatory
action research study. Int J Nurs Stud 2009
Jan;46(1):13-22. PMID: 18950767
Fujita T. Better sleep in the ICU for

preventing postoperative delirium.
J Am Coll Surg 2009 Dec;209(6):792-3.
D-116
131.
Leslie DL, Zhang Y, Holford TR, et al.

Premature death associated with delirium at
1-year follow-up. Arch Intern Med 2005
Jul 25;165(14):1657-62. PMID: 16043686
132.
Naughton BJ, Saltzman S, Ramadan F, et al.

A multifactorial intervention to reduce
prevalence of delirium and shorten hospital
length of stay. J Am Geriatr Soc 2005
Jan;53(1):18-23. PMID: 15667371
133.
Papaioannou A, Fraidakis O,
Michaloudis D, et al. The impact of the type
of anaesthesia on cognitive status and
delirium during the first postoperative days
in elderly patients. Eur J Anaesthesiol 2005
Jul;22(7):492-9. PMID: 16045136
134.
Dautzenberg PL, Mulder LJ,

Olde Rikkert MG, et al. Delirium in elderly
hospitalised patients: protective effects of
chronic rivastigmine usage. Int J Geriatr
Psychiatry 2004 Jul;19(7):641-4.
PMID: 15254920
135.
Flaherty JH, Tariq SH, Raghavan S, et al.

A model for managing delirious older
inpatients. J Am Geriatr Soc 2003
Jul;51(7):1031-5. PMID: 12834527
136.
Morita T, Tei Y, Inoue S. Agitated terminal

delirium and association with partial opioid
substitution and hydration. J Palliat Med
2003 Aug;6(4):557-63. PMID: 14516497
137.
Robinson SB, Rosher RB. Can a beverage

cart help improve hydration? Geriatr Nurs
2002 Jul-Aug;23(4):208-11.
138.
Sommer BR, Wise LC, Kraemer HC.

Is dopamine administration possibly a risk
factor for delirium? Crit Care Med 2002
Jul;30(7):1508-11. PMID: 12130971
139.
Han L, McCusker J, Cole M, et al. Use of

medications with anticholinergic effect
predicts clinical severity of delirium
symptoms in older medical inpatients.
Arch Intern Med 2001 Apr 23;161(8):1099105. PMID: 11322844
140.
Milisen K, Foreman MD, Abraham IL, et al.

A nurse-led interdisciplinary intervention
program for delirium in elderly hip-fracture
patients. J Am Geriatr Soc 2001;49(5):52332.
141.
Patten SB, Williams JV, Petcu R, et al.

Delirium in psychiatric inpatients: a casecontrol study. Can J Psychiatry 2001
Mar;46(2):162-6. PMID: 11280086
D-117
Evidence Tables for Chapter 21. Preventing In-Facility Pressure Ulcers

Table 1, Chapter 21. Multi-component pressure ulcer prevention initiatives conducted in acute care settings in the United States
Author/Year
Description of PSP
Study
Design
Lynch and
1a
Vickery 2010
Zero-tolerance
philosophy
Target safety
problem: PU
Key elements:
Multidisciplinary
team; educate
staff/resident/family,
streamline
documentation,
wound-care
workshops,
case studies,
setting goals,
identify and address
barriers
Pre-post
Theory or
Logic
Model
NS
Description
of
Organization
166-bed
acute
rehabilitation
Contexts
Outcomes:
Benefits
External:
New CMS reimbursement
Organizational
Characteristics:NS
Teamwork, Leadership,
Culture:
After reviewing 2007 data
on PUs, the team was
dismayed at the number of
misidentification of PUs
at admission; skin
assessments incomplete
and inconclusive;
incorrect staging;
incorrect documentation
(e.g., document denuded
skin as PU); documentation
fragmented; definition of
thorough skin assessment;
inconsistent documentation
of interventions; incorrectly
transcribing interventions to
appropriate documentation.
Interdisciplinary team
Education
at orientation, annually,
and one-on-one,
via web
Documentation
streamlined to 1 form
Wound care workshops
for nurses
at orientation; after
2 months
Report cards
Length: 1 year
Process:
Multidisciplinary team
reviews current processes
of care and finds errors with
assessment and
documentation; education
of staff is quickly put in
place; staff is encouraged to
report HAPUs and view as
an opportunity to learn; rate
goals are set for hospital
and by unit; report cards are
posted so units can track
their progress.
Successes:
Due to streamlining
documentation, timely and
accurate completion of
documentation increases
from 60% to 90% in
90 days; patients on a
neurobehavioral stroke unit
did not develop PU
Barriers:
Patients dissatisfied with
off-loading boots
Addressing Barriers:
Trial initiated to evaluate
use of pillow; leads to
improved outcomes
Sustainability:
Quarterly newsletter
attached to paychecks
PU Rates:
Pre: 2.8%
Post: 0.48%
(-82.8%)
D-118
Influence of
Contexts on
Outcomes
NS
Author/Year
Description of PSP
Study
Design
Young et al.
2a
2010
Clinician-led
task force
leads prevention
initiatives
Target safety
problem: PU
Key elements:
Clinician-led
task force;
skin champions;
adoption of
Save our Skin logo;
education/training;
revise policies and
procedures based
on CPGs; integrate
new documentation
and assessment
tools
Pre-post
Theory or
Logic
Model
Shared
governance
approach
(decisions
made
at point of
care)
Description
of
Organization
540-bed
acute care
facility
(3 campuses)
in Indiana
Contexts
Outcomes:
Benefits
External:NS
Organizational
Characteristics: NS
Culture:
A clinician-led skin care
team replaces
administration-led
Save our Skin (SOS)
logo adopted and
appears on educational
forms and t-shirts worn
during audits and
educational in-services
Laminated SOS logo
tool placed on doors of
patients at-risk
Revise online policies
and procedures; 1 new
policy remains
Adopt Braden Scale
(electronically)
Body map assessment
tool
Mandatory training
includes presentation
of case studies
Educational brochure
for residents and
families
Feedback on
educational
presentations
Updating of core
orientation, uploaded to
hospital Web site
Posters depicting PU
rates, examples of new
forms, and revised
policies/procedures
Length: 2 years
Process:
Members of clinician-led
task force include director of
Clinical Care and Oncology
Nursing Services, manager
of Wound Care Institute,
and nursing representatives
from 15 hospital units;
task force members appoint
skin champions from each
unit; champions invited to
join team; task force
members join
subcommittees of choice to
develop logo, policy and
procedures and other
program components;
after comparing practices to
CPGs 7 existing policies are
combined into 1; manager
of the Wound Care Institute
works with the Director of
Informatics on revising
online policies and
procedures; monthly quality
audits
Successes:
Revised policies reduced
from 7 to 1; documentation
of skin care reduced from
8 to 3
Barriers:
Time constraints,
insufficient computer
resources, competing goals
Clinicians were allocated
4 paid hours to carry out
responsibilities; web access
to library resources were
Incidence:
Pre:
Campus 1:
12.5%
Campus 2:
8.7%
Campus 3:
NR
Post:
Campus 1:
9.1%
Campus 2:
2.8%
Campus 3:
NR
D-119
Influence of
Contexts on
Outcomes
NS
Author/Year
Bales et al.
3
2009
Description of PSP
Implementation of
evidenced-based
prevention
strategies
Target safety
problem: PU
Key elements:
Increase hours of
certified wound,
ostomy and
continence nurses
(CWOCNs) to fulltime to provide
24-hour support to
staff, provide
mandatory
education and
resources for
staging and treating
wounds, increase
wound monitoring
and reporting
efforts, purchase
pressureredistribution beds,
add musical alarms
to remind nurses to
turn patients, and
identify at-risk
surgical patients.
Study
Design
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
300-bed
community
hospital, USA
Contexts
placed on each unit

Hospital-wide
standardizing of
patient-turning
schedules
Flip-chart algorithm
placed bedside to
differentiate between
old and new skin care
products
Audit tool
External:
To comply with principles of
the Magnet program, which
validates excellence in
nursing practices
Organizational
Characteristics:
Magnet hospital that serves
mostly adult and geriatric
patients
Culture:
Decentralized decisionmaking in which shared
decision-making
is encouraged prevails,
feedback and participation
of all staff is actively
encouraged
CWOCNs, computers to aid
in staging wounds and
treatment information, and
external alarms to remind
nurses to turn patients
added to intranet;
Sustainability:
RNs and LPNs must
demonstrate
competency annually
Monthly updates
provided via intranet to
nursing personnel by
unit champions/ team
members; includes
product changes
Length: 1 year
Process:
24-hour support
provided by CWOCNs
Mandatory education
Strict oversight of
monitoring and
reporting
Periodical motivational
campaigns that
included staff and unit
incentives.
Successes:
Patients received
optimal care
Institution avoided the
cost of treating stage 3
or 4 ulcers
Barriers:
Staff motivation and lack of
proper reporting and
documentation
D-120
Monthly to quarterly
campaigns are launched to
maintain staff motivation.
Nursing units that had zerohospital acquired PUs are
recognized and awarded
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Hospital
acquired
prevalence
rates:
Pre: 4.20%
Post: 0%
The hospitals
managerial
style
encouraged
staff
involvement
in decisionmaking about
the process of
developing a
program and
the leadership
team gave
strong
support to the
program and
promoted it to
both other
leaders in the
team and
hospital staff.
Author/Year
Description of PSP
Study
Design
Theory or
Logic
Model
Description
of
Organization
Contexts
during campaigns.
Sustainability:
Success requires
awareness of
key management skills and
priorities, such as
strong leadership,
involvement of staff in
decision-making and a
desire to foster
interdisciplinary
relationships.
D-121
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Author/Year
Description of PSP
Study
Design
Chicano &
Droishagen,
4
2009
Implement
strategies to lower
the incidence of
hospital-acquired
pressure ulcers
Target safety
problem: PU
Key elements:
Developed a
protocol for
assessment and
documentation of
wounds,
implemented
procedures for
CWOCN to work
with staff and
patients to initiate
appropriate
treatment,
implemented the
Braden Scale for
Pressure Sore Risk,
conducted a
literature review on
the use of
thromboembolic
device stockings
and compression
devices and revised
practice standards
for use of devices
based on findings of
review
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
25-bed
intermediate
care unit in
the
United States
Contexts
Outcomes:
Benefits
External:
Quarterly HAPU data
indicate increased
prevalence
Organizational
Characteristics: NS
Culture:
The care unit supported a
self-governing nursing
council.
Survey to identify practices
regarding skin assessment,
documentation, and
nursing intervention and
opportunities for education,
CWOCNs, and
Braden Scale
Length: 23 months
Process:
Phase 1 took 5 months and
involved developing
protocols and procedures to
assess and treat wounds;
Phase 2 took 3 months to
complete and involved
educating staff and
implementing the Braden
Scale,
Phase 3 took 15 months to
complete and involved a
literature review and
revision of practice
standards on use of
compression devices.
Successes:
Staff participation in survey,
continued adherence to
implemented prevention
practices, development of
educational materials, and
staff acceptance of shared
governance
Barriers:
Engaging staff as council
members in the planning
and implementation of the
project.
Updating staff of progress
and continual
encouragement to
participate from other
council members
Sustainability:
Commitment and diligence
from the quality improvement team and selfgovernance council.
Hospital
acquired
incidence
rate:
Pre:
6 occurrences
during a
12 month
period,
5 during
subsequent
5 months
D-122
Post:
1 occurrence
within
12 months
following
implementation, 0 at
latest
assessment
covering
2 month
period.
Influence of
Contexts on
Outcomes
Commitment
and diligence
from the
quality
improvement
team and
from the
members of
the staffs
selfgovernance
councils
played a
significant
factor in
achieving our
goal of
reducing
HAPU
prevalence in
our
intermediate
care unit.
Author/Year
Description of PSP
Study
Design
Walsh et al.
5a
2009
Implementation of
evidence-based
practices
Target Safety
problem: PU
Key elements:
revision of skin
management
program; use of
CPGs; educating
clinician/nurse;
multidisciplinary
team; add certified
WOCN to
management;
replace risk
assessment tool;
replace wound care
products
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
1 acute care
facility in
northwest CT
Bed size: 371
Contexts
Outcomes:
Benefits
External: CMS
Organizational
Characteristics:
Regional medical center
and community teaching
hospital; primary provider
for 350,000
Culture: NS
Add WOCN nurse to
team
Rely on EBPs
(AHRQ CPGs, IHI,
WOCN Society)
Rely on The National
Database of Nursing
Quality Indicators PU
presentation for reeducation on wound
etiology and staging
Clinician and staff
education (computerbased and classroom
presentations)
Replace risk
assessment tool with
Braden Scale
Update skin
management policies/
procedures
Assessed wound care
products
Multidisciplinary team
Alert system (POA
sticker)
Visual turning clocks,
laminated pocket cards
including CPG
information
Bed surface algorithms
Length: 18 months
Process:
Clinical education relies on
6 essential elements of
prevention; in 2007,
Braden scale risk
assessment tool replaces
current un-validated tool
form (not research based);
each unit assigns an
interdisciplinary team;
purchases of new beds,
stretchers and curtains
followed by new skin lotion
and incontinence care
products
Successes:
Reduction in prevalence;
increased focused
communication among
patient caregivers; buy-in
from clinicians improves
behavior
Barriers: NS
Addressing Barriers: NS
Sustainability:
Remains current regarding
initiatives for improved
patient safety, changes in
regulatory mandates, and
changes in EBP.
Prevalence:
Baseline:
12.8%
Postimplementation:
0.6%
D-123
Influence of
Contexts on
Outcomes
NS
Author/Year
Dibsie, L. 2008
Description of PSP
Implementation of
evidence-based
protocol and
practices for
preventing and
treating PUs
Target safety
problem: PU
Key elements:
Development of
protocol for
monitoring,
preventing, and
treating PUs based
on recent evidence,
standardization of
all products related
to prevention and
treatment of PUs,
and education for
nurses on the
protocol and use of
products.
Study
Design
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
4 adult critical
care units
(54 beds
total) at
2 academic
medical
centers in the
United States
Contexts
Illustrative wound
reference guides with
recommended
treatment modalities
Standardize wound
care products
External:
Two significant events
occurred and there was an
overall lack of reporting and
communication of issues
related to skin breakdown
(specifics of events not
reported)
Organizational
Characteristics: NS
Culture: NS
Staff nurse skin committee
and skin champions
D-124
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Length: >1 year

Process:
Change began with
becoming educated about
current practices and
equipment in wound care.
Once educated, the nursing
skin committee began
purchasing equipment,
developing procedures for
monitoring and
documenting skin
breakdown, and educating
staff on monitoring,
reporting, and treating PUs.
Successes:
Decrease in the rate of
hospital-acquired stage 2 or
greater pressure ulcers.
Barriers:
Coordinating efforts
between 2 sites
Coordinating and
identifying skin
committee members
and staff champions
Scheduling staff
education
Continuation of efforts
Cost of purchasing new
equipment
Communication, active
involvement of clinical
managerial leaders, and
Surgical ICU
acquired:
Pre: 6.1%
Post: 6.1%
Facility-wide
overall rate:
Pre:4.2%
Post: 3.2%
The changes
in the climate
and practice
related to skin
care and
prevention of
breakdown
are the direct
result of
nursing taking
ownership of
their practice
with the
support of
nursing
leaders at all
levels.
Author/Year
Description of PSP
Study
Design
Theory or
Logic
Model
Description
of
Organization
Contexts
constant education support.

Sustainability:
Organization commitment
remains strong and next
steps for success, such as
developing aggressive
indicators of success and
having staff identify practice
issues are in the works.
D-125
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Author/Year
Description of PSP
Study
Design
McInerney, J.
7
2008
To implement
multiple strategies
for decreasing the
prevalence of
hospital-acquired
PUs
Target safety
problem: PU
Key elements:
Electronic medical
records, risk
assessment
measures, pressure
relief measures
(new equipment
and personnel
augmentation, and
interdisciplinary
team to develop
protocols.
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
Two-hospital
system with
548 beds in
United States;
548 bed; nonprofit
Contexts
Outcomes:
Benefits
External: NS
Organizational
Characteristics: NS
Culture: NS
WOCN oversaw
implementation of
strategies.
Length:
18 months to implement
program; follow-up reported
for over 5 years
Process:
Electronic medical
records (EMRs) to
assess and document
skin care needs.
Risk assessment
measures (e.g., Braden
Scale)
Automated consults
and orders through
EMRs
Pressure relief
measures
Staff education
Hiring of second
WOCN
Successes:
Reduction in PUs, cost
savings and elimination of
pain and suffering for the
patients
Barriers: NS
Sustainability: NS
Hospital
acquired
prevalence
rates:
Pre:
12.8% all
PUs;
6.7% PUs on
heel.
Post
(4.5 years
after
implementation):
1.9% all PUs;
1.1% PUs on
heel.
D-126
Influence of
Contexts on
Outcomes
With the
assistance of
the
automated
consults and
orders, the
addition of
another WOC
nurse, the
appropriate
equipment,
the interdisciplinary
task force,
continuing
education,
and
monitoring,
the hospital
system was
able to reduce
the hospitalacquired
pressure ulcer
prevalence
rate by 81%,
and the rate
for heel ulcers
alone was
reduced by
90%.
Estimated
annual cost
savings:
$11,466,000
Author/Year
Description of PSP
Study
Design
Ballard et al.
8
2007
Implementation of
multiple strategies
in an intensive care
unit (ICU) to reduce
the rate of PUs
Target safety
problem: PU
Key elements:
Strategies included:
restructured risk
assessment and
documentation,
translated numeric
data into easy-tounderstand graphs
of PU rates,
increased staff
awareness,
implemented
turn rounds,
increased
prevalence
assessments and
redesigned
skin team, used
evidence-based
practices for
monitoring and
treating PUs, and
created an access
database to track
weekly prevalence
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
Two-unit ICU
with a total of
44 beds
located in
two separate
geographical
locations in
the
United States.
Contexts
Outcomes:
Benefits
External:
Joining the National
Database of Nursing Quality
Indicators (NDNQI) and
realizing that ICU had high
prevalence of PUs.
Organizational
Characteristics: NS
Culture:
Primary nursing model
CWOCN conducted a
needs assessment, creation
of user friendly reports to
show rate of PUs, posted
data of PU rates for staff to
see, skin teams (consisted
of nursing staff who
performed weekly
prevalence assessments
and provided education),
and Access database
Length: 1 year
Process:
Conducted needs
assessment to identify
areas of weakness in
identifying, monitoring,
treating, and reporting PUs;
made revisions to protocol
based on results of needs
assessment; created user
friendly reports to display
PU rates; increased staff
awareness through
displaying PU rates and
providing education;
implemented turn rounds
every two hours;
redesigned skin team,
implemented evidencebased practices to assess
risk and monitor PUs, and
implemented Access
database to track PUs.
Successes:
Reduction in rate of PUs
and improved patient
outcomes.
Barriers: NS
Sustainability:
Staff commitment to
implementing strategies and
maintaining quality care.
Hospital
acquired
incidence
rate: NS
Hospital
acquired
prevalence
rates:
Pre: 34%
Post: 8.0%
D-127
Influence of
Contexts on
Outcomes
Utilizing
benchmark
data helped
the ICU focus
on pressure
ulcer
prevention,
which led to
improved
patient
outcomes.
Author/Year
Description of PSP
Study
Design
Catania et al.
9
2007
Design and
implementation of
the Pressure Ulcer
Prevention Protocol
Interventions
(PUPPI): a nursing
initiative to prevent
PUs
Target safety
problem: PU
Key elements:
The PUPPI involves
assessing risk and
nutritional status,
providing skin care,
documenting, and
giving referrals
as needed.
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
All 5 inpatient
units in one
hospital in the
United States.
Units included
2 medical
units,
2 surgical
units, and
one critical
care unit.
Contexts
Outcomes:
Benefits
External:
2 stage IV ulcer identified;
evidence from the NDNQI
survey that the prevalence
of PUs in the hospital in the
study exceeded the national
benchmark by close to
50%.
Organizational
Characteristics:
Dedicated cancer hospital
Culture: NS
Quality improvement team
that consisted of a quality
manager, nursing director,
certified nurse aids (CNSs),
nursing staff developmental
specialists, and an
enterostomal therapy nurse
to develop and implement
protocol.
Length:
6-months to implement;
follow-up data reported for
18 months
Process:
Initial efforts started in 2003
and involved having clinical
nurse specialists assess
patient risk using the
Braden Scale. These efforts
led to the development of a
quality-improvement team
in 2004 and the
development of the
PUPPIs. The PUPPI was
implemented in
September 2004 and
included a systematic
process for monitoring and
educating staff.
Successes:
Reduction in rates of PUs
Barriers: NS
Sustainability:
Proactive nursing staff who
have adopted initiatives in
protocol into their daily
routine.
Hospital
acquired
incidence
rate: NS
Hospital
acquired
prevalence
rates:
Pre:
11.11% all
ulcers;
6.67%
hospital
acquired
Post:
4.08% all
ulcers;
1.36%
hospital
acquired
D-128
Influence of
Contexts on
Outcomes
While the
unit CNSs
have
championed
this process
and continue
to monitor the
program, it
has been the
nursing staff
who have
embraced
evidencebased nursing
practice and
brought it to
the bedside
by adopting
the initiative
into daily
practice.
Author/Year
Description of PSP
Study
Design
LeMaster, K.
10
2007
Pressure Ulcer
Prevention Project
implemented on
targeted units
pulmonary unit and
oncology unit
Target safety
problem: PU
Key elements:
Turning at-risk
patients every
two hours
minimally, placing a
pressure-reducing
overlay on the bed
of every patient at
risk, and elevating
bony prominences
at risk
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
Pulmonary
and oncology
unit of the
largest
hospital
campus
within one
healthy
system;
a 502-bed
hospital in the
United States.
Contexts
Outcomes:
Benefits
External:
Selection of study units was
based on unit having a
higher hospital-acquired PU
rate than the NDNQI
database mean for similar
units and having higher
hour-of-care ratios than the
NDNQI mean for similar
units.
Organizational
Characteristics: NS
Culture: NS
Manual containing
information about wounds
and wound care,
instructions on the use of
the Braden Scale to assess
risk for developing PUs,
patient turn schedule, and
cues to use as reminders to
turn patients.
Length:
Summer 2004April 2005
Process:
The first phase of
implementation involved
assessing and establishing
baseline knowledge of unit
staff nurses for assessing
risk. Staff then identified
resources and studied the
manual. Nursing staff began
assessing and documenting
risk and implementing other
aspects of the protocol
(placing pressure-reducing
overlay on bed). A CNS
provided consultation and
oversight throughout
implementation period.
Successes:
Reduction of PUs in
targeted units and
successful duplication of
intervention in other medical
units.
Barriers:
Braden scores were not
documented at 100% per
policy. Patients were
missed because of failure in
communication between
two different electronic
documentation systems.
Barrier to be eliminated with
transition to a single,
universal electronic record
system within hospital.
Sustainability:
Manual and cues to help
maintain consistent and
complete practice patterns.
Hospital
acquired
incidence
rate: NS
Hospital
acquired
prevalence
rates:
Pre
Pulmonary
Unit: 9.0%
Post
Pulmonary
Unit: 0.0%
Pre Oncology
Unit: 12.0%
Post
Oncology
Unit: 0.0%
D-129
Influence of
Contexts on
Outcomes
NS
Author/Year
Description of PSP
Study
Design
Courtney et al.
11
2006
To develop and
implement Save
Our Skin program to
reduce the rate of
PUs
Target safety
problem: PU
Key elements:
Updating pressurerelieving mattress,
introducing skin
breakdown
prevention
protocols, clarifying
staff roles and
responsibilities
(introduced a skin
champion), and
improving
measurement a
communication of
PU performance
data
Time
series
Gibbons et al.
12
2006
To develop and
implement best
practice guidelines,
known as the
SKIN bundle
(Surfaces, Keep the
patient turning,
Incontinence
management,
Time
series
Theory or
Logic
Model
Used
procedures
of the
Six Sigma
method, a
datafocused,
decisionmaking
process
that utilizes
a five
phase
process
called
DMAIC:
Defining
the
problem,
Measuring
the
performance,
Analyzing
the
problem,
Improving
the
situation,
and
Initiating
change.
NS
Description
of
Organization
710-licensed
bed, multisite,
not-for-profit
facility that
serves a
37-county
area; is
Magnet
designated
Contexts
Outcomes:
Benefits
External:
Results using the Nursing
Care Quality Initiative
guidelines that revealed
high prevalence of PUs
(13%) and lack of
documentation and
management. Revitalized
interest in treatment and
prevention shown by
American Nurses
Association and AHRQ in
developing new guidelines.
Organizational
Characteristics:
Magnet designated hospital
Culture: NS
Implemented guidelines for
the prevention and
management of PUs from
the Wound, Ostomy, and
Continence Nurses Society
and assessed risk using the
Braden Scale.
Hospital
acquired
incidence
rate:
Pre: 9.4%
Post:
st
1 quarter
implementation 3.1%;
last follow-up
1.8%
Hospital
acquired
prevalence
rates: NS
Large
528-bed
hospital, part
of nations
largest
Catholic and
non-profit
health system
External:
Development of the SKIN
bundle is part of the
Ascension Health Care
systems initiative to
reduce/eliminate
preventable hospital-related
injuries and deaths.
Organizational
Length:
Follow-up 3 years
Process:
Adopted Six Sigma
procedures, assessed
potential causes of high
incidence of PUs and lack
of staff coordination and
management of PUs, and
introduction of solutions:
staff training and
awareness, development
and implementation of
Skin Breakdown Prevention
protocol, replacement of
pressure mattresses,
designation of
Skin champion,
clarification of staff roles,
and implementing
monitoring procedures
Successes:
Reduced incidence of PUs
and cost savings
Barriers: NS
Sustainability:
Defining staff
responsibilities, monitoring
performance, using data to
inform staff performance,
and making data public
Length:
10 months to implement;
follow-up 2 years
Influence of
Contexts on
Outcomes
This project
refocused
efforts on
traditional
direct nursing
care and
problem
solving
procedures
from the
Six Sigma
method to
implement the
Save Our
Skin program.
Hospital
acquired
incidence
rate:
Pre: 5.7%
Post: 0.448
No new
Stage III or IV
HAPU
Of eight
priorities
identified for
action by
Ascension
Health;
St. Vincents
Medical
Center was
D-130
Process:
Began with engaging
leadership, forming an
interdisciplinary team, and
providing protected time to
Author/Year
Description of PSP
Nutrition) to prevent
PUs
Target safety
problem: PU
Key elements:
Developed a
synergistic group
of interventions that
includes appropriate
surface selection
(e.g., pressure
mattress), regular
turning of patients,
incontinence
management,
nutrition and
hydration, ongoing
monitoring and staff
training.
Study
Design
Theory or
Logic
Model
Description
of
Organization
Contexts
Outcomes:
Benefits
Characteristics:
Faith-based, non-profit
hospital
Culture: NS
Regular assessment and
documentation on flow
sheets, skin risk alert
reminders, weekly team
meetings, and ongoing
performance monitoring and
reporting.
work on project. The project

moved toward identifying
best practices, assessing
current practices, and
developing the SKIN
bundle. Lastly, the project
involved educating staff and
piloting the SKIN bundle.
Successes:
90% reduction in incidence
of PUs.
Barriers:
Educating staff,
communication, motivation,
and hard- to-treat patients
(patients whose treatment
involves hours of sitting or
lying down, such as
radiology or dialysis)
Keep educational offerings
basic, short and focused,
and available at multiple
times; make sure key staff
organizing initiative have
good communication skills
and plan for times and
methods of communication,
celebrate successes and
provide tangible incentives,
make a plan for hard-totreat patients.
Sustainability:
Being open to suggestions
from staff, continually
focusing on education,
monitoring outcomes, and
promoting free exchange of
information.
occurred
between
August 2004
and February
2006
D-131
Influence of
Contexts on
Outcomes
selected to
develop the
PU process.
The hospital
leadership
welcomed
the
opportunity to
develop this
nursing-driven
program as a
means of
establishing
pride in
professional
nursing
practice.
67 acute care
facilities in the
Ascension
health system
agreed to
implement the
SKIN bundle
plus common
measures of
quality and
performance.
Author/Year
Description of PSP
Study
Design
Hiser, et al.
13
2006
To implement a
team approach to
performance
improvement and
develop an
education plan for
clinical staff to
better prevent and
treat PUs
Target safety
problem:
PU
Key elements:
Education,
policy changes,
development of
evidence-based
protocols, cost
improvement
strategies,
implementing new
support surfaces,
and improved
reporting and
monitoring through
quarterly
prevalence studies
and improved risk
assessment using
the Braden scale.
Time
series
Theory or
Logic
Model
NS
Description
of
Organization
580-bed
regional
medical
facility in the
U.S.
Contexts
Outcomes:
Benefits
External: NS
Organizational
Characteristics: NS
Culture: NS
Created a Wound Care
Team that consisted of
CWOCNs and an advanced
registered nurse practitioner
to implement changes and
educate staff; replaced the
Norton Scale with the
Braden scale to assess risk.
Length:
2 years follow-up.
Process:
Implementation started with
a review of the literature of
best practices for
prevention and treatment of
PUs.
Successes:
Reduced prevalence of PUs
and annual cost savings.
Barriers: NS
Sustainability:
Ongoing education and
newsletters reporting
progress and positive
feedback to staff.
Hospital
acquired
prevalence
rates:
Pre: 9.2%
Post: 6.6%
(measured at
2 years
follow-up)
D-132
Influence of
Contexts on
Outcomes
NS
Author/Year
Description of PSP
Study
Design
Lyder et al.
14
2004
To implement a
multihospital
collaboration to
increase the
identification of
patients at high risk
of PUs and to
promote the use of
preventive
measures among
hospitalized
Medicare patients
Target safety
problem: PU
Key elements:
Increase in the
following: tracking
of PUs,
performance and
documentation of
risk assessment,
use of prevention
protocol (includes
education and
oversight of staff),
scheduled
repositioning, use of
pressure-reducing
devices, nutritional
consults, and
accuracy of staging.
Pre-post
Theory or
Logic
Model
Plan-DoStudy-Act
(PDSA)
Description
of
Organization
17 hospitals
ranging from
200 to 800
beds with
9 located in
urban and
8 in rural
settings in the
U.S.
Contexts
Outcomes:
Benefits
External:
In response to Centers for
Medicare & Medicaid
Services charge to improve
quality of care to Medicare
patients.
Organizational
Characteristics: NS
Culture:
Hospitals needed to
develop a team approach to
implementing changes.
Qualidigm, the Connecticut
QIO,
Computerized charting
system for tracking PUs,
creation of skin care task
force, and pocket-sized
wound staging card.
Length:
9 months implementation,
2 years follow-up.
Process:
The PDSA framework
involved 1) identifying
problem to be changed and
designing an intervention;
2) implementing this
intervention; 3) evaluating
the impact of the
intervention, and
implementing what was
learned from evaluation.
Successes:
Significant increases in
identifying high-risk
patients, repositioning bed
and chair bound patients,
use of nutritional consults,
and staging of acquired
Stage II or greater PUs.
Barriers:
View that PU prevention
was a nursing issue.
Re-educating various
disciplines about their role
in PU prevention
Sustainability:
Hospitals found that the
most sustainable
interventions were
institutionalized, such as
change in pressurerelieving mattress.
Interventions that depended
more on sufficient staff
such as turn schedules
were less sustainable.
Hospital
acquired
incidence
rate:
No
statistically
significant
change from
baseline to
follow-up
(20.6 to 20.8,
p = 0.90)
Hospital
acquired
prevalence
rates: NS
Decrease in
median length
of hospital
stay (8.0 days
to 7.0 days,
p = 0.05)
D-133
Influence of
Contexts on
Outcomes
Focusing
pressure ulcer
prediction and
prevention
programs on
the nursing
staff is limited
insofar as
effective
pressure ulcer
prevention
requires a
multidisciplina
ry effort. The
PDSA model
assists
hospitals in
working in
multidisciplina
ry teams and
places the
onus for
improvement
on the team
rather than on
a particular
discipline.
Author/Year
Stier et al.
15
2004
Description of PSP
Study
Design
Theory or
Logic
Model
NS
Description
of
Organization
A large notfor-profit
health care
system in the
U.S. with over
5,600 beds
and more
than 33,000
employees.
The system is
composed of
18 hospitals,
4 skilled
nursing
facilities,
1 certified
home health
agency, and
2 hospice
agencies.
The focus of
current study
is on
implementing
skin care
initiatives in
acute care
hospitals.
To implement a
Time
system wide
series
multidisciplinary
skin care initiative to
standardize care to
reduce the
incidence and
severity of PUs
Target safety
problem: PU
Key elements:
The project involved
standardization of
risk assessment
methods,
delineating
timeframes for
patient assessment,
and reassessment,
developing a
uniform skin care
formulary,
negotiating a
system-wide
contract for
therapeutic support
surfaces, and
providing staff
education.
a
Not included in the Soban 2009 review16
CMS:
Centers for Medicare and Medicaid Services
CNS:
Clinical nurse specialist
CPG:
Clinical practice guidelines
CWOCN:
Certified Wound, Ostomy, and Continence Nurse
DMAIC: Defining, measuring, analyzing, improving, initiating change
EBP:
Evidence-based practice
EMR:
HAPU: Hospital-acquired pressure ulcer
ICU:
Intensive care unit
IHI:
LPN:
Licensed practical nurse
Contexts
Outcomes:
Benefits
External: NS
Organizational
Characteristics: NS
Culture: NS
Implemented Braden scale
to standardize risk
assessment
Length: 2 years follow-up

Process:
Convened a
multidisciplinary team of
experts to develop an
implementation plan. The
first initiative implemented
was the Braden scale of risk
assessment. The second
was working closely with
Materials Support Services
to develop a formulary for
skin care products. The final
steps involved staff
education and implementing
quality control measures.
Successes:
Reduction in the inpatient
incidence of PUs.
Barriers: NS
Sustainability:
Valid and reliable
measurement system that
allows for ongoing
assessment and evaluation
of performance and ongoing
education.
Hospital
acquired
incidence
rate:
Pre: 2.2%
Post: 1.3%
D-134
Influence of
Contexts on
Outcomes
A
standardized
approach to
patient
assessment/
reassessment
through the
use of
evidencebased
guidelines
and
educational
programs led
to a common
understanding
of pressure
ulcer
management,
improved
communicatio
n among care
providers, and
sustained
improvement
in patient
outcomes.
MICU:
NDNQI:
NR:
NS:
PDSA:
POA:
PSP:
PU:
PUPPI:
QIO:
RN:
SKIN:
SOS:
WOC:
WOCN:
Medical intensive care unit

National Database of Nursing Quality Indicators
Not reported
Not stated
Plan-Do-Study-Act
Present on admission
Patient safety practices
Pressure ulcer
Pressure Ulcer Prevention Protocol Intervention
Quality Improvement Organization
Registered nurse
Surfaces, Keep the patient turning, Incontinence management, Nutrition
Save our skin
Wound, ostomy and continence
Wound, Ostomy and Continence Nurses Society
D-135
Table 2, Chapter 21. Multi-component pressure ulcer prevention initiatives conducted in long-term care settings in the United States
Author/
Year
Description
of PSP
Study
Design
Horn et al.
17
2010
Real-Time
Time series
Program
(renamed
On-Time Quality
Improvement for
Long Term Care
[On-Time])
Target safety
problem: PU
Key elements:
CNA assist in
redesigning
documentation to
include core data
elements to help
identify high-risk
patients;
facilitators
provide feedback
on weekly clinical
decision-making
reports; staff
educated on QI
methods and
smooth
integration of
these CNA
documentation
and clinical
reports into dayto-day flow
Theory or Logic
Model
Description of
Organization
Based on best
practices from
AHRQ and AMDA
guidelines, and
findings from the
National Pressure
Ulcer Long-term
Care Study
(NPULS)
11 not-for-profit
facilities in 7 states
Bed size:
44432 beds
13 highest-risk
units per facility
participated
Contexts
Influence of
Contexts on
Outcomes
External: AHRQ
Length: 9 months
CMS HRPrU QM
Facility B which
funded
Process:
prior to
had the highest
Organizational
implementation
Facilitators work
reduction in PU
Characteristics: NS
(k = 7): 13.0%
with a
(-82.4%) was the
Teamwork,
CMS HRPrU QM
multidisciplinary
only facility that:
Leadership,
12 months after
team from each
had 100%
implementation
Culture: NS
facility.
participation of
Implementation tools: Redesigned CNA
(k = 7): 8.7%
residents
HRPrU QM %
documentation
CNA
(n = 75)
incorporating core change
documentation
Facility B was 1 of
(5 facilities using
data elements
processes and
3 facilities who
2 reports)
including nutrition
timely reports to
incorporated all 4
-25% to -82.4%
identify patients at and incontinence
clinical reports for
High Risk PrU QM care planning.
variables.
risk
% change
CNAs coached to
Two facilities with
A project leader
(2 facilities using 1 the lowest reduction
improve
(e.g., DON) and
report)
documentation.
in PUs did not
ongoing team
Sites fax scannable +8.3%, +14.3%
involve a
identified
Average number of multidisciplinary
forms to project
Educate staff on
in-house acquired team.
office.
QI methods and
PU (all stages) per
Clinical reports
use of
facility prereturned
within
documentation
implementation
forms and reports 24 hours and
vs postdisplayed.
Feedback includes implementation:
inconsistencies and 12.1 to 4.6
(62% reduction)
completeness of
Average number of
CNA
documentation per CNA
documentation
unit/unit over
forms reduced by
time/shift.
After reviewing with 53.2%.
CNAs, need for
additional
education noted.
Conference calls
(bi-weekly), allfacility meetings
(every 6 months)
and on-site
D-136
Implementation
Details
Outcomes:
Benefits
Author/
Year
Description
of PSP
Study
Design
Theory or Logic
Model
Description of
Organization
Contexts
Implementation
Details
meetings were
scheduled with
facilitators, project
leaders and
frontline staff.
Successes:
CNAs widely
accept revised
forms and increase
productivity.
Documentation
completeness rates
increase from
80%90% to
mid-90%.
Barriers:
EMR system used
by 1 facility could
only export data
elements and
create 1 report
Issues raised with
preparing the CNA
documentation
forms needing
the residents
study ID number
and
faxing forms for
report generation
Staff turnover
especially by DON
slowed project
momentum.
Addressing
Barriers:
Add new CNA
documentation
process into
orientation
programs
D-137
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Author/
Year
Rantz et al.
18
2010
Description
of PSP
Bedside EMR
(OEMR, Irvine,
CA) and
statewide on-site
clinical
consultation
services
(QIPMO
Quality
Improvement
Program for
Missouri)
Target safety
Study
Design
RCT 4-group
comparison
Group 1:
EMR plus
consult
Group 2:
EMR
Group 3:
Consult
Group 4:
Control
Theory or Logic
Model
NS
Description of
Organization
18 facilities in 3
U.S. states
Group 1:
4 facilities
Bed size range,
98240,
total 668
Group 2:
4 facilities
Bed size range,
105218,
total 635
Group 3:
Contexts
Implementation
Details
Outcomes:
Benefits
Phase in use of
documentation.
Develop a strong
multidisciplinary
team to lead
improvement
efforts and not
rely on one
project leader.
Sustainability:
HIT needed to
capture CNA
documentation and
generate reports.
Managing the
manual data
collection, faxing
forms to the project
office and creating
clinical reports for
distribution back to
the facilities on a
weekly basis could
not be maintained
over the long term
for many facilities.
Program expanded
throughout the U.S.
External: CMS funds Length: 2 years
Relative
OEMR hardware,
improvement in
Process:
software and
high risk pressure
ongoing tech support Project
sores (negative
Organizational
scores indicate
coordinator
Characteristics:
improvement)
works with
12 months
Mix of for-profit,
OEMR staff
not-for-profit, and
Staff works with Group 1: -53%
Group 2: -12%
governmental
QIPMO nurses
facilities
at least monthly Group 3: -5%
Teamwork,
Group 4: +435%
QIPMO nurses
Leadership, Culture:
encourage staff 24 months
NS
Group 1: -3%
to focus on
D-138
Influence of
Contexts on
Outcomes
Total costs for the

3-year evaluation
for the groups of
facilities
implementing
technology
increased $15.11
(12.5%) for Group 1
and $16.89 (9.6%)
for Group 2, while
those for the
comparison groups
did not.
Author/
Year
Description
of PSP
problem:
Comprehensive
Key elements:
Mandatory
OEMR training,
QIPMO nurses
Study
Design
Theory or Logic
Model
Description of
Organization
Contexts
5 facilities
Bed size range,
90123,
total 543
Group 4:
5 facilities
Bed size range,
120310,
total 890
Group 1, 3, 4 from
Missouri
Group 2:
Other States
clinical care and
improving care
Project
systems to be
coordinator
enabled by
assigned at
OEMR
intervention
Successes:
facility
Group 1, 2 and 3
QIPMO nurses
showed
improvements at
12 months;
Group 1 and 2
sustained at
24 months
Barriers: NS
Addressing
Barriers: NS
Sustainability:
Improvement
sustained during
Year 2 for Group 1
and 2
D-139
Implementation
Details
Outcomes:
Benefits
Group 2: -8%
Group 3: +59%
Group 4: +105%
Influence of
Contexts on
Outcomes
Cost increases
were most likely
attributable to the
cost of technology,
maintaining and
supporting the
technology, and ongoing staff training
to use the EMR and
not increase direct
care staffing or
turnover.
Author/
Year
Description
of PSP
Study
Design
Milne et al.
19
2009
LTACH care
Time series
process
improvement
Target safety
problem: PU
Key elements:
Nursing
association
consults; team
training; improve
assessment and
documentation
methods; EMR
revised; formal
and informal staff
education;
wound care
product reviews
Theory or Logic
Model
Description of
Organization
Contexts
Failure mode and

effects analysis*
Long-term acute
care facility in CT
Bed size, 108
External: NS
Organizational
Characteristics:
Above average PU
prevalence
Teamwork,
Faulty EMR
Inconsistent use
of EMR by
clinicians
Deficient risk
assessment
documentation
Training by
nursing
association
APN appointed
in-house leader
APN and nursing
supervisor
become WCC
Team clinicians
trained in
prevalence data
collection
EMR revised;
PUSH tool added
Staff educated via
formal clinical
rounds, interactive
sessions and oneon-one bedside
sessions
Immediate
feedback given on
training
D-140
Implementation
Details
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Length: 13 months Mean facilityData on PU
facility wide
acquired PU
prevention
Process:
prevalence:
implementation in a
LTACH is spare.
Roles for new
Pre: 41%
Two LTACH units
skin team
Post: 4.2%
members
Pulmonary-focused however were able
to reduce PUs to
defined
unit:
<3% due to
Team meets
Pre: 25%
increased
weekly to review Post: <3%
diligence by the
failure modes
SCI/trauma unit:
team.
and develop new Pre: 33.8%
The authors noted
care processes
Post: 2.9%
an increased
Revamping of
collaboration among
policies and
disciplines with
procedures after
regard to wound
review of CPGs
prevention and
Wound care
treatment as well as
product reviews
a tendency for early
Successes:
intervention when
PU reduced to <3%
wounds are newly
on two units due to
discovered.
increased
monitoring of
modified nasal
cannula (pulmonary
unit) and increased
attentiveness to
heel offloading,
support surfaces
and proper
positioning
(SCI/trauma unit);
of the 396 charts
reviewed, <1% had
missing data;
staging and wound
etiology were
consistently
determined by
wound team in
greater than 90% of
Author/
Year
Description
of PSP
Study
Design
Theory or Logic
Model
Description of
Organization
Contexts
Implementation
Details
cases (based on a
review of 45 patient
charts)
Barriers:
Rates climbed once
strict monitoring
was leveled off
Addressing
Barriers:
Increase in unit
presence, chart
monitoring,
feedback to staff,
and biweekly
prevalence rounds
Sustainability:
CWCN
certification of 2
team members
provide in-house
expertise
Monthly review
of documentation
and PU
prevention
interventions
Early
intervention
D-141
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Author/
Year
Description
of PSP
Study
Design
Tippet A.
20
2009
Physician
Time series
consultant leads
deficient nursing
home to zero
facility-acquired
PUs
Target safety
problem: PU
Key elements:
Physician wound
consultant,
multidisciplinary
team, education,
weekly informal
feedback, wound
care protocols
based on AHRQ
CPG, wound
coordinator
sustains program
Theory or Logic
Model
Description of
Organization
Contexts
Based on AHRQ
CPG
Midwest skilled
facility
Bed size: 151
External: G-level
citation (actual harm
deficiency) and state
survey deficiencies
Organizational
Characteristics: NS
Teamwork,
NS
Physician
consultant
Multi-disciplinary
team
Braden Scale,
AHRQ CPG
Incentive
programs
Informal feedback
Simplified wound
care formulary
Equipment
evaluation (Delphi
process used to
evaluate
products)
D-142
Implementation
Details
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Length: 6 years
Average preEstimated cost
Process:
initiative incidence: savings per PU/per
5.19%
month:
Physician
Average post$1,617
consultant
initiative incidence: Monthly savings:
educates staff
0.73%
$10,187
and conducts
Yearly savings:
yearly follow-up (p<0.0001)
4 year post>$122,000
training (all
initiative incidence:
mandatory)
0.06%
Team forms
goals and meets (p<0.0001)
weekly
Select members
conduct wound
rounds
Follow-up
training through
in services, and
yearly follow-up
Nursing
supervisors
conduct one-onone with staff
and weekly
informal
feedback
Preventive care
plans created
Protocols
discussed in
classes, become
part of routine
shift reporting
and charting
All nursing staff
made
accountable for
care and
reporting
Successes:
Goal of zero facility
Author/
Year
Description
of PSP
Study
Design
Theory or Logic
Model
Description of
Organization
Contexts
Implementation
Details
acquired ulcers
reached after
6 months
Facility citation free
Accolades from
surveyors for
wound program
Judged
competitions
between floors
promote teamwork
and buy-in
Barriers: NS
Addressing
Barriers: NS
Sustainability:
Wound care
coordinator position
established to
supervise, train,
provide clinical
support and track
wounds.
Permanent decline
after 6 months
through study end
D-143
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Author/
Year
Description
of PSP
Study
Design
Theory or Logic
Model
Rosen et al.
21
2006
Ability,
Incentives, and
Management
feedback (AIM
system)
Target safety
problem: PU
Key elements:
Staff ability
enhancement
(skin care
training, use of
penlights and
TAP card),
real-time
management
feedback,
financial
incentives
Longitudinal
NS
time series
study; four
12-week
periods
(baseline
assessment,
intervention,
and two postintervention
periods)
Description of
Organization
Contexts
Implementation
Details
Outcomes:
Benefits
Not-for-profit
nursing home in
U.S.
Bed size, 136
External: AHRQ
funded
Organizational
Characteristics:
Received multiple
Department of
Health citations due
to persistently high
PU rates
Teamwork,
Lack of management
to oversee earlier
processes
Implementation
tools:
Research team
contacts
administrators
responsible for
overseeing
implementation.
Mandatory
skin care
training (a
40-minute
computerbased,
interactive-video
education
program).
Penlights
Caregivers wear
plastic TAP
(turn and
position card) to
remind all
hospital
personnel the
direction
residents should
Length: 48 weeks
Process:
One skin care
nurse assessed
patients upon
admission or
notification by staff
of any skin
changes.
During the postintervention
periods, no weekly
reports were
provided to the
administrators, no
established targets
or goals were
established, and
there were no
financial incentives
offered to staff.
Only 3 of 29 new
hires completed
training.
Sustainability:
The intervention
was not sustained
over the two postintervention periods
however Rosen et
al. indicated that a
highly motivated
administrator could
have maintained
the 3 program
components.
Significant
reduction in
emergence of
stage 14 PUs
Pre-intervention:
28.3%
Intervention:
9.3%
(z[I] = 2.64,
p<0.001)
Total ulcers
Stage 1 and
beyond
Pre-intervention
(n = 134):
38% (28.3)
Intervention
(n = 107):
10% (9.3)
Post-Intervention I:
19% (17.7)
Post-Intervention II:
19% (17.7)
Total ulcers
Stage 2 and
beyond
Pre-intervention:
31% (23.1)
Intervention:
10% (9.3)
Post-Intervention I:
15% (14.0)
Post-Intervention II:
17% (15.9)
D-144
Influence of
Contexts on
Outcomes
With a mean cost of
$2700 of treating a
single stage II PU,
[26] reducing the
incidence of these
ulcers by
approximately 15
over 12 weeks
yields a potential
savings of more
than $40,000 while
distributing less
than $10,000 as
incentives. This
does not take into
consideration the
added savings in
fewer personal
injury lawsuits.
The primary
management
feedback tool was
adherence to the
mandated training
(not emergence of a
new PU). Additional
real-time feedback
was provided to
staff in the form of a
visual
thermometer of
PU occurrences
each week. All a
nonfinancial
incentive, it served
as a supplementary
motivating factor as
the incidence of
PUs was visually
perceived as
declining.
Author/
Year
Description
of PSP
Study
Design
Theory or Logic
Model
Description of
Organization
Contexts
D-145
be facing every
2 hours.
Administrators
receive a
weekly report of
staff that had
completed
training.
A graphic
thermometer
of PU incidence
was also
updated weekly
and displayed in
the staff lounge.
Each staff
member
received $75
if the PU
incidence was
below target
goal (incidence
<3%) set by
administration.
Staff
reprimanded for
non-completion.
Staff terminated
for not
completing
training during
extension
period.
Implementation
Details
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Author/
Year
Description
of PSP
Study
Design
Abel et al.
22
2005
Process of care
Pre-post
system changes
in collaboration
with a state QIO
Target safety
problem:
PUs
Key elements:
Collaborative
with a state QIO,
intervention tool
kit, nurses aid
and licensed staff
training
Theory or Logic
Model
Description of
Organization
Contexts
Implementation
Details
Outcomes:
Benefits
NS
20 facilities in
Texas
Average residents:
100
Average Medicare
beds: 15
External: Identified
from 143 Medicarecertified skilled
nursing facilities as
having high rates of
PUs and a high
volume of residents
receiving preventive
care
Organizational
Characteristics:
Selected due to
accessibility to state
QIO (Texas Medical
Foundation [TMF])
Teamwork,
NS
TMF provides
tools
o Nurses Station
Reference
Cards
o Pocket
Assessment
Card
o Mobility
Program
o Fax
Communication
Form
o Care Planning
Tool
o Resident
Patient and
Family
Education
Brochure
Tool kit components
based on information
Length: 2 years
Process:
Monthly onsite
visits by TMF
Tools modified
Periodic
progress
assessment
Successes:
Performance
significantly
improved on 8 of
12 QIs
Management
maintains
autonomy which
promoted
continued
commitment and
a sense of
ownership
Barriers:
Staff resistance
Staff turnover
and variation in
new staff
orientation often
contributed to
clinical or
operational
practices that
were
inconsistent with
their protocol
requirements.
Incomplete risk
assessments
Monitoring
systems not
appropriately
used
Incidence rate:
Pre: 13.6%
Post: 10.0%
Significant
improvements in
8 QIs (baseline vs.
re-measurement):
Proportion of
residents with
appropriate risk
assessment
completed within
2 days of
admission
(2.2% vs.
15.3%;
p<0.0001)
Proportion of
high-risk
residents with
appropriate care
plan for ALL
selected triggers
for high-risk
residents
(10.1% vs.
21.8%;
p<0.0001)
Proportion of
high-risk
residents whose
care reflects the
triggered care
plan
interventions
(2.0% vs. 9.8%;
p<0.0001)
Proportion of
residents with
PUs that receive
weekly skin
D-146
Influence of
Contexts on
Outcomes
Although there are
areas for
improvement, the
implementation of
process of care
system changes by
NHs in a
collaborative
relationship with a
QIO may yield
improvements in
measures of patient
outcomes (e.g., PU
incidence).
Abel et al. also
indicated that the
10 facilities with the
highest [QI] scores
at re-measurement
showed a trend
toward a lower [PU]
incidence rates than
the 10 facilities with
the lowest [QI]
indicator scores at
re-measurement
(S = 125.5,
p = 0.07).
Facilities with the
highest QI scores
versus facilities with
the lowest QI
scores (baseline vs.
re-measurement;
PU incidence rate,
%):
High scoring group:
12.3% vs. 7.7%
Low scoring group:
14.8% vs. 12.2%
Facilities with the
Author/
Year
Description
of PSP
Study
Design
Theory or Logic
Model
Description of
Organization
Contexts
Implementation
Details
Outcomes:
Benefits
from the AHRQ

CPGs, Rhode Island
Quality Partners, and
regulatory
requirements (federal
and state)
Nursing staff
internally
responsible
TMF externally
responsible
QA committee
Documented risk
factors not acted
upon
Addressing
Barriers: Monthly
visits by TMF and

improving
performance
Sustainability: NS
D-147
Influence of
Contexts on
Outcomes
assessments
greatest
(12.6% vs.
improvement versus
32.8%;
facilities with the
p<0.0001)
least improvement
in QI scores
Proportion of
facility-acquired (baseline vs. reand community- measurement; PU
incidence rate, %):
acquired PUs
with appropriate High scoring group:
ulcer description 13.1% vs. 7.1%
within 24 hours Low scoring group:
14.0% vs. 12.8%
of ulcer
recognition
(53.5% vs.
68.9%;
p = 0.035)
Proportion of
residents with
PUs and mobility
issues using a
pressure relief
mattress/overlay
(50.7 vs. 76.7;
p<0.0001)
Proportion of
residents
identified as
high risk (as per
MDS) using a
pressure relief
mattress/overlay
(33.0% vs.
53.4%;
p = 0.003)
Proportion of
residents whose
treatment orders
and care plan
interventions for
PUs reflect
facility wound
Author/
Year
Rantz et al.
23
2001
Description
of PSP
Statewide
implementation
of Show-Me QI
report
Target safety
problem:
Comprehensive
Key elements:
Workshops,
Minimum Data
Set (MDS)
Quality Indicator
(QI) feedback
reports, clinical
consultation
Study
Design
Theory or Logic
Model
RCT
NS
Group 1:
Workshop
plus feedback
reports
Group 2:
Workshop,
feedback
reports and
clinical
consult
Group 3:
Control
Description of
Organization
87 nursing homes
in Missouri
Bed size:
160:
10
61120: 52
120+:
25
Contexts
External: NS
Organizational
Characteristics:
Adequate experience
with transmitting
MDS data
electronically
Teamwork,
NS
Educational
workshop
RAI manual
RAPs
CPG (AHRQs)
Comparative
feedback ShowMe QI report
(quarterly)
GCNS consult
D-148
Implementation
Details
Length: 1 year
Process:
Core group
receives ShowMe QI report in
workshop;
subsequent
quarterly reports
sent to
administrator
and DON
GCNS help
interpret report,
assess resident
problems, and
document care
15 facilities
(Group 2) had
1 on-site visits
and GCNS calls
18 facilities
(Group 2) had
only 1 call and
limited GCNS
calls
13 quality
indicator
outcome
measures were
evaluated
Successes:
Reduction in
pressure ulcers
(overall and lowrisk) for residents in
facilities using
GCNS
Barriers:
Outcomes:
Benefits
care protocol
(1.3% vs. 4.9%;
p = 0.0505)
Secondary
regression
analysis:
MDS QI 29
Pressure Ulcers
(overall):
Case mix: 0.156
Time Pre-Post:
0.240
Intervention: 0.026
Group X Time:
0.085 (p0.10)
MDS QI 29lr
Pressure ulcers
low risk:
Case mix: 0.417
Time Pre-Post:
0.037
Intervention: 0.064
Group X Time:
0.057 (p0.10)
Influence of
Contexts on
Outcomes
A subset of Group
2 nursing homes
that were intensely
involved with the
intervention
showed
improvement in
MDS QI scores for
five outcome
measures including
MDS QI 29
(pressure ulcers).
Nursing homes
that did have
continuous quality
improvement
systems in place
were often part of
larger health care
systems that have
ongoing support
from a quality
improvement
expert.
Author/
Year
Description
of PSP
Study
Design
Theory or Logic
Model
Description of
Organization
Contexts
Implementation
Details
Short staff
Staff turnover
especially nurse
RAI coordinator
Taking care
themselves
Cancelled site
visit at last
minute
Additional time
needed to
correct
inaccurate MDS
assessments
Teams mired in
the MDS
assessment
process and
coding issues
Difficulty
convincing staff
to use
continuous QI
principles
Addressing
Barriers:
Stronger
incentives to use
GCNS
GCNS more
local
More flexible site
visit times
Extend time to
implement
change
Use teams to
address
problems
Post accomplishments
D-149
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
Author/
Year
Ryden et al.
24
2000
Description
of PSP
Protocol
implementation
by APNs
Target safety
problem:
Comprehensive
Key elements:
APNs assist staff
to implement
care plan;
APNs provide
direct care to
residents
Study
Design
Theory or Logic
Model
Controlled
Havelocks (1974)
before-andmodel of effective
after
research utilization
APN
treatment
(2 facilities)
vs. usual care
(1 facility)
Description of
Organization
3 privately-owned
facilities located in
suburban
MinneapolisSt. Paul area;
certified for
Medicare
Contexts
Implementation
Details
Multiple nurses
responsible for
RAI process
Use of quality
manager on staff
to support care
delivery
improvements
Leadership buy
in to QI
Sustainability: NS
External: NS
Length: 6 months
Organizational
Process:
Characteristics:
RAs assess
APNs work with head
risk/collect data
nurse who works
2 APNs reassess
with physician or
risk, analyze data
GNP
(10 hrs/week per
Teamwork,
facility)
Leadership, Culture: APNs meet with
NS
residents
15-30 min/wk
AHRQ CPG
Successes:
Staff education
6 months of APN
Work with nursing treatment
assistants
significantly
APNs participate improved 3 of 4
clinical problems
in conferences
compare to usual
and wound care
care
rounds
Barriers:
High turnover of
unlicensed staff
Addressing
Barriers: NS
Sustainability:
A wound care
committee was
established at
1 facility.
D-150
Outcomes:
Benefits
Influence of
Contexts on
Outcomes
APN Treatment
(n = 86)
Pre: 19.8
Post: 3.5
2
x = 3.01(1),
p = 0.04, one-tailed
Usual Care
(n = 111)
Pre: 17.3
Post: 10.0
The relatively
short time (10 hr
per week in each
nursing home) and
the high turnover
rates of unlicensed
staff (range of
11%-45%) reduced
opportunities for
each APN to
establish
relationships with
staff.
APNs: Advanced practice gerontological nurses

CMS:
Centers for Medicare and Medicaid
CPG:
Clinical practice guidelines
DON:
Director of Nursing
EMR:
GCNS: Gerontological clinical nurse specialist
GNP:
General nurse practitioner
GP:
General Practitioner
HRPrU: High-risk PU quality measure
Int:
Intervention
LPN:
Licensed practical nurses
LTACH: Long-term acute care hospital
NS:
Not stated
PT:
Physical therapist
PU:
Pressure ulcer
QI:
Quality indicator
QM:
Quality measure
RA:
Resident assistants
RAI:
Resident assessment instrument
RAP RAI:
Resident assessment protocols
RCT:
Randomized controlled trial
SCI:
Spinal cord injury
WCC: Wound Care Certified
D-151
References
1.
Lynch S, Vickery P. Steps to reducing

hospital-acquired pressure ulcers. Nursing
2010 Nov;40(11):61-2. PMID: 20975436
2.
Young J, Ernsting M, Kehoe A, et al.

Results of a clinician-led evidence-based
task force initiative relating to pressure ulcer
risk assessment and prevention. J Wound
Ostomy Continence Nurs 2010 SepOct;37(5):495-503. PMID: 20736858
3.
Bales I, Padwojski A. Reaching for the

moon: achieving zero pressure ulcer
prevalence. J Wound Care 2009
Apr;18(4):137-144. PMID: 19349933
4.
Chicano SG, Drolshagen C. Reducing

hospital-acquired pressure ulcers. J Wound
Ostomy Continence Nurs 2009 JanFeb;36(1):45-50. PMID: 19155823
5.
Walsh NS, Blanck AW, Barrett KL.

Pressure ulcer management in the acute care
setting: a response to regulatory mandates. J
Wound Ostomy Continence Nurs 2009 JulAug;36(4):385-8. PMID: 19609158
6.
Dibsie LG. Implementing evidence-based

practice to prevent skin breakdown. Crit
Care Nurs Q 2008 Apr-Jun;31(2):140-9.
PMID: 18360144
7.
McInerney JA. Reducing hospital-acquired

pressure ulcer prevalence through a focused
prevention program. Adv Skin Wound Care
2008 Feb;21(2):75-8. PMID: 18349734
8.
Ballard N, McCombs A, Deboor S, et al.

How our ICU decreased the rate of hospitalacquired pressure ulcers. J Nurs Care Qual
2008 Jan-Mar;23(1):92-6. PMID: 18281882
9.
Catania K, Huang C, James P, et al. Wound

wise: PUPPI: the Pressure Ulcer Prevention
Protocol Interventions. Am J Nurs 2007
Apr;107(4):44-52; quiz 53. PMID:
17413732
10.
LeMaster KM. Reducing incidence and

prevalence of hospital-acquired pressure
ulcers at Genesis Medical Center. Jt Comm J
Qual Patient Saf 2007 Oct;33(10):611-6,
585. PMID: 18030863
D-152
11.
Courtney BA, Ruppman JB, Cooper HM.

Save our skin: initiative cuts pressure ulcer
incidence in half. Nurs Manage 2006
Apr;37(4):36, 38, 40 passim. PMID:
16603946
12.
Gibbons W, Shanks HT, Kleinhelter P, et al.

Eliminating facility-acquired pressure ulcers
at Ascension Health. Jt Comm J Qual
Patient Saf 2006 Sep;32(9):488-96. PMID:
17987872
13.
Hiser B, Rochette J, Philbin S, et al.

Implementing a pressure ulcer prevention
program and enhancing the role of the
CWOCN: impact on outcomes. Ostomy
Wound Manage 2006 Feb;52(2):48-59.
PMID: 16464994
14.
Lyder CH, Grady J, Mathur D, et al.

Preventing pressure ulcers in Connecticut
hospitals by using the plan-do-study-act
model of quality improvement. Jt Comm J
Qual Saf 2004 Apr;30(4):205-14. Also
available:
www.mizuhosi.com/pressuremanagement/L
yder_Preventing_04.pdf. PMID: 15085786
15.
Stier L, Dlugacz YD, OConnor LJ, et al.

Reinforcing organizationwide pressure ulcer
reduction on high-risk geriatric inpatient
units. Outcomes Manag 2004 JanMar;8(1):28-32. PMID: 14740581
16.
Soban LM, Hempel S, Munjas BA, et al.

Preventing pressure ulcers in hospitals: a
systematic review of nurse-focused quality
improvement interventions. Jt Comm J Qual
Patient Saf 2011 Jun;37(6):245-52.
PMID: 21706984
17.
Horn SD, Sharkey SS, Hudak S, et al.

Pressure ulcer prevention in long-term-care
facilities: a pilot study implementing
standardized nurse aide documentation and
feedback reports. Adv Skin Wound Care
2010 Mar;23(3):120-31. PMID: 20177165
18.
Rantz MJ, Hicks L, Petroski GF, et al.

Cost, staffing and quality impact of bedside
electronic medical record (EMR) in nursing
homes. J Am Med Dir Assoc 2010
Sep;11(7):485-93. PMID: 20816336
19.
Milne CT, Trigilia D, Houle TL, et al.

Reducing pressure ulcer prevalence rates in
the long-term acute care setting. Ostomy
Wound Manage 2009 Apr;55(4):50-9.
PMID: 19387096
20.
Tippet AW. Reducing the incidence of

pressure ulcers in nursing home residents:
a prospective 6-year evaluation. Ostomy
Wound Manage 2009 Nov 1;55(11):52-8.
PMID: 19934464
21.
Rosen J, Mittal V, Degenholtz H, et al.

Ability, incentives, and management
feedback: organizational change to reduce
pressure ulcers in a nursing home. J Am
Med Dir Assoc 2006 Mar;7(3):141-6.
PMID: 16503306
22.
Abel RL, Warren K, Bean G, et al. Quality

improvement in nursing homes in Texas:
results from a pressure ulcer prevention
project. J Am Med Dir Assoc 2005 MayJun;6(3):181-8. PMID: 15894247
23.
Rantz MJ, Popejoy L, Petroski GF, et al.

Randomized clinical trial of a quality
improvement intervention in nursing homes.
Gerontologist 2001 Aug;41(4):525-38.
PMID: 11490051
24.
Ryden MB, Snyder M, Gross CR, et al.

Value-added outcomes: the use of advanced
practice nurses in long-term care facilities.
Gerontologist 2000 Dec;40(6):654-62.
PMID: 11131082
D-153
Evidence Tables for Chapter 22. Inpatient Intensive Glucose Control Strategies To
Reduce Death and Infection (NEW)
Table 1, Chapter 22. Large trials (n > 500) evaluating the health outcome effects of IIT
Patient population
Single or multi-center
Country
Implementation/
Context
Diabetes
mellitus (%)
Glucose
target, T v
C (mg/dL)
SICU
Single center
10
Belgium
Insulin protocol
was developed
and use overseen
by study
investigators.
13
80-110
v 180-200
Neurosurgical ICU
Single center
58
Italy
Efforts made to
limit nursing
turnover. New
nursing staff
worked with
experienced staff.
Study conducted
in a hospital that
had already
conducted similar
IIT study in SICU
patients. Authors
note the nurse:bed
ratio of 2.5 was
not changed for
study.
NR
80-110 v
180-200
16
80-110 v
180-200
MICU
Single center
59
Belgium
Inpatient BG Mortality and

Hypoglycemia
achieved,
T v C (RR,
Definition (mg/dL),
TvC
95% CI)
rate T v C, RR
(mg/dL)
(95%CI)
103 v 153
ICU mortality <40, 5 v 0.76%, RR
6.65 (2.83-15.62)
(p<0.001)
4.6 v 8%
(p=0.005
unadjusted)
RR 0.42 (95%
CI 0.22-0.62);
Hospital
mortality: 7.2
v 10.9%
(p=0.01)
RR 0.66; 95%
CI 0.48-0.92
92 v 143
6-month
<50, 93.8 v 62.8%,
(p<0.001)
mortality: 74.0
p<0.001
v 72.0%
(p=0.82)
111 v 153
(p<0.001)
D-154
ICU mortality:
24.2 v 26.8%
(p=0.31)
Hospital
mortality: 37.3
v 40.0%
(p=0.33)
RR 0.93; 95%
CI 0.81-1.08
90d mortality:
35.9 v 37.7%
(p=0.53)
<40, 18.7 v 3.1%
Other reported
outcomes*
TvC
Quality
Renal
replacement
4.8 v 8.2%
(p=0.007)
Sepsis
4.2 v 7.8%
(p=0.0003)
Fair
Sepsis 2.9 v
3.3% (p=NS)
Long-term
disability:
40.2 v 41.1%
(p=0.98)
Infection 0.7 vs
0.8% (p=NS)
Renal
replacement
20.8 v 22.7%
(p=0.50)
Fair
MICU
Single center
59
Belgium
Patient population
Country
Implementation/
Context
Diabetes
mellitus (%)
Glucose
target, T v
C (mg/dL)
MICU
Multicenter
20
Germany
No details
provided
30
80-110
v 180-200
MICU/SICU
Multicenter
60
Europe
Characteristics
from each study
site were reported.
Median nurse:bed
ratio was 2. ICUs
ranged widely in
size, patient
volume, and
number of
glucometers per
ICU.
24/7 ICU coverage
by intensivists.
Protocol designed
by
multidisciplinary
team at study site.
Physicians and
nurses attended
training sessions
before and during
study.
Three month staff
training period
before study.
17 T, 22 C
(p=0.031)
80-110
v 140-180
32 T,
48 C
(p<0.001)
80-110
v 180-200
115 v 171
(p<0.001)
13 T,
12 C
(p=NS)
80-110 v
180-200
120 v 149
(p,0.001)
MICU/SICU
Single center
18
Saudi Arabia
MICU/SICU
Single center
53
Colombia

Hypoglycemia
achieved,
T v C (RR,
Definition (mg/dL),
TvC
95% CI)
rate T v C, RR
(mg/dL)
(95%CI)
112 v 151
28d mortality:
<40, 17 v 4.1%
(p<0.001)
24.7 v 26%
RR 4.11 (95% CI
(p=0.74)
2.21-7.63)
RR 0.95, 95%
CI 0.70-1.28
90d mortality:
39.7 v 35.4%
(p=0.31)
117 v 144 ICU mortality:
< 40, 8.7 v 2.7%
(p<0.001)
17.2 v 15.3%
(p=0.41)
Hospital
mortality: 23.3
v 19.4%
(p=0.11)
28d mortality:
18.7 v 15.3%
(p=0.14)
D-155
ICU mortality: < 40, 28.6 v 3.1%, p

13.5 v 17.1%
< 0.001
(p=0.70)
RR 1.09
(0.70-1.72)
Hospital
mortality: 27.1
v 32.3%
(p=0.19)
RR 0.84
(0.64-1.09)
ICU mortality:
<40, 8.3 v 0.8%
33.1 v 31.2%;
RR 1.06
(0.82-1.37)
28d mortality:
36.6 v 32.4%;
RR 1.1 (0.851.42)
Other reported
outcomes*
TvC
Quality
Renal
replacement
27.5 v 22.5%
(p=0.001)
MICU
Multicenter
20
Germany
Renal
replacement
(patient days)
519 v 523
(p=0.75)
Fair
Renal
replacement
11.7 v 12.1%
(p=0.89)
Sepsis 36.9 v
40.9% (p=0.35)
Fair
Infection 27.2 v
33.2% (p=NS)
Renal
replacement
10.8 v 13%
(p=0.45)
Fair
Patient population
Country
Implementation/
Context
MICU/SICU
Pre-trial pilot
54
Multicenter International
studies carried out
to test/improve
insulin protocol.
Final
computerized
insulin protocol
algorithm
accessible to
study sites
through a central
Web site. No clear
explicit training
prior to study.
Acute MI
No details
Multicenter CCU
provided
61
Sweden
Acute MI
62
Multicenter Europe
No details
provided
Diabetes
mellitus (%)
Glucose
target, T v
C (mg/dL)
20
80-108 v
<180
39
126-198 v
NR
77
established
DM; 23 new
DM
of < 1y

Hypoglycemia
achieved,
T v C (RR,
Definition (mg/dL),
TvC
95% CI)
rate T v C, RR
(mg/dL)
(95%CI)
115 v 144
28d mortality:
<40, 6.8 v 0.5%
(p<0.001)
22.3 v 20.8% OR 14.7 (9.0-25.9)
(p=0.17)
RR 1.09
(0.96-1.23)
90d mortality:
27.5 v 24.9%
(p=0.02)
RR 1.14
(1.02-1.28)
24 hours:
T: 172.8
(59.4)
C: 210.6
(73.8)
p < .001
group 1 and
24 hours:
2: 126-180
group 1:
group 3:
163.8 (54.0),
NR
group 2:
163.8 (50.4),
group 3:
180.0 (64.8)
p = .0001
D-156
<54, 15.0 v 0% (p <

3 month
.001)
mortality:
12.4% v
15.6%, p =
NS
1 year
mortality:
18.6% v 26.1
%,
RR 0.69; 95%
CI 0.49-0.96
Adjusted 2< 54, Gr 1 v Gr2 v
year mortality: Gr3: 12.7 v 9.6 v 1.0
Group 1 v 3 =
1.19 (0.86 1.64)
Group 2 v 3 =
1.23 (0.89 1.69)
Other reported
outcomes*
TvC
Renal
replacement
15.4 v 14.5%
(p=0.34)
Sepsis
12.8 v 12.4%
(p=0.57)
Quality
Fair
Fair
Poor
Patient population
Country
Implementation/
Context
Diabetes
mellitus (%)
Glucose
target, T v
C (mg/dL)

Hypoglycemia
Other reported
Quality
achieved,
T v C (RR,
Definition (mg/dL),
outcomes*
TvC
95% CI)
rate T v C, RR
TvC
(mg/dL)
(95%CI)
Stroke
Conducted as a
17
72-126 v 24 hour mean
90-day
< 72 for > 30 mins,
Poor
63
Mutlicenter Britain
pragmatic trial as
<306
difference I v
mortality:
15.7, control group
part of routine
C (95% CI):
30.0% v
rate NR
clinical care. No
10.3 (4.9 27.3%,
OR (95% CI)
clear explicit
15.5), p <
= 1.14 (0.86training prior to
.0001
1.51)
study.
90 day severe
disability:
35.1% v
36.0%,
OR (95% CI)
= 0.96 (0.701.32)
Abbreviations: BG = Blood glucose; d = day; CCU = coronary care unit ; ICU = intensive care unit; MICU = medical intensive care unit; SICU = surgical intensive care unit;
C = comparator; DM = diabetes mellitus; NR = not reported; NS = not statistically significant; RR = relative risk; T = treatment
Other reported outcomes include renal replacement, infection, cardiovascular events, and long-term disability.
Quality was assessed using criteria from the US Preventive Services Task Force.
SI unit conversion for glucose: 1 mg/dL x 0.0555 = 1 mmol/L.
* Infection includes wound infection, urinary tract infection, or pneumonia; or a combination of these.
Morning blood glucose.
Average of blood glucose measurements, not otherwise specified.
Time weighted mean blood glucose.
Adjusted for chronic liver disease, traumatic brain injury, APACHE II and international normalized ratio.
D-157
References
1.
Umpierrez GE, Isaacs SD, Bazargan N, et

al. Hyperglycemia: an independent marker
of in-hospital mortality in patients with
undiagnosed diabetes.[see comment]. J Clin
Endocrinol Metab. 2002;87(3):978-82.
11889147
2.
Levetan CS, Passaro M, Jablonski K, et al.

Unrecognized diabetes among hospitalized
patients. Diabetes Care. 1998;21(2):246-9.
9539990
3.
Furnary AP, Gao G, Grunkemeier GL, et al.

Continuous insulin infusion reduces
mortality in patients with diabetes
undergoing coronary artery bypass grafting.
[see comment]. J Thorac Cardiovasc Surg.
2003;125(5):1007-21. 12771873
4.
Pomposelli JJ, Baxter JK, 3rd, Babineau TJ,

et al. Early postoperative glucose control
predicts nosocomial infection rate in
diabetic patients. JPEN J Parenter Enteral
Nutr. 1998;22(2):77-81. 9527963
5.
6.
7.
8.
Bochicchio GV, Sung J, Joshi M, et al.

Persistent hyperglycemia is predictive of
outcome in critically ill trauma
patients.[erratum appears in J Trauma. 2005
Nov;59(5):1277-8.] J Trauma.
2005;58(5):921-4. 15920404
Capes SE, Hunt D, Malmberg K, et al.
Stress hyperglycaemia and increased risk of
death after myocardial infarction in patients
with and without diabetes: a systematic
overview. [see comment]. Lancet.
2000;355(9206):773-8. 10711923
Capes SE, Hunt D, Malmberg K, et al.
Stress hyperglycemia and prognosis of
stroke in nondiabetic and diabetic patients: a
systematic overview. [see comment]. Stroke.
2001;32(10):2426-32. 11588337
Clement S, Braithwaite SS, Magee MF, et
al. Management of diabetes and
hyperglycemia in hospitals.[erratum appears
in Diabetes Care. 2004 Mar;27(3):856 Note:
Hirsh, Irl B [corrected to Hirsch, Irl B];
dosage error in text]. Diabetes Care.
2004;27(2):553-91. 14747243
D-158
9.
Institute for Health Care Improvement.

Establish a glycemic control policy in your
ICU. 2008 Available at:
www.ihi.org/IHI/Topics/CriticalCare/Intensi
veCare/Changes/IndividualChanges/Establis
haGlycemicControlPolicyinYourICU.htm.
10.
van den Berghe G, Wouters P, Weekers F, et

al. Intensive insulin therapy in the critically
ill patients. N Engl J Med.
2001;345(19):1359-67. 11794168
11.
Ingels C, Debaveye Y, Milants I, et al. Strict

blood glucose control with insulin during
intensive care after cardiac surgery: impact
on 4-years survival, dependency on medical
care, and quality-of-life. Eur Heart J.
2006;27(22):2716-24. 16608860
12.
Kansagara D, Fu R, Freeman M, et al.

Intensive insulin therapy in hospitalized
patients: a systematic review. Ann Intern
Med. 2011;154(4):268-82. 21320942
13.
Green DM, OPhelan KH, Bassin SL, et al.

therapy in critically ill neurologic patients.
Neurocrit Care. 2010;13(3):299-306.
20697836
14.
Coester A, Neumann CR, Schmidt MI.

Intensive insulin therapy in severe traumatic
brain injury: a randomized trial. J Trauma.
2010;68(4):904-11. 20032790
15.
Cao SG, Ren JA, Shen B, et al. Intensive

versus conventional insulin therapy in type 2
diabetes patients undergoing D2
gastrectomy for gastric cancer: a
randomized controlled trial. World J Surg.
2011;35(1):85-92. 20878324
16.
Umpierrez GE, Smiley D, Jacobs S, et al.

patients with type 2 diabetes undergoing
general surgery (RABBIT 2 surgery).
Diabetes Care. 2011;34(2):256-61.
21228246
17.
Krinsley JS. Association between

hyperglycemia and increased hospital
mortality in a heterogeneous population of
critically ill patients.[see comment]. Mayo
Clin Proc. 2003;78(12):1471-8. 14661676
18.
Arabi YM, Dabbagh OC, Tamim HM, et al.

therapy: a randomized controlled trial in
medical and surgical critically ill patients.
Crit Care Med. 2008;36(12):3190-7.
18936702
19.
Devos P, Preiser JC, Melot C. Impact of

tight glucose control by intensive insulin
therapy on ICU mortality and the rate of
hypoglycaemia: final results of the
GLUCONTROL study. Intensive Care
Medicine. 2007;33(Supplement 2)(S189)
20.
Brunkhorst FM, Engel C, Bloos F, et al.

Intensive insulin therapy and pentastarch
resuscitation in severe sepsis. N Engl J Med.
2008;358(2):125-39. 18184958
21.
Hermanides J, Bosman RJ, Vriesendorp TM,

et al. Hypoglycemia is associated with
intensive care unit mortality. Crit Care Med.
2010;38(6):1430-4. 20386307
22.
Dortch MJ, Mowery NT, Ozdas A, et al. A

computerized insulin infusion titration
protocol improves glucose control with less
hypoglycemia compared to a manual
titration protocol in a trauma intensive care
unit. JPEN J Parenter Enteral Nutr.
2008;32(1):18-27. 18165443
23.
Toschlog EA, Newton C, Allen N, et al.

Morbidity reduction in critically ill trauma
patients through use of a computerized
insulin infusion protocol: a preliminary
study. J Trauma. 2007;62(6):1370-5;
discussion 5-6. 17563651
24.
Treggiari MM, Karir V, Yanez ND, et al.

Intensive insulin therapy and mortality in
critically ill patients. Crit Care.
2008;12(1):R29. 18312617
25.
Taylor BE, Schallom ME, Sona CS, et al.

Efficacy and safety of an insulin infusion
protocol in a surgical ICU. J Am Coll Surg.
2006;202(1):1-9. 16377491
26.
27.
Caddell KA, Komanapalli CB, Slater MS, et

al. Patient-specific insulin-resistance-guided
infusion improves glycemic control in
cardiac surgery. Ann Thorac Surg.
2010;90(6):1818-23. 21095317
Wilson M, Weinreb J, Hoo GW. Intensive
insulin therapy in critical care: a review of
12 protocols. Diabetes Care.
2007;30(4):1005-11. 17213376
D-159
28.
Meijering S, Corstjens AM, Tulleken JE, et

al. Towards a feasible algorithm for tight
glycaemic control in critically ill patients: a
systematic review of the literature. Crit
Care. 2006;10(1):R19. 16469124
29.
Newton CA, Smiley D, Bode BW, et al. A

comparison study of continuous insulin
infusion protocols in the medical intensive
care unit: computer-guided vs. standard
column-based algorithms. J Hosp Med.
2010;5(8):432-7. 20945468
30.
Juneja R, Roudebush C, Kumar N, et al.

Utilization of a computerized intravenous
insulin infusion program to control blood
glucose in the intensive care unit. Diabetes
Technol Ther. 2007;9(3):232-40. 17561793
31.
Barletta JF, McAllen KJ, Eriksson EA, et al.

The effect of a computer-assisted insulin
protocol on glycemic control in a surgical
intensive care unit. Diabetes Technol Ther.
2011;13(4):495-500. 21355721
32.
Lee A, Faddoul B, Sowan A, et al.

Computerisation of a paper-based
intravenous insulin protocol reduces errors
in a prospective crossover simulated tight
glycaemic control study. Intensive Crit Care
Nurs. 2010;26(3):161-8. 20430622
33.
Holzinger U, Warszawska J, Kitzberger R,

et al. Real-time continuous glucose
monitoring in critically ill patients: a
prospective randomized trial. Diabetes Care.
2010;33(3):467-72. 20007948
34.
Yatabe T, Yamazaki R, Kitagawa H, et al.

The evaluation of the ability of closed-loop
glycemic control device to maintain the
blood glucose concentration in intensive
care unit patients. Crit Care Med.
2011;39(3):575-8. 21178768
35.
Atkin SH, Dasmahapatra A, Jaker MA, et al.

Fingerstick glucose determination in shock.
Ann Intern Med. 1991;114(12):1020-4.
2029097
36.
Critchell CD, Savarese V, Callahan A, et al.

Accuracy of bedside capillary blood glucose
measurements in critically ill patients.
Intensive Care Med. 2007;33(12):2079-84.
17763842
37.
Tanvetyanon T, Walkenstein MD, Marra A.

Inaccurate glucose determination by
fingerstick in a patient with peripheral
arterial disease. Ann Intern Med.
2002;137(9):W1. 12416976
38.
Juneja D, Pandey R, Singh O. Comparison

between arterial and capillary blood glucose
monitoring in patients with shock. Eur J
Intern Med. 2011;22(3):241-4.21570641
39.
40.
41.
47.
Dickerson LM, Ye X, Sack JL, et al.

Glycemic control in medical inpatients with
type 2 diabetes mellitus receiving sliding
scale insulin regimens versus routine
diabetes medications: a multicenter
randomized controlled trial. Ann Fam Med.
2003;1(1):29-35.15043177
48.
Korytkowski MT, Salata RJ, Koerbel GL, et

al. Insulin therapy and glycemic control in
hospitalized patients with diabetes during
enteral nutrition therapy: a randomized
controlled clinical trial. Diabetes Care.
2009;32(4):594-6.19336639
49.
Desachy A, Vuagnat AC, Ghazali AD, et al.

Accuracy of bedside glucometry in critically
ill patients: influence of clinical
characteristics and perfusion index. Mayo
Clin Proc. 2008;83(4):400-5.18380985
Datta S, Qaadir A, Villanueva G, et al.

Once-daily insulin glargine versus 6-hour
sliding scale regular insulin for control of
hyperglycemia after a bariatric surgical
procedure: a randomized clinical trial.
Endocr Pract. 2007;13(3):225-31.17599852
50.
Pidcoke HF, Wade CE, Mann EA, et al.

Anemia causes hypoglycemia in intensive
care unit patients due to error in singlechannel glucometers: methods of reducing
patient risk. Crit Care Med. 2010;38(2):4716.19789438
Furnary AP, Wu Y, Bookin SO. Effect of

hyperglycemia and continuous intravenous
insulin infusions on outcomes of cardiac
surgical procedures: the Portland Diabetic
Project. Endocr Pract. 2004;10 Suppl 2:2133.15251637
51.
Krinsley JS. Effect of an intensive glucose

management protocol on the mortality of
critically ill adult patients.[see
comment][erratum appears in Mayo Clin
Proc. 2005 Aug;80(8):1101]. Mayo Clinic
Proceedings. 2004;79(8):992-1000
52.
Aragon D. Evaluation of nursing work effort

and perceptions about blood glucose testing
in tight glycemic control. Am J Crit Care.
2006;15(4):370-7.16823014
53.
De La Rosa Gdel C, Donado JH, Restrepo

AH, et al. Strict glycaemic control in
patients hospitalised in a mixed medical and
surgical intensive care unit: a randomised
clinical trial. Crit Care.
2008;12(5):R120.18799004
54.
NICE-SUGAR Study, Finfer S, Chittock

DR, et al. Intensive versus conventional
glucose control in critically ill patients. N
Engl J Med. 2009;360(13):128397.19318384
55.
Van den Berghe G, Schetz M, Vlasselaers

D, et al. Clinical review: Intensive insulin
therapy in critically ill patients: NICESUGAR or Leuven blood glucose target? J
Clin Endocrinol Metab. 2009;94(9):316370.19531590
Kanji S, Singh A, Tierney M, et al.

Standardization of intravenous insulin
therapy improves the efficiency and safety
of blood glucose control in critically ill
adults. Intensive Care Med. 2004;30(5):80410.15127193
42.
de Azevedo JR, de Araujo LO, da Silva WS,

et al. A carbohydrate-restrictive strategy is
safer and as efficient as intensive insulin
therapy in critically ill patients. J Crit Care.
2010;25(1):84-9.19327317
43.
Chase JG, Shaw G, Le Compte A, et al.

Implementation and evaluation of the
SPRINT protocol for tight glycaemic control
in critically ill patients: a clinical practice
change. Crit Care. 2008;12(2).18412978
44.
Metchick LN, Petit WA, Jr., Inzucchi SE.

Inpatient management of diabetes mellitus.
Am J Med. 2002;113(4):317-23.12361818
45.
Cheung NW, Chipps DR. Sliding scale

insulin: will the false idol finally fall? Intern
Med J. 2010;40(9):662-4.20862784
46.
Umpierrez GE, Smiley D, Zisman A, et al.

patients with type 2 diabetes (RABBIT 2
trial).[see comment]. Diabetes Care.
2007;30(9):2181-6.17513708
D-160
56.
Furnary AP, Zerr KJ, Grunkemeier GL, et

al. Continuous intravenous insulin infusion
reduces the incidence of deep sternal wound
infection in diabetic patients after cardiac
surgical procedures. Ann Thorac Surg.
1999;67(2):352-60; discussion 602.10197653
57.
Qaseem A, Humphrey LL, Chou R, et al.

Use of intensive insulin therapy for the
management of glycemic control in
hospitalized patients: a clinical practice
guideline from the American College of
Physicians. Ann Intern Med.
2011;154(4):260-7.21320941
58.
Bilotta F, Caramia R, Paoloni FP, et al.

Safety and efficacy of intensive insulin
therapy in critical neurosurgical patients.
Anesthesiology. 2009;110(3):6119.19237874
59.
Van den Berghe G, Wilmer A, Hermans G,

et al. Intensive insulin therapy in the medical
ICU. N Engl J Med. 2006;354(5):44961.16452557
D-161
60.
Preiser JC, Devos P, Ruiz-Santana S, et al.

A prospective randomised multi-centre
controlled trial on tight glucose control by
intensive insulin therapy in adult intensive
care units: the Glucontrol study. Intensive
Care Med. 2009;35(10):1738-48.19636533
61.
Malmberg K, Ryden L, Efendic S, et al.

Randomized trial of insulin-glucose infusion
followed by subcutaneous insulin treatment
in diabetic patients with acute myocardial
infarction (DIGAMI study): effects on
mortality at 1 year. J Am Coll Cardiol.
1995;26(1):57-65.7797776
62.
Malmberg K, Ryden L, Wedel H, et al.

Intense metabolic control by means of
insulin in patients with diabetes mellitus and
acute myocardial infarction (DIGAMI 2):
effects on mortality and morbidity. Eur
Heart J. 2005;26(7):650-61.15728645
63.
Gray CS, Hildreth AJ, Sandercock PA, et al.

Glucose-potassium-insulin infusions in the
management of post-stroke hyperglycaemia:
the UK Glucose Insulin in Stroke Trial
(GIST-UK). Lancet Neurol. 2007;6(5):397406.17434094
Evidence Tables for Chapter 23. Interventions To Prevent Contrast-Induced Acute

Kidney Injury
Table 1, Chapter 23. Included studies
Study, publication
date
Brar, 2010
Number of trials
Sample size
AMSTAR criteria
Bicarbonate
N-acetylcysteine and
bicarbonate
14
2290
10
Evidence of
benefit for
intervention
N
10
1594
12/2009
Iso-osmolar contrast
36
7166
9/2004
N-acetylcysteine
22
2746
11
11
N
N
8/2007
25
3270
11
4/2008
10/2007
2/2009
Not stated
11/2008
11/2006
9/2008
12/2008
1/2008
11/2008
Bicarbonate
Bicarbonate
Bicarbonate
Bicarbonate
Bicarbonate
N-acetycysteine
Bicarbonate
Bicarbonate
Bicarbonate
Renal replacement
therapy
Bicarbonate
N-acetylcysteine
Statins
Bicarbonate
4
7
18
9
17
26
7
17
12
16
573
1307
3055
2043
2448
3352
1734
2633
1854
2763
8
9
11
7
9
9
6
11
8
10
N
Y
N
N
N
Y
Y
Y
Y
Y
N
751
10
16
12
23
1090
1677
1194
3563
11
9
8
11
Y
Y
N
N
Literature
search end date
11/2008
Brown, 2009
2,3
2/2009
From, 2010
Gonzales, 2007
Heinrich, 2009
6
Ho, 2008
7
Hogan, 2008
8
Hoste, 2009
9
Joannidis, 2008
10
Kanbay, 2009
11
Kelly, 2008
12
Kunadian, 2010
13
Meier, 2009
14
Navaneethan, 2009
15
Reed, 2009
Song, 2010
16
17
Trivedi, 2009
18
Trivedi, 2010
19
Zhang, 2011
20
Zoungas, 2009
6/2010
10/2008
1/2008
7/2010
12/2008
Intervention evaluated
D-162
References
1.
Brar SS, Hiremath S, Dangas G, Mehran R,

Brar SK, Leon MB. Sodium bicarbonate for
the prevention of contrast induced-acute
kidney injury: a systematic review and metaanalysis. Clin J Am Soc Nephrol
2009;4:1584-92.
2.
Brown JR, Block CA, Malenka DJ,

OConnor GT, Schoolwerth AC, Thompson
CA. Sodium bicarbonate plus Nacetylcysteine prophylaxis: a meta-analysis.
JACC Cardiovasc Interv 2009;2:1116-24.
3.
From AM, Al Badarin FJ, McDonald FS,

Bartholmai BJ, Cha SS, Rihal CS. Iodixanol
versus low-osmolar contrast media for
meta-analysis of randomized, controlled
trials. Circ Cardiovasc Interv 2010;3:351-8.
4.
5.
6.
7.
8.
Gonzales DA, Norsworthy KJ, Kern SJ, et

al. A meta-analysis of N-acetylcysteine in
contrast-induced nephrotoxicity:
unsupervised clustering to resolve
heterogeneity. BMC Med 2007;5:32.
Heinrich MC, Haberle L, Muller V, Bautz
W, Uder M. Nephrotoxicity of iso-osmolar
iodixanol compared with nonionic lowosmolar contrast media: meta-analysis of
randomized controlled trials. Radiology
2009;250:68-86.
Ho KM, Morgan DJ. Use of isotonic sodium
bicarbonate to prevent radiocontrast
nephropathy in patients with mild preexisting renal impairment: a meta-analysis.
Anaesth Intensive Care 2008;36:646-53.
Hogan SE, LAllier P, Chetcuti S, et al.
Current role of sodium bicarbonate-based
preprocedural hydration for the prevention
meta-analysis. Am Heart J 2008;156:414-21.
Hoste EA, De Waele JJ, Gevaert SA,
Uchino S, Kellum JA. Sodium bicarbonate
for prevention of contrast-induced acute
kidney injury: a systematic review and metaanalysis. Nephrol Dial Transplant
2010;25:747-58.
D-163
9.
Joannidis M, Schmid M, Wiedermann CJ.

Prevention of contrast media-induced
nephropathy by isotonic sodium
bicarbonate: a meta-analysis. Wien Klin
Wochenschr 2008;120:742-8.
10.
Kanbay M, Covic A, Coca SG, Turgut F,

Akcay A, Parikh CR. Sodium bicarbonate
nephropathy: a meta-analysis of 17
randomized trials. Int Urol Nephrol
2009;41:617-27.
11.
Kelly AM, Dwamena B, Cronin P, Bernstein

SJ, Carlos RC. Meta-analysis: effectiveness
of drugs for preventing contrast-induced
nephropathy. Ann Intern Med
2008;148:284-94.
12.
Kunadian V, Zaman A, Spyridopoulos I,

Qiu W. Sodium bicarbonate for the
a meta-analysis of published clinical trials.
Eur J Radiol 2011;79:48-55.
13.
Meier P, Ko DT, Tamura A, Tamhane U,

Gurm HS. Sodium bicarbonate-based
hydration prevents contrast-induced
nephropathy: a meta-analysis. BMC Med
2009;7:23.
14.
Navaneethan SD, Singh S, Appasamy S,

Wing RE, Sehgal AR. Sodium bicarbonate
therapy for prevention of contrast-induced
nephropathy: a systematic review and metaanalysis. Am J Kidney Dis 2009;53:617-27.
15.
Reed M, Meier P, Tamhane UU, Welch KB,

Moscucci M, Gurm HS. The relative renal
safety of iodixanol compared with lowosmolar contrast media: a meta-analysis of
randomized controlled trials. JACC
16.
Song K, Jiang S, Shi Y, Shen H, Shi X, Jing

D. Renal replacement therapy for prevention
meta-analysis of randomized controlled
trials. Am J Nephrol 2010;32:497-504.
17.
Trivedi H, Daram S, Szabo A, Bartorelli

AL, Marenzi G. High-dose N-acetylcysteine
nephropathy. Am J Med 2009;122:874 e915.
18.
Trivedi H, Nadella R, Szabo A. Hydration

with sodium bicarbonate for the prevention
of contrast-induced nephropathy: a metaanalysis of randomized controlled trials.
Clin Nephrol 2010;74:288-96.
19.
Zhang T, Shen LH, Hu LH, He B. Statins

nephropathy: a systematic review and metaanalysis. Am J Nephrol 2011;33:344-51.
20.
Zoungas S, Ninomiya T, Huxley R, et al.

Systematic review: sodium bicarbonate
treatment regimens for the prevention of
contrast-induced nephropathy. Ann Intern
Med 2009;151:631-8.
D-164
Evidence Tables for Chapter 24. Rapid Response Systems (NEW)

Table 1, Chapter 24. RRS evidence table: effectiveness
Author,
year
Anwar ul,
1
2010
Description of
PSP
Study Design Theory Description

or
of
Sample Size Logic Organization
Multi-component
Model
PICU physicians Pre-post
NA
600 bed
(Pediatric MET)
tertiary
9340
teaching
hospital (75
pediatric beds)
in Pakistan
Contexts
Implementation
Details
Education
sessions with
quarterly
reinforcement
Outcomes:
Benefits
Mortality : ICU
mortality of
patients
admitted to ICU
from floor (total
sample 77)
Results:50% to
15%
Statistics:
p=0.001 OR
0.18 (0.09-0.35)
Cardiac arrest:
Results: 5.2 to
2.7/
1000 admits
Statistics:
p=0.004
OR=.52 (0.122.26)
D-165
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Author,
year
Bader,
2
2009
Description of
PSP

Contexts
Implementation
Outcomes: Outcomes: Influence
Comments
or
of
Details
Benefits
Harms of Contexts
on
Multi-component
Model
Outcomes
Nurse led. Had
Pre-post
NA
304 bed acute Organizational
12 month review Mortality : nonAuthors do not give
Critical care
care noncharacteristics: Director and development ICU arrests
denominator data for
of
outreach
not given
teaching
of RRT, activation Results:61% to
cardio-respiratory arrest
component as well
hospital, part quality<br>Leadership: criteria, integration 26%
nor mortality data though
(proactive rounding
into ED nursing, Statistics:
of large health Leadership team
they do give denominator
on ICU discharged
development of
p<0.05
system
data for number of RRS
patients and also
CCOT component
including 13
calls.
other hospitals
responded to ED
followed by rapid Cardiac arrest:
in US.
(most RRS dont
cycle pilot test
no denominator
go to the ED.)
then full
Results: 36 to
implementation. 17/
RRT model with
year
Statistics no
CCOT function
value given
though stated to
be statistically
significant
suggesting
p<0.05
Transfer to ICU
per RRT call
Results:21% to
14%
Statistics:
p<0.05
D-166
Author,
year
Benson,
3
2008
Description of
PSP

or
of
Multi-component
Model
1 of 4 advanced
Pre-post
NA
350-bed
practice nurses
teaching
(APN) responded Not reported
hospital, US
to nurse initiated
calls with
intensivists and
other disciplines
involved as
needed by the
APN; if two calls
received
simultaneously ICU
physician served
as back up
(RRT model with
physician back-up)
Contexts
Implementation
Details
Outcomes:
Benefits
Credentialing,
information and
education
interventions
(email, newsletter
articles, rounding,
informational
sessions at
meetings), clinical
practice protocols
developed
Mortality :
average
mortality per
month
Results:9%
decrease (no
actual rates or
stats reported)
Statistics: NR
Cardiac arrest:
58.7% reduction
in codes per
1000
admissions
Results: 9.41
vs. 3.89
Statistics p =
.0065
National
Database of
Nursing Quality
Indicators
(NDNQI)
Failure to
Rescue rate
Results:19.5%
reduction (no
actual rates or
stats reported)
Statistics: NR
D-167
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Author,
year
Description of
PSP

or
of
Multi-component
Model
Campello, MET consists of
Pre-post
NA
470 bed non4
2009
ICU physician and
teaching
ICU nurse
88407
hospital in
Portugal
admissions
RRS and
implementation
Contexts
Implementation
Details
Outcomes:
Benefits
Trained all staff in Mortality : InBLS then widened hospital total

emergency call
Results:5.35
criteria (code) to (4.3-6.4) to 5.65
include standard (4.9-6.4)
RRS criteria for
1000 admits
deteriorating
Statistics:
patients.
p=0.152
Simulation training
with mannequins, Cardiac arrest:
education,
Results: 4.21
information
(3.3-5.2) to 3.38
posters.
(2.8-4.0)
1000 admits
Statistics
p=0.037
cardiac arrest
mortality
Results:3.65
(2.8-4.5) to 3.18
(2.6-3.8)
1000 admits
Statistics:
p=0.014
D-168
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Two data sets, one in the

first 2 years after RRS
and then 4 years post.
Results in outcomes are
for the 2-year follow-up;
none of the significant
differences were present
at the 4-year follow-up.
Author,
year
Chan,
5
2008
Description of
PSP

or
of
Multi-component
Model
respiratory therapy Pre-post
NA
404 bed
and 2 ICU nurses
tertiary care
(RRT model)
49171
academic
urban medical
RRS and
center in US
education program
Contexts
Implementation
Details
Outcomes:
Benefits
education program Mortality :

but otherwise
hospital wide
Results: 3.22 to
limited info
3.09/
100 admits
Statistics: AOR
0.95 (0.81-1.11)
p=0.52
Cardiac arrest:
non-ICU codes
Results: 6.08 to
3.08/
1000 admits
Statistics: 0.59
(0.40-0.89)
p=0.01
Hospital wide
codesResults: 11.2 to
7.5/
1000 admits
Statistics: AOR
0.76(0.57-1.01)
p=0.06
D-169
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Chose as a primary
outcome total hospital
code rate (including ICU
codes) and found no
benefit. ICU patients are
not part of RRS exposure
group. Their non-ICU
(general ward) codes did
drop significantly.
Author,
year
Gerdik,
6
2010
Description of
PSP

or
of
Multi-component
Model
respiratory
Pre-post
NA
696 bed
therapists and
academic
medical center
critical care nurses not given
in US
(RRT model)
RRT and education
Contexts
Implementation
Details
Outcomes:
Benefits
Pilot program
Mortality : total
followed by
Results: 32.5
campus wide
vs.
implementation 8 31.0/1000
months later.
admits
Worked with UHC Statistics: ns
collaborative in
developing
Cardiac arrest:
implementation. Results: 25.2
Secured
vs. 17.4/
stakeholders, then month
added patient and Statistics: none
family activation given
ICU
readmisssion
Results: no data
given
Statistics: ns
change
D-170
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
ICUs contributed FTEs to

structure team. gave
mortality data/1000
admissions but gave code
data per month
Author,
year
Hanson,
7
2009
Description of
PSP

or
of
Multi-component
Model
Peds MET consists Pre-post
NA
136 bed
of PICU fellow,
pediatric
PICU resident,
approximately
university
PICU nurse and
11,800
affiliated
respiratory therapy
hospital in US
RRS and
education
Contexts
Implementation
Details
Outcomes:
Benefits
Criteria
Mortality : ward
development,
(not total) but
Collaborative
included those
participation (IHI), with DNR (i.e.
planning,
expected and
education, hospital unexpected)
wide
Results: 1.5 vs.
implementation
0.45/
1000 admits
Statistics: RR=
0.30 (0-1.04) p=
0.07
Cardiac arrest:
ward
Results: 1.27
vs. 0.45/
1000 admits
Statistics RR=
0.35 (0-1.24)
p=0.126
time between
codes
Results: 2512 to
9418 patient
days
Statistics: not
given
Total hospital
mortality
Results: 9.64
vs. 7.31/
1000 admits
Statistics:
RR=0.076 (01.03) p= 0.078
D-171
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Author,
year
Hatler,
8
2009
Konrad,
9
2010
Description of
PSP

or
of
Multi-component
Model
ICU nurse and
Pre-post
NA
620 bed notrespiratory therapy
for profit urban
non-teaching
(RRT model)
50209
hospital in US
RRT and education
MET consists of
Pre-post
ICU nurse and ICU
277717
physician
admissions
afferent and
efferent limbs,
education
NA
900 bed
teaching
hospital in
Sweden
Contexts
Implementation
Details
Outcomes:
Benefits
Cardiac arrest:
Results: 0.93
vs. 0.63/
1000 discharge
Statistics not
given, may be
ns
direct and online Mortality :
intranet education, adjusted total
pocket cards for Results: RR 0.9
alert criteria with Statistics:
an education
p=0.003
period during the
initial
Cardiac arrest:
implementation
Results: 1.12
vs. 0.83/
1000
admissions
Statistics
p=0.035
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Team structure,
alert criteria,
documentation
development,
education
180 day
mortality
Results: 37%
vs. 15.8%
Statistics: NR
LOS
Results: no
change
D-172
Adjusted
Only study to report longmortality
term mortality
was
significantly
decreased
in both
medical and
surgical
patients
Author,
year
Kotsakis,
10
2011
Description of
PSP

Contexts
or
of
Multi-component
Model
Peds MET consists Pre-post
NA
4 tertiary level External : Funded by
of Peds ICU
pediatric
Ministry of Health
attending and/or 111432
hospitals
fellow, respiratory hospital
Canada.
therapists and ICU admissions
Hospital sizes
nurse available to
not given.
inpatients on
general wards via
paging. Had family
activation. MET
and Code team
were same group
of people (unified
team)
Implementation
Details
Outcomes:
Benefits
3 phases,
Mortality: total
development
hospital
1.education phase mortality: 10 vs.
2. pilot phase
9.6/1000 admits
when team only Statistics: NS
avail M-F during
Cardiac arrests:
day
Results: 1.9
3.Full 24/7
vs.1.8/ 1000
7d/week
implementation. admits
Statistics: NS
MET and Code
Blue Team were
the same group ICU mortality
Results: 0.3 vs.
(unified team)
0.1/ 1000
hospital admits
Statistics:
p=0.05
ICU
readmission
Results: NR
Statistics: NR
D-173
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Prospectively collected
after implementation
Author,
year
Laurens,
11
2011
Description of
PSP
Multi-component
MET consisted of
anesthesiologist,
medical house
officer and ICU/ED
nurse. responds to
any patient outside
ICU
describes the alert
criteria, education
process and RRS
process

or
of
Model
Pre-post
NA
150 bed
regional
96000
teaching
admissions
hospital in
Australia
Contexts
Implementation
Details
Outcomes:
Benefits
One month
Mortality:
education program unadjusted
prior to
hospital
introduction of the Results: 9.9 vs.
MET with ongoing 7.5/ 1000
admissions
education.
Formal training for Statistics:
MET team
RRR=24.2%
members and
p=0.003
index cards for
staff with alert
Cardiac arrests:
criteria
Results: 77 vs.
42/1000 admits
Statistics:
RRR=45.5%
p=0.0025
ICU admissions
Results: 22.4 to
17.6/ 1000
admissions
Statistics:
RRR=21.4%
p=0.003
D-174
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Decline in cardiac arrests

may have been affected
by increase in number of
patient deemed Do Not
Resuscitate by the team;
use of MET was low,
denominator based on
average annual admits,
did not give the exact
number. Did not give
confidence intervals. Did
not present cardiac arrest
data/1000 admits in text,
only in graph.
Author,
year
Lighthall,
12
2010
Description of
PSP

or
of
Multi-component
Model
MET consisted of Pre-post
NA
150 bed VA
ICU fellow,
hospital
anesthesiologist, unclear
affiliated with
ICU tending, ICU
a university
nurse, pharmacist,
medical
respiratory
school
therapist available
24/7 to general
ward patients
Contexts
Implementation
Details
Outcomes:
Benefits
Implemented after Mortality : all

a 4 month
Results: 2.71
education period vs. 2.24/100
discharges
Statistics:
p=0.04
Mortality: nonDNR
Results: 0.68
vs.0.39/ 100
discharges
Statistics:
p=0.003
Cardiac arrest:
Results: 10.1
vs. 4.36/100
discharges
Statistics
p<0.01
D-175
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Results for mortality were

no longer significant after
adjusting for secular
trends in mortality;
reduction in arrests was
not significant until 10
months after RRS
implementation; potential
underutilization of the
team; gives annual
admissions but not the
actual number of
discharges/admissions as
a denominator
Author,
year
Rothberg,
13
2011
Description of
PSP

Contexts
or
of
Multi-component
Model
Hospitalist-led
Time series NA
670-bed
Implementation tools :
MET including
tertiary
In accordance with the
critical care nurse, 154,382
teaching
IHI program
respiratory
admissions
hospital in US
therapist,
intravenous
therapist, and
patients physician
(ICU physician
served as back up)
Implementation
Details
Outcomes:
Benefits
Outcomes: Influence
Comments
Harms of Contexts
on
Outcomes
Initial
Mortality :
Stratified
Codes called for medical
implementation on Overall hospital
analyses by crises declined for units
2 med floors then mortality
codes within outside critical care only;
critical care Rate of MET activation
spread to entire Results: 22
deaths/
vs. codes (18 calls/1000
hospital over 3
months; Education 1000
outside
admissions)
included meetings, admissions
critical care:
e-mails, and
across study
posters; anyone period
could
Statistics: NS
activate;75% calls
from med, 20%
Cardiac arrest:
from surgical
Cardiac arrests
did not change
significantly
Results: 7.3 to
4.2/1000
admissions
Statistics
p<0.0001
Rate of fatal
codes/1000
admissions
Results: Delta =
0.06 (no specific
pre
post rate
reported in text,
graphed in
figure 4 only)
Statistics: p =
.65
D-176
Author,
year

Contexts
or
of
Multi-component
Model
Santamaria, MET consists of
Other
NA
400 bed
14
2010
ICU registrar,
controlled
tertiary
Part of the MERIT
general medical
study (see
teaching
study
registrar and the comments)
hospital
ICU nurse.
Between
separate code
14,838 and
team
26,575
Describes a MET admissions,
program
depending on
sample point
Sarani*,
15
2011
Description of
PSP
2 separate METs Pre-post

one for surgery
and one for
140,583
medicine. Both
discharges
teams have critical
care nurse,
pharmacy, reps
therapy, resident
from primary team,
ICU attending or
fellow during
daytime and a
telemedicine ICU
attending at night.
NA
Academic
hospital in US.
Size not given
Implementation
Details
Outcomes:
Benefits
Created MET as
part of MERIT
study, they were a
MET hospital in
that study
Mortality :
unexpected
Results: 0.58
vs. 0.30/1000
admits in last
time period
Statistics:
p<0.05
Cardiac arrest:
Results: 0.78
vs. 0.25/1000
admits in last
time period
Statistics
p<0.001
Limited, states
cardiac surgical
service did not
participate but
nothing beyond
that
Unanticipated
ICU admission
Results: 0.65
vs. 0.89/1000
admits in last
time period
Statistics: ns
Mortality:
hospital
mortality
Results:
Medical: 4.29
vs. 3.23%,
p<0.001;
Surgical: 1.21
vs. 1.11%
Statistics: ns
Cardiac arrest
Results: 4.07
vs. 2.32/1000
discharges
Statistics:
p<0.001
describes criteria
for RRS and the
structure
D-177
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Was one of the MERIT

study MET hospitals but
this data includes time
periods beyond the
MERIT study. They have
several sample epochs for
comparison of the
longitudinal long term
effects of MET -rates of
calling the MET increased
over each time period, as
cardiac arrest and
mortality rates fell
Surgical vs. Significantly higher

medical
reduction in cardiac arrest
rate in medical (40%) vs.
surgical (32%) (p<0.001);
mortality decreased
significantly only on
medical service; medical
service had 3 times higher
cardiac arrest rate otherwise, few
differences. Describes
case-mix but does not
explicitly state there was
adjustment.
Author,
year
Scott,
16
2009
Shah*,
17
2011
Description of
PSP

or
of
Multi-component
Model
ICU nurse and
Pre-post
NA
640 bed
respiratory therapy
tertiary
(RRT model)
not given
teaching
hospital
RRT and education
ICU nurse and
respiratory
therapist (RRT
model)
describes criteria
and what
constitutes a code
Pre-post
231,305
patient days,
61,389
admissions
NA
3 affiliated
academic
hospitals in
the US.
Contexts
Implementation
Details
Outcomes:
Benefits
1 month pilot
Cardiac arrest:
followed by house- Results: 7 vs. 2/
wide
1000 patient
implementation
days
Statistics:
unknown
Pre-intervention Mortality: Inperiod followed by hospital:
a 9 month roll-out Results: 2.4%
followed by full
vs. 2.06%,
intervention period 1.94%, 2.46% in
subsequent
postimplementation
period,
respectively
Statistics:
p=0.03,
0.01,and 0.83
respectively for
each postimplementation
period.
Cardiac arrests
Results: 0.83
vs. 0.98/
1000 final
period
Statistics: p=0.3
D-178
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
No sample size and no

statistical analysis
Existing in-house code

team could have affected
effectiveness- physicians
are already available;
RRT call rate was 26.7
per 1000 hospital
admissions
Author,
year
Tibballs,
18
2009
Description of
PSP

or
of
Multi-component
Model
Directed by
Pre-post
NA
215-bed
hospitals
tertiary care
resuscitation
104780
pediatric
officer, RN
admissions
hospital,
coordinating
pre, 138424
Australia
position, MET
post
included ICU
physician and RN,
ED physician and
RN
Contexts
Implementation
Details
Outcomes:
Benefits
Included intensive
education, hiring
additional ICU
nurses
Mortality: total
in-hospital
Results; 4.38
vs. 2.87/
1000 admits
Statistics: RR=
0.65 (0.57-0.75)
p<0.0001
Mortality:
unexpected
general ward
Results: 0.12
vs. 0.04/
1000
Statistics:
RR=0.35 (0.130.92) p=0.03
Cardiac arrest:
unexpected
non-ICU
Results: 0.19
vs. 0.17/
1000 admits
Statistics
RR=0.91 (0.501.64) p=0.75
Cardiac arrest:
preventable
non-ICU
Results: 0.16
vs. 0.07/1000
admits
Statistics:
RR=0.45 (0.20.97) p=0.04
* one reviewer had indicated that the article did not apply but a subsequent reviewer included and data was available
D-179
Outcomes: Influence
Harms of Contexts
on
Outcomes
Comments
Article also discussed

issues with definitions of
cardiac arrest,
preventable arrest
References
1.

SR. Experience of pediatric rapid response
team in a tertiary care hospital in Pakistan.
Indian J Pediatr. 2010; 77:273-6.
2.

Comm Rescue me: saving the vulnerable
non-ICU patient population. J Qual Patient
Saf. 2009; 35:199-205.
3.

Fisher Using an advanced practice nursing
model for a rapid response team. J. Jt Comm
J Qual Patient Saf. 2008; 34:743-7.
4.

Azevedo LF, Costa-Pereira A. Immediate
and long-term impact of medical emergency
teams on cardiac arrest prevalence and
mortality: a plea for periodic basic lifesupport training programs. Crit Care Med.
2009; 37:3054-61.
5.

RA, Kosiborod M, Spertus JA. Hospitalwide code rates and mortality before and
after implementation of a rapid response
team. JAMA. 2008; 300:2506-13.
6.

level 1 trauma center. Resuscitation. 2010;
81:1676-81.
7.

al. A reduction in cardiac arrests and
duration of clinical instability after
implementation of a paediatric rapid
response system. Qual Saf Health Care.
2009; 18:500-4.
8.
9.
Hatler C, Mast D, Bedker D et al.

Implementing a rapid response team to
decrease emergencies outside the ICU: one
hospitals experience. Medsurg Nurs. 2009;
18:84-90, 126.
Ekbom A, Martling CR. Reducing inhospital cardiac arrests and hospital
mortality by introducing a medical
emergency team. Intensive Care Med. 2010;
36:100-6.
D-180
10.

Implementation of a Multicenter Rapid
Response System in Pediatric Academic
Hospitals Is Effective. Pediatrics. 2011;
128:72-8.
11.
Laurens N, Dwyer T. The impact of medical

emergency teams on ICU admission rates,
cardiopulmonary arrests and mortality in a
regional hospital. Resuscitation. 2011;
82:707-12.
12.

Wagner TH. Introduction of a rapid
response system at a United States veterans
affairs hospital reduced cardiac arrests.
Anesth Analg. 2010; 111:679-86.
13.

Friderici J, Keyes M. Four years experience
with a hospitalist-led medical emergency
team: An interrupted time series. Journal of
Hospital Medicine. 2011.
14.
Santamaria J, Tobin A, Holmes J. Changing

cardiac arrest and hospital mortality rates
through a medical emergency team takes
time and constant review. Crit Care Med.
2010; 38:445-50.
15.

Clinical emergencies and outcomes in
patients admitted to a surgical versus
medical service. Resuscitation. 2011;
82:415-8.
16.
Scott SS, Elliott S. Implementation of a

rapid response team: a success story. Crit
Care Nurse. 2009; 29:66-75; quiz 76.
17.

G. Rapid response team in an academic
institution: does it make a difference? Chest.
2011; 139:1361-7.
18.
Tibballs J, Kinney S. Reduction of hospital

mortality and of preventable cardiac arrest
and death on introduction of a pediatric
medical emergency team. Pediatric Critical
Care Medicine. 2009; 10:30612+423+424+425.
Table 2, Chapter 24. Implementation tableRRS

Author, year
Adelstein,
1
2011
Buist,* 2007
Calzavacca,
3
2010
Chen, 2010
Cretikos,
5
2007
Description of
RRS
offered in
Appendix A but
appendix not with
pdf. uses two
tiered mechanism
for calling for
assistance
Study Design
prospective
evaluation of
breaches of
PACE system
before and after
changes
Senior ICU nurse,

senior ICU
registrar and
medical ward
registrar.
before after
design
MET system with

ICU registrar and
ICU nurse, 24/7
coverage for
inpatients on
general wards.
cohort
comparison
(early MET time
period and
another time
period several
years later)
physician led MET clusterconsisting of

randomized
senior ICU
registrar, general
med registrar and
ICU nurse (MERIT
study)
ICU registrar, ICU prospective
nurse, general
medicine registrar
(MERIT trial MET
hospitals)
Main Study
objectives
to assess if new
750 bed tertiary university
strategies could
affiliated hospital
improve the time to
delivery of MET
components as
compared to previous
MET system
too assess impact of
change programs
(education program
for new interns, nurse
liaisons, and
development
programs for
hosuestaff) on
incidence of cardiac
arrest
Does maturation of a
RRS improve the
failure to rescue rate
(recognition of
deterioration) and the
associated outcomes
Implementation
Comments
Themes
centralized activation
quantitative
system, review of all
events, automatic
escalation to code team if
MET did not respond
within 30 min, institution
of nurse educator for
training and compliance
400 bed suburban teaching
nurse liaison, career
hospital (first one in the world development and
to have a true MET)
education/oreintation
400 bd teaching hospital with

several years of having a
MET program (one of the
earliest hospitals to have one)
to assess reasons for multiple (MERIT study

calling emergency
hospitals)
help between
hospitals with a MET
and those without
To assess the
process components
of MET
implementation that
correlated with
utilization
12 hospitals of varying sizes

(the 12 MET hospitals in the
MERIT trial)
D-181
change in delayed
activations (late
recognition),
unanticipated ICU
admission
institution of NFR (DNR)
orders
effect of teaching
hospital, metropolitan
hospital, patient location
and time of activation
knowledge of activation
Quantitative but only to
criteria, understanding of utilization rates not
MET purpose,
outcomes
perceptions of readiness
for change, overall
attitude to MET program
Author, year
Donaldson,
6
2009
Foraida,
7
2003 ,
DeVita,
8
2004
Genardi,
9
2008
Description of
Study Design
RRS
not known as it
multi-modal
involved multiple
(qualitative using
hospitals, probably interviews
varied
ICU registrar,
prospective
Anesthesia, ICU
nurse, resp
therapy; 8 defined
roles-Team leader
airway manager,
airway assistant,
procedure
physician, chest
compressions,
runs medication/
equipment chart,
recorder, bedside
nursing
not given
prospective
Main Study
objectives
Identify factors
associated with
successful
implementation,
develop plans to help
others replicate such
success, standardize
process measures,
evaluate impact
through nurse
perceptions.
to determine if
specific educational
and feedback
interventions would
increase MET
utilization
to revitalize their
existing RRT and
improve on code
reductions
multiple (>500 hospitals,
nested within 9 multihospital
grantee organizations)
567 bed tertiary urban

teaching hospital
community hospital (size not

given)
Implementation
Themes
Extra resources, rapid
transfer, communication
enhancement, one stop
shopping(single team
assessment), robust
early adopters vs. late or
poor functioning RRS
immediate review of all

stat sequential paging
events, feedback to
those involved in
delaying MET activation,
creating better objective
alert criteria,
dissemination and
education for those new
criteria.
Increase MET calls, and
decrease multiple
primary service stat
sequential pages.
education, support for
nurses, critical thinking
skills, increase access to
RRT, change to
centralized paging
rewards program
(recognition of effort),
improved documentation,
alter alert criteria, ensure
competencies
D-182
Comments
Very qualitative, did not
define successful RRSs by
any objective criteria
Quantitative data on
utilization and incidence of
cardiac arrest but not
mortality
Quantitative, gives change

in codes and mortality
before and after change (%
decrease only, no statistics
reported)
Author, year
Jones,
10
2006
Jones,
11
2006
Jones,
12
2010
Shapiro,
13
2010
Description of
RRS
Pre-intervention
had a unified
code/MET team
with
anesthesiology,
ICU and
cardiology
registrars, ICU
nurse and primary
service physician,
post intervention
separate the
functions dropping
the cardiology and
anesthesiology
members from the
separate MET
ICU registrar, ICU
nurse and
receiving unit
medical registrar.
Separate from the
code team
Rapid response
nurse (2 dedicated
positions),
patients on-call
physician
various, different
hospitals
Study Design
prospective
before after trial
Main Study
objectives
to assess whether
systems changes in
existing MET would
increase utilization
rate
affiliated hospital
Implementation
Comments
Themes
Team composition
Quantitative data for
(separation of unified
utilization rates and
code/MET into separate incidence of true code calls
teams with separate
activations), Method of
activation (changing the
activation methods to
separate the teams),
Triggers (changing alert
criteria for calling a MET)
re-education on purpose
of MET, criteria, and the
changes
prospective
interventional but
continuous as
opposed to
before after with
defined
intervention
change
prospective
mixed, mostly
semi-structured
focus groups
assess education
program to increase
utilization of existing
MET

affiliated hospital
education, improved
communication, on-thejob aids (e.g., posters,
observational charts),
differences in MET usage
for medical vs. surgical
admissions
to determine if
mandatory activation
of MET improves
outcomes compared
to elective activation
872 bed academic hospital
conversion from elective Quantitative data on

MET activation to
utilization and incidence of
mandatory based on alert cardiac arrest.
criteria
to determine nurses
perceptions of RRS
impact on practice
and what constitutes
a successful RRS
multiple
D-183
Almost all METs/RRTs

are not mandatory
activation by staff despite
alert criteria being met
impact on practice,
characteristics of
successful teams
Quantitative data on
utilization rate but it is
continuous so may wish to
exclude
Author, year
Williams,*
14
2011
Description of
RRS
RRT model with
ICU nurse, ED
nurse, reps
therapist
Study Design
focus group
methodology
Main Study
objectives
clarify nurses
perceptions of RRS
156 bed community hospital
Implementation
Themes
experience with
activation, composition of
teams, concerns about
activating a RRT
advantage of RRT to
nurses and patients
D-184
Comments
References
1.
Adelstein BA, Piza MA, Nayyar V,

Mudaliar Y, Klineberg PL, Rubin G. Rapid
response systems: A prospective study of
response times. J Crit Care. 2011.
2.
Buist M, Harrison J, Abaloz E, Van Dyke S.

Six year audit of cardiac arrests and medical
emergency team calls in an Australian outer
metropolitan teaching hospital. BMJ. 2007;
335:1210-2.
3.
4.
Calzavacca P, Licari E, Tee A et al. The

impact of Rapid Response System on
delayed emergency team activation patient
characteristics and outcomes--a follow-up
study. Resuscitation. 2010; 81:31-5.
Chen J, Bellomo R, Hillman K, Flabouris A,
Finfer S. Triggers for emergency team
activation: a multicenter assessment. J Crit
Care. 2010; 25:359 e1-7.
5.
Cretikos MA, Chen J, Hillman KM,

Bellomo R, Finfer SR, Flabouris A. The
effectiveness of implementation of the
medical emergency team (MET) system and
factors associated with use during the
MERIT study. Crit Care Resusc. 2007;
9:206-12.
6.
Donaldson N, Shapiro S, Scott M, Foley M,

Spetz J. Leading successful rapid response
teams: A multisite implementation
evaluation. J Nurs Adm. 2009; 39:176-81.
7.
Foraida MI, DeVita MA, Braithwaite RS,

Stuart SA, Brooks MM, Simmons RL.
Improving the utilization of medical crisis
teams (Condition C) at an urban tertiary care
hospital. J Crit Care. 2003; 18:87-94.
8.
DeVita MA, Braithwaite RS, Mahidhara R,

Stuart S, Foraida M, Simmons RL. Quality
& Safety in Health Care: Use of medical
emergency team responses to reduce
hospital cardiopulmonary arrests. 2004; 2514.
9.
Genardi ME, Cronin SN, Thomas L. Dimens

Revitalizing an established rapid response
team. Crit Care Nurs. 2008; 27:104-9.
10.
Jones DA, Mitra B, Barbetti J, Choate K,

Leong T, Bellomo R. Increasing the use of
an existing medical emergency team in a
teaching hospital. Anaesth Intensive Care.
2006; 34:731-5.
D-185
11.
Jones D, Bates S, Warrillow S et al. Effect

of an education programme on the
utilization of a medical emergency team in a
teaching hospital. Intern Med J. 2006;
36:231-6.
12.
Jones CM, Bleyer AJ, Petree B. Joint

Commission journal on quality and patient
safety/Joint Commission Resources:
Evolution of a rapid response system from
voluntary to mandatory activation. 2010;
36:266-70, 241.
13.
Shapiro SE, Donaldson NE, Scott MB.

Rapid response teams seen through the eyes
of the nurse. Am J Nurs. 2010; 110:28-34;
quiz 35-6.
14.
Williams DJ, Newman A, Jones C, Woodard

B. Nurses perceptions of how rapid
response teams affect the nurse, team, and
system. J Nurs Care Qual. 2011; 26:265-72.
Table 3, Chapter 24. Patient safety-RRT: risk of bias

Author, year
Was the
allocation
sequence
adequately
generated?
Was the
allocation
adequately
concealed?
Were baseline
characteristics similar?
Were
incomplete
outcome data
adequately
addressed?
No
Were
baseline
outcome
measurements
similar?*
Unclear
Unclear
Was knowledge
of the allocated
interventions
adequately
prevented during
the study?
No
Anwar ul,
1
2010
2
Bader, 2009
Benson,
3
2008
Campello,
4
2009
5
Chan, 2008
6
Gerdik, 2010
Hanson,
7
2009
8
Hatler, 2009
9
Konrad, 2010
Kotsakis,
10
2011
Laurens,
11
2011
Lighthall,
12
2010
Rothberg,
13
2011
Santamaria,
14
2010
Sarani*,
15
2011
16
Scott, 2009
17
Shah*, 2011
Tibballs,
18
2009
No
No
No
No
No
No
Unclear
Unclear
No
No
No
No
No
Yes
Yes
No
No
Unclear
Unclear
No
No
Yes
Unclear
Unclear
No
Yes
Unclear
Unclear
No
Yes
Yes
No
No
No
Yes
Unclear
Unclear
No
No
No
Unclear
Unclear
Unclear
No
No
No
Yes
Yes
Yes
No
Yes
Yes
No
No
No
No
No
No
Unclear
Yes
Yes
No
No
Yes
Unclear
Unclear
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
Yes
No
Unclear
No
Yes
Yes
No
No
Yes
Unclear
Unclear
No
Yes
Yes
No
Unclear
Unclear
No
Yes
No
Unclear
Yes
No
No
Yes
Unclear
Unclear
No
Yes
Yes
No
No
No
Unclear
Unclear
No
Yes
Yes
No
No
No
No
No
No
Unclear
Yes
Unclear
No
Yes
No
Unclear
Unclear
Yes
No
No
No
Yes
Unclear
Yes
Unclear
Yes
Yes
D-186
Was the study

adequately
protected
against
contamination?
Was the study

free from
selective
outcome
reporting?
References
1.

SR. Experience of pediatric rapid response
team in a tertiary care hospital in Pakistan.
Indian J Pediatr. 2010; 77:273-6.
2.

Comm Rescue me: saving the vulnerable
non-ICU patient population. J Qual Patient
Saf. 2009; 35:199-205.
3.
4.
5.
6.
7.
8.
9.

Fisher Using an advanced practice nursing
model for a rapid response team. J. Jt Comm
J Qual Patient Saf. 2008; 34:743-7.
Azevedo LF, Costa-Pereira A. Immediate
and long-term impact of medical emergency
teams on cardiac arrest prevalence and
mortality: a plea for periodic basic lifesupport training programs. Crit Care Med.
2009; 37:3054-61.
RA, Kosiborod M, Spertus JA. Hospitalwide code rates and mortality before and
after implementation of a rapid response
team. JAMA. 2008; 300:2506-13.
level 1 trauma center. Resuscitation. 2010;
81:1676-81.
al. A reduction in cardiac arrests and
duration of clinical instability after
implementation of a paediatric rapid
response system. Qual Saf Health Care.
2009; 18:500-4.
Hatler C, Mast D, Bedker D et al.
Implementing a rapid response team to
decrease emergencies outside the ICU: one
hospitals experience. Medsurg Nurs. 2009;
18:84-90, 126.
Ekbom A, Martling CR. Reducing inhospital cardiac arrests and hospital
mortality by introducing a medical
emergency team. Intensive Care Med. 2010;
36:100-6.
D-187
10.

Implementation of a Multicenter Rapid
Response System in Pediatric Academic
Hospitals Is Effective. Pediatrics. 2011;
128:72-8.
11.
Laurens N, Dwyer T. The impact of medical

emergency teams on ICU admission rates,
cardiopulmonary arrests and mortality in a
regional hospital. Resuscitation. 2011;
82:707-12.
12.

Wagner TH. Introduction of a rapid
response system at a United States veterans
affairs hospital reduced cardiac arrests.
Anesth Analg. 2010; 111:679-86.
13.

Friderici J, Keyes M. Four years experience
with a hospitalist-led medical emergency
team: An interrupted time series. Journal of
Hospital Medicine. 2011.
14.
Santamaria J, Tobin A, Holmes J. Changing

cardiac arrest and hospital mortality rates
through a medical emergency team takes
time and constant review. Crit Care Med.
2010; 38:445-50.
15.

Clinical emergencies and outcomes in
patients admitted to a surgical versus
medical service. Resuscitation. 2011;
82:415-8.
16.
Scott SS, Elliott S. Implementation of a

rapid response team: a success story. Crit
Care Nurse. 2009; 29:66-75; quiz 76.
17.

G. Rapid response team in an academic
institution: does it make a difference? Chest.
2011; 139:1361-7.
18.
Tibballs J, Kinney S. Reduction of hospital

mortality and of preventable cardiac arrest
and death on introduction of a pediatric
medical emergency team. Pediatric Critical
Care Medicine. 2009; 10:30612+423+424+42
Evidence Tables for Chapter 25. Medication Reconciliation

Supported by Clinical Pharmacists (NEW)
Evidence Tables for Chapter 26. Identifying Patients at Risk

for Suicide: Brief Review (NEW)
Evidence Tables for Chapter 27. Strategies To Prevent

Stress-Related Gastrointestinal Bleeding (Stress Ulcer
Prophylaxis): Brief Update Review
Evidence Tables for Chapter 28. Prevention of Venous

Thromboembolism: Brief Update Review
Evidence Tables for Chapter 29. Preventing Patient Death or

Serious Injury Associated With Radiation Exposure from
Fluoroscopy and Computed Tomography: Brief Review
(NEW)
Evidence Tables for Chapter 30. Ensuring Documentation of

Patients Preferences for Life-Sustaining Treatment: Brief
Update Review
Evidence Tables for Chapter 31. Human Factors and

Ergonomics
D-188
Evidence Tables for Chapter 32. Promoting Engagement by Patients and Families To
Reduce Adverse Events (NEW)
Table 1, Chapter 32. Evidence table: patients engagement
Author,
year
Weingart,
1
2004
McGuckin,
2
2004
Description of
PSP
Study
design
Multiple
Sample
interventions
size
or multifaceted
interventions
Proving patients RCT
with
personalized
209
medication list
to help prevent
medication
errors
Asking all health Pre-post

care workers
who had direct 35
contact with
them, Did you
wash/sanitize
your hands?
logic
organization
model
No
No
Boston
teaching
hospital
A 24-bed
inpatient
rehabilitation
unit located in
an acute care
university
hospital
Contexts
Implementatio Measurement
n details
tool
Organizational
characteristics:
a 40-bed unit;
The unit used
paper
medication
order forms that
were faxed to
the pharmacy
and entered into
the hospitals
electronic
pharmacy
information
system; CPOE
not available at
time of study
Teamwork,
leadership,
culture: Nurse
manager was
member of
research team
Patient surveys;
identification of
med incidents
through
interviews of
pharmacists,
housestaff,
electronic
review
Visit with patient soap/sanitizer

by premed to
usage per
discuss hand
resident-day
hygiene (HH); before, during,
education
and after the
brochure;
intervention
prompt to ask
providers re HH;
video; visual aid
prompt
D-189
Outcomes:
Benefits
Outcomes: Influence of
Harms
contexts on
outcomes
Comments
adverse drug
rate between
intervention and
control
8.4% versus
2.9%, p=0.12
close-call rate
between
intervention
patients and
controls
(7.5% versus
9.8%) p=0.57
patients aware of
drug-related
mistakes during
the
hospitalization11%
Hand Hygiene
per resident day
5 to 9.7 during
intervention, 6.7
at 6 weeks, 7.0
at 3 months.
p<0.001 for all
timepoints
Patients
asked
physicians
40% of time,
nurses 95%
of time
% of patients
comfortable
asking - 75%
% of HCWs
washing hands
when asked by
patient-60%
Author,
year
Description of
PSP
Study
design
Multiple
Sample
interventions
size
or multifaceted
interventions
3
Stone 2007 Patient
Pre-post
empowerment
(materials
187 acute
telling patients hospitals
to ask HCWs to
clean their
hands).
Included other
interventions as
well as patient
engagement:
bedside alcohol
hand rub, ward
posters
changed
monthly, pts
encouraged to
ask HCWs to
clean their
hands). An
optional
component was
six-monthly
audit and
feedback of
hand hygiene
logic
organization
model
No
187 acute
hospitals
Contexts
Implementatio Measurement
n details
tool
Implementation
Tools: National
Patient Safety
Agencys Clean
Your Hands
Campaign
(CYHC) seeks
to improve 293
healthcare
workers
(HCWs) handhygiene
behaviour in
England and
Wales
Outcomes:
Benefits
Monthly median
alcohol hand rub
(AHR) use: 44
pre to 56 post;
p<0.001
Combined
median use of
AHR and soap:
13.2 to 31
ml/patient bedday;
Health careassociated
infection rates:
No changes
apart from
seasonal
changes in
norovirus and
CDAD
D-190
Harms
contexts on
outcomes
increase may
have been
confounded
by a change
in soap/AHR
provider
Comments
limitations of
self-reported
data; high
response rate;
targeting use of
AHR, changed
many aspects of
hand-hygiene
behaviour,
increasing AHR
use in particular,
across the acute
sector of the
NHS without
reducing soap
usage. Audit and
feedback, a
component
emphasized
much less than
AHR and
posters, was
less widely
implemented.
Table 2, Chapter 32. Evidence table: patients engagement, risk of bias
Weingart,
1
2004
McGuckin,
2
2004
Stone
3
2007
Was the
allocation
sequence
adequately
generated?
Was the
allocation
adequately
concealed
?
Were baseline
outcome
measurement
s similar?*
Were baseline
characteristics
similar?
Were incomplete
outcome data
adequately
addressed?*
Yes
Unclear
Yes
No
Unclear
Yes
No
Yes
No
No
Yes
Yes
No
No
No
No
No
No
Unclear
not applicable
Unclear
No
Yes
Yes
D-191
Was knowledge
of the allocated
interventions
adequately
prevented during
the study? *
Was the study

adequately
protected
against
contamination?
Was the study free

from selective
outcome
reporting?
References
1.
Weingart SN, Toth M, Eneman J, et al. Int J

Qual Health Care: Lessons from a patient
partnership intervention to prevent adverse
drug events. 2004; 16:499-507.
2.
McGuckin M, Taylor A, Martin V, et al. Am

J Infect Control: Evaluation of a patient
education model for increasing hand
hygiene compliance in an inpatient
rehabilitation unit. 2004; 32:235-8.
3.
D-192
Stone S, Slade R, Fuller C, et al. Journal of

Hospital Infection: Early communication:
Does a national campaign to improve hand
hygiene in the NHS work? Initial English
and Welsh experience from the NOSEC
study (National Observational Study to
Evaluate the CleanYourHandsCampaign).
2007; 66:293-6.
Evidence Tables for Chapter 33. Promoting a Culture of Safety

Table 1, Chapter 33. Patient safety culture: evidence table
Author,
year
Abstoss,
1
2011
Description of
PSP
Study
design
Multiple
interventions or
multifaceted
interventions
7 interventions:
-Culture:
Feedback on
errors (posters
and emails), QI
education and
training (TV
channel and
curriculum)
-System: CPOE,
medication
management
(pharmacist), pt
safety report
form revisions
Sample
size
Pre-post
Post 2009
(n = 85,
resp. rat =
90%)
logic
organization
model
Contexts
University of
Organizational:
Michigans
characteristics2007C.S. Mott
2009
Childrens
Hospital PICU
Implementation
details
Measurement
tool
Outcomes:
Benefits
Cannot tell how

much training
staff got and who
received
it//Poster tracking
Days since last
medication error
with harm and
detailed emails
Safety Attitudes
Questionnaire
(SAQ) *only 13
items related to
medication
error/reporting
are reported in
this study
Culture
survey:
Teamwork
climate:
52.8% to
71.8%
agreement;
Safety
climate:
54.6% to
63.4%
agreement
(not sig))
(p < .01) and
(p = .13)
4 cultural & 3
system-level
interventions
Reported
errors
resulting in
harm
0.56 to 0.15
events/10,000
doses
p<0.01
Overall error
reporting rate
3.16 to
3.95/10,000
doses
ns
D-193
Outcomes:
Harms
Influence
of
contexts
on
outcomes
comment
Comments
**Still
abstracting
Table 1, Chapter 33. Patient safety culture: evidence table (continued)

Author,
year
AdamsPizarro,
2
2008
Description of
PSP
Study
design
Multiple
interventions or
multifaceted
interventions
Regional
improvement
collaboratives;
ICU initiatives
included
rounding and
daily goals,
ventilator bundle
Sample
size
Theory or Description
logic model
of
organization
IHI
20 ICUs
breakthrough
429 in 14 series
hospitals model,
(individual Culture
response Improvement
Guide
rate =
65%)
Pre-post
Contexts
Implementation
details
External:
Collaborative,
state safety
organization
Measurement Outcomes:
tool
Benefits
IHI model, facilitated AHRQ

workshops and
Hospital
coordination through Survey on
outside safety
Patient Safety
organizations//Culture (HSOPS)
Organizational: improvement guide
characteristics: toolkit - resources for
understanding
Average 272
inpatient beds, culture, planning
culture interventions
86% urban,
36% teaching based on the initial
culture assessment in
each unit; OR
improved more
dimensions than ICU,
ED
D-194
Culture
survey:
only 3 of 12
domains
showed
improvement;
decrease by
13.6 in one
overall
measure
(Safety
Grade) and
Overall
Perception of
Safety
decreased
1%
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Comments
Also included
EDs and ORs
- only
including ICU
data here.
Only 56% of
ICUs
completed
both surveys
and are
included

Author,
year
Blegen,
3
2010
Description of
PSP
Study
design
Multiple
interventions or
multifaceted
interventions
Triad for Optimal
Patient Safety
(TOPS): (1) team
training, (2) unit
based safety
team, (3) patient
engagement in
daily goals
Sample
size
Pre-post
-3 inpatient
medical
units nested
in 3 different
hospitals;
post n = 368
(response
rate = 81%
logic
organization
model
1 academic, 1
community, 1
integrated
healthcare
system; 26-34
beds; 1nurse:45 patients
Contexts
Implementation
details
Organizational:
characteristics:
All hospitals
located in San
Francisco Bay
area, CA;
Differed in
physician care
model,
pharmacy
presence on
the unit, and
use of
information
technology,
2006-2007
(1) 4hr. team

training (2) Unit
safety team;
identified safety
concerns, model
effective team
behavior, small
group learning
sessions (3)
Nurses worked
with patient/family
daily on daily
goals card
Leadership:
leadership
provided
protected time
for a unit-level
project
champion on
each unit
D-195
Measurement Outcomes: Outcomes: Influence of

tool
Benefits
Harms
contexts on
outcomes
comment
AHRQ Hospital
Survey on
Patient Safety
(HSOPS)
Culture
survey:
Increases in
mean
dimension
score
significant
for 5 of 10
dimensions
p < .05 for
all 5
Site x time
interaction
on 6
dimensions;
Post diffs
among
professional
groups, 3
culture
dimensions
Comments
-Significant site x
time interactionsculture scores
for one hospital
did not change
or changed in
negative
direction without that unit,
the analysis in
the other 2
hospitals
showed changes
in all 10
dimensions (no
indication of
which unit was
dropped)
-Differences also
found between
professional
groups on 3
dimensions,
overall
perceptions of
safety, and
frequency of
event reporting;
nurses tended to
score culture
most positively
post intervention
compared to
physicians and
pharmacists
Author,
year
Cooper,
4
2008
Description of
PSP
Study
design
Multiple
interventions or
multifaceted
interventions
Simulation-based
anesthesia crisis
resource
management
training.
Sample
size
pre-post
with control
hospitals
293,
response
rate 38%
logic
organization
model
Contexts
Implementation
details
4 academic
Not reported
medical
centers
associated with
Harvard. 2
control
academic
medical
centers. in
Massachusetts.
One-day, 6 to 7
hour simulationbased anesthesia
crisis resource
management
training session in
4 hospitals. 2
control hospitals,
staff did not
receive training.
D-196
Measurement Outcomes: Outcomes: Influence of

tool
Benefits
Harms
contexts on
outcomes
comment
Patient Safety
Climate in
Healthcare
Organizations
(PSCHO)
Culture
survey:
No
significant
differences
Comments

Author,
year
Description of
PSP
Study design
Sample size
Multiple
interventions or
multifaceted
interventions
Donahue, prepare
5
2011
paraprofessionals
to communicate
changes in
patient status
using structured
communication,
including
reducing cultural
barriers to
interdisciplinary
communication
Pre-post
logic model
of
organization
IHI Spread
for Change
111 (41%) post
Framework,
survey
Crew
(paraprofessionals Resource
only)
Management
techniques
Danbury
Hospital,
Danbury,
CT; Not
described
Contexts
Leadership:
Chief Nursing
Executive,
Chief Medical
Officer involvement
and
messaging
Implementation
details
Measurement
tool
Outcomes:
Benefits
IHI Spread for

Change
Framework//Meetings
with stakeholder
groups; unit-based
champions,
education and
training (SBAR,
communication
focused); executive
walk rounds
AHRQ
Hospital
Survey on
Patient Safety
(HSOPS)
Culture survey:
Reported
change
(improvement
of >10%) on
4/42 survey
items
Use of
structured
communication
(SBAR)
74% to 90%
increase
Not reported
Rapid
response
events
reported
Increase from
351 to 487;
Decrease in
number of
RRS events
that led to
code events
(29% pre, 22%
post)
Not reported
D-197
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Comments
Also includes
some
qualitative data
from focus
groups
describing
change in
communication
after
implementation
Author,
year
Frankel,
7
2008
Description of
PSP
Study design
Sample size
Multiple
interventions or
multifaceted
interventions
rigorous
Pre-post
WalkRounds
702 of all
professions, 21
patient care
areas, resp rate
60% (n =1, 256;
45% RN)
logic model
of
organization
Contexts
Implementation
details
1 academic, Organizational: Quality and safety

1 community characteristics: personnel
teaching, US US, 2002responsible, training,
2005,
feedback//Hospital
senior executive
Leadership:
rounds on unit;
Senior leaders database of safety
were core
concerns,
participants in recommendations,
intervention
and actions taken to
planning and address the issues
execution
D-198
Measurement
tool
Outcomes:
Benefits
Safety
Attitudes
Questionnaire
(SAQ)
Culture survey:
62% to 77% in
1 hospital,
46% to 56% in
2nd hospital
p=0.03 and
0.06 in the 2
hospitals
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Comments
Only 2 of 7
hospitals
complied fully
with approach
& only those
results are
reported; also
only SAQ
results from
units with
>50%
response rate
are reported
(about half of
the units). 2
hospitals did
not implement
Walk Rounds
rigorously -no
significant
change in
those units
*Table 1, Chapter 33. Patient safety culture: evidence table (continued)

Author,
Description of
Study
Theory or Description Contexts
year
PSP
design
logic
of
model organization
Multiple
Sample
interventions or
size
multifaceted
interventions
Structured
OLeary,
Concurrent
2 teaching
Not reported
9
interdisciplinary control
2010
service units
rounds, regular
(30 beds
interdisciplinary n = 147
each) in a
meetings
(response
897-bed
rate = 92%)
tertiary care
Structured
teaching
interdisciplinary
hospital
rounding format,
regular
interdisciplinary
meetings
Implementation Measurement Outcomes: Benefits Outcomes: Influence

details
tool
Harms of contexts
on
outcomes
comment
Structured
interdisciplinary
rounds each
weekday using
structured
communication
tool;
Interdisciplinary
working group
met for 12 weeks
before
implementation,
developed format,
frequency, timing
Safety
Attitudes
Questionnaire
(SAQ)
Culture survey:
Teamwork climate:
Mean, control = 77.3,
intervention = 82.4;
Safety climate, mean:
Control = 75.4,
Intervention = 76.5
Teamwork (p = .01);
Safety (n.s., p = .90))
Length of Stay
adjusted LOS was
0.19 days longer for
the intervention unit
vs. control
n.s. (p = 0.17)
Cost
adjusted cost was
$24.05 less for the
intervention unit vs.
control
n.s. (p = 0.94)
D-199
Comments
Teamwork
climate was
significantly
higher for
intervention
unit nurses
(83.5 vs. 74.2,
p = .005), but
there was no
significant
difference on
safety climate.
Teamwork and
safety climate
were rated
higher by
intervention
Physicians but
not statistically
significant
Author,
year
OLeary,
8
2011
Description of
PSP
Study
design
Multiple
Sample
interventions or
size
multifaceted
interventions
Structured Inter- concurrent
Disciplinary
control
Rounds
49 nurses in
both units
(84%
response
rate)
logic
of
model organization
Contexts
large tertiary Not reported

care teach
hospital,
hospitalist
unit, 30 beds
Implementation Measurement Outcomes: Benefits Outcomes: Influence

details
tool
Harms of contexts
on
outcomes
comment
every weekday,
30 minutes, led
by nurse manager
and unit medical
director, used
structured
communication
tool
Safety
Attitudes
Questionnaire
(SAQ)
(teamwork
and safety
domains)
Culture survey:
median 75
intervention, 61.1
control
p=0.03
Rating of quality of
communication and
collaboration with
hospitalizes
80% intervention vs.
54% control
p<0.01
Teamwork climate
median 85.7
intervention, 61.6
control
p=0.008
D-200
Comments
Author,
year
Paine,
10
2010
Description of
PSP
Study
design
Multiple
interventions or
multifaceted
interventions
Comprehensive
Unit-Based
Safety Program,
hospital-wide
training, culture
score goal setting
Sample
size
Hospital-wide,
multiple
interventions
Pettker,
11
2009
Pre-post
5461
surveys
post (144
units), 79%
response
rate
protocol
Pre-post
standardization,
creation of
not
patient safety RN reported
position and
patient safety
committee, team
skills training
risk-reduction
clinical practices
and creation of a
comprehensive
culture of safety.
Theory Description
or logic
of
model organization
Contexts
Large urban Organizational:

academic
characteristics:
center
Substantial
safety
infrastructure,
event reporting
Leadership:
Leadership
representative
for each CUSP,
Board of
Trustees
engagement
Tertiary-level
academic
medical
center. OB
service
averages
5500
admissions
per year.
Organizational:
characteristics:
academic
medical center,
OB service,
urban/suburban
Implementation Measurement
details
tool
Outcomes:
Benefits
CUSP includes Safety Attitudes

patient safety
Questionnaire
office coach;
(SAQ)
units with low
safety climate
encouraged to
implement
CUSP
Culture survey:
Improvements in
all SAQ domains
except stress
recognition
p<0.001
Incremental
Safety Attitudes
2004-2006:
Questionnaire
(SAQ)
expert review,
protocol
standardization,
patient safety
RN and
committee, team
skills training,
fetal heart
monitoring
certification,
crew resource
management
team training
Culture survey:
%reporting good
teamwork climate
& good safety
climate improved
from 38.5% to
55.4% and 33.3%
to 55.4%,
respectively.
D-201
Adverse
outcomes index
indicators
3.3% pre to 1.6%
post
P= .011
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Comments
Culture scores
decreased in
17 of 144 units,
details not
reported,
informal
interviews
suggested that
manager
turnover, unit
construction,
and
implementation
of new IT may
have
contributed to
lower scores
Author,
year
Pettker,
12
2011
Riley,
14
2011
Description of
PSP
Study
design
Multiple
interventions or
multifaceted
interventions
NOT
ABSTRACTEDthis article
already
abstracted - just
culture results are
here
Sample
size
TeamSTEPPS
didactic training
program,
TeamSTEPPS
with in-situ
simulation
training exercises
Didactic training
with in-situ
simulation v.
didactic only
Theory
or logic
model
Description
of
organization
Contexts
details
tool
Not
recorded
post 183 response
rate 72%
Reasons 3
rural/suburban
model
US
community
not reported
labor and
how many
delivery units,
completed
50-66 beds
survey
Cluster
RCT
Organizational:
characteristics:
midwest, 20052008
Outcomes:
Benefits
Culture survey:
Statistically
significant
improvements on
4 domains,
worsening 1
(Perceptions of
favorable working
conditions, no
change 1
<p 0.05 for 5 (4
improved, 1
worse)
Local tailoring of Safety
Culture survey:
TeamSTEPPS;
Attitudes
perinatal
simulation
Questionnaire outcomes Weighted Adverse
included detailed (SAQ)
Outcomes Score
debriefing// 1
%change -37%
hospital full, -1% didactic,
TeamSTEPPS
+43% control
didactic training
p<0.05 for full
(condensed); 1 intervention
TeamSTEPPS
with series of insitu training
exercises,
repeated until
staff targets were
met
D-202
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Comments
NOT
ABSTRACTEDthis study
already
abstracted for
the other article
reporting on
this - just
culture results
are here. NO
ROB done
either.
cluster RCT
Author,
year
Sexton,
15
2011
Description of
PSP
Study
design
Multiple
interventions
or multifaceted
interventions
CUSP
(Comprehensive
Unit-Based
Safety
Program), as
part of Keystone
ICU project
Sample
size
Pre-post
Pre n =
4,260
(overall
res. rate =
71%; 99
ICUS);
Post n =
3,533
(overall
res. rate =
73%; 71
ICUs)
Theory
or logic
model
Description
of
organization
71 Michigan
hospitals
71 ICUS
-68%
Teaching
-31% Faithbased
-27% bed sz.
>=500
-25% bed sz.
200-299
Contexts
Implementation
details
Measurement
tool
Organizational: CUSP
Safety Attitudes
characteristics: Intervention_4Steps: Questionnaire
see description (1) Educate staff on (SAQ)
science of safety;
(2) Identify
errors/defects, (3)
partner with senior
leadership, (4) Use
tool to learn from
one defect per
month//ICU project
teams created. In
addition to CUSP,
also implemented
intervention to
reduce CLABSI
and/or VAP, and
Daily Goals
checklist.
Implementation
period for each
intervention approx.
3 months.
D-203
Outcomes:
Benefits
Culture survey:
The overall mean
percent positive
scores increased
significantly from
baseline to
follow-up. Nine
ICUs met 60%
positive criteria
for success in
2004
Mean safety
culture %positive
(pre = 42.5%,
post = 52.2%, p <
.001); (Results
for specific
questions
reported in Table
2)
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Hosp. size,
faithbased:
Gain
higher for
faith-based
and small
h., though
diffs. not
tested
directly
Comments
Only reported
data from ICUs
that reported
culture surveys
at both time
points (71 of
127 total ICUs)
Author,
year
Thomas,
16
2005
Tiessen,
17
B., 2008
Description of
PSP
Study
design
Multiple
Sample
interventions
size
or
multifaceted
interventions
Executive
Cluster
walkrounds
RCT
evaluate and
reevaluate
patient safety
culture,
encourage
patient safety
learning, share
stories, weekly
executive
walkrounds,
prioritize
improvement
efforts, identify
staff safety
concerns,
implement
improvements
8 point patient
safety plan
Pre-post
not stated;
35%
response
rate
Theory
or logic
model
Description
of
organization
Contexts
details
tool
711 bed
Not reported
academic
tertiary care
hospital
inpatient units.
In units
randomized to
receive
executive walk
rounds
intervention,
executives
rounded once
every 4 weeks
for 3 visits per
unit.
88-bed, acute External:
The patient
care, rural
Canadian Council safety practices
on Health
community
were rolled out,
Services
hospital in
hospital-wide
Accreditation new over a 2 year
Ontario,
Patient Safety
Canada.
period (2005Required
2007).
Practices,
upcoming
accreditation
survey
Organizational:
characteristics:
hospital financial
situation issues,
staffing changes
and reductions
Leadership:
created Patient
Safety Action
Plan
D-204
Outcomes:
Benefits
Harms
contexts on
outcomes
comment
Comments
Safety Attitudes Culture survey:

Questionnaire 78 in both types
(SAQ)
of units
NS
Patient Safety
Climate in
Healthcare
Organizations
(PSCHO)
Culture survey:
significant
improvement on
only 2 of 30
statements,
significant
decrease on one
statement
one
statement
did show
decrease 46
to 29%
p=0.01
Hospital
financial
issues may
have
impacted
effectiveness
Significant
improvement
on 2
statements:
asking for help
not a sign of
incompetence,
and If I make a
mistake that
has serious
consequences,
I tell someone
about it.
Author,
year
Timmel,
18
2010
Description of
PSP
Study
design
Multiple
interventions
or
multifaceted
interventions
CUSP; video,
identify safety
hazards, learn
from defects,
teamwork and
communication
tools
Sample
size
Pre-post
n = 28
(100%
response
rate)
Theory
or logic
model
Description
of
organization
Contexts
1 surgical unit
in a large,
urban
academic
medical center
Leadership:
Senior hospital
executive
participated as
part of CUSP
team
Implementation
details
Measurement
tool
CUSP team met Safety Attitudes

monthly, science Questionnaire
of safety training, (SAQ)
staff safety
assessment,
learning from 1
defect per month;
baseline 2006,
follow up in 2007
and 2008
Outcomes:
Benefits
Culture survey:
Safety climate:
80% to 90%;
Teamwork
climate: 56% to
80%; hospital
management:
39% to 47%; Unit
Management:
62% to 68%;
Working
conditions: 48%
to 55%; Stress
recognition: 45%
to 46%
all composite
scores except
stress recognition
significantly
improved from 06
to 08 (p < .001)
Nurse turnover
3/12 FTEs left in
2006; 0/16 left in
2008 and 2009
D-205
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Comments
Author,
year
Description of
PSP
Multiple
interventions or
multifaceted
interventions
Edwards, Ad-hoc safety
6
rounds,
2008
enhancements to
error reporting
system and
related
education,
standardized
communication
protocols
(SBAR), transfer
of care checklist,
implementation of
EMR system
Study
design
Sample
size
Pre-post
n = 428
(Response
rate =
32%)
Theory Description
or
of
logic organization
model
2 metropolitan
pediatric
tertiary care
hospitals in
same health
system; 1
academic, 1
community
Contexts
Implementation
details
Organizational: 2 initiatives to
see description address error &
perceptions of
safety: Safety
Leadership:
rounds, reporting
supported
system updates; 3
survey
initiatives designed
to improve hand
offs and transitions
across units:
SBAR, transfer
checklist, EMR
5 components,
staggered
implementation
over time
D-206
Measurement Outcomes: Outcomes: Influence

tool
Benefits
Harms
of
contexts
on
outcomes
comment
AHRQ
Hospital
Survey on
Patient Safety
(HSOPS)
Culture
survey
results:
6/11
domains
significantly
increased, 1
significantly
decreased,
2 no
significant
change.
P< 0.01
community
hospital
scored
higher on 3
domains
compared
to
academic
hospital
Comments
Physicians were not

surveyed, significant
pre-intervention
differences reported
between hospitals on
two domains of
culture, but not
accounted for in
analyses; 3
interventions
designed to improve
hand-offs and care
transitions-however,
this domain score
significantly
decreased postimplementation; 2
interventions to
improve overall
perceptions, but no
change postimplementation.
Significant
differences in post
scores between
hospitals; hospital x
time interaction likely
but not tested
Author,
year
Description of
PSP
Study
design
Multiple
interventions
or
multifaceted
interventions
Pronovost, Comprehensive
13
2005
unit based
safety program:
(1) assess
culture, (2)
science of
safety
education, (3)
staff
identification of
safety
concerns, (4)
senior
executive
involvement, (5)
improvements
implemented
from #3, (6)
efforts
documented,
(7) results
shared, (8)
reassessment
of culture
Sample
size
Quasistepped
wedge
design
WICU n =
64
(response
rate =
86%);
SICU n =
23
(Response
rate =
84%)
Theory
or logic
model
Description
of
organization
Contexts
2 ICU units in
a large
metropolitan
tertiary care
hospital
Leadership:
Each unit was
adopted by a
senior level
executive;
dedicated
improvement
team to support
intervention
implementation
details
tool
Forms for
Safety climate
collecting/sharing scale (SCS)
improvement
success stories,
daily goals
sheets, tool for
medication errors.
Science of safety
education, staff
identify how next
patient will be
harmed and how
to prevent
Outcomes:
Benefits
Culture survey
results (nurse
turnover and
length of stay)
WICU : 35%
positive climate to
52% positive
climate post; SICU
35% positive
climate to 68%
positive climate
post
Stats reported for
individual
questions, but not
for overall domain
changes; 8/10 p
<0.05 in WICU,
3/10 in SICU had p
< .05
WICU: decreased
from 9% to 2%;
SICU: Decreased
from 8% to 2%
P = NS
CUSP (8 step
version)
WICU: decreased
from 2 days to 1
day; SICU:
Decreased from 3
days to 2.3 days
P< 0.05
D-207
Outcomes: Influence
Harms
of
contexts
on
outcomes
comment
Comments
pre-post study
in both units
References
1.
Abstoss KM, Shaw BE, Owens TA, et al.

Increasing medication error reporting rates
while reducing harm through simultaneous
cultural and system-level interventions in an
intensive care unit. BMJ Qual Saf 2011.
2.
Adams-Pizarro I, Walker ZA, Robinson J, et

al. Using the AHRQ Hospital Survey on
Patient Safety Culture as an Intervention
Tool for Regional Clinical Improvement
Collaboratives. 2008.
3.
4.
5.
6.
7.
Blegen MA, Sehgal NL, Alldredge BK, et

al. Improving safety culture on adult
medical units through multidisciplinary
teamwork and communication interventions:
the TOPS Project. Qual Saf Health Care
2010; 19(4):346-50.
Cooper JB, Blum RH, Carroll JS, et al.
Differences in safety climate among hospital
anesthesia departments and the effect of a
realistic simulation-based training program.
Anesth Analg 2008; 106(2):574-84, table of
contents.
Donahue M, Miller M, Smith L, et al. A
leadership initiative to improve
communication and enhance safety. Am J
Med Qual 2011; 26(3):206-11.
Edwards P, Scott T, Richardson P, et al.
Using staff perceptions on patient safety as a
tool for improving safety culture in a
pediatric hospital system. Journal of Patient
Safety 2008; 4(2):113-8.
Frankel A, Grillo SP, Pittman M, et al.
Revealing and resolving patient safety
defects: the impact of leadership
WalkRounds on frontline caregiver
assessments of patient safety. Health Serv
Res 2008; 43(6):2050-66.
8.
OLeary KJ, Haviley C, Slade ME, et al.

Improving teamwork: impact of structured
interdisciplinary rounds on a hospitalist unit.
J Hosp Med 2011; 6(2):88-93.
9.
OLeary KJ, Wayne DB, Haviley C, et al.

Improving teamwork: impact of structured
interdisciplinary rounds on a medical
teaching unit. J Gen Intern Med 2010;
25(8):826-32.
D-208
10.
Paine LA, Rosenstein BJ, Sexton JB, et al.

Assessing and improving safety culture
throughout an academic medical centre: a
prospective cohort study. Qual Saf Health
Care 2010; 19(6):547-54.
11.
Pettker CM, Thung SF, Norwitz ER, et al.

Impact of a comprehensive patient safety
strategy on obstetric adverse events. Am J
Obstet Gynecol 2009; 200(5):492.e1-8.
12.
Pettker CM, Thung SF, Raab CA, et al. A

comprehensive obstetrics patient safety
program improves safety climate and
culture. Am J Obstet Gynecol 2011;
204(3):216.e1-6.
13.
Pronovost P, Weast B, Rosenstein B, et al.

Implementing and validating a
comprehensive unit-based safety program.
Journal of Patient Safety 2005; 1(1):33-40.
14.
Riley W, Davis S, Miller K, et al. Didactic

and simulation nontechnical skills team
training to improve perinatal patient
outcomes in a community hospital. Jt Comm
J Qual Patient Saf 2011; 37(8):357-64.
15.
Sexton JB, Berenholtz SM, Goeschel CA, et

al. Assessing and improving safety climate
in a large cohort of intensive care units. Crit
Care Med 2011; 39(5):934-9.
16.
Thomas EJ, Sexton JB, Neilands TB, et al.

The effect of executive walk rounds on
nurse safety climate attitudes: a randomized
trial of clinical units [ISRCTN85147255].
BMC Health Serv Res 2005; 5(1):28.
17.
Tiessen B. On the journey to a culture of

patient safety. Healthc Q 2008; 11(4):58-63.
18.
Timmel J, Kent PS, Holzmueller CG, et al.

Impact of the Comprehensive Unit-based
Safety Program (CUSP) on safety culture in
a surgical inpatient unit. Jt Comm J Qual
Patient Saf 2010; 36(6):252-60.
Evidence Tables for Chapter 34. Effect of Nurse-to-Patient Staffing Ratios on Patient Morbidity
and Mortality
Author, Year
Unruh and
1
Zhang, 2012
Description Study Design

of PSP
Sample Size
Multicomponent
Not a study of Retrospective
an
cohort
intervention
124 Florida
hospitals
between 1996
2004
Needleman, et Not a study of Longitudinal

2
al., 2011
an
assessment of
intervention changes in
nurse staffing,
and mortality in
one hospital
197,961 patient
admissions
176,696 nursing
shifts
Theory or Logic Contexts

Model
A conceptual
model is
presented that
relates case mix,
location,
ownership, size,
and payer mix
with changes in
nurse staffing
over time
None
Implementation Outcomes: Benefits

Details
Outcomes:
Harms
124 Florida hospitals
Not relevant, not Higher RN FTE were None

a study of an
associated with better mentioned
USA
intervention
outcome for most, but
not all AHRQ patient
Academic status not reported
safety indicators,
including failure-toNo assessment of existing
rescue
quality / safety infrastructure
No assessment of
organizational complexity
SCTL: Not assessed
A single tertiary academic
hospital recognized for
exemplary care
USA
Nurse: careful assessment of
actual nurse workload for
specific patients
Academic status assessed
Existing quality and safety
infrastructure and
inferred from recognition by
authorities as a high quality
hospital
D-209
Not relevant, not Exposure to each shift None

a study of an
with a RN staffing
mentioned
intervention
level below target
increased risk of death
by 2%
In non-ICU patients,
risk increased 4%
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
Case mix,
urban status,
Medicaid, and
HMO days of
care were all
positively
related to
changes to
failure-torescue
High
High patient
Low
turnover also
associated with
worse
outcomes
Author, Year
Twigg et al.,
3
2011
Description Study Design Theory or Logic

of PSP
Model
Sample Size
Multicomponent
Not a study of Described as an None
an
interrupted time
intervention series, but
per se,
presented as a
Western
pre / post
Australia
assessment
ordered the
introduction 236,454 patients
of a new
150,925 nurse
staffing
staffing records
method for
nurses
Contexts

Details
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
3 adult tertiary teaching

hospitals
Not relevant, not

a study of an
intervention per
se
For all
patients and for
medical and surgical
patients the death
rate decreased
significantly
None
mentioned
None
mentioned
High
Not relevant, not

a study of an
intervention per
se
Lower patient-to-nurse
staffing ratios were
associated with lower
30-day mortality and
failure to rescue
25% of RNs
None
reported they mentioned
perceived
decreased
support from
LVNs, 34% of
RNs reported
decreased
support from
unlicensed
personnel
High
Australia
Nurse hours of care and skill
mix
Academic states assessed
No assessment of existing
Organizational infrastructure
described in terms of
comprehensive clinical
services being provided
Aiken et al.,
4
2010
Not a study of
an
intervention
per se, rather
California
legislation
mandated
certain nurseto-patient
ratios
Cross-sectional None
22,336 hospital
nurses in
California
SCTL: Not assessed

California staff nurses
USA
RN staffing, patient-nurse
workload
SCTL: Not assessed
D-210
Author, Year
Harless and
5
Mark, 2010

of PSP
Model
Sample Size
Multicomponent
Not a study of 283 California
None
an
hospitals
intervention
11,945,226 adult
Longitudinal inpatients
analysis of
changes in
nurse staffing
in California
between
1996 - 2001
Contexts

Details
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
283 acute care hospitals
Not relevant
None
mentioned
None
mentioned
High
Not relevant, not Each additional RN

None
a study of an
FTA per patient-day
mentioned
intervention
was associated with a
0.25% decrease in
mortality
None
mentioned
High
USA
Numerous financial and
economic payer variables
Each increase in one

RN FTE per 1,000
patient days was
associated with a
4.3% decrease in
mortality
No assessment of quality /
safety infrastructure
No assessment of
organizational infrastructure
SCTL: Not assessed
Schilling et al., Not a study of Retrospective

None
6
2010
an
cohort
intervention
166,920 adults
admitted to
Michigan
hospitals in 2003
- 2006 with an
emergency
department
admission for
any of 4
diagnosis
39 Michigan hospitals
USA
Nurse staffing estimated by
taking the ratio of each
hospitals FTE for RN and
dividing by patient-days
No assessment of
SCTL: Not assessed
D-211
Author, Year
Aiken et al.,
7
2008

of PSP
Model
Sample Size
Multicomponent
None
an
cross-sectional
intervention
10,184 nurses
(50% random
sample,
response rate
52%) and
232,342 surgical
patients
Contexts

Details
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
168 acute care hospitals in

Pennsylvania 1998-1999
Not relevant, not 30-day mortality rate

a study of an
for general surgical
intervention
patients reported as
19.5 per 1,000
admissions (1.95%).
None
mentioned
Better care
environments
were found to
be associated
with lower 30
day mortality
and lower 30day failure-torescue.
High
None
mentioned
High
USA
RN staffing mean patients
per nurse, Nurse education ,
A composite score of the care
environment, encompassing
subscales from the Nursing
Work Index
quality safety infrastructure
High vs. low technology
assessed
Cho et al.,
8
2008

an
cross-sectional
intervention
27,372 ICU
patients
None
30-day-failure-toresuce rate reported

as 84.4 patients per
1,000 admissions
(8.4%).
More nurse staffing
and higher nurse
education levels were
found to be associated
with lower 30 day
mortality and lower 30day failure-to-rescue.
SCTL:
Practice Environment Scale
(PES) of the Nursing Work
Index (NWI) gets at a related
concept.
236 hospitals
Not relevant, not Each additional
None
a study of an
assigned patient per mentioned
Korea
intervention
RN in secondary
hospitals was
Mean years of ICU nurse
associated with a 9%
experience, RN staffing
increase in odds of
dying; there was no
SCTL:
statistically significant
Nonassociation of nursing
US/UK/Canada/Australia/New
experience with
Zealand study
mortality.
D-212
Author, Year

of PSP
Sample Size
Multicomponent
Kiekkas et al., Not a study of Observational
9
an
2008
prospective
intervention cross-sectional

Model
None
Convenience
sampling of 396
patients

Details
A general tertiary 14-bed

Not relevant, not
a study of an
academic hospital between
October 2005 and September intervention
2006
Outcomes:
Harms
No statistically
None
significant
mentioned
associations were
found in risk-adjusted
ICU mortality.
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
None
mentioned
High
None
mentioned
High
Greece
Daily nursing workload/
workload exposure
Hamilton et al., Not a study of Prospective

10
2007
an
cohort
intervention
2,636 low birth
weight or
preterm infants
None
SCTL:
NonUS/UK/Canada/Australia/New
Zealand study
54 neonatal ICUs.
Not relevant, not Higher specialist
None
a study of an
nursing provision was mentioned
UK
intervention
statistically
significantly
Total number of RNs per
associated with a
shift, Nursing provision ratio
lower risk-adjusted
per shift, Specialist nursing
observed mortality
provision ratio per shift
rate.
No assessment of
SCTL:
Not assessed
D-213
Author, Year
Mark et al.,
11
2007

of PSP
Model
Sample Size
Multicomponent
None
an
cross-sectional
intervention
3.65 million
pediatric patient
discharges
Contexts

Details
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
No relationship was
None
found between inmentioned
hospital pediatric
death and nurse
staffing for
hospitalized California
pediatric patients.
None
mentioned
High
Not relevant, not The highest quartile of None

a study of an
patient-to-nurse ratios mentioned
intervention
was associated with a
26% higher mortality
rate and 29% higher
failure-to-rescue rate
than the lowest
quartile of patient-tonurse ratios.
None
mentioned
High
286 general acute care and Not relevant, not

a study of an
childrens hospitals in
California between 1996 and intervention
2001.
USA
Outcomes:
Harms
RN staffing, Licensed
vocational nurse (LVN)
staffing, Unlicensed hours of
care provided per patient day
quality/safety infrastructure
Presence of pediatric ICU or
NICU
Rafferty et al., Not a study of Retrospective

None
12
an
2007
cross-sectional
intervention
118,752 surgical
patients and
3,984 nurses
(mostly RNs)
(response rate =
49.4%)
SCTL:
Not assessed
30 acute trusts in 1998.
UK
Mean hospital patient-nurse
ratio derived from survey of
nurses
No assessment of
SCTL:
Not assessed
D-214
Author, Year
Stone et al.,
13
2007

of PSP
Sample Size
Multicomponent
an
cross-sectional
intervention
Convenience
sample of
15,846 ICU
patients and 1,
095 RN were
surveyed
(average
response rate =
60%)

Model

Details
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
A conceptual
framework was
presented that
related the
potential
contributions of
patient
characteristic,
structures of
care, and
administrative
processes
including
organizational
climate, staffing,
overtime and
wages on patient
outcomes.
Not relevant, not Patients admitted to

a study of an
ICUs with more RN
intervention
hours per patient day
had significantly lower
30-day mortality.
An increase in
catheterassociated
bloodstream
infections in
organization
with a more
positive
organizational
climate
No significant High
relationship
was observed
between
organizational
climate and 30day mortality.
51 ICUs in 31 acute care

hospitals.
USA
Nursing staffing measured by
RN hours per patient day in
the ICU, Overtime use
measured as proportion of
overtime to regular hours,
Organizational climate in ICU
measured as composite
score of Perception of Nurse
Work Environment (Choi et
al., 2004).
Financial status assessed
No assessment of
SCTL:
Organizational climate
assessed with the
perceptions of Nurse Work
Environment
D-215
No significant
relationship was
observed between
overtime use and 30day mortality.
Overall
Risk of
Bias
Author, Year
Tourangeau,
Doran, et al.,
14
2006

of PSP
Sample Size
Multicomponent
an
cross-sectional
intervention
49,993 patients
with four
diagnoses:
acute
myocardial
infarction,
stroke,
pneumonia, or
septicemia and
3,886 nurses
(response rate =
65%)

Model
A conceptual
framework was
presented that
included
numerous
variables in six
categories:
condition of the
hospital practice
environment,
nurse staffing,
physician
expertise, nurse
and nurse
employment
characteristic,
care
management
processes (use
of care
maps/protocols),
and hospital
type/location on
30-day mortality.

Details
Outcomes:
Harms
Not relevant, not Lower 30-day mortality None

a study of an
rates found to be
mentioned
intervention
associated with:
higher proportion of
registered nursing
Canada
staff, higher proportion
of baccalaureateNursing staff mix, Nursing
educated nurses,
staff dose, Percentage of full
lower total dose of all
time nursing staff, Years
categories of nursing
experience on unit,
staff, higher nursePercentage of nurses with
reported adequacy of
baccalaureate or higher,
staffing and resources,
Overall health nurse level,
higher use of care
Hours of missed work in
maps/protocols, higher
preceding 3 months, Quality
nurse-reported quality
of nurse-physician
of care, lower nurserelationships, Nurse-rated
reported manager
manager ability and support,
ability and support,
Nurse-rated adequacy of
and higher nurse
staffing and resources,
burnout
Amount of teamwork, Overall
nurse job satisfaction, Nursereported quality of care,
Nurse burnout.
75 Ontario teaching and
community acute care
hospitals in 2002-2003.
Amount of professional role

support available for nursing
staff.
Frequency of use of care
maps/protocols to guide
patient care (one nurse
survey item with 5-point
frequency response options).
SCTL:
Teamwork, nurse burnout,
nurse-physician relationship
all explicitly measured via
nurse survey
D-216
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
Teamwork and High

physician
relationship
were not
associated with
differences in
mortality, but
higher nurse
burnout was
associated with
lower 30-day
mortality.
Author, Year
Estabrooks et
15
al., 2005

of PSP
Model
Sample Size
Multicomponent
None
an
cross-sectional
intervention
18,142 patients
with an acute
medical
diagnosis of
acute
myocardial
infarction,
congestive heart
failure, chronic
obstructive
pulmonary
disease,
pneumonia, or
stroke
4,799 nurses
working
(response rate =
52.8%)
Contexts

Details
Outcomes:
Harms
Not relevant, not Four factors were

None
a study of an
found in a
mentioned
intervention
multivariable
regression to be
Canada
30-day mortality rates:
Nurse: education level, Skill
a higher proportion of
mix, employment status,
baccalaureate
Nurse-reported autonomy, job
prepared nurses; a
satisfaction, perception of
staffing adequacy, perception
RNs in nursing staff
of unmet patient needs,
mix; a higher
Amount of non-nursing
proportion of
permanent RNs; and a
activities performed, support
higher reported nursefor float policy, Nursephysician
physician relationship,
collaboration.
Frequency of emotional
abuse experience.
49 Alberta acute care
hospitals during fiscal year
1998-1999.

use of high technology
SCTL:
Nursing Work Index and
Maslach Burnout Inventory
assessed
D-217
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
NurseHigh
physician
collaboration
was associated
with lower 30
day mortality
rates.
Author, Year
Halm et al.,
16
2005

of PSP
Model
Sample Size
Multicomponent
None
an
cross-sectional
intervention
2,709 generalsurgical patients
and 140 staff
RNs
Contexts

Details
One large Midwestern acute

care hospital.
Not relevant, not Nurse staffing was not None

a study of an
statistically
mentioned
intervention
significantly
associated with 30day mortality or
inpatient failure-torescue.
USA
RN staffing
No assessment of
SCTL:
Maslach Burnout Inventory
assessed
D-218
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
None
mentioned
High
Author, Year
Person et al.,
17
2004

of PSP
Sample Size
Multicomponent
an
cross-sectional
intervention
118,940 acute
myocardial
infarction (AMI)
patients

Model
None

Details
4,401 acute care hospitals in Not relevant, not

a study of an
1994 1995.
intervention
USA
Ratio of full-time equivalent
RNs to average daily census,
Ratio of full-time equivalent
licensed practical and
vocational nurses per
average daily census, Parttime nursing staff estimated
as 0.5 full-time equivalent.
Types of cardiac services
offered
SCTL:
Not assessed
D-219
Outcomes:
Harms
Lower in-hospital
None
mortality rates were
mentioned
associated with higher
RN staffing in
hospitals.
Higher in-hospital
mortality rates were
licensed
vocational/practical
staffing in hospitals.
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
None
mentioned
High
Author, Year
Aiken et al.,
18
2003

of PSP
Model
Sample Size
Multicomponent
None
an
cross-sectional
intervention
10,184 nurses
(50% random
sample,
response rate
52%) and
232,342 surgical
patients
Contexts

Details

USA
Registered nurse education
level, Nursing workload,
Mean years of nurse
experience
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
Not relevant, not A higher proportion of None

a study of an
baccalaureate
mentioned
intervention
educated nurses and
lower nursing
workload were
associated with a
lower risk-adjusted
mortality and failure to
rescue rates.
None
mentioned
High
Not relevant, not A higher nurse-patient None

a study of an
ratio was significantly mentioned
intervention
inpatient unit mortality
rates.
None
mentioned
High

quality or safety
assessed
SasichayNot a study of Retrospective

None
Akkadechanunt an
cross-sectional
19
et al., 2003
intervention
2,531 medicalsurgical patients
with principal
diagnoses in
following groups:
disease of the
heart, malignant
neoplasms,
hypertension,
cerebrovascular
diseases, and
pneumonia/other
lung diseases.
SCTL:
Not assessed
17 inpatient units in one
university hospital
Thailand
Ratio of total nursing staff to
patients, Proportion of RN to
total nursing staff, Mean
years RN experience,
Percentage of baccalaureateeducated nurses.
SCTL:
Non- US/UK/Canada/
Australia/New Zealand study
D-220
Author, Year
Aiken et al.,
20
2002

of PSP
Model
Sample Size
Multicomponent
None
an
cross-sectional
intervention
10,184 nurses
(50% random
sample,
response rate
52%) and
232,342 surgical
patients
Contexts

Details

None
Not relevant, not Higher patient-toa study of an
nurse ratio found to be mentioned
intervention
30-day mortality (p <
.001). Odds of patient
death increased by
7% for every
additional patient in
nurse workload.
USA
RN staffing
quality or safety
assessed
SCTL:
Not assessed
D-221
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
None
mentioned
High
Author, Year

of PSP
Sample Size
Multicomponent
Needleman et Not a study of Retrospective
21
an
al., 2002
cross-sectional
intervention
5,075,969
medical patient
discharges and
1,104,659
surgical patient
discharges

Model
None

Details
Outcomes:
Harms
None
Not relevant, not No statistically
a study of an
significant
mentioned
intervention
relationships were
found between inhospital mortality rates
USA
and nurse staffing
Eight nurse staffing indicators
indicators.
were assessed: Number of
RN hours of nursing care per
Two statistically
patient day, Number of
significant
licensed practical nurse hours
relationships were
per patient day, Number of
found between lower
aid hours of care per patient
hospital failure-today, Total hours of nursing
rescue rates and
care per patient day,
nurse staffing:
Proportion of RN hours of all
For medical patients, a
hours of nursing care,
Proportion of licensed
hours of care provided
practical nurse hours of all
by RNs.
hours of nursing care,
Number of hours of care
For surgical patients, a
provided by licensed nurses
greater number of
(RN + practical nurse) per
hours of care provided
patient day, RN hours as a
by RNs.
proportion of licensed nurse
hours.
799 hospitals in 11 states
from 1997 and 1998 fiscal
years
Acute care hospitals

No assessment of
SCTL:
Not assessed
D-222
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
None
mentioned
High
Author, Year

of PSP
Sample Size
Multicomponent
Tourangeau et Not a study of Retrospective
22
an
al., 2002
cross-sectional
intervention
49,993 patients
with four
diagnoses:
acute
myocardial
infarction,
stroke,
pneumonia, or
septicemia and
3,988 RNs
(response rate =
57%)

Model

Details
A conceptual
framework was
presented that
included eight
domains: nurse
staffing, nurse
skill mix,
professional role
support, nurse
characteristic,
nurse practice
environment
condition,
continuity of
registered nurse
care provider,
and other
determinants on
30-day mortality.
Not relevant, not Lower 30-day mortality None

a study of an
was found to be
mentioned
intervention
significantly
associated with:
higher proportions of
RN staffing, more
years if nurse
experience on the
clinical unit, and
higher number of
shifts missed by
nurses in the
preceding 3 months
75 Ontario teaching and

community acute care
hospitals in 1998-1999.
Canada
No assessment of
organizational complexities
SCTL:
Canadian Practice
Environment Index assessed
(drawn from the Nursing
Work Index NWI-R)
D-223
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
Condition of
High
nursing practice
environment
was not
associated with
lower 30-day
mortality.
Author, Year
Tarnow-Mordi
23
et al., 2000

of PSP
Sample Size
Multicomponent
an
cross-sectional
intervention
1,050 patient
episodes

Model

Details
None
Not relevant, not Higher hospital

None
a study of an
mortality was
mentioned
intervention
significantly
associated with
patients exposure to
high versus moderate
overall ICU workload.
One medical-surgical ICU in

Scottish acute care hospital
between 1992 and 1995.
UK
Average and peak values of
nursing requirements per ICU
shift were calculated for each
patients day of stay in the
ICU.
Outcomes:
Harms
Influence of
Contexts on
Outcomes*
Overall
Risk of
Bias
None
mentioned
High

No assessment of
organizational complexities
SCTL:
Not assessed
Abbreviations: CI = confidence interval; FTR = failure-to-rescue; ICU = intensive care unit; OR = odds ratio; SCTL = Safety/Culture/Teamwork/Leadership
*Since there are no interventional studies in this section, we used this column to report results of context variables other than nurse staffing or workload as a modifier of the effect of nurse staffing on
patient outcome.
D-224
References
1.
Unruh LY, Zhang NJ. Nurse Staffing and

Patient Safety in Hospitals New Variable
and Longitudinal Approaches. Nursing
Research. 2012 Jan-Feb;61(1):3-12. PMID
WOS:000298158000002.
2.
Needleman J, Buerhaus P, Pankratz VS, et

al. Nurse Staffing and Inpatient Hospital
Mortality. New England Journal of
Medicine. 2011 Mar;364(11):1037-45.
PMID WOS:000288439000009.
3.
Twigg D, Duffield C, Bremner A, et al. The

impact of the nursing hours per patient day
(NHPPD) staffing method on patient
outcomes: A retrospective analysis of
patient and staffing data. International
May;48(5):540-8. PMID
WOS:000291713900003.
4.
Aiken LH, Sloane DM, Cimiotti JP, et al.

Implications of the California Nurse Staffing
Mandate for Other States. Health Services
Research. 2010 Aug;45(4):904-21. PMID
WOS:000279734400002.
5.
Harless DW, Mark BA. Nurse Staffing and

Quality of Care With Direct Measurement of
Inpatient Staffing. Medical Care. 2010
Jul;48(7):659-63. PMID
WOS:000279428200014.
6.
Schilling PL, Campbell DA, Englesbe MJ,

et al. A Comparison of In-hospital Mortality
Risk Conferred by High Hospital
Occupancy, Differences in Nurse Staffing
Levels, Weekend Admission, and Seasonal
Influenza. Medical Care. 2010
Mar;48(3):224-32. PMID
WOS:000275198200006.
7.
Aiken LH, Clarke SP, Sloane DM. Effects

of hospital care environment on patient
mortality and nurse outcomes. Journal of
Nursing Administration. 2008
May;38(5):223-9. PMID
WOS:000255996100012.
D-225
8.
Cho SH, Hwang JH, Kim J. Nurse staffing

and patient mortality in intensive care units.
Nursing Research. 2008 Sep-Oct;57(5):32230. PMID WOS:000259472200004.9.
Kiekkas P, Sakellaropoulos GC, Brokalaki
H, et al. Association between nursing
workload and mortality of intensive care
unit patients. J Nurs Scholarsh.
2008;40(4):385-90. PMID 19094155.
10.
Hamilton KE, Redshaw ME, Tarnow-Mordi

W. Nurse staffing in relation to risk-adjusted
mortality in neonatal care. Arch Dis Child
Fetal Neonatal Ed. 2007 Mar;92(2):F99F103. PMID 17088341.
11.
Mark BA, Harless DW, Berman WF. Nurse

staffing and adverse events in hospitalized
children. Policy Polit Nurs Pract. 2007
May;8(2):83-92. PMID 17652626.
12.
Rafferty AM, Clarke SP, Coles J, et al.

Outcomes of variation in hospital nurse
staffing in English hospitals: cross-sectional
analysis of survey data and discharge
records. Int J Nurs Stud. 2007
Feb;44(2):175-82. PMID 17064706.
13.
Stone PW, Mooney-Kane C, Larson EL, et

al. Nurse working conditions and patient
safety outcomes. Med Care. 2007
Jun;45(6):571-8. PMID 17515785.
14.
Tourangeau AE, Doran DM, Hall LM, et al.

Impact of hospital nursing care on 30-day
mortality for acute medical patients. Journal
of Advanced Nursing. 2006 Jan;57(1):32-44.
PMID WOS:000243297400003.
15.
Estabrooks CA, Midodzi WK, Cummings

GG, et al. The impact of hospital nursing
characteristics on 30-day mortality. Nurs
Res. 2005 Mar-Apr;54(2):74-84. PMID
15778649.
16.
Halm M, Peterson M, Kandels M, et al.

emotional exhaustion, and job
dissatisfaction. Clin Nurse Spec. 2005 SepOct;19(5):241-51; quiz 52-4. PMID
16179855.
17.
Person SD, Allison JJ, Kiefe CI, et al. Nurse

staffing and mortality for Medicare patients
with acute myocardial infarction. Med Care.
2004 Jan;42(1):4-12. PMID 14713734.
18.
Aiken LH, Clarke SP, Cheung RB, et al.

Educational levels of hospital nurses and
surgical patient mortality. JAMA. 2003 Sep
24;290(12):1617-23. PMID 14506121.
21.
Needleman J, Buerhaus P, Mattke S, et al.

Nurse-staffing levels and the quality of care
in hospitals. N Engl J Med. 2002 May
30;346(22):1715-22. PMID 12037152.
19.
Sasichay-Akkadechanunt T, Scalzi CC,

Jawad AF. The relationship between nurse
staffing and patient outcomes. J Nurs Adm.
2003 Sep;33(9):478-85. PMID 14501564.
22.
20.
Aiken LH, Clarke SP, Sloane DM, et al.

nurse burnout, and job dissatisfaction.
JAMA. 2002 Oct 23-30;288(16):1987-93.
PMID 12387650.
Tourangeau AE, Giovannetti P, Tu JV, et al.

Nursing-related determinants of 30-day
mortality for hospitalized patients. Can J
Nurs Res. 2002 Mar;33(4):71-88. PMID
11998198.
23.
Tarnow-Mordi WO, Hau C, Warden A, et

al. Hospital mortality in relation to staff
workload: a 4-year study in an adult
intensive-care unit. Lancet. 2000 Jul
15;356(9225):185-9. PMID 10963195.
D-226
Evidence Tables for Chapter 35. Patient Safety Practices Targeted at Diagnostic
Errors (NEW)
Author, Year
Diagnostic Error
Single Intervention Type

Additional Review Methods
Diagnostic
Galasko,
(51)
interpretation of
1971
radiographs
CarewMcColl,
(49)
1983
Diagnostic
interpretation of
radiographs
Robson,
(58)
1985
Diagnostic
interpretation of
radiographs
Ciatto,
(40)
1995
Diagnostic errors
during readings of
mammograms
Lind, 1995
(53)
Diagnostic errors in
surgical pathology
reports
Experimental Intervention
Review of x-ray films of

outpatients attending
accident services by
radiologist and other staff
within a short turn around
(24hrs)
Patient or Related
Outcome
Identification of a missed
injury
Review of x-rays in an
accident and emergency
department
Number of patients allowed

home with serious injuries
which were radiologically
apparent but which have
been overlooked
Review of x-ray films by
Diagnostic accuracy of the
radiologist and other staff
interpretation of fractures
impacting patient treatment
and prognosis
Independent double read of Breast cancer detection
mammograms by
rates and referral rates
experienced radiologists
Review of surgical
diagnostic pathology
biopsies by a second
pathologist prior to release
of final reports
Major diagnostic errors in

surgical pathology reports
that could directly affect
patient care
D-227
Study Design: Result
Notes
Other: Retrospective review of radiographs

24 hrs after initial interpretation. In 0.6% of
cases, review identified a missed injury. In
0.4% of cases, the radiologist failed to
identify the injury while the senior
houseman on duty did. In 0.6% of cases,
review identified a missed injury, and in
only 2 of 4,665 (0.04%) of cases, both
review sessions missed the injury.
Other: The majority (85%) of abnormalities
were identified. Most overlooked
abnormalities were not clinically significant.
Missed diagnosis;
misdiagnosis
Other: Diagnostic accuracy is correlated

with seniority and experience (the casualty
officer was more accurate than the
students, second only to the radiologist).
Other: The mean increase in referral rate
for double reading compared with single
reading was 15.1%, and increased cancer
detection by 4.6%.
Other: 380 errors in 2,694 cases. 32 major
errors with a potential for inappropriate
patient care, 104 diagnostic discrepancies,
192 minor errors and 52 clerical errors.
Missed diagnosis;
misdiagnosis
Missed diagnosis;
misdiagnosis;
proof of concept
Missed diagnosis
Misdiagnosis;
delayed
diagnosis
Author, Year
Bruner,
(47)
1997
Dudley,
(29)
1997
Thiesse,
(59)
1997
Lufkin,
(54)
1998
Espinosa,
(50)
2000
Diagnostic Error
Patient or Related
Outcome
Diagnostic
Review of brain or spinal
Substantial and serious
discrepancies in brain cord biopsy results by a
neuropathology diagnostic
and spinal biopsy
neuropathology consultation errors
reports
service
Other: Disagreement between original and

review diagnoses in 42.8% cases, with 8%
serious errors in diagnosis. 96 cases (44%)
less serious, but still clinically substantial.
31.9% disagreements occurred in patients
referred directly compared to 51.0% of
disagreements in review done solely based
on pathology consultation.
Serious errors of
Provision of an (ECG)
Serious errors of ECG
Other: Serious errors reduced by 59%
electrocardiograph
report by a cardiac
interpretation
when there was a prior technical report
(ECG) interpretation
technician at the time of
provided by an ECG technician. Many of
these errors led to worse clinical outcomes.
in an accident and
recording, before senior
Independent review of ECG revealed
emergency
house officers ECG
moderate agreement between technicians
department
interpretation
and senior officers (kappa = 0.45) and
between senior officers from different
departments (kappa = 0.42).
Incorrect radiologic
Independent ascertainment Response status of cancer Other: Major disagreements occurred in
evaluation of overall
of therapy response status patients - major and minor
43% and minor in 8% of reviewed files.
response (to therapy) of cancer patients by review disagreements between trial Number of tumor responses to therapy
status in oncologic
of radiologic findings by an investigators and evaluation designated as significant was reduced by
patients participating evaluation committee (EC) committee (that could
23.2% after review by EC.
impact patient
in multi-center trials
management)
Incorrect radiologic
Radiologists review of
Other: Emergency department physicians
diagnoses by
radiographs interpreted by
discordant radiographic
were confident in their interpretations in
emergency
emergency room physicians interpretations that alter
9,599/16,410 cases (58%). Review of the
department
patients treatment
118 discordant interpretations in the
physicians
confident group demonstrated 11 were
significant. Discordant interpretations were
higher in cases when emergency
department physicians were not confident.
Radiograph
Pre/Post: Longitudinal study; after
Review of radiograph
Reduction of errors,
interpretation errors in discrepancies at monthly
implementation, false negative error rate
including missed fractures
emergency
decreased from 3% to 1.2%. Review
meetings by radiologists;
or foreign bodies.
department
process revised; patient satisfaction
team redesigned the review
improved, as did turnaround time for
process after intervention
interpretations.
D-228
Notes
Misdiagnosis
Missed diagnosis
Diagnostic
discrepancy:
proof of concept
Misdiagnosis
Missed diagnosis;
misdiagnosis;
delayed
diagnosis
Author, Year
Nam, 2001
(30)
Westra,
(60)
2002
Canon,
(48)
2003
Kwek, 2003
Nordrum,
(44)
2004
*Howard,
(42)
2006
(52)
Diagnostic Error
Patient or Related
Outcome
Misdiagnosis due to
Repeat colonic transit study Success rate post
inaccurate
in patients with chronic
colectomy for chronic
interpretation of
constipation and suspected constipation
colonic transit study in colonic inertia to confirm the
patients with chronic diagnosis prior to colectomy
constipation
Missed diagnosis of
A secondary review of
Treatment modification
head and neck
histopathologic diagnoses
based on changes in
cancers
diagnoses
Detection of polyps
Secondary reading of
Diagnostic accuracy of
and/or colorectal
barium enemas
polyps and carcinomas
carcinomas
Experimental Design: The success rate of

colectomy for colonic inertia was
significantly higher in patients who
underwent a repeat transit study confirming
inertia than in patients who underwent
colectomy based on a single study.
Other: Retrospective review; in 87% of
diagnoses changes treatment was
modified.
Other: Prospective study; double reading
of barium enemas did not improve
sensitivity and increased false positive
rate.
Breast cancer
Double reading of
Other: Retrospective review; doubledetection
mammograms in Singapore mammography screening
reading mammography intervention led to
Breast Screening Project
results
cancer detection improvement. Double
reading increased the number of patient
recalls. Positive predictive value (PPV)
decreased from 8.2% to 6.1%.
Incorrect histologic
Use of still images sent via Discordant diagnoses
Other: Agreement 67.8% of the time with
diagnoses
electronic network (from
expected to have clinical
still images, and 68.9% of the time with
glass slides of paraffinimplications
reviewing glass slides, when compared to
embedded histologic
an original second opinion diagnosis. The
material) to obtain second
cause of error was interpretation for 15 (of
pathologists opinion
90 cases), both image selection and
diagnosis
interpretation for 9 cases, image selection
alone in 3 cases, and image quality,
selection and interpretation in 2 cases.
37.9% of still image discrepancies are
likely to have had significant clinical
implications.
Missed injuries in
Implementation of a trauma Missed injuries in trauma
Other: 14% of patients had one or more
trauma patients in a
tertiary survey (reevaluation patients in a Level II trauma injuries missed in primary and secondary
Level II trauma center of laboratory studies) within center
examinations that were captured during
24 hours of admission
tertiary examination.
D-229
Notes
Misdiagnosis
Missed diagnosis;
misdiagnosis
Missed diagnosis;
misdiagnosis
Missed diagnosis;
misdiagnosis
Misdiagnosis;
delayed
diagnosis
Missed diagnosis
Author, Year
(45)
Raab, 2006
Singh, 2006
(46)
Duijm, 2007
(41)
Manion,
(43)
2008
Parameswara
(56)
n, 2008
Diagnostic Error
Patient or Related
Outcome
Pre/post: Double viewing did not lower the
Pre sign-out double viewing Incorrect diagnoses (that
interpretation of
frequency of cytologic-histologic correlation
of all pulmonary cytology
could impact patient
pulmonary cytology
slides
management and outcome) false-negative errors. Double viewing
slides (based on
detects errors in up to 1 of every 37 cases.
correlation of cytology
While the double cytology slide viewing
and surgical
was helpful at some project sites in
specimens histology
detecting pre-sign out error, the
results)
intervention did not significantly reduce
error frequencies at any of three study
sites. Agreement with subsequent surgical
diagnosis was moderate when definitive
diagnoses were made.
Diagnostic error for
Urgent referral; timingPresence of cancer
Other: Retrospective review and audit;
head and neck cancer based referral to reduce
diagnosis, and the time
86% of urgent patients were seen within 2
delay in diagnosis and
delay to reach diagnosis
weeks. 24% had oral squamous cell
influence on diagnostic
carcinoma.
pathway
Missed breast cancer Independent double reading Cancer detection rates and Other: Additional reading by technologists
diagnosis
of mammograms by two
referral rates of women with increased the cancer detection rate by 0.36
cancers per 1,000 women and the referral
mammography
positive screening results
rate by 0.13%.
technologists beyond the
from any reader
standard double reading by
two radiologists
Second opinion in pathology Diagnostic accuracy no
Other: Retrospective chart review of major
diagnostic errors
to expose clinically
major disagreement in
diagnostic disagreements (2.3% of
(varied clinical
significant errors
pathology that would impact reviewed cases). Second opinion for
conditions)
treatment or prognosis
clinically significant error validated in 34 of
132 cases reviewed.
Missed abnormal
Sampling the remaining
Identification of pathologic
Other: Review sampling showed pathologic
findings in histology
tissue of colorectal biopsies abnormalities in remaining
abnormality in 3.9% cases. New diagnostic
specimens
originally diagnosed as
biopsy tissues (originally
information identified in 1.7% of cases, but
normal- with additional step reported as normal)
lost in 1.3% of cases (present in initial
sections- to reveal
sections but not in remaining tissue).
pathologic abnormalities
D-230
Notes
Missed diagnosis;
misdiagnosis
Delayed
diagnosis
Missed diagnosis
Missed diagnosis;
misdiagnosis
Missed diagnosis;
delayed
diagnosis
Author, Year
(57)
Raab, 2008
Murphy,
(55)
2010
Hamady,
(71)
2005
Diagnostic Error
Patient or Related
Outcome
Incorrect surgical
Two diagnostic error
Although this was a
pathology diagnoses detection processes:
retrospective review of
targeted review of a random surgical pathology
5% of surgical pathology
specimens, the study
specimens, and focused
evaluated: a) impact
review by 3 subspecialty
identified diagnostic errors
pathologists of cases with a could have had on patient
perceived higher level of
outcomes, including
diagnostic uncertainty or
management, and b)
lack of standardization in
whether the patient
terminology
experienced subsequent
harm
Missed colonic and
Double reporting by two
Identification of clinically
extra-colonic lesions
radiologists of minimal
relevant colonic and extrain minimal preparation preparation CT of colon
colonic lesions that could
Computer
(MPCTC) in elderly patients impact future patient
Tomography (CT) of
management
colon
Incorrect
Pathology reports that
Malignancy status of tumor,
interpretation of
received discrepant
indicated course of
pathology reports for interpretations from a
treatment and expected
thyroid cancer
referring and receiving
prognosis
clinician were reviewed by a
third clinician blinded to the
thyroid cancer diagnosis
Educational Interventions
McCarthy,
Incorrect diagnosis by Teach parents Acute Illness Number of infants with
(86)
1990
parents of symptoms Observation Scale (AIOS)
serious illnesses
of serious illness
to detect childs illness vs.
3-point global scoring
system for evaluating
chance of serious illness
D-231
Notes
Other: Targeted review process identified

Missed diagnosis
195 errors, (2.6% of reviewed cases) and
focused review process identified 50 errors
(13.2%). The number of major errors
detected was 27 (0.36%) and 12 (3.2%),
respectively. In secondary review of major
errors (follow-up range from 8 months to
5.5 years), subsequent harm to the patient
was observed in 11 cases (41%) from the
targeted review and 7 cases (58%) from
the focused review.
Other: Double reporting of colonic
identified one extra-colonic cancer, at the
expense of 5 unnecessary endoscopies.
The positive predictive value for colon
cancer was 69% for single reporting and
54.5% for double reporting.
Other prospective design: Of 66 patients
with thyroid cancer referred from general
hospitals to specialty clinics for a second
opinion on diagnosis, 12 cases (18%)
received disagreement between initial and
second review of the pathology report,
resulting in re-review by a third, blinded
reviewer. Five cases involved strong
disagreement leading to change in both
prognosis and treatment strategy. All 12
cases involved a change in prognosis:
worsened in 8 (67%) and improved in 4
(33%). There were two cases each where
re-review resulted in a switch from benign
to malignant and vice versa.
Missed diagnosis
RCT: Judgments of the intervention group

were more reliable than those of the
control group (weight kappa = 0.50 vs.
0.26). Sensitivity, positive and negative
predictive values not statistically different.
Misdiagnosis;
proof of concept
Misdiagnosis
Author, Year
Diagnostic Error
Fridriksson,
(87)
2001
Misdiagnoses of
sudden onset
headache (an early
sign of ruptured
aneurysm)
Thaler,
(88)
2010
Errors in ECG
readings due to
switched electrode
cables
A community teaching
program on educating local
physicians about sudden
onset of headache in
subarachnoid hemorrhage
(SAH); continuous
interaction between
neurosurgeons and local
physicians including
seminars on SAH, individual
follow-up of all referred
patients
A 45 min teaching session
for ICU nurses and
physicians about correct
ECG recording and errors
resulting from improper
electrode placements
Personnel Changes
Incorrect diagnoses of Comparison of diagnostic
De Lacey,
(84)
radiographs in
1980
accuracy between casualty
accident and
officers and radiologists
emergency
departments
(37)
Sakr, 1999
errors, including
errors in the diagnosis
pathway (i.e., history,
physical examination,
and radiographic
interpretation errors)
Use of junior doctor or

nurse practitioner (NP) in
providing care in the
Emergency Department
Patient or Related
Outcome
Early misdiagnoses of
ruptured aneurysms;
aneurysm surgery rates,
surgery outcomes and
morbidity and mortality
outcomes at 6 months post
SAH
Reduction of cable reversal

rates (which could lead to
incorrect ECG diagnoses
and unnecessary
subsequent tests and
hospitalizations)

Other: An initial misdiagnosis was
identified in 12% of patients, and
diagnostic error decreased by 77% with
intervention.
D-232
Missed diagnosis
Pre/post: Frequency of electrode cable

Missed diagnosis;
misplacements was 4.8% pre-intervention misdiagnosis
and 1.2% post-intervention. This translates
to a 75% reduction in ECG errors due to
electrode cable reversals.
Other: Prospective study to compare

radiograph interpretation between casualty
officers and radiologists. Uncertain or
incorrect interpretation led to 6.8% of all
patients receiving unnecessary procedures
(e.g., casting an unbroken limb), 1.7%
unnecessary return to X-ray, and 0.6%
unnecessary outpatient referral.
Clinically important errors in RCT: There was no difference between the
history, examination,
clinically important radiographic diagnostic
radiograph interpretation,
errors made by NPs and by junior doctors
treatment and/or advice
(e.g., 89.8% of patients seen by junior
and/or follow-up
doctors reported improvement in condition,
while 91.1% of patients seen by NPs
reported improvement). 15% of patients
seen by junior doctors required follow-up
visits within 28 days while 9.7% of patients
seen by NPs required follow-up visits
within 28 days.
radiograph interpretation
Notes
Missed diagnosis;
misdiagnosis
Missed diagnosis;
misdiagnosis
Author, Year
Diagnostic Error
Structured Process Changes

Enderson,
Missed injuries
(91)
1990
associated with
trauma
Klassen,
(92)
1993
Missed positive
radiographic findings
(fracture, dislocation
or effusion) after
trauma
Schriger,
(39)
2001
Occult mental illness
(93)
Tertiary Survey to capture

missed trauma injuries
Brand protocol (protocol for

ordering radiographs of
injured extremities in
patients >15 years old)
applied by triage nurses to
determine the need for a
radiograph in the pediatric
emergency department
Implementation of
computerized psychiatric
interview (PRIME-MD)
Patient or Related
Outcome
Notes
Diagnostic accuracy;
improvement of patient
outcomes (mortality and
morbidity) by identifying
missed injuries
Number of positive
radiographic findings;
number of missed positive
radiographic findings and
long-term clinical
importance thereof, in
pediatric trauma patients
Pre/Post: 41 missed injuries were identified Missed diagnosis

in 37 patients (N = 399) with Tertiary
survey.
RCT: Brand group ordered 81.9%

radiographs; control 87.1%. Positive
radiograph percentage was 40.8% vs.
42.6%, respectively. 3.2% were missed in
Brand compared to 0% in control.
Missed diagnosis
Detection of occult mental

illness (upon admission to
emergency department)
RCT: PRIME-MD survey, completed by

emergency department patients, provided
to emergency physician did not improve
the frequency of diagnosing psychiatric
conditions. 42% of patients within the study
were identified as high risk for occult
psychiatric illness according to PRIME-MD.
Physicians reached psychiatric diagnosis
5% and offered psychiatric consultations to
3%.
Pre/post: Missed injuries decreased from
2.4% to 1.5% overall, and from 5.7% to
3.4% in Trauma ICU patients after TS
implementation. Missed injuries occurred
more often in older patients, those that
were admitted and those with high injury
severity scores.
Other: Prospective study; 13 missed
injuries identified in 12 of 76 pediatric
trauma patients. Fractures were the most
common missed injury. Children involved
in motor vehicle incidents were most likely
to have missed injuries.
Missed diagnosis
Biffl, 2003
Missed injuries in
Level I trauma center
Routine trauma survey (TS) Missed injuries in level I

in trauma intensive care unit trauma center
patients
Soundappan,
(94)
2004
Missed injuries
associated with
trauma (pediatric)
Extended tertiary survey in

pediatric trauma patients
Incidence of missed
diagnosis
D-233
Missed diagnosis;
delayed
diagnosis
Missed diagnosis
Author, Year
Diagnostic Error
*Perno,
(25)
2005
Delayed diagnosis of
injury in a Pediatric
Trauma Center
Ursprung,
(95)
2005
Diagnostic errors
related to laboratory
tests or radiologic
studies; delays in
patient care or
information transfer/
communication
(additional errors
probed)
Incorrect
interpretation of
frozen sections of
pathology specimens
(98)
Raab, 2006
(96)
Raab, 2006
Improved diagnostic
accuracy from Pap
test
Patient or Related
Outcome
Implementation of a
Delayed diagnoses of injury Other: DDI occurred in 15 (0.46%) of
trauma patients. Previous study by same
Pediatric Trauma Response (DDI) in admitted pediatric
group revealed 4.3% DDI, an almost 10team and trauma service for trauma patients
fold decrease between the two studies.
severely injured children in
Among the 15 DDI cases in the latter
Pediatric Trauma Centers
study, 13 diagnoses were identified by
tertiary examination, and 2 patients were
discharged without diagnosis and returned
to the hospital after worsening symptoms.
Real time patient safety
Impact of errors (i.e., delays Other: 338 errors detected; 27 of 36 items
auditing during routine
in patient services or errors on checklist detected >1 error. Significant
clinical work in the ICU (36- in information transfer) on
safety errors were detected promptly and
item patient safety checklist patient clinical management rapid changes in policy and practice
focused on several errors
and on adverse outcomes
ensued.
including diagnostic errors)
Continuous monitoring over

time of data correlation
between frozen sections
and permanent sections via
the Q-Tracks Quality
Improvement Program
Number of frozen
permanent section
discordant results and
deferred diagnoses (that
management and outcome)
Toyota production system

redesign to improve
workflow by 1-by-1
continuous flow process
Decrease in additional Pap

test or surgical procedure,
increase in diagnostic
accuracy
D-234
Notes
Missed diagnosis;
delayed
diagnosis
Missed diagnosis;
misdiagnosis;
delayed
diagnosis
Other: Mean frozen-permanent section

Missed diagnosis;
discordant frequencies 1.36%. Longer
misdiagnosis
participation in Q-Tracks significantly
associated with lower discordant
frequencies; 4- or 5-year participation
showed decrease in discordant frequency
of 0.99%; 1-year was 0.84%. Median
discordant rates increased with increased
bed size of institution. Government-owned
institutions exhibited lower deferred
diagnoses than non-government
institutions.
Experimental Design; Pre/Post: 8-month
Missed diagnosis
non-concurrent cohort study; the number of
correlating Pap tests and surgical
pathology specimens increased from 42 in
pre-intervention to 51 in the intervention
group. Slight, but not significant, decrease
in diagnostic discrepancies between preintervention/post-intervention.
Author, Year
(97)
Raab, 2006
(99)
Diagnostic Error
Patient or Related
Outcome
Thyroid gland fine
Standardized terminology
Diagnostic accuracy of the
needle aspiration
scheme (Toyota Production FNA interpretation; surgery
(FNA) diagnostic error System Process Redesign) rates and repeated FNA
for reporting of cytologic
rates
results from thyroid fine
needle aspirations (FNA)
Raab, 2008
Incorrect Pap test

cytologic diagnoses
Mueller,
(100)
2010
Geriatric health
problems previously
unknown to a general
practitioner (GP) and
overlooked treatment
needs
De Vries,
(101)
2011
Surgical diagnosis
accuracy
Pre-post: Separate cohorts/interventions

analyzed; post intervention significantly
fewer patients had surgery, received noninterpretable results, or repeated FNA.
False-negative diagnosis rate decrease
from 41.8% to 19.1% (p = .006), FNA
sensitivity increased from 70.2% to 90.6%
(p < .001), and atypical diagnoses rate
decreased from 8.2% to 3.7% (p < .001).
The false positive rate increased slightly
and FNA specificity decreased but neither
difference was significant.
Continuous monitoring of
Pap test diagnostic
Other: Longer participation in program by
the correlation of Pap test
accuracy and detection of
an institution associated with higher Pap
cytologic-histologic data pre-neoplastic lesions (that test sensitivity and higher proportion of
via the Q-Tracks Quality
positive histologic diagnoses for a Pap test
Improvement Program
management and outcome) of atypical squamous cells (ASC). Longer
participation also associated with higher
proportion of women with follow-up positive
histologic diagnoses for ASC. Compared to
government-owned institutions, nongovernment institutions exhibited slightly
higher predictive value of positive tests.
Larger institutions had significantly lower
sensitivity, but time of participation in the
quality improvement program remained a
significant factor in all analyses.
Standardized Assessment
Further management
Other: STEP intervention helped GPs
for Elderly Patients in a
interventions planned by
identify missing or unknown
Primary Care Setting (a 44- GPs for previously
immunizations, anxiety in patients, chest
item STEP instrument
overlooked geriatric health pain, depression, urinary incontinence,
based on self-reporting and problems and treatment
breathlessness, smoking habits as well as
standardized patient
needs (that could affect
claudication, abnormal clock drawing test,
interview), to explore
patient outcome)
and thyroid dysfunction. Patients had a
conditions new to GPs
median of 11 health problems identified by
STEP, of which 2 were new to the GP.
Surgical checklist, SURgical Morbidity, mortality, level of Other: Retrospective review; cognitive,
PAtient Safety System
patient disability and need
system, technical or unknown
(SURPASS); review of
for additional operations;
categorization of errors determined
claims records to see if
malpractice claims
postoperatively.
checklist could have
prevented claims
D-235
Notes
Missed diagnosis
Missed diagnosis;
misdiagnosis
Missed diagnosis;
delayed
diagnosis
Missed diagnosis;
misdiagnosis;
delayed
diagnosis; proof
of concept
Author, Year
Technique
Attard,
(72)
1992
Diagnostic Error
Patient or Related
Outcome
Incorrect diagnosis in
patients presenting
with abdominal pain
Pain relief with paraveretum Incorrect management

for acute abdominal pain
decision (to operate or not)
and incorrect discharge
diagnoses
Resnick,
(73)
1996
Incorrect diagnosis of
urinary incontinence
in nursing home
women
Stress test combined with

cystometry to diagnose
urinary incontinence
Misdiagnoses of urinary
stress incontinence
Borgstein,
(74)
1997
Incorrect appendicitis
diagnosis
Correct diagnoses post

laparoscopy (and postappendectomy when
surgery was performed)
Vermeulen,
(75)
1999
Incorrect appendicitis
diagnosis
Diagnostic laparoscopy for

female patients of childbearing age with clinical
signs of acute appendicitis,
prior to appendectomy
The influence of pain
medication administration
on diagnosis of appendicitis
Prieto,
(76)
2003
Incorrect indication of
surgical margins of
melanocytic lesions in
en face frozen
compared to
permanent paraffinembedded sections
Use of en face frozen

sections (i.e., sections cut
parallel to the surgical
margin) for evaluation of
surgical margins of
melanocytic lesions
Although no direct patient

outcomes studied,
evaluation of the diagnostic
accuracy of a rapid method
to identify the surgical
margins of melanocytic
lesions could have had
impact on patient
management and outcome
Diagnostic accuracy;
whether surgery was
deemed necessary or not
D-236
RCT: Reduction in pain after paraveretum,

without reducing diagnostic accuracy.
Subsequent decision to operate or observe
was considered incorrect in fewer cases
treated with paraveretum vs. the saline
group (2/50 vs. 9/50; p=0-051, Fishers
exact test).
Other: Combining cystometry with stress
test improved diagnostic accuracy
drastically. Of the 77% of women in whom
the results of both tests were congruent, all
were correctly classified (vs. videourodynamic evaluation). No woman with
stress incontinence was missed by the
two-test strategy, nor was anyone with DH
misclassified. Neither test was more
accurate in cases where the test results
diverged.
Other: The negative appendectomy rate
after laparoscopy was 5%. In the group of
fertile females without laparoscopy the
negative appendicectomy rate was 38%.
Notes
Misdiagnosis
Missed diagnosis;
misdiagnosis
Misdiagnosis;
delayed
diagnosis
Experimental Design: Emergency

Misdiagnosis
department patients presenting with pain in
lower right abdominal quadrant were
randomized to receive morphine or
placebo. The morphine cohort had a higher
positive predictive value, and lower
negative predictive value; differences
between morphine and placebo group
were not statistically significant.
Other: Poor overall agreement by frozen v. Misdiagnosis;
permanent analysis (kappa = .03).Better
proof of concept
agreement between frozen and permanent
section diagnoses for the non-melanocytic
lesions (NML) than for the malignant
melanomas (MM) cases. Within-physician
agreement ranged from poor to moderate
(kappa range from -.1 to .4).
Author, Year
Diagnostic errors
based on altered
physical examination
findings
Thomas,
(26)
2003
Kokki, 2005
Hewett,
(78)
2010
Diagnostic Error
(77)
Delay in diagnosis or
decrease in
diagnostic accuracy of
physical examination
findings for
appendicitis
Patient or Related
Outcome
Morphine sulfate (MS)
Patient disposition and
administered for pain during ultimate diagnosis (including
diagnostic process
presence and severity of
physical findings)
Oxycodone for pain relief in

children presenting to the
emergency department with
moderate to severe
abdominal pain
Notes
RCT: No differences between control and

MS group with respect to disposition from
the emergency department, ultimate need
for operation, ultimate diagnosis (according
to medical records and patients follow up)
and need for repeat physician visit within a
week for abdominal pain.
RCT: Prospective, double-blind, and
placebo-controlled clinical trial; there was
significantly greater reduction in pain
reported on a visual analog scale among
patients that received oxycodone than
those administered saline placebo. From
before drug or placebo administration to
after administration, diagnostic accuracy
increased from 72% to 88% in those
treated with oxycodone and remained at
84% in the placebo group. The rate of
negative exploratory laparotomy was
similar in both groups.
Experimental Design: Patients undergoing
cap-fitted colonoscopy had significantly
lower miss rate for all adenomas compared
with regular colonoscopy (21% vs. 33%),
but there was no difference when analyzed
at the patient level rather than number of
adenomas.
Missed diagnosis;
misdiagnosis
Diagnostic accuracy; time to Experimental Design: In control group,

reach diagnosis
MEDITEL identified correct diagnosis in
85% of cases; physicians reached the
correct diagnosis in 65% of cases. In the
experimental group, MEDITEL reached
correct diagnosis in 58% of cases, and
physicians in 83%. Time to diagnosis
reduced in the experimental group, but did
not reach statistical significance.
Discharge diagnosis
RCT: Randomized trials with prospective
accuracy
data collection; predictive accuracy of CAD
was 48% initially, but rose to 69% with
decision aids, computers and performance
feedback.
Missed diagnosis;
misdiagnosis;
delayed
diagnosis
Pain relief and diagnostic

accuracy of physical
examination findings and
clinical outcomes
Missed colorectal
Cap-fitted colonoscopy,
Missed colorectal adenoma
adenoma diagnosis in which allows for flattening of diagnosis in colonoscopy
colonoscopy
haustral folds and/or
improves mucosal exposure
Technology-based Systems Interventions

Wexler,
Time to correct
Computer-assisted system
(103)
1975
diagnosis
of diagnosis (MEDITEL)
Wellwood,
(104)
1992
Appendicitis diagnosis Computer-aided diagnostic

accuracy
(CAD) tool; abdominal pain
interpretation
D-237
Missed diagnosis;
misdiagnosis;
delayed
diagnosis
Missed diagnosis
Missed diagnosis;
misdiagnosis
Author, Year
Diagnostic Error
Patient or Related
Outcome
Implementation of a
Diagnostic accuracy (proxy
computerized acute cardiac CCU or telemetry unit
ischemia time-insensitive
admission)
predictive instrument (ACITIPI)
Notes
Missed diagnosis
or delayed
diagnosis
(presumed from
appropriate
admissions)
Implementation of an acute
cardiac ischemia predictive
instrument, similar to acute
cardiac ischemia timeinsensitive predictive
instrument (ACI-TIPI),
calculated and delivered in
hardcopy to clinicians
Experimental Design: controlled clinical

trial. Appropriate admission to CCU or
telemetry unit did not change for patients
with acute MI or unstable angina when
ACI-TIPI implemented. Use of ACI-TIPI
reduced CCU admissions from 14% to
10%, telemetry unit admissions from 39%
to 31% and increased discharges to home
from 45% to 65% for non- AMI patients.
Among patients with stable angina, use of
ACI-TIPI reduced CCU admissions from
26% to 13% and increased discharges
from 20 to 22%. Telemetry unit admissions
decreased from 68% to 59%.
Experimental Design: interrupted time
series. Of the 2,320 patients seen across
six emergency departments, diagnostic
accuracy (83.4% vs. 79.6%, p = .002) and
specificity (78.1% vs. 73.2%, p = .002), but
not sensitivity (94.5% vs. 95.3%), were
significantly improved by providing
physicians with predictive instrument
results. False-positive rate among patients
with a low probability of ischemia dropped
significantly (47% vs. 60%, p = .002), and
admissions to CCU significantly decreased
from 44% to 33% (p = .001) among
patients without ischemia when physicians
had access to predictive instrument results.
RCT: Prospective, randomized controlled
trial; intervention group had a 38% shorter
median time interval between receipt of
critical laboratory result and action with
patient. However, the time until alerting
condition was resolved did not reach
clinical significance.
Selker,
(105)
1998
Missed diagnosis of
acute cardiac
ischemia
Pozen,
(117)
1984
Missed diagnosis of
acute cardiac
ischemia
Kuperman,
(28)
1999
Time interval between Computer system to detect Interval from when a critical
laboratory results and critical conditions and notify result was available for
clinical action
the physician
review until appropriate
treatment administered
Diagnostic accuracy and

proxy of CCU admission
D-238
Missed diagnosis,
misdiagnosis
Delayed
diagnosis
Author, Year
Bogusevicius,
(27)
2002
Major, 2002
(32)
*Poon,
(106)
2002
(107)
Gur, 2004
Diagnostic Error
Acute mechanical
small bowel
obstruction
Patient or Related
Outcome
Computer-aided diagnosis
Time to diagnosis;
(CAD) and contrast
morbidity, mortality,
radiography for diagnosis of sensitivity, specificity and
acute mechanical small
positive/negative predictive
bowel obstruction
values
Diagnostic errors from

omission of laboratory
alerts and physiologic
condition alerts
Computer system coupled

to an alert engine to reduce
errors of omission for critical
care units. Patients
randomly assigned to (1)
alerts group, or (2) no alerts
group.
Inadequate
Result Notification via
communication to
Alphanumeric pagers
physicians of patients (ReNAP) feature in clinical
laboratory test results information system for realtime laboratory notification
of physicians via pagers
Recall and breast
Introduction of computercancer detection rates aided detection (CAD) and
mammography diagnosis
system
Kakeda,
(108)
2004
Lung cancer detection Computer-aided diagnosis

(CAD) system to detect
nodules from lung cancers
Cupples,
(109)
2005
Breast cancer
detection rates (from
mammography
screening program)
Unsafe workups
Ramnarayan,
(110)
2006
Fenton,
(111)
2007
Breast cancer
detection rates
Implementation of
computer-aided detection
(CAD) program
Implementation of
(CAD) program
Implementation of
(CAD) technology to assist
in the interpretation of
mammography
Mortality
Although no patient
outcomes studied (only
usage patterns and users
satisfaction studied) related
to unnecessary delays in
patient care
breast cancer
Notes
RCT: Prospective, randomized clinical trial;

CAD had no significant advantage over
contrast radiography in terms of diagnostic
accuracy, but reduced time to diagnosis.
Mean time to diagnosis was 1 hour for
CAD and 16 hours for contrast
radiography.
Experimental Design /Other: Prospective
data collection; patients in alerts group had
a higher mortality rate in both SICU and
ward compared to no alerts. Critical alerts
for ICU patients increased morbidity and
mortality.
Missed diagnosis;
misdiagnosis;
delayed
diagnosis
Pre/post: Improved ReNAP usage patterns Delayed

and satisfaction.
diagnosis
Other: Retrospective review; recall rates

were 11.39% and 11.40% for without CAD
and with CAD, respectively. Cancer
detection rates were 3.49% and 3.55%
without CAD and with CAD, respectively.
Diagnostic accuracy of lung Other: Retrospective review; CAD system
cancer
improved the detection of lung nodules by
improving area under ROC curve from
0.924 to 0.986.
Screening results
Other: Prospective study; cancer detection
(diagnostic accuracy) of
increased 16.3%, with invasive cancer
breast cancer detection
detection increasing 164% while in situ
rates
cancer detection declined 6.7%.
Diagnostic accuracy
Pre/Post: Prospective study; CAD reduced
the number of unsafe workups from
45.2% to 32.7%.
Diagnostic accuracy
Other: Comparative study; cancer
detection rate did not improve with use of
CAD in mammography screening.
Specificity decreased from 90.2% to
87.2%.
D-239
Missed diagnosis;
misdiagnosis
Missed diagnosis;
misdiagnosis
Missed diagnosis
Missed diagnosis;
misdiagnosis
Missed diagnosis;
misdiagnosis
Author, Year
(112)
Diagnostic Error
Patient or Related
Outcome
Starting in 2005, SMS text
Time delay between receipt
message notifications with
of clinically significant
patient critical values sent to information and appropriate
clinician. From 2001-, a
course of patient treatment
callback system had been in
place to send patient critical
values to clinicians
Park, 2008
Interval between
results and clinical
action
*Piva,
(113)
2009
Failure to adequately Computerized notification

communicate a critical system for reporting critical
laboratory value
values
*Singh,
(114)
2009
Inadequate
communication of
abnormal cancerscreening test results
in electronic health
records
Misdiagnosis of noninfectious conditions
as cellulitis
David,
(115)
2011
Etchells,
(33)
2011
Diagnostic errors due

to lack of timely
information of
physicians for critical
laboratory
abnormalities

Pre/post / Other: Comparative study; time
to action for critical hyperkalemia in ICUs
and general wards in 2001 was 213
minutes and 476 minutes, respectively. In
2005, with SMS, times dropped to 74.5
minutes and 241 minutes, respectively.
Clinical response to callback alerts was
73.3%, and was 79.3% for SMS texts.
Other: The computerized system improved
communications within 1 hour timeframes
as compared to the traditional phone
process for all hospital services except
medical specialties.
Pre/post: Lack of timely follow-up
decreased immediately from 29.9% to
5.4% and was sustained at month 4 after
implementing the intervention.
Although no patient
outcomes studied, timely
physician notification could
have had impact on patient
events
Timely follow-up of
alert for positive fecal occult abnormal cancer screening
blood (FOBT) cancer
test results (FOBT) to
screening test results
reduce missed or delayed
diagnoses of colorectal
cancer
Visually-based
Number of patients admitted Other: In 18/28 of misdiagnoses, VCDDSS
to the hospital with an
computerized diagnostic
included the correct diagnosis, while in
incorrect cellulitis diagnosis only 4/28 cases did the physician identify
decision support system
(VCDDSS, also named
the correct diagnosis.
VisualDx) to generate an
improved differential
diagnosis
Real-time clinical alerting
Clinical actions completed
Experimental Design: Based on laboratory
systems for critically
in response to the alerts
values, 50% of potential clinical actions
abnormal laboratory values (that could affect patient
occurred when the alert system was on as
via text messages sent to
outcome) and patients
well as 50% while off. Adverse events
physicians using
adverse events
within 48h were actually higher in cases
alphanumeric pagers or
while alert system was on (42%) than while
smart phones (decision
off (33%) but this difference only
support also provided via
approached significance (p = .06).
smart phones or hospital
intranet)
D-240
Notes
Delayed
diagnosis
Delayed
diagnosis
Missed diagnosis;
delayed
diagnosis
Misdiagnosis
Delayed
diagnosis
Author, Year
Diagnostic Error
Patient or Related
Outcome
Fitzgerald,
Errors during
Real time computerRCT: Error free resuscitation in 16% of
Patient morbidity and
(34)
2011
reception and
prompted evidence-based
mortality; including length of baseline controls and 21.8% intervention.
resuscitation of
decision support system
Predicted mortality rate 11%, but actual of
hospital stay
severely injured adult (with decision and action
5.2%, meant insufficient power for
trauma patients
algorithms) during reception
analyzing a true mortality difference
(including errors in the and resuscitation of
statistically. No significant reduction in
diagnosis pathway)
severely injured adults in
sepsis or adult respiratory distress
Level I adult trauma center
syndrome, but aspiration pneumonia was
reduced from 5.3% (control group) to 2.5%
(intervention).
Olsson,
Missed or delayed
Neural network-based
Diagnostic accuracy (proxy Other prospective design: Compared to
(116)
2006
diagnosis of acute
decision making tool added for indicated treatment for
cardiology attending, interns regularly
cardiac ischemia
to ECG results to
ST-segment elevation
treating chest pain patients in the
recommend statistical
myocardial infarction)
emergency department classified 68%
likelihood that results
ECGs indicating ischemia and 92% of
normal ECGs correctly without the decision
indicated thrombolytic
aid. After switching to the decision aid two
agents and
weeks following baseline, the interns rates
revascularization
changed to 93% and 87%, respectively,
with significant increases in sensitivity and
decreases in specificity.
Multiple Intervention Types
Additional Review Methods and Educational Interventions
Interpretation of
Film review process and
Diagnostic accuracy
Seltzer,
Other: Retrospective review; seniority
(61)
radiographs
education sessions with
1981
positively correlated to diagnostic
medical students
accuracy. 80% of abnormalities were
thought to be of clinical importance. First
year residents had an omission rate of
6.1% while second and third year residents
had 4.8%.
Interpretation of
Thomas,
Red star report reminds or
Diagnostic accuracy
Other: Internal audit system; Red star
(64)
radiographs
1992
indicates something
reports issued in 2.8% of cases. 0.7% of
possibly missed or
patients needed to return for follow-up due
incorrectly interpreted.
to incorrect interpretations. Less than 50%
Educational conference
required an alteration to treatment.
held to discuss results of
reports
D-241
Notes
Misdiagnosis;
delayed
diagnosis
Missed diagnosis,
misdiagnosis
Missed diagnosis;
misdiagnosis;
proof of concept
Missed
diagnosis;
misdiagnosis
Author, Year
Diagnostic Error
Patient or Related
Outcome
Additional Review Methods, Educational Interventions, and Structured Process Changes
Kundel,
Diagnostic accuracy
Visual (gaze-duration
Proper interpretation of
(62,63)
1990
of pulmonary nodule threshold algorithm)
chest radiographs
interpretation
feedback to radiologists
based upon eye-position
recordings. Re-review of
radiographs
Additional Review Methods, Educational and Technology-based Systems Interventions
McPhee,
Missed cancer
Cancer screening
Cancer screening test
(62)
1989
diagnosis
reminders, audit with
performance
feedback or control (no
intervention). Half the cohort
was also provided with
educational course; 6
intervention cohorts with
medical residents randomly
assigned.
Additional Review Methods and Personnel Changes
Trotter,
Diagnostic errors from Interpretation of skin
Clinical importance of
(67)
2003
interpretation of skin
biopsies by general
discrepant skin biopsy
biopsies by general
pathologists (vs.
results between general
pathologists (vs.
dermatopathologists)
pathologists and
dermatopathologists)
dermatopathologists
Additional Review Methods, Personnel Changes, and Structured Process Changes

(69)
Tsai, 2005
Incorrect diagnosis of Interpretation of urine
Correct diagnosis of acute
acute renal failure
analysis by a nephrologist
renal failure based on urine
(ARF) based on urine for patients with kidney
analysis interpretation (that
analysis interpretation disease; Urinalysis
can impact patient
conducted and report
management and outcome)
written by a nephrologist
rather than clinical
laboratory
Additional Review Methods and Structured Process Changes
D-242
Notes
Other: The more time spent looking at a

certain section of a radiograph, the higher
the chance for error. Feedback cohort
outperformed the control group. Feedback
led to more confident true-positive
diagnoses. 42% of nodules missed initially
were identified after feedback.
Missed diagnosis;
misdiagnosis;
proof of concept
RCT: 20% of patients had active colorectal

symptoms, 37% had one or more
colorectal cancer risk factors, and 48% had
one or more cervical cancer risk factors.
Cancer screening reminders increased
performance the most, followed by audit
with feedback.
Missed diagnosis;
misdiagnosis;
delayed
diagnosis
Other: Agreement in 93.5% of cases;

blinded review of skin biopsies by
dermatopathologists had a sensitivity of
100% in review of general pathologist
identification of lesions. 1.4% of biopsies
had discrepancies that were of potential
clinical significance.
Missed diagnosis;
misdiagnosis
Other: The first nephrologist (A) provided Missed diagnosis

correct cause of ARF in 24 of 26 cases
(92.3%) when performing urinalysis
directly. However, diagnosis was correct by
nephrologist A in only 23.1% and by a
second nephrologist (B) in 19.2% when
analyzing clinical laboratory-generated
urinalysis reports. Diagnosing from
nephrologist As direct urinalysis report,
nephrologist B increased diagnostic
accuracy to 69.3%. Nephrologists were
more likely to recognize presence of RTE
cells, granular casts, and dysmorphic
RBCs in urine.
Author, Year
Ross, 1996
Goodyear,
(70)
2008
(65)
Diagnostic Error
Incorrect vertebral
fracture diagnoses
Laboratory error
Patient or Related
Outcome
Blinding of X-ray readings to Incorrect vertebral fracture
film sequence and patient
diagnoses (that can impact
identity for the detection of patient management and
vertebral fractures
outcome)
Daily supervisory review of

culture reports in
microbiology laboratory
Other: Blinding x-rays to sequence offers

no advantages, increases frequency of
errors and may inflate incidence rates.
Incidents in this study are when there
was no fracture at index x-ray but fracture
was present at follow-up x-ray (average =
3.6 yrs follow-up).
Proper treatment; if
Other: Prospective assessment; review of
microorganism susceptibility culture results and antibiotic susceptibility
is mistaken, incorrect
were found to correct errors in 0.8% of
antibiotic prescribed
cases, and in 0.3% of cases the
corrections were clinically significant. Most
clinical significance was related to the
susceptibility issues concerning culture
results. 302 positive cultures / 101,703
were considered potentially clinically
significant.
Additional Review Methods and Technique

(31)
Beigi, 2007
Incorrect diagnosis of Re-examination of patients Epiphora status at 12
lacrimal duct
scheduled for
months follow-up; and
obstruction/stenosis/f dacryocystorhinostomy
surgery rates
unctional block
based on lacrimal duct
syringing with four tests
Additional Review Methods and Technology-based Systems Interventions
(66)
Jiang, 2001
Breast cancer
Computer-aided diagnosis
Diagnostic accuracy, as
detection
(CAD) program
measured by interobserver
variability, of breast cancer
via mammogram
interpretation
Peldschus,
(68)
2005
Lung lesions/cancer
detection
Missed diagnosis;
misdiagnosis
Missed diagnosis;
misdiagnosis
Other: Re-examination resulted in 18% not Misdiagnosis

requiring previously scheduled major
surgical intervention.
Other, Pre/Post: Prospective review;

access to the tool improved radiologist
agreement and reduced the occurrence of
substantial disagreements. Among
attending radiologists, and residents, the
reductions were statistically significant at
63% and 28%, respectively.
Reevaluation of chest CT
Diagnostic accuracy of lung Other: Retrospective review; CAD detected
studies for focal lung lesions lesions
significant lung lesions in an additional
with the computer-aided
33% of patients.
detection (CAD) system as
a second reader
D-243
Notes
Missed diagnosis;
misdiagnosis;
proof of concept
Missed diagnosis;
misdiagnosis
Author, Year
Diagnostic Error
(9)
Delayed sepsis
detection in surgical
intensive care
Moore, 2009
For early identification of

sepsis, utilized routine
bedside nursing
measurements taken every
12 hours to determine
whether a patient met
threshold for escalating
further assessment by
nurse practitioner or
resident physician. If one of
these providers identified a
source of infection, an
intensivist was then
included to determine
whether treatment for
sepsis was initiated.
Educational Interventions and Structured Process Changes
Gleadhill,
Diagnostic error in
Casualty officers
(89)
1987
radiograph
interpretation reviewed by
interpretation
radiologist, who was
considered to have the
correct report. Clinical
guidelines introduced to
standardize patients
selected for referral
Chern,
Feedback to physicians of
(35)
2005
high-risk patients
outcomes for high-risk
discharged from the
patients discharged from
emergency room
the emergency department
according to telephone
follow-up and review of 3day return emergency
department visits; residents
educated about uncertain
presentations of serious
diseases
Patient or Related
Notes
Outcome
Mortality as a proxy of
Other Prospective Design, Pre/Post: Of
Delayed
delayed diagnosis of sepsis 4,991 sepsis screens with 920 patients
diagnosis
across 927 admissions to the surgical ICU,
the sepsis early identification tool and
protocol yielded a sensitivity of 96.5%,
specificity of 96.7%, positive predictive
value of 80.2%, and negative predictive
value of 99.5%. Compared to the year
before implementing the sepsis tool,
mortality from severe sepsis and septic
shock decreased from 35.1% to 24.2%.
The authors reported that mortality in the
medical and cardiovascular ICUs did not
decrease notably at the same location
during the study period.
Reduction in clinically
significant errors; late error
detection
Experimental Design, Pre/Post: Number of Missed diagnosis;

referrals to Radiology dropped significantly delayed
from 59% to 48%, while rate of late error
diagnosis
detection was unchanged.
Return visits to the

emergency department and
clinically significant adverse
events (including return
visits with serious
misdiagnoses)
Pre/post: Intervention reduced adverse

Misdiagnosis;
events (diagnostic and other) from 4.1% to delayed
1.5%, and return emergency department
diagnosis
visits from 10.1% to 4.9%. Of the 54
patients across both study periods that
experienced adverse events, 40 had
misdiagnoses.
D-244
Author, Year
Diagnostic Error
Educational and Technology-based Systems Interventions

Linver,
Breast cancer
Dedicated mammography
(90)
1992
detection
computer system.
Educational mammography
courses dedicated to
radiologists.
Patient or Related
Outcome
Diagnostic accuracy
Personnel Changes and Technology-based Systems Interventions

Jacobs,
Telemedicine system
Facial fractures
(85)
2002
compared to plain
facial fractures
radiography and diagnosis
by oral and maxillofacial
surgeons (OMFS) and
accident and emergency
department doctors
Personnel Changes and Structured Process Changes
Interval between
Vernon,
Development of a formal
Mortality, time to receiving
(36)
emergency
1999
trauma response team
necessary medical attention
department arrival
(CT scan, etc.)
and critical tests
Structured Process Changes and Technology-based Systems Interventions

Mental illness; referral Three cohort intervention:
Lewis,
Clinical outcome; referral to
(102)
to mental health
1996
(1) no additional
mental health specialist
specialist
information, (2) results of
12-item General Health
Questionnaire (GHQ), and
(3) results of selfadministered computerized
assessment (PROQSY) of
common mental disorders
D-245
Pre/Post: Breast cancer diagnoses

increased 50% pre-training and posttraining, sensitivity increased from 80 to
86%. Positive predictive value remained
32%. Surgical consultations increased
significantly.
Notes
Missed diagnosis;
misdiagnosis
Other: Comparative study; sensitivity and

Missed diagnosis;
specificity of diagnosis by OMFS and A&E misdiagnosis;
higher while viewing plain radiography than Proof of concept;
telemedicine system.
Experimental: prospective, case-control

study; patients treated by trauma response
team had shorter wait times for
computerized tomography scanning,
operation room and overall time within the
emergency department. Mortality rate was
similar for both groups, but better for
severely injured children treated by
response team in comparison to reference
population.
Missed diagnosis;
misdiagnosis;
delayed
diagnosis
Other: GPs given varying levels of

information to accurately diagnose mental
disorders. Those given computerized
assessment results saw modest clinical
improvements in patients. No increase in
referral rates to mental health
professionals in computerized results
group.
Missed diagnosis;
misdiagnosis
Author, Year
Diagnostic Error
Depression screening PRIME-MD survey with 3

and diagnosis
levels of electronic medical
record feedback: (1) active
care, (2) passive care, and
(3) usual care
Rollman,
(38)
2002
Structured Process Changes and Technique

Brossner,
Prostate biopsies and Ultrasound-guided prostate
(79)
2000
cancer detection
biopsy technique;
comparative study of two
techniques to ascertain
which is more accurate at
identifying prostate cancer
Naughton,
Prostate biopsies and 12 vs. 6 biopsy cores taken
(80)
2000
cancer detection
via transrectal ultrasound
Presti, 2000
Ravery,
(82)
2000
(81)
Patient or Related
Outcome
Diagnosis; treatment plan
RCT: Patient depression score on

Missed diagnosis;
misdiagnosis
Hamilton Rating Scale for Depression
decreased similarly regardless of
physicians level of feedback. Screening for
major depression, assisted diagnostic
tools, and exposure to evidence-based
treatment guidelines did not influence
treatment plan.
Other: Comparative study; diagnostic
Missed diagnosis
accuracy did not differ between
approaches; morbidity and duration of pain
increased with 12-core biopsy procedure.
prostate cancer
Other: Comparative study; no difference in

overall prostate cancer detection rate or in
pain assessment.
Other: Trends did not achieve statistical
significance between 8- and 10-biopsy
regimens. Routine sextant biopsies
detected 82% of cancers, and 77% of
missed cancers were detected by lateral
peripheral zone biopsies. Performing 10
biopsies of peripheral zone increased
cancer detection rates by 14%.
Other: Prospective study; protocol had a
6.6% improvement in prostate cancer
detection rate. DRE significantly influenced
detection rate of each protocol.
prostate cancer
Prostate cancer
detection
prostate cancer
Technique and Technology-based Systems Interventions

*Weatherburn, Overall rate of
Picture Archiving and
(83)
2000
misdiagnoses and
Communications System
rate of serious
(PACS) in the accident and
misdiagnoses leading emergency department
to patient recall and
treatment change
Notes
Cancer detection rate;

morbidity differences
between techniques
Prostate biopsies and Adding additional biopsies

cancer detection
to the diagnostic process
Extensive biopsy protocol

implemented
Overall rate of
misdiagnoses and rate of
serious misdiagnoses
leading to patient recall and
treatment change
Missed diagnosis;
misdiagnosis
Missed diagnosis
Missed diagnosis
Experimental Design / Pre/post: Significant Missed diagnosis

reduction in misdiagnosis when PACS was
used (1.5% for film vs. 0.7% for PACS), but
the rate of serious misdiagnoses involving
patient recall did not change significantly.
PACS reduced false negative
interpretations but not rate of serious
misdiagnosis.
* The evaluations of interventions (n=6) with evaluations that were identified in the Singh 2012 systematic review(23).
D-246
Table 2, Chapter 35. Summary of randomized trials

Author
Diagnostic Error
Diagnostic Accuracy Outcome

(72)
Attard, 1992
patients presenting with
abdominal pain
Thomas,
Diagnostic errors based
(26)
2003
on altered physical
examination findings
Type of
Intervention
Compared
intervention
Description of
Outcome
Effect Size (95%

CI)*
Pain relief with papaveretum for

acute abdominal pain
Placebo
Wrong Diagnosis
0.22 (0.05-0.98)
Morphine sulfate administered for Placebo

pain during diagnostic process
0.96 (0.73-1.27)
Cap-fitted colonoscopy (allows

for flattening of haustral folds
and/or improves mucosal
exposure)
Teaching parents an Acute
Illness Observation Scale (AIOS)
to detect childs serious illness
Diagnostic accuracy
(based on information
from follow-up
visits/hospital
discharges)
Missed adenoma
diagnoses (per
adenomas)
False positives
False negatives
0.24 (P < 0.0001)

1.78 (not
statistically
significant )
Patients with false

negative radiograph
interpretations
33.33 (2.01-554.09)
Hewett,
(78)
2010
Missed colorectal
adenoma diagnosis in
colonoscopy
McCarthy,
(86)
1990
Incorrect diagnosis by
parents of symptoms of
serious illness
EI
Klassen,
(92)
1993
Missed positive
(fracture, dislocation or
effusion) after trauma
Misdiagnosis of
appendicitis
SPC
Wellwood,
(104)
1992
TBS
Regular high
resolution
colonoscopy
3-point global
scoring system for
evaluating the
chance of serious
illness
Physicians carrying
out standard
procedures
Triage nurses using the Brand

protocol (for ordering X-rays of
injured extremities) in the
pediatric emergency department
Diagnostic aid with a
No diagnostic aid
standardized data collection form
for abdominal pain interpretation
Standardized data
standardized data collection form collection forms only
and computer-aided diagnostic
tool for abdominal pain
interpretation
D-247
0.63 (0.41-0.99)
Diagnostic accuracy for P = 0.66

appendicitis
Diagnostic accuracy for P = 0.66
appendicitis
Author
Bogusevicius,
(27)
2002
Diagnostic Error
Type of
Intervention
Missed acute mechanical TBS
small bowel obstruction
Computer-aided diagnosis for
diagnosis of acute mechanical
small bowel obstruction (SBO)
Compared
intervention
Contrast
radiography
Description of
Outcome
False positives for
complete SBO
False negatives for
complete SBO
False positives for
partial SBO
False negatives for
SBO
Further Diagnostic Test Use Outcome

(37)
Sakr, 1999
PC
errors, including errors in
the diagnosis pathway
(i.e., history, physical
examination, and
radiographic
Klassen,
Missed positive
SPC
(92)
1993
(fracture, dislocation or
effusion) after trauma
Effect Size (95%

CI)*
Relative risk could
not be calculated (0
events)
0.54 (0.11-2.77)
0.54 (0.11-2.77)
Relative risk could
not be calculated (0
events)
Use of nurse practitioner in

Use of junior doctors Inappropriate radiologic 0.94 (0.75-1.18)
providing care in the emergency in the emergency
follow-up (unnecessary
department
department
request or failure to
request)
Triage nurses using the Brand

Physicians carrying
protocol (for ordering X-rays of
out standard
injured extremities) in the
procedures
pediatric emergency department
D-248
Patients with
radiographs ordered
0.94 (0.75-1.18)
Author
McPhee,
(62)
1989
Diagnostic Error
Type of
Intervention
Missed cancer diagnosis ARM, EI, and
TBS
Therapeutic Use Outcome

(72)
Attard, 1992
patients presenting with
abdominal pain
Diagnostic errors based
Thomas,
(26)
on altered physical
2003
examination findings
Computer generated list of
overdue tests at patients visits
(cancer screening reminders)
Compared
intervention
No intervention
Audit with feedback
No intervention
Patient education
No intervention
Pain relief with papaveretum for

acute abdominal pain
Placebo
Morphine sulfate administered for Placebo

pain during diagnostic process
D-249
Description of
Outcome
Further cancer
screening (Results
given as postintervention compliance
scores relative to
standards according to
the American Cancer
Society
recommendations)
Further cancer
screening (results given
as post-intervention
compliance scores
relative to standards
according to the
American Cancer
Society
recommendations)
Further cancer
screening (results given
as postintervention
compliance scores
relative to standards
according to the
American Cancer
Society
recommendations)
Effect Size (95%

CI)*
Statistically
significant
Inappropriate
management (surgery
or patient observation)
Admissions for
observation or
discharge home
Surgeries
Repeat physician visit
for abdominal pain
within 7 days
Possible incorrect
surgical management
0.22 (0.05-0.98)
Statistically
significant
Statistically
significant**
P = 0.50
P = 0.51
2.84 (0.31-26.08)
2.84 (0.31-26.08)
Author
Diagnostic Error
Kuperman,
(28)
1999
Delays between
laboratory results and
clinical action
(37)
Sakr, 1999
Wellwood,
(104)
1992
Rollman,
(38)
2002
Type of
Intervention
TBS
PC
examination, and
radiographic
Misdiagnosis of
TBS
appendicitis
Missed depression
diagnosis
SPC and TBS
Compared
intervention
Computer system to detect
No automatic
critical laboratory conditions and notification for alerts
notify the physician via Hospitals
paging system
Use of Junior
providing care in the emergency Doctors in the
department
emergency
department
Description of
Outcome
Time to appropriate
treatment
Effect Size (95%

CI)*
P = 0.003
Unplanned follow-up
visits
0.65 (0.45-0.96)

Standardized data collection form
Standardized data collection form
+ computer-aided diagnostic tool
Active care: Primary care
providers (PCPs) were exposed
to advisory messages on the
paper encounter-based upon
AHCPRs guidelines AND advise
to click on the computer desk top
icon to obtain further treatment
advise from the EMR intranet site
Admissions
Surgeries
0.91 (0.84-0.99)
0.98 (0.82-1.16)
Standardized data Admissions

collection forms only Surgeries
1.01 (0.91-1.12)
1.09 (0.90-1.32)
Passive care: PCPs

provided with a
reminder of their
patients depression
dx on the paper
encounter form to
treat depressive
episodes, but
offered no details on
how to do so
Usual care
0.95 (0.49-1.87)
No diagnostic aid
Passive care: PCPs provided

with a reminder of their patients
depression diagnosis on the
paper encounter form to treat
depressive episodes, but offered
no details on how to do so
Active care: PCPs were exposed Usual care
D-250
PCP counsels patient

1.25 (0.67-2.33)
for depression
Mental health referral
0.74 (0.45-1.23)
suggested
Antidepressant
1.25 (0.67-2.33)
medications prescribed

for depression
suggested
Antidepressant
for depression
suggested
Antidepressant
1.01 (0.64-1.59)
0.95 (0.49-1.87)
1.19 (0.63-2.25)
0.75 (0.44-1.25)
1.19 (0.63-2.25)
Author
Diagnostic Error
Type of
Intervention
Patient Outcomes
(37)
Sakr, 1999
PC
examination, and
radiographic
Bogusevicius, Missed acute mechanical TBS
(27)
2002
Fitzgerald,
(34)
2011
Kuperman,
(28)
1999
Errors during reception

and resuscitation of
severely injured adult
trauma patients
(including errors in the
diagnosis pathway)
Delays between
clinical action
TBS
TBS
Compared
intervention
Description of
Outcome
Effect Size (95%

CI)*

Use of junior doctors Non improvement in
condition
department
department
0.94 (0.68-1.30)
Computer-aided diagnosis for

diagnosis of acute mechanical
Real time computer-prompted
evidence-based decision support
system (with decision and action
algorithms) during reception and
resuscitation of severely injured
adults in Level I adult trauma
center
Contrast
radiography
Mortality
Morbidity outcome
5 (0.25-100.97)
1.33 (0.32-5.58)
Control (without
computer-aided
decision support
system)
Error rate
Morbidity from shock
management
Aspiration pneumonia
Sepsis
0.89 (0.79-1.00)
P = 0.03
P = 0.046
Not statistically
significant
Not statistically
significant
ARDS (acute
respiratory distress
syndrome)
Functional
Not statistically
independence measure significant
score
Hospital length of stay Not statistically
significant
P < 0.001
Transfusion of blood
productions
Mortality
1.15 (0.65-2.03)
P = 0.11
No automatic
Time to resolution of
critical laboratory conditions and notification for alerts alerting conditions
notify the physician via hospitals
P = 0.41
Adverse events
paging system
D-251
Author
Diagnostic Error
Rollman,
(38)
2002
Missed depression
diagnosis
Type of
Intervention
SPC
Active care: PCPs were exposed
Compared
intervention
Passive care: PCPs
provided with a
reminder of their
patients depression
diagnosis on the
paper encounter
form to treat
depressive
episodes, but
offered no details on
how to do so
Usual care
Passive care: PCPs provided

with a reminder of their patients
depression dx on the paper
encounter form to treat
depressive episodes, but offered
no details on how to do so
Active care: PCPs were exposed Usual care
Composite Clinical Outcomes
(37)
Sakr, 1999
PC
examination, and
radiographic
Schriger,
Misdiagnosis of occult
SPC
(39)
2001
mental illness
Description of
Outcome
Nonimprovement of
depressive symptoms
Effect Size (95%

CI)*
1.06 (0.78-1.44)
Nonimprovement of
depressive symptoms
0.88 (0.65-1.19)
Nonimprovement of
depressive symptoms
0.93 (0.70-1.25)

Use of junior doctors DAO+ TUO: Clinically 0.86 (0.63-1.18)
important errors
department
department
(composite outcome for
diagnostic errors,
treatment/follow-up
errors)
Report of a computerized
psychiatric interview (PRIMEMD) given to the physician
D-252
PRIME-MD report
not given to the
Physician
Consultation or referral 1.60 (0.47-5.48)

for mental illness plus
other (psychiatric
diagnosis)
Author
Diagnostic Error
Type of
Intervention
TBS
Compared
intervention
No automatic
critical laboratory conditions and notification for alerts
notify the physician via hospitals
paging system
Description of
Effect Size (95%
Outcome
CI)*
Delays between
Kuperman,
TUO+PO: Adverse
1.20 (0.78-1.84)
(28)
1999
events
clinical action
(cardiopulmonary
arrest, ICU admissions,
strokes, acute renal
failure, death, need for
surgery)
Abbreviations: AHCPR = Agency for Health Care Policy and Research; ARM = additional review methods; DAO = diagnostic accuracy outcome; EI = educational intervention;
EMR = electronic medical record; ICU: intensive care unit; nss = not statistically significant; PC = personnel change; PCP = primary care physician; PO = patient outcomes;
PRIME-MD: Primary Care Evaluation of Mental Disorders; SBO = small bowel obstruction; SPC = structured process change; ss = statistically significant; T= technique; TBS =
technology-based systems intervention; TUO = therapeutic use outcome.
*Effect size is relative risk except for Fitzgerald et al. where error rate was used; McPhee et al,where difference in scores post intervention was used and Kuperman et al.al where
time to appropriate treatment was used.
Results were significant for: stool occult blood testing, rectal examination, sigmoidoscopy, pelvic exam, breast exam, mammography AND non-significant for Pap smear.
Results were significant for: breast exam, mammography AND non-significant for: occult blood test, rectal exam, sigmoidoscopy, Pap smear, pelvic exam
**Results were significant for breast exam AND non-significant for mammography.
D-253
References
1.
Graber ML. Next steps: envisioning a

research agenda. Advances in health
sciences education : theory and practice.
2009;14 Suppl 1:107-12. Epub 2009/08/12.
doi: 10.1007/s10459-009-9183-1. PubMed
PMID: 19669917.
2.
Schiff GD, Hasan O, Kim S, Abrams R,

Cosby K, Lambert BL, et al. Diagnostic
error in medicine: analysis of 583 physicianreported errors. Archives of internal
medicine. 2009;169(20):1881-7. Epub
2009/11/11. doi:
10.1001/archinternmed.2009.333. PubMed
PMID: 19901140.
3.
Shojania KG, Burton EC, McDonald KM,

Goldman L. Changes in rates of autopsydetected diagnostic errors over time: a
systematic review. JAMA.
2003;289(21):2849-56. Epub 2003/06/05.
doi: 10.1001/jama.289.21.2849 [doi]
289/21/2849 [pii]. PubMed PMID:
12783916.
4.
Newswire P. Americans Are Concerned

About Hospital-Based Medical and Surgical
Errors. Wall Street Journal Onlines Health
Industry Edition. 2004.
5.
Schiff GD, Kim S, Abrams R, Cosby K,

Lambert B, Elstein AS, et al. Diagnosing
Diagnosis Errors: Lessons from a Multiinstitutional. In: Henriksen K, Battles JB,
Marks ES, Lewin DI, editors. Advances in
Implementation (Volume 2:. Advances in
Patient Safety. Rockville (MD)2005.
6.
Singh H, Daci K, Petersen LA, Collins C,

Petersen NJ, Shethia A, et al. Missed
opportunities to initiate endoscopic
evaluation for colorectal cancer diagnosis.
Am J Gastroenterol. 2009;104(10):2543-54.
Epub 2009/06/25. doi: ajg2009324 [pii]
10.1038/ajg.2009.324. PubMed PMID:
PMC2758321.
D-254
7.
Singh H, Hirani K, Kadiyala H, Rudomiotov

O, Davis T, Khan MM, et al. Characteristics
and predictors of missed opportunities in
lung cancer diagnosis: an electronic health
record-based study. J Clin Oncol.
2010;28(20):3307-15. Epub 2010/06/10.
doi: JCO.2009.25.6636 [pii]
10.1200/JCO.2009.25.6636. PubMed PMID:
PMC2903328.
8.
Singh H, Thomas EJ, Wilson L, Kelly PA,

Pietz K, Elkeeb D, et al. Errors of diagnosis
in pediatric practice: a multisite survey.
2010/06/23. doi: peds.2009-3218 [pii]
10.1542/peds.2009-3218. PubMed PMID:
PMC2921702.
9.
Moore LJ, Jones SL, Kreiner LA, McKinley

B, Sucher JF, Todd SR, et al. Validation of a
screening tool for the early identification of
sepsis. The Journal of trauma.
2009;66(6):1539-46; discussion 46-7. Epub
2009/06/11. doi:
10.1097/TA.0b013e3181a3ac4b00005373200906000-00005 [pii]. PubMed PMID:
19509612.
10.
Phillips RL, Jr., Bartholomew LA, Dovey

SM, Fryer GE, Jr., Miyoshi TJ, Green LA.
Learning from malpractice claims about
negligent, adverse events in primary care in
the United States. Quality & safety in health
care. 2004;13(2):121-6. Epub 2004/04/08.
PMCID: PMC1743812.
11.
Selbst SM. Pediatric emergency medicine:

legal briefs. Pediatric emergency care.
2005;21(3):214-8. Epub 2005/03/04.
12.
Tehrani AS, Lee H, Mathews S, Shore A,

Frick KD, Makary M, et al., editors. 20-year
Summary of U.S. Malpractice Claims for
Diagnostic Errors from 1985-2005. 33rd
Annual Meeting of the Society for Medical
Decision Making; 2011; Chicago, IL.
13.
Bishop TF, Ryan AM, Casalino LP. Paid

malpractice claims for adverse events in
inpatient and outpatient settings. JAMA.
2011;305(23):2427-31. Epub 2011/06/16.
doi: 305/23/2427 [pii]
10.1001/jama.2011.813. PubMed PMID:
21673294.
21.
Winters BD, Aswani MS, Pronovost PJ.

Commentary: Reducing diagnostic errors:
another role for checklists? Acad Med.
2011;86(3):279-81. Epub 2011/02/25. doi:
10.1097/ACM.0b013e3182082692 [doi]
00001888-201103000-00011 [pii]. PubMed
PMID: 21346432.
14.
Ely JW, Graber ML, Croskerry P. Checklists

to reduce diagnostic errors. Acad Med.
2011;86(3):307-13. Epub 2011/01/21. doi:
10.1097/ACM.0b013e31820824cd [doi].
22.
15.
Cosby KS. A framework for classifying

factors that contribute to error in the
emergency department. Annals of
Epub 2003/11/25. doi:
10.1016/S0196064403006358. PubMed
PMID: 14634609.
Wahls TL, Cram P. Proposed interventions

to decrease the frequency of missed test
results. Advances in health sciences
Suppl 1:51-6. Epub 2009/08/12. doi:
10.1007/s10459-009-9180-4. PubMed
PMID: 19669920.
23.
Singh H, Graber M, Kissam SM, Sorensen

AV, Lenfestey NF, Tant EM, et al. Systemrelated interventions to reduce diagnostic
errors: a narrative review. BMJ Qual Saf.
2011. doi: 10.1136/bmjqs-2011-000150.
24.
Graber ML, Kissam S, Payne VL, Meyer

AN, Sorensen A, Lenfestey N, et al.
Cognitive interventions to reduce diagnostic
error: a narrative review. BMJ Qual Saf.
2012;21(7):535-57. Epub 2012/05/01. doi:
bmjqs-2011-000149 [pii] 10.1136/bmjqs2011-000149. PubMed PMID: 22543420.
25.
Perno JF, Schunk JE, Hansen KW, Furnival

RA. Significant reduction in delayed
diagnosis of injury with implementation of a
pediatric trauma service. Pediatric
emergency care. 2005;21(6):367-71. Epub
2005/06/09. doi: 00006565-20050600000003 [pii]. PubMed PMID: 15942513.
26.
Thomas SH, Silen W, Cheema F, Reisner A,

Aman S, Goldstein JN, et al. Effects of
morphine analgesia on diagnostic accuracy
in Emergency Department patients with
abdominal pain: a prospective, randomized
trial. J Am Coll Surg. 2003;196(1):18-31.
12517545.
27.
Bogusevicius A, Maleckas A, Pundzius J,

Skaudickas D. Prospective randomised trial
of computer-aided diagnosis and contrast
radiography in acute small bowel
obstruction. Eur J Surg. 2002;168(2):78-83.
Epub 2002/07/13. doi:
10.1080/11024150252884287. PubMed
PMID: 12113275.
16.
Tversky A, Kahneman D. Judgment under

Uncertainty: Heuristics and Biases. Science.
1974;185(4157):1124-31. Epub 1974/09/27.
doi: 185/4157/1124 [pii]
10.1126/science.185.4157.1124. PubMed
PMID: 17835457.
17.
Metcalfe J, Shimamura AP. Metacognition :

knowing about knowing. Cambridge, Mass.
; London: MIT Press; 1994.
18.
Reason J. Human error: models and

management. BMJ. 2000;320(7237):768-70.
PMC1117770.
19.
Berner ES. Diagnostic error in medicine:

introduction. Advances in health sciences
Suppl 1:1-5. Epub 2009/08/12. doi:
10.1007/s10459-009-9187-x. PubMed
PMID: 19669914.
20.
Schiff GD, Bates DW. Can electronic

clinical documentation help prevent
diagnostic errors? N Engl J Med.
2010;362(12):1066-9. Epub 2010/03/26.
doi: 362/12/1066 [pii]
10.1056/NEJMp0911734 [doi]. PubMed
PMID: 20335582.
D-255
28.
Kuperman GJ, Teich JM, Tanasijevic MJ,

MaLuf N, Rittenberg E, Jha A, et al.
Improving response to critical laboratory
results with automation: results of a
randomized controlled trial. J Am Med
Inform Assoc. 1999;6(6):512-22. Epub
1999/12/01. PubMed PMID: 10579608;
29.
Dudley M, Channer KS. Assessment of the

value of technician reporting of
electrocardiographs in an accident and
1997;14(5):307-10. Epub 1997/10/08.
PMCID: PMC1343097.
30.
31.
32.
33.
34.
Nam YS, Pikarsky AJ, Wexner SD, Singh

JJ, Weiss EG, Nogueras JJ, et al.
Reproducibility of colonic transit study in
patients with chronic constipation. Dis
Colon Rectum. 2001;44(1):86-92. Epub
2002/01/24. PubMed PMID: 11805568.
Beigi B, Uddin JM, McMullan TF, Linardos
E. Inaccuracy of diagnosis in a cohort of
patients on the waiting list for
dacryocystorhinostomy when the diagnosis
was made by only syringing the lacrimal
system. Eur J Ophthalmol. 2007;17(4):4859. Epub 2007/08/03. PubMed PMID:
17671919.
Major K, Shabot MM, Cunneen S. Wireless
clinical alerts and patient outcomes in the
surgical intensive care unit. Am Surg.
2002;68(12):1057-60. Epub 2003/01/09.
Etchells E, Adhikari NK, Wu R, Cheung M,
Quan S, Mraz R, et al. Real-time automated
paging and decision support for critical
laboratory abnormalities. BMJ Qual Saf.
2011;20(11):924-30. Epub 2011/07/05. doi:
bmjqs.2010.051110 [pii]
10.1136/bmjqs.2010.051110. PubMed
PMID: 21725046.
Fitzgerald M, Cameron P, Mackenzie C,
Farrow N, Scicluna P, Gocentas R, et al.
Trauma resuscitation errors and computerassisted decision support. Arch Surg.
2011;146(2):218-25. Epub 2011/02/23. doi:
146/2/218 [pii] 10.1001/archsurg.2010.333.
D-256
35.
Chern CH, How CK, Wang LM, Lee CH,

Graff L. Decreasing clinically significant
adverse events using feedback to emergency
physicians of telephone follow-up outcomes.
Annals of emergency medicine.
2005;45(1):15-23. Epub 2005/01/07. doi:
S0196064404012880 [pii]
10.1016/j.annemergmed.2004.08.012 [doi].
36.
Vernon DD, Furnival RA, Hansen KW,

Diller EM, Bolte RG, Johnson DG, et al.
Effect of a pediatric trauma response team
on emergency department treatment time
and mortality of pediatric trauma victims.
1999/01/26. PubMed PMID: 9917434.
37.
Sakr M, Angus J, Perrin J, Nixon C, Nicholl

J, Wardrope J. Care of minor injuries by
emergency nurse practitioners or junior
doctors: a randomised controlled trial.
Lancet. 1999;354(9187):1321-6. Epub
1999/10/26. doi: S0140673699024472 [pii].
38.
Rollman BL, Hanusa BH, Lowe HJ, Gilbert

T, Kapoor WN, Schulberg HC. A
randomized trial using computerized
decision support to improve treatment of
major depression in primary care. J Gen
Intern Med. 2002;17(7):493-503. Epub
2002/07/23. doi: jgi10421 [pii]. PubMed
PMC1495078.
39.
Schriger DL, Gibbons PS, Langone CA, Lee

S, Altshuler LL. Enabling the diagnosis of
occult psychiatric illness in the emergency
department: A randomized, controlled trial
of the computerized, self-administered
PRIME-MD diagnostic system. Annals of
PubMed PMID: ISI:000166803800002.
40.
Ciatto S, Del Turco MR, Morrone D, Catarzi

S, Ambrogetti D, Cariddi A, et al.
Independent double reading of screening
mammograms. J Med Screen. 1995;2(2):99101. Epub 1995/01/01. PubMed PMID:
7497164.
41.
42.
43.
44.
45.
Duijm LE, Groenewoud JH, Fracheboud J,

de Koning HJ. Additional double reading of
screening mammograms by radiologic
technologists: impact on screening
performance parameters. J Natl Cancer Inst.
2007;99(15):1162-70. Epub 2007/07/27.
doi: djm050 [pii] 10.1093/jnci/djm050 [doi].
48.
Canon CL, Smith JK, Morgan DE, Jones

BC, Fell SC, Kenney PJ, et al. Double
reading of barium enemas: is it necessary?
2003;181(6):1607-10. Epub 2003/11/25.
49.
Carew-McColl M. Radiological
interpretation in an accident and emergency
department. Br J Clin Pract. 1983;37(1112):375-7. Epub 1983/11/01. PubMed
PMID: 6671078.
50.
Espinosa JA, Nolan TW. Reducing errors

made by emergency physicians in
interpreting radiographs: longitudinal study.
BMJ. 2000;320(7237):737-40. Epub
2000/03/17. PubMed PMID: 10720354;
51.
Nordrum I, Johansen M, Amin A, Isaksen

V, Ludvigsen JA. Diagnostic accuracy of
second-opinion diagnoses based on still
images. Hum Pathol. 2004;35(1):129-35.
Epub 2004/01/28. doi: S0046817703004982
[pii]. PubMed PMID: 14745735.
Galasko CS, Monahan PR. Value of reexamining x-ray films of outpatients

attending accident services. Br Med J.
1971;1(5750):643-4. Epub 1971/03/20.
PMCID: PMC1795440.
52.
Raab SS, Stone CH, Jensen CS, Zarbo RJ,

Meier FA, Grzybicki DM, et al. Double
slide viewing as a cytology quality
improvement initiative. Am J Clin Pathol.
2006;125(4):526-33. Epub 2006/04/22. doi:
TKNM9GRGPL2JT95Y [pii]
10.1309/TKNM-9GRG-PL2J-T95Y [doi].
Kwek BH, Lau TN, Ng FC, Gao F. Nonconsensual double reading in the Singapore
Breast Screening Project: benefits and
limitations. Annals of the Academy of
Medicine, Singapore. 2003;32(4):438-41.
12968545.
53.
Lind AC, Bewtra C, Healy JC, Sims KL.

Prospective peer review in surgical
pathology. Am J Clin Pathol.
1995;104(5):560-6. Epub 1995/11/01.
54.
Lufkin KC, Smith SW, Matticks CA,

Brunette DD. Radiologists review of
radiographs interpreted confidently by
emergency physicians infrequently leads to
changes in patient management. Annals of
Epub 1998/02/24. PubMed PMID: 9472181.
55.
Murphy R, Slater A, Uberoi R, Bungay H,

Ferrett C. Reduction of perception error by
double reporting of minimal preparation CT
colon. Br J Radiol. 2010;83(988):331-5.
Epub 2009/08/05. doi: 65634575 [pii]
10.1259/bjr/65634575 [doi]. PubMed
PMID: 19651707.
Howard J, Sundararajan R, Thomas SG,

Walsh M, Sundararajan M. Reducing missed
injuries at a level II trauma center. J Trauma
Nurs. 2006;13(3):89-95. Epub 2006/10/21.
Manion E, Cohen MB, Weydert J.
Mandatory second opinion in surgical
pathology referral material: clinical
consequences of major disagreements. The
American journal of surgical pathology.
2008;32(5):732-7. Epub 2008/03/25. doi:
10.1097/PAS.0b013e31815a04f5. PubMed
PMID: 18360282.
46.
Singh P, Warnakulasuriya S. The two-week

wait cancer initiative on oral cancer; the
predictive value of urgent referrals to an oral
medicine unit. British dental journal.
2006;201(11):717-20; discussion 4. Epub
2006/12/13. doi: 10.1038/sj.bdj.4814304.
47.
Bruner JM, Inouye L, Fuller GN, Langford

LA. Diagnostic discrepancies and their
clinical impact in a neuropathology referral
practice. Cancer. 1997;79(4):796-803. Epub
1997/02/15. doi: 10.1002/(SICI)10970142(19970215)79:4<796::AIDCNCR17>3.0.CO;2-V [pii]. PubMed PMID:
9024718.
D-257
56.
57.
58.
59.
60.
61.
62.
Parameswaran L, Prihoda TJ, Sharkey FE.

Diagnostic efficacy of additional stepsections in colorectal biopsies originally
diagnosed as normal. Hum Pathol.
2008;39(4):579-83. Epub 2008/02/22. doi:
S0046-8177(07)00476-5 [pii]
10.1016/j.humpath.2007.08.015 [doi].
63.
Kundel HL, Nodine CF, Krupinski EA.

Computer-displayed eye position as a visual
aid to pulmonary nodule interpretation.
Invest Radiol. 1990;25(8):890-6. Epub
1990/08/01. PubMed PMID: 2394571.
64.
Raab SS, Grzybicki DM, Mahood LK,

Parwani AV, Kuan SF, Rao UN.
Effectiveness of random and focused review
in detecting surgical pathology error. Am J
Clin Pathol. 2008;130(6):905-12. Epub
2008/11/21. doi: R8245P6P760X8402 [pii]
10.1309/AJCPPIA5D7MYKDWF [doi].
Thomas HG, Mason AC, Smith RM,

Fergusson CM. Value of radiograph audit in
an accident service department. Injury.
1992;23(1):47-50. Epub 1992/01/01. doi:
0020-1383(92)90126-D [pii]. PubMed
PMID: 1541500.
65.
Robson N, van Benthem PP, Gan R, Dixon

AK. Casualty X-ray reporting: a student
survey. Clin Radiol. 1985;36(5):479-81.
4075715.
Ross PD, Huang C, Karpf D, Lydick E, Coel

M, Hirsch L, et al. Blinded reading of
radiographs increases the frequency of
errors in vertebral fracture detection. J Bone
Miner Res. 1996;11(11):1793-800. Epub
1996/11/01. doi: 10.1002/jbmr.5650111124
66.
Jiang Y, Nishikawa RM, Schmidt RA,

Toledano AY, Doi K. Potential of computeraided diagnosis to reduce variability in
radiologists interpretations of
mammograms depicting microcalcifications.
Radiology. 2001;220(3):787-94. Epub
2001/08/30. PubMed PMID: 11526283.
67.
Trotter MJ, Bruecks AK. Interpretation of

skin biopsies by general pathologists:
diagnostic discrepancy rate measured by
blinded review. Arch Pathol Lab Med.
2003;127(11):1489-92. Epub 2003/10/22.
doi: OA3050 [pii] 10.1043/15432165(2003)127<1489:IOSBBG>2.0.CO;2
68.
Peldschus K, Herzog P, Wood SA, Cheema

JI, Costello P, Schoepf UJ. Computer-aided
diagnosis as a second reader: spectrum of
findings in CT studies of the chest
interpreted as normal. Chest.
2005;128(3):1517-23. Epub 2005/09/16.
doi: 10.1378/chest.128.3.1517. PubMed
PMID: 16162752.
69.
Tsai JJ, Yeun JY, Kumar VA, Don BR.

Comparison and interpretation of urinalysis
performed by a nephrologist versus a
hospital-based clinical laboratory. Am J
Kidney Dis. 2005;46(5):820-9. Epub
2005/10/29. doi: S0272-6386(05)01050-4
[pii] 10.1053/j.ajkd.2005.07.039 [doi].
Thiesse P, Ollivier L, Di Stefano-Louineau

D, Negrier S, Savary J, Pignard K, et al.
Response rate accuracy in oncology trials:
reasons for interobserver variability. Groupe
Francais dImmunotherapie of the
Federation Nationale des Centres de Lutte
Contre le Cancer. J Clin Oncol.
1997;15(12):3507-14. Epub 1997/12/13.
Westra WH, Kronz JD, Eisele DW. The
impact of second opinion surgical pathology
on the practice of head and neck surgery: a
decade experience at a large referral
hospital. Head & neck. 2002;24(7):684-93.
Epub 2002/07/12. doi: 10.1002/hed.10105.
Seltzer SE, Hessel SJ, Herman PG,
Swensson RG, Sheriff CR. Resident film
interpretations and staff review. AJR
American journal of roentgenology.
1981;137(1):129-33. Epub 1981/07/01.
McPhee SJ, Bird JA, Jenkins CN, Fordham
D. Promoting cancer screening. A
randomized, controlled trial of three
interventions. Archives of internal medicine.
1989;149(8):1866-72. Epub 1989/08/01.
D-258
70.
Goodyear N, Ulness BK, Prentice JL,

Cookson BT, Limaye AP. Systematic
assessment of culture review as a tool to
assess errors in the clinical microbiology
laboratory. Arch Pathol Lab Med.
2008;132(11):1792-5. Epub 2008/11/04.
doi: 2008-0037-OA [pii] 10.1043/15432165-132.11.1792. PubMed PMID:
18976017.
71.
Hamady ZZ, Mather N, Lansdown MR,

Davidson L, Maclennan KA. Surgical
pathological second opinion in thyroid
malignancy: impact on patients
management and prognosis. Eur J Surg
Oncol. 2005;31(1):74-7. Epub 2005/01/12.
doi: S0748-7983(04)00244-6 [pii]
10.1016/j.ejso.2004.08.010. PubMed PMID:
15642429.
72.
Attard AR, Corlett MJ, Kidner NJ, Leslie

AP, Fraser IA. Safety of early pain relief for
acute abdominal pain. BMJ.
1992;305(6853):554-6. Epub 1992/09/05.
PMCID: PMC1883284.
73.
Resnick NM, Brandeis GH, Baumann MM,

DuBeau CE, Yalla SV. Misdiagnosis of
urinary incontinence in nursing home
women: prevalence and a proposed solution.
Neurourol Urodyn. 1996;15(6):599-613;
discussion -8. Epub 1996/01/01. doi:
10.1002/(SICI)15206777(1996)15:6<599::AIDNAU2>3.0.CO;2-A [pii]
10.1002/(SICI)15206777(1996)15:6<599::AIDNAU2>3.0.CO;2-A [doi]. PubMed PMID:
8916113.
74.
75.
Borgstein PJ, Gordijn RV, Eijsbouts QA,

Cuesta MA. Acute appendicitis--a clear-cut
case in men, a guessing game in young
women. A prospective study on the role of
laparoscopy. Surg Endosc. 1997;11(9):9237. Epub 1997/09/19. PubMed PMID:
9294274.
Vermeulen B, Morabia A, Unger PF,
Goehring C, Grangier C, Skljarov I, et al.
Acute appendicitis: influence of early pain
relief on the accuracy of clinical and US
findings in the decision to operate--a
randomized trial. Radiology.
1999;210(3):639-43. Epub 1999/04/20.
D-259
76.
Prieto VG, Argenyi ZB, Barnhill RL, Duray

PH, Elenitsas R, From L, et al. Are en face
frozen sections accurate for diagnosing
margin status in melanocytic lesions? Am J
Clin Pathol. 2003;120(2):203-8. Epub
2003/08/23. doi: 10.1309/J1Q0-V35EUTMV-R193 [doi]. PubMed PMID:
12931550.
77.
Kokki H, Lintula H, Vanamo K, Heiskanen

M, Eskelinen M. Oxycodone vs placebo in
children with undifferentiated abdominal
pain: a randomized, double-blind clinical
trial of the effect of analgesia on diagnostic
accuracy. Arch Pediatr Adolesc Med.
2005;159(4):320-5. Epub 2005/04/06. doi:
159/4/320 [pii] 10.1001/archpedi.159.4.320.
78.
Hewett DG, Rex DK. Cap-fitted

colonoscopy: a randomized, tandem
colonoscopy study of adenoma miss rates.
Gastrointest Endosc. 2010;72(4):775-81.
Epub 2010/06/29. doi: S00165107(10)01558-0 [pii]
10.1016/j.gie.2010.04.030 [doi]. PubMed
PMID: 20579648.
79.
Brossner C, Madersbacher S, Bayer G,

Pycha A, Klingler HC, Maier U.
Comparative study of two different TRUSguided sextant biopsy techniques in
detecting prostate cancer in one biopsy
session. European urology. 2000;37(1):6571. Epub 2000/02/15. PubMed PMID:
10671788.
80.
Naughton CK, Miller DC, Mager DE,

Ornstein DK, Catalona WJ. A prospective
randomized trial comparing 6 versus 12
prostate biopsy cores: impact on cancer
detection. The Journal of urology.
2000;164(2):388-92. Epub 2000/07/14.
81.
Presti JC, Jr., Chang JJ, Bhargava V,

Shinohara K. The optimal systematic
prostate biopsy scheme should include 8
rather than 6 biopsies: results of a
prospective clinical trial. The Journal of
urology. 2000;163(1):163-6; discussion 6-7.
10604337.
82.
Ravery V, Goldblatt L, Royer B, Blanc E,

Toublanc M, Boccon-Gibod L. Extensive
biopsy protocol improves the detection rate
of prostate cancer. The Journal of urology.
2000;164(2):393-6. Epub 2000/07/14.
83.
Weatherburn G, Bryan S, Nicholas A, Cocks

R. The effect of a picture archiving and
communications system (PACS) on
diagnostic performance in the accident and
2000;17(3):180-4. Epub 2000/05/20.
PMCID: PMC1725383.
84.
85.
86.
87.
88.
de Lacey G, Barker A, Harper J, Wignall B.

An assessment of the clinical effects of
reporting accident and emergency
radiographs. Br J Radiol. 1980;53(628):3049. Epub 1980/04/01. PubMed PMID:
7378697.
Jacobs MJ, Edmondson MJ, Lowry JC.
Accuracy of diagnosis of fractures by
maxillofacial and accident and emergency
doctors using plain radiography compared
with a telemedicine system: a prospective
study. Br J Oral Maxillofac Surg.
2002;40(2):156-62. Epub 2002/08/16.
McCarthy PL, Sznajderman SD, LustmanFindling K, Baron MA, Fink HD,
Czarkowski N, et al. Mothers clinical
judgment: a randomized trial of the Acute
Illness Observation Scales. J Pediatr.
1990;116(2):200-6. Epub 1990/02/01.
Fridriksson S, Hillman J, Landtblom AM,
Boive J. Education of referring doctors
about sudden onset headache in
subarachnoid hemorrhage. A prospective
study. Acta Neurol Scand. 2001;103(4):23842. Epub 2001/05/01. doi: ane065 [pii].
Thaler T, Tempelmann V, Maggiorini M,
Rudiger A. The frequency of
electrocardiographic errors due to electrode
cable switches: a before and after study. J
Electrocardiol. 2010;43(6):676-81. Epub
2010/07/02. doi: S0022-0736(10)00203-7
[pii] 10.1016/j.jelectrocard.2010.05.007
D-260
89.
Gleadhill DN, Thomson JY, Simms P. Can

more efficient use be made of x ray
examinations in the accident and emergency
department? Br Med J (Clin Res Ed).
1987;294(6577):943-7. Epub 1987/04/11.
PMCID: PMC1246007.
90.
Linver MN, Paster SB, Rosenberg RD, Key

CR, Stidley CA, King WV. Improvement in
mammography interpretation skills in a
community radiology practice after
dedicated teaching courses: 2-year medical
audit of 38,633 cases. Radiology.
1992;184(1):39-43. Epub 1992/07/01.
91.
Enderson BL, Reath DB, Meadors J, Dallas

W, DeBoo JM, Maull KI. The tertiary
trauma survey: a prospective study of
missed injury. The Journal of trauma.
1990;30(6):666-9; discussion 9-70. Epub
1990/06/01. PubMed PMID: 2352294.
92.
Klassen TP, Ropp LJ, Sutcliffe T, Blouin R,

Dulberg C, Raman S, et al. A randomized,
controlled trial of radiograph ordering for
extremity trauma in a pediatric emergency
department. Annals of emergency medicine.
1993;22(10):1524-9. Epub 1993/10/01. doi:
S0196-0644(05)81252-1 [pii]. PubMed
PMID: 8214829.
93.
Biffl WL, Harrington DT, Cioffi WG.

Implementation of a tertiary trauma survey
decreases missed injuries. The Journal of
trauma. 2003;54(1):38-43; discussion -4.
Epub 2003/01/25. doi:
10.1097/01.TA.0000046379.29204.CE
94.
Soundappan SV, Holland AJ, Cass DT. Role

of an extended tertiary survey in detecting
missed injuries in children. The Journal of
trauma. 2004;57(1):114-8; discussion 8.
15284560.
95.
Ursprung R, Gray JE, Edwards WH, Horbar

JD, Nickerson J, Plsek P, et al. Real time
patient safety audits: improving safety every
day. Quality & safety in health care.
2005;14(4):284-9. Epub 2005/08/04. doi:
14/4/284 [pii] 10.1136/qshc.2004.012542.
PMCID: PMC1744058.
96.
97.
98.
99.
100.
101.
Raab SS, Andrew-Jaja C, Condel JL, Dabbs

DJ. Improving Papanicolaou test quality and
reducing medical errors by using Toyota
production system methods. Am J Obstet
Gynecol. 2006;194(1):57-64. Epub
2006/01/04. doi: S0002-9378(05)00972-5
[pii] 10.1016/j.ajog.2005.06.069. PubMed
PMID: 16389010.
Raab SS, Grzybicki DM, Sudilovsky D,
Balassanian R, Janosky JE, Vrbin CM.
Effectiveness of Toyota process redesign in
reducing thyroid gland fine-needle
aspiration error. Am J Clin Pathol.
2006;126(4):585-92. Epub 2006/08/30. doi:
NJQ1L7KA10UKV93Q [pii]
10.1309/NJQ1L7KA10UKV93Q [doi].
Raab SS, Tworek JA, Souers R, Zarbo RJ.
The value of monitoring frozen sectionpermanent section correlation data over
time. Arch Pathol Lab Med.
2006;130(3):337-42. Epub 2006/03/08. doi:
CP5722 [pii] 10.1043/15432165(2006)130[337:TVOMFS]2.0.CO;2
Raab SS, Jones BA, Souers R, Tworek JA.
The effect of continuous monitoring of
cytologic-histologic correlation data on
cervical cancer screening performance. Arch
Pathol Lab Med. 2008;132(1):16-22. Epub
2008/01/10. doi: 2007-0197-CPR1 [pii]
10.1043/15432165(2008)132[16:TEOCMO]2.0.CO;2
Mueller CA, Klaassen-Mielke R, Penner E,
Junius-Walker U, Hummers-Pradier E,
Theile G. Disclosure of new health problems
and intervention planning using a geriatric
assessment in a primary care setting. Croat
Med J. 2010;51(6):493-500. Epub
2010/12/17. PubMed PMID: 21162161;
AM, Smorenburg SM, Gouma DJ,
Boermeester MA. Prevention of surgical
safety checklist. Ann Surg.
2011;253(3):624-8. Epub 2011/01/07. doi:
10.1097/SLA.0b013e3182068880. PubMed
PMID: 21209590.
D-261
102.
Lewis G, Sharp D, Bartholomew J, Pelosi

AJ. Computerized assessment of common
mental disorders in primary care: effect on
clinical outcome. Fam Pract.
1996;13(2):120-6. Epub 1996/04/01.
103.
Wexler JR, Swender PT, Tunnessen WW,

Jr., Oski FA. Impact of a system of
computer-assisted diagnosis. Initial
evaluation of the hospitalized patient. Am J
Dis Child. 1975;129(2):203-5. Epub
1975/02/11. PubMed PMID: 1091140.
104.
Wellwood J, Johannessen S, Spiegelhalter

DJ. How does computer-aided diagnosis
improve the management of acute
abdominal pain? Ann R Coll Surg Engl.
1992;74(1):40-6. Epub 1992/01/01. PubMed
PMC2497469.
105.
Selker HP, Beshansky JR, Griffith JL,

Aufderheide TP, Ballin DS, Bernard SA, et
al. Use of the acute cardiac ischemia timeinsensitive predictive instrument (ACI-TIPI)
to assist with triage of patients with chest
pain or other symptoms suggestive of acute
cardiac ischemia. A multicenter, controlled
clinical trial. Ann Intern Med.
1998;129(11):845-55. Epub 1998/12/29.
106.
Poon EG, Kuperman GJ, Fiskio J, Bates

DW. Real-time notification of laboratory
data requested by users through
alphanumeric pagers. J Am Med Inform
Assoc. 2002;9(3):217-22. Epub 2002/04/25.
PMCID: PMC344581.
107.
Gur D, Sumkin JH, Rockette HE, Ganott M,

Hakim C, Hardesty L, et al. Changes in
breast cancer detection and mammography
recall rates after the introduction of a
computer-aided detection system. J Natl
Cancer Inst. 2004;96(3):185-90. Epub
2004/02/05. PubMed PMID: 14759985.
108.
Kakeda S, Moriya J, Sato H, Aoki T,

Watanabe H, Nakata H, et al. Improved
detection of lung nodules on chest
radiographs using a commercial computeraided diagnosis system. AJR American
journal of roentgenology. 2004;182(2):50510. Epub 2004/01/23. PubMed PMID:
14736690.
109.
Cupples TE, Cunningham JE, Reynolds JC.

Impact of computer-aided detection in a
regional screening mammography program.
2005;185(4):944-50. Epub 2005/09/24. doi:
10.2214/AJR.04.1300. PubMed PMID:
16177413.
110.
Ramnarayan P, Winrow A, Coren M,

Nanduri V, Buchdahl R, Jacobs B, et al.
Diagnostic omission errors in acute
paediatric practice: impact of a reminder
system on decision-making. BMC medical
Epub 2006/11/08. doi: 10.1186/1472-69476-37. PubMed PMID: 17087835; PubMed
Central PMCID: PMC1654143.
111.
Fenton JJ, Taplin SH, Carney PA, Abraham

L, Sickles EA, DOrsi C, et al. Influence of
computer-aided detection on performance of
screening mammography. N Engl J Med.
2007;356(14):1399-409. Epub 2007/04/06.
doi: 10.1056/NEJMoa066099. PubMed
PMC3182841.
112.
Park HI, Min WK, Lee W, Park H, Park CJ,

Chi HS, et al. Evaluating the short message
service alerting system for critical value
notification via PDA telephones. Ann Clin
Lab Sci. 2008;38(2):149-56. Epub
2008/05/13. doi: 38/2/149 [pii]. PubMed
PMID: 18469361.
113.
114.
Piva E, Sciacovelli L, Zaninotto M,

Laposata M, Plebani M. Evaluation of
effectiveness of a computerized notification
system for reporting critical values. Am J
Clin Pathol. 2009;131(3):432-41. Epub
2009/02/21. doi: 131/3/432 [pii]
10.1309/AJCPYS80BUCBXTUH. PubMed
PMID: 19228648.
Singh H, Wilson L, Petersen LA, Sawhney
MK, Reis B, Espadas D, et al. Improving
follow-up of abnormal cancer screens using
electronic health records: trust but verify test
result communication. BMC medical
Epub 2009/12/17. doi: 1472-6947-9-49 [pii]
10.1186/1472-6947-9-49. PubMed PMID:
PMC2797509.
D-262
115.
David CV, Chira S, Eells SJ, Ladrigan M,

Papier A, Miller LG, et al. Diagnostic
accuracy in patients admitted to hospitals
with cellulitis. Dermatol Online J.
2011;17(3):1. Epub 2011/03/24. PubMed
PMID: 21426867.
116.
Olsson SE, Ohlsson M, Ohlin H, Dzaferagic

S, Nilsson ML, Sandkull P, et al. Decision
support for the initial triage of patients with
acute coronary syndromes. Clin Physiol
Funct Imaging. 2006;26(3):151-6. Epub
2006/04/28. doi: CPF669 [pii]
10.1111/j.1475-097X.2006.00669.x.
117.
Pozen MW, DAgostino RB, Selker HP,

Sytkowski PA, Hood WB, Jr. A predictive
instrument to improve coronary-care-unit
admission practices in acute ischemic heart
disease. A prospective multicenter clinical
trial. N Engl J Med. 1984;310(20):1273-8.
Epub 1984/05/17. doi:
10.1056/NEJM198405173102001. PubMed
PMID: 6371525.
118.
Thomas DC, Spitzer WO, MacFarlane JK.

Inter-observer error among surgeons and
nurses in presymptomatic detection of breast
disease. J Chronic Dis. 1981;34(12):617-26.
Epub 1981/01/01. PubMed PMID: 7309826.
119.
Davis Giardina T SH. Should patients get

direct access to their laboratory test results?:
An answer with many questions. JAMA:
The Journal of the American Medical
Association. 2011;306(22):2502-3. doi:
10.1001/jama.2011.1797.
120.
Casalino LP, Dunham D, Chin MH, Bielang

R, Kistner EO, Karrison TG, et al.
Frequency of Failure to Inform Patients of
Clinically Significant Outpatient Test
Results. Arch Intern Med.
2009;169(12):1123-9.
121.
Callen JL, Westbrook JI, Georgiou A, Li J.

Failure to Follow-Up Test Results for
Ambulatory Patients: A Systematic Review.
J Gen Intern Med. 2011. Epub 2011/12/21.
doi: 10.1007/s11606-011-1949-5. PubMed
PMID: 22183961.
122.
Molins E, Comas M, Roman R, RodriguezBlanco T, Sala M, Macia F, et al. Effect of

participation on the cumulative risk of falsepositive recall in a breast cancer screening
programme. Public Health.
2009;123(9):635-7. Epub 2009/09/08. doi:
S0033-3506(09)00198-X [pii]
10.1016/j.puhe.2009.07.007 [doi]. PubMed

PMID: 19733372.
123.
D-263
Singh H, Vij M. Eight recommendations for

policies for communicating abnormal test
results. Jt Comm J Qual Patient Saf.
2010;36:226-32
Evidence Tables for Chapter 36. Monitoring Patient Safety Problems (NEW)
The layout of this evidence table is customized based on the data reported by the included studies. Some columns in the evidence
tables for other PSP topics are not included in this table or merged with other columns. For example, the Description of
Organization column is merged into the Context column. There is no Theory or Logic Model column in this table because none
of the included studies reported such data. The customized layout allows the data collected to fit into the table appropriately.
Table 1, Chapter 36. Evidence from studies comparing methods for detecting patient safety problems
Author/
Year
Olsen
1
2007
Description of PSP
Study Design
Contexts
Outcomes: Benefits or Harms
Three methods for detecting adverse events (AEs)

were studied:
1) Incident reports
At the time of data collection risk managers
encouraged reporting of AEs and near misses but
provided no further criteria or guidelines for
reporting except that it was mandatory to supply
details of incidents in which security staff are
involved. Reporting is confidential but not
anonymous. The forms contain both mandatory
data fields and space for free text. During the
periods of data collection there were neither
additional incentives nor specific encouragements
to enhance reporting.
2) Active surveillance of prescription charts by
pharmacists
Hospital pharmacists attend the wards on
weekdays during normal working hours to ensure
continuity of pre-admission medications and to
detect prescribing errors. After discussion with ward
doctors errors and omissions are corrected on the
prescription charts. For each intervention a brief
record is made on a standardized form. The forms
related to the care of the 288 patients entered into
the study were collected and analyzed centrally in
the pharmacy.
3) Record review at time of discharge
Specialist registrars (senior residents) monitored by
external reviewers assessed all case records within
10 days of discharge of consecutively discharged
or deceased patients from the participating firms.
The occurrence of an adverse event or potential
This is a prospective
observational study. Data on AEs
were collected on 288 patients
discharged from adult medical
and surgical units in an acute care
hospital
External: In the UK, several

initiatives had been established
by the Department of Health to
promote patient safety. The
National Reporting and Learning
System (NRLS) established by
the National Patient Safety
Agency (NPSA) is one of the
initiatives. Incidents reported to
nine types of the National Health
System (NHS) Trust (ranging from
acute general hospitals to
community optometry), are
relayed centrally for classification
and analysis.
Organizational Characteristics: A
district general acute care hospital
in the NHS in the UK. The
hospital had an 850-bed and
received around 40 000
admissions per year. The hospital
trust covers a full range of
medical and general surgical
specialties backed up by full
intensive care facilities.
Teamwork: Safety data were
collected from three general
medical and three general
surgical teams. The teams were
selected by the head of risk
management.
Leadership: None mentioned
Record review detected 26 (9%)

AEs and 40 (14%) potential
adverse events (PAEs) occurring
during the index admission. Three
adverse events and 11 potential
adverse events were associated
with medications. Other
commonly occurring events
included inadequate clinical
monitoring and management
(17/66), technical problems with a
procedure (9/66), infection-related
problems (8/66) and failure to
arrange adequate follow-up or
care at discharge (7/66).
Incident reporting detected
11 PAEs and no AEs. These
PAEs included delay in crossmatching blood for a patient
requiring surgery; poor clinical
hand over of a patient from
accident and emergency to ward
staff; a fall causing a bruised head
that required medical
assessment, an intravenous
cannula misplaced in the brachial
artery, five concerned falls without
significant injury and two episodes
in which security staff were called
in relation to absconded or
aggressive patients.
Pharmacy surveillance found
D-264
Author/
Year
Wetzels
2
2008
Description of PSP
adverse event was determined for each case.
Each event was classified according to the stage of
care and a mutually exclusive problem category
(diagnosis, overall assessment of patients
condition including comorbidities, technical
problems occurring during a procedure, infectionrelated, general problems with ongoing monitoring
and management of patients and medicationrelated problems). Record review was also carried
out by members of the clinical team caring for the
patients. But in this report only the data collected by
the external assessors were used.
Five methods for identifying adverse events in
general practice:
1) Physician reported adverse events
The physicians recorded all events using a
simplified computerized registration form based on
an existing international taxonomy for errors in
general practice. The physicians registered event
date, birth date of patient, gender, event category
(practice administration (archive; medical record;
appointment; other), diagnostic (wrong diagnosis;
delayed diagnosis; missed diagnosis; other),
therapeutic (wrong, incomplete; delayed; none,
though it should be; other), communication (with
patients; with caregivers; other), and additional
remarks and/or context.
2) Pharmacist reported adverse events
The pharmacist recorded events from her point of
view using an adjusted form developed for this
purpose. Event date, birth date of patient, gender,
practice, event category (prescribing error (wrong
prescription, wrong administration; wrong dose;
other), adverse reaction (adverse reaction; allergic
reaction; overdose; interaction; contra-indication;
other), dispensing error (too late; wrong medicine;
wrong dose; other), and additional remarks or
context were recorded.
3) Patients experiences of adverse events
In the waiting room of the two practices samples of
50 patients, consecutively visiting the practice,
were invited to complete a questionnaire on
Study Design
A prospective observational
study, comparing the five
methods in two general practices
in a period of five months (May to
October 2006)
A total of approximately
8,250 patients were registered
with the two practices
D-265
Contexts
Culture: In this hospital, as

throughout the NHS, risk
managers encourage clinical staff
to report, on printed forms,
incidents that may affect patients
adversely.
Implementation tools: Various
forms were used for data
collection.
30 medication errors all of which

were PAEs. The most common
problems related to failure to
prescribe regular or indicated
medication (15/30) and failure to
prescribe the correct dose of a
drug (9/30).
There was little overlap in the
nature of events detected by the
three methods.
External: None mentioned

Organizational Characteristics:
Two general practices in the
Netherlands; no other detail
provided
Teamwork: Multiple physicians,
pharmacists, or researchers were
involved in the study collecting or
reviewing data.
Culture: None mentioned
Implementation tools: The
physicians recorded all events
using a simplified computerized
registration form based on an
existing international taxonomy for
errors in general practice. The
pharmacist recorded events from
her point of view using an
adjusted form developed for this
purpose. A questionnaire was
used to collect data patients
experience. Refer to Description
of PSP for more description.
A total of 68 events were

identified using these methods.
The events detected in four
categories: 1) Events in office
administration, 2) Events in
diagnosis, 3) Treatment events,
and 4) Events in Communication.
All five methods proved to identify
a number of adverse events.
Each of the methods provided
events that were not found with
other methods. There was no
overlap between the methods
regarding the identified events.
The patient survey accounted for
the highest number of events and
the pharmacist reports for the
lowest number. All methods
resulted in a variety of events,
except for the pharmacist reports,
which only referred to
pharmaceutical treatment. The
identified events referred to adult
male and female patients of all
ages, but events on children were
very seldom reported.
Author/
Year
Ferranti
3
2008
Description of PSP
experienced problems with safety of their health
care in the previous six months. A drop box was
used to collect the completed questionnaires.
Questions were derived from items of the Medical
Harvard Study, and from questions of two survey
studies. The questionnaire guaranteed anonymity
of participating patients.
4) Assessment of a random sample of medical
records
Thirty medical records per physician were randomly
selected from patients who had visited their general
practice in the observation period. Anonymous
medical records, containing the information from
this period were printed out. Two clinical
researchers examined the information
independently. They scrutinized the records for
indications of events and, when found, categorized
the event (errors in office administration, diagnosis,
treatment or communication with their
subcategories); and added demographic data of the
patient. Subsequently the physicians discussed
their findings and reached consensus.
5) Assessment of all deceased patients
One physician examined the medical records of the
patients who had died in the period of the study for
events. The same registration form and analysis
procedure as for the audits of medical records was
used.
Two ADE detection systems were studied:
1) Voluntary reporting
The safety reporting system was developed as a
home-grown web application to provide a single
point of entry for voluntary reporting and allow
standardized evaluation of safety events across
Duke University Health System (DUHS). All DUHS
employees may access the reporting system and
are encouraged to report any safety events
witnessed, including near misses. Although
anonymous reporting is possible, DUHS policy
supports a non-punitive culture of safety. Safety
reporting system captures a myriad of event types
including medication/intravenous-related, blood
Study Design
Contexts
The study retrospectively

analyzed all ADEs detected using
the two independent system in
adults treated in the hospital (all
inpatients receiving service on
23 adult care nursing units
between December 1st, 2006 and
June 30th, 2007).
Adult, inpatient ADEs were
evaluated and scored using a
standardized methodology. ADEs
per 1,000 patient days were
calculated.

Organizational Characteristics: It
is a large, tertiary care academic
medical center in the DUHS
Teamwork: For both voluntary
reporting and computerized
surveillance, multidisciplinary
teams were used for investigating
reviewing and confirming the
findings. Refer to Description of
PSP for more detail.
Leadership: For voluntary
reporting, a multidisciplinary
leadership team reviewed and
Computerized surveillance
detected 710 ADEs (6.93/1,000
patient days), whereas voluntary
reporting identified 205 ADEs
(1.96/1,000 patient days). For
each major drug category
(anticoagulants, hypoglycemia,
narcotics and benzodiazepines,
and miscellaneous), surveillance
and voluntary reporting detected
significantly different event rates.
Most surveillance events were
hypoglycemia-related, whereas
most voluntarily-reported events
D-266
Author/
Year
Levinson
4
2010
Description of PSP
transfusions, surgical, falls, treatment/testing,
dissatisfied patient, and others. Each
medication/intravenous-related report was
investigated by a team of 4 medication safety
pharmacists and scored for severity before
submission to a multidisciplinary leadership team
for review and confirmation. All events with a
severity score were deemed adverse drug events
(ADEs).
2) Computerized surveillance
The DUHs computerized ADE-S system was
deployed by an internal team of technical and
safety experts. Each evening, ADE-S evaluates
medication, laboratory, and patient demographic
information against a set of clinical rules or triggers
to detect potential ADEs or evolving unsafe
conditions. Nearly 130 rules have been deployed
since the systems inception, but only 14 high-risk
rules with high true-positive rates were considered
in surveillance. These 14 rules span 3 main
categories: abnormal laboratory results, use of
antidotes, and drug-lab combinations. Adverse drug
event surveillance delivers an electronic daily report
to a web-based surveillance application that details
all triggers fired by the system. This list was
evaluated by 3 clinical pharmacists who perform a
chart review to determine whether an ADE
occurred. Pharmacists identified all possible
medications involved in the event and assigned a
causality score using the Naranjo algorithm and a
severity score using the DUH 7-point scale. All
events scored with causality Q5 and a severity Q3
were considered ADEs. Pair wise inter-rater
reliability scores (J statistic) exceeded 0.88 for each
rater pair.
The following safety problem monitoring methods
were assessed:
1) Nurse Reviews
Contracted registered nurses reviewed medical
records for each sampled 278 Medicare
beneficiarys hospitalization. Nurses used a
standardized review process developed by the
Study Design

evaluated the usefulness of five
methods for identifying patient
safety events in 278 Medicare
beneficiary hospitalizations
selected from all Medicare
discharges from acute care
D-267
Contexts
confirmed the findings.

Culture: There is a highly vigilant,
non-punitive culture of safety at
DUH
Implementation tools: Business
intelligence software was used to
provide real time access to event
reports from both the For both
voluntary reporting and
computerized surveillance
systems to empower caregivers
with safety data originating from
their clinical care areas.
were in the miscellaneous

category. The 2 systems detected
statistically different ADE rates
when stratified by nursing station.
Of all unique ADEs (875), only
40 (5.6%) were common between
the systems.

None mentioned
Teamwork: None mentioned
Implementation tools: IHIs GTT
Nurse reviews and POA analysis

identified the greatest number of
safety events. Nurse reviews
identified 93 of the 120 safety
events in the case study and POA
analysis identified 61 events.
Beneficiary interviews identified
Author/
Year
Description of PSP
Study Design
Contexts
Institute for Healthcare Improvement (IHI) as part of

its Global Trigger Tool (GTT) protocol. The nurse
review used IHIs GTT worksheet that listed 54
triggers that could be found within a medical
record to indicate the possibility of an event. When
a trigger was found, the nurse reviewer explored
the medical record further to identify possible
events and associated level of harm.
2) Analysis of present-on-admission (POA)
Indicators
Administrative billing data directly from hospitals for
each of the 278 sample Medicare beneficiary
hospitalizations was analyzed. POA indicators in
the billing data was used to identify hospitalizations
that may have had events. When the POA indicator
showed that a diagnosis was not present upon
admission, the investigator concluded that the
condition developed during the hospital stay and
might have been the result of an event.
3) Beneficiary Interviews
The investigators conducted telephone interviews
with 220 of the 278 Medicare beneficiaries or their
family members to learn about the medical care
experienced during sampled hospitalizations. The
interview protocol was designed to determine
whether beneficiaries experienced any episodes
while in the hospital that might have involved
events. It also included questions about such topics
as medications, procedures, infections, and falls.
4) Hospital Incident Reports
The investigators requested that hospitals provide
any internal incident reports, such as submissions
to any hospital incident-reporting systems, adverse
drug reaction reports, complaints, peer reviews,
and mortality and morbidity reviews associated with
the 278 sample Medicare beneficiary
hospitalizations. Reports provided by hospitals
included issues related to risk management,
hospital infections, surgical management, and
others.
5) Analysis of Patient Safety Indicators
The investigators applied the Agency for
hospitals in two selected counties

during a 1-week period in August
2008. The investigators compared
events flagged by each method to
the 120 events identified and/or
confirmed through physician
reviews.
worksheet, POA indicators, and

AHRQs PSI software program
were used in the study. Detailed
description of these tools was
provided in the appendix of the
4
Levinson study.
22 events, and the remaining 2

screening methods identified 8
events each. Of the 120 events,
55 (46 percent) were identified by
only 1 screening method. Nurse
reviews identified 35 events
(29 percent of the 120 events)
not flagged by any other
screening method. POA analysis
alone flagged 14 events
(12 percent of the 120 events).
Although the five screening
methods were useful in identifying
events, 406 of the 662 flags
generated by the methods were
not associated with any of the
120 events identified in the case
study. The POA analysis
generated the most flags that
were not associated with events
(183 flags) and PSI analysis
generated the fewest (4 flags).
D-268
Author/
Year
LevtzionKorach
5
2010
Description of PSP
Healthcare Research and Qualitys (AHRQs)
Patient Safety Indicator (PSI) software program to
hospital administrative billing data for the 278
sample Medicare beneficiary hospitalizations.
AHRQ developed the PSI software to monitor
health care quality using administrative data, such
as patient demographics (e.g., age, gender), and
diagnoses and procedure codes. The PSI software
is based upon a series of algorithms that detect 20
provider-level complications that indicate possible
events (e.g., death of a low-risk patient).
The following safety problem monitoring methods
were assessed:
1) An incident reporting system
The hospital used a commercially available Webbased incident reporting system. Hospital
personnel could report confidentially through any
hospital computer using a secure login and could
report anything that they perceive might be an
issue. Each adverse event report contains the
reporters initial comments and a section for the
departmental manager to clarify issues further and
add comments and actions. The manager is
responsible for reviewing each report and assigning
one or more contributing factors from a drop-down
list of 50 potential contributing factors. For the most
important reports, management will have direct
conversation with the reporters after the evaluation
is complete
2) Reports to hospital risk management
A nurse-lawyer leads the risk management team.
Physicians and nurses, in about equal numbers,
call the team to report adverse events and poor
patient outcomes. Risk management staff members
investigate each case and determine on the basis
of the estimated risk whether to report the case to
the malpractice carrier. This information is collected
manually with no systematic categorization and is
entered in an electronic index. Risk management
also provides information back to managers or
frontline individuals so that risks can be mitigated.
3) A patient complaints database
Study Design
Contexts
This is a prospective
observational study, comparing
the five safety problem detection
methods. Data were collected for
a 22-month period from
May 10, 2004, to
February 28, 2006.
8,616 incident reports (involving
13,255 contributing factors),
1,003 risk management reports,
4,722 patient complaints
(involving 6,617 specific
problems), 61 walk rounds
(involving 572 comments), and
322 malpractice claims (involving
949 issues) were evaluated.

This study was performed at
Brigham and Womens Hospital,
a 747-bed tertiary care academic
medical center affiliated with
Harvard Medical School.
There are approximately
52,000 inpatient admissions and
950,000 outpatient visits annually.
The hospital employs more than
12,000 people, of whom
approximately 3,000 are
physicians.
The hospital had more
independent data sources than is
the norm.
Teamwork: The study mentioned
multidisciplinary team efforts for
some methods used
(e.g., hospital risk management,
handling patient complaints
databases, and executive walk
rounds). Refer to Description of
PSP for more description.
Leadership: For executive walk
round, the deep involvement by
the top-level executives was
mentioned in the study. Refer to
Description of PSP for more
description.
Across the five methods, the

leading categories of safety
problems were communication,
11.6%; technical skills, 10.9%;
and clinical judgment, 9%. Each
of the methods had a different
category that was most frequent.
Clinical judgment was the leading
category in the malpractice claims
data (24.3%) but was barely
represented in the incident
reporting system (1.1%) and not
represented at all in executive
walk rounds.
Communication played an
important role both in the
malpractice claims (17.1%) and
the patient complaints data
(21.8%) but not in the hospitals
risk management data (3%).
Provider behavior accounted for
19% of complaints in the patient
complaints system, second only
to communication (clearly the two
are closely related). However,
provider behavior represented
only 1.1% of the malpractice
claims and 2.1% of reports to
risk management and was not
represented in the executive walk
rounds or incident reporting
D-269
Author/
Year
Description of PSP
Study Design
The hospitals Family and Patient Relations

Department responds to patient and family
complaints (concerns), suggestions, and
compliments. The departments coordinators
receive the complaints, assign them to one of
20 categories and one or more of
118 subcategories, and process them into a
database. The department works directly with the
hospital risk management team and safety team,
which includes a physician, nurses, and safety
analysts; although the analysts mostly do not have
a medical background, they are trained in patient
safety.
4) Executive walk rounds
Executive leadership walk rounds began at the
hospital in January 2001. Semiweekly, a member of
the hospital leadership (hospital chief executive
officer, chief medical officer, chief nursing officer,
chief operating officer) accompanied by the
hospitals safety officer, a safety analyst, and a
pharmacy representative visits a different service in
the hospital and engages with the staff (mainly
nurses but occasionally also physicians) about
safety concerns. In stimulated discussions, staff is
encouraged to speak freely and make suggestions
for improvement. The staff comments (negative and
positive) are assigned one or more (out of 51)
contributing factors and a priority score, which are
then recorded in an electronic database. Analyses
of the comments are then compiled into action
items that are discussed with the accountable vice
president.
5) Malpractice claims
The hospital used a data collection system called
CMAPS (Claims Management, Analysis, and
Processing System) from the malpractice insurer,
CRICO/Risk Management Foundation (RMF;
Cambridge, MA). Initial information is obtained from
potential claim reports, hospital risk managers, or
from formal malpractice claims and suits. Further
information is added as it becomes available (for
example, depositions, expert reviews, medical
D-270
Contexts
Culture: The institution has

a history of patient safety
awareness, and was willing to
allow all its defect data to be
closely examined.
Implementation tools: The
hospital used a commercially
available Web-based incident
reporting system. The hospital
used CMAPS to collect data on
malpractice claims. Refer to
Description of PSP for more
description about the incident
reporting system and the CMAPS
system.
system.
Equipment (15.7%),
electronic records (12.2%), and
environment/infrastructure
(12.1%) were the leading
categories in executive walk
rounds but were ranked low in the
other systems.
In the incident reports,
identification issues (24.4%) and
falls (16.8%) were the leading
categories but were barely
represented in the other systems.
Overall, there is a low level of
consistency across the five
methods. The highest correlations
between the different categories
across the methods were
between malpractice claims,
reports to risk management, and
patient complaints. The adverse
event reporting system and
executive walk rounds had low
and negative correlation with the
other four systems.
Author/
Year
Tinoco
6
2011
Description of PSP
records, adjustor notes). Nurse coders assign one
or more (from 170) risk management issues,
factors that may have contributed to the allegation,
injury, or initiation of the claim/suit. There are clear
definitions, standardized coding algorithms, and
collaboration between coders leading to high interrater reliability. The data are stored in an electronic
database that is available for querying, analysis,
and generation of reports. There are about
30 claims per year.
Two methods for detecting inpatient adverse drug
events (ADEs) and hospital-associated infections
(HAIs) were studied: A computerized surveillance
system (CSS) or manual chart review (MCR)
For CCS, the HELP (Health Evaluation through
Logical Processing) system was used. This
electronic system manages billing and
administrative codes for each hospital admission,
as well as information from several clinical
domains: admission, discharge, and transfer
(ADT)/registration, pharmacy, laboratory,
microbiology, nurse charting, and physician
narratives, etc. The physician narratives stored in
the HELP system as freetext documents include
emergency department report, admission history
and physical report, consultant note, radiology
report, surgical procedure note, and discharge
summary.
The HAI detection criteria used by CSS were
originally based on the guidelines from the Study of
the Efficacy of Nosocomial Infection Control and
the Centers for Disease Control and Prevention
(CDC). In addition to routine HAI surveillance, daily
urine samples from all catheterized patients were
obtained as part of an existing, hospital-wide
urinary catheter surveillance program. The ADE
detection criteria used by CSS include various
clinical triggers such as medication discontinuation
orders, dose decrease orders, antidote orders,
laboratory test orders, abnormal laboratory test
results and vital signs. Suspected cases are
flagged by CSS and reported to surveillance
Study Design
Contexts

analyzed inpatient ADEs and
HAIs detected either by CSS or
MCR.
Data were collected from
2,137 unique, prescreened
admissions to the medical and
surgical services of the LDS
Hospital between October 1, 2000
and December 31, 2001.
Descriptive analysis was
performed for events detected
using the two methods by type of
AE, type of information about the
AE, and sources of the
information.

The study was performed at LDS
Hospital, a major teaching
hospital in Salt Lake City, Utah.
Teamwork: None mentioned
Implementation tools: For CSS,
the HELP system was used,
which has an integrated CSS that
prospectively screens electronic
patient data for indicators of AEs,
including HAIs and ADEs. Refer
to Description of PSP for more
description about the HELP
system.
CSS detected more HAIs than

MCR (92% vs. 34%); however,
a similar number of ADEs was
detected by both systems
(52% vs. 51%). The agreement
between systems was greater for
HAIs than ADEs (26% vs. 3%).
The CSS missed events that
did not have information in coded
format or that were described only
in physician narratives. The MCR
detected events missed by CSS
using information in physician
narratives. Some ADEs found by
MCR were detected by CSS but
not verified by a clinician.
D-271
Author/
Year
Description of PSP
Study Design
Contexts
personnel for validation. An infection preventionist

or a clinical pharmacist verifies each HAI or ADE,
respectively, using information from the patient
record, direct bedside observations, and interviews
with patients and their providers.
The MCR data were from a previous multiinstitutional research investigation of AEs
(workload study). No other detail was provided
about how MCR was performed.
D-272
References
1.
Olsen S, Neale G, Schwab K, et al.

Hospital staff should use more than one
method to detect adverse events and
potential adverse events: incident reporting,
pharmacist surveillance and local real-time
record review may all have a place. Qual Saf
Health Care 2007 Feb;16(1):40-4. Also
available:
http://www.ncbi.nlm.nih.gov/pmc/articles/P
MC2464933/pdf/40.pdf. PMID: 17301203
2.
Wetzels R, Wolters R, van Weel C, et al.

Mix of methods is needed to identify
adverse events in general practice: a
prospective observational study. BMC Fam
Pract 2008;9:35. PMID: 18554418
3.
Ferranti J, Horvath MM, Cozart H, et al.

A multifaceted approach to safety: the
synergistic detection of adverse drug events
in adult inpatients. J Patient Saf 2008;4:18490. Also available:
http://analytics.dhts.duke.edu/wysiwyg/dow
nloads/Ferranti_JPS_adults.pdf.
D-273
4.
Levinson DR. Adverse events in hospitals:

methods for indentifying events [OEI-06-0800221]. Washington (DC): Department of
Health and Human Services, Office of
Inspector General; 2010 Mar. 60 p.
5.
Van Der Linden W, Warg A, Nordin P.

National register study of operating time and
outcome in hernia repair. Arch Surg 2011
Oct;146(10):1198-203. PMID: 22006880
6.
Tinoco A, Evans RS, Staes CJ, et al.

Comparison of computerized surveillance
and manual chart review for adverse events.
J Am Med Inform Assoc 2011 JulAug;18(4):491-7. PMID: 21672911
Evidence Tables for Chapter 37. Interventions To Improve Care Transitions at

Table 1, Chapter 37. Included studies and interventions
Sample Size
Study
Design
Study, Year
Adler et al, 2009
Al-Rashed et al, 2002
Intervention
Readmission
144
156
CCT
General medical
Geriatric/General
medical
44
45
RCT
General medical
49
47
RCT
124
119
CCT
General medical
Geriatric/General
medical
Geriatric/Mixed
Diagnoses
Geriatric/Mixed
Diagnoses
Geriatric/General
medical
83
83
1235
158
371
379
64
64
626
581
43
30
Brand et al, 2004
Coleman et al, 2004
CCT
Coleman et al, 2006
RCT
RCT
Courtney et al, 2009

9
Cowan et al, 2006
CCT
General medical
229
243
RCT
General medical
186
175
RCT
Geriatric
Geriatric/General
medical
200
200
201
199
General medical
Mixed patient
population
59
18
315
316
117
115
66
66
376
373
13
14
RCT
15
CCT
16
17
Hogan and Fox, 1990

19
CCT
Graumlich et al, 2009
Jack et al, 2009
RCT
Forster et al, 2005
Hellstrom et al, 2011
11
12
Gillespie et al, 2009
10
Einstadler et al, 1996
Gallagher et al, 2011
General medical
Geriatric/General
medical
Dellasega et al, 2000
Gow et al, 1999
Costs
Balaban et al, 2008

Bolas et al, 2004
Control
CCT
Adverse Event
Acute care/ED
utilization
Patient Population
18
RCT
CCT
CCT
RCT
Geriatric
Geriatric/General
medical
General medical and
surgical
D-274
Sample Size
Study
Design
Study, Year
20
Koehler et al, 2009

Lim et al, 2003
21
Intervention
Readmission
21
20
314
340
Geriatric
Mixed patient
population
Geriatric/General
medical
356
350
231
221
25
29
746
792
186
177
20
20
181
181
RCT
General medical
Geriatric/Mixed
Diagnoses
Geriatric/General
medical
Geriatric/General
medical
Geriatric/General
medical
185
181
CCT
General medical
1014
1248
RCT
Geriatric/Mix
Diagnoses
49
49
RCT
General medical
250
250
RCT
22
RCT
23
CCT
Makowsky et al, 2009

24
Martin et al, 1994
Naylor et al, 1999
RCT
25
CCT
26
RCT
Mudge et al, 2006
27
RCT
Nazareth et al, 2001

Nikolaus et al, 1999
Palmer et al, 2002
Parry et al, 2009
Control
RCT
Lipton and Bird, 1994
Naylor, 1990
Patient Population
Geriatric/General
medical
Geriatric/General
medical
28
29
30
31
Schillig et al, 2011
32
RCT
Adverse Event
Costs
General medical
84
92
Scullin et al, 2007
RCT
General medical
391
371
Scullin et al, 2011
35
CCT
84
749
176
178
CCT
General medical
Geriatric/General
medical
Geriatric/General
medical
469
355
RCT
Mixed Diagnoses
453
166+
190 (2
groups)
453
43 (0-4 day
stay)+ 137 (5+
day stay)
64
68
36
RCT
Steeman et al, 2006

Stewart et al, 1998
37
38
39
CCT
40
RCT
Styrborn et al, 1995

Thomas et al, 1993
Geriatric/General
medical
Geriatric/General
medical
D-275
RCT
Siu et al, 1996
34
Schnipper et al, 2006
33
Acute care/ED
utilization
Sample Size
Study, Year
Voss et al, 2011
41
42
Walker et al, 2009
43
Study
Design
Patient Population
Control
Intervention
Readmission
CCT
General medical
736
257
CCT
General medical
366
396
695
Weinberger et al, 1996

RCT
Mixed Diagnoses
701
Abbreviations: CCT=controlled clinical trial; RCT=randomized controlled trial
D-276
Adverse Event
Costs
Acute care/ED
utilization
Table 2, Chapter 37. Study quality criteria
Study, Year
1
Adler et al, 2009
2
Al-Rashed et al, 2002
3
Balaban et al, 2008
4
Bolas et al, 2004
5
Brand et al, 2004
6
Coleman et al, 2004
7
Coleman et al, 2006
8
Courtney et al, 2009
9
Cowan et al, 2006
10
Dellasega et al, 2000
11
Einstadler et al, 1996
12
Forster et al, 2005
13
Gallagher et al, 2011
14
Gillespie et al, 2009
15
Gow et al, 1999
16
Graumlich et al, 2009
17
Hellstrom et al, 2011
Hogan and Fox,
18
1990
19
Jack et al, 2009
20
Koehler et al, 2009
21
Lim et al, 2003
22
Lipton and Bird, 1994
Makowsky et al,
23
2009
24
Martin et al, 1994
25
Mudge et al, 2006
26
Naylor et al, 1999
27
Naylor, 1990
28
Nazareth et al, 2001
29
Nikolaus et al, 1999
30
Palmer et al, 2002
31
Parry et al, 2009
32
Schillig et al, 2011
33
Schnipper et al, 2006
34
Scullin et al, 2007
Study
Design
CCT
CCT
RCT
RCT
CCT
CCT
RCT
RCT
CCT
RCT
CCT
RCT
RCT
RCT
CCT
RCT
CCT
Allocation
Sequence
Random?
No
Unclear
Unclear
Yes
Unclear
No
Yes
Yes
No
Yes
No
Yes
Yes
Yes
No
Unclear
No
Allocation
Concealed?
No
Unclear
Unclear
Unclear
Unclear
No
Unclear
Yes
Yes
Yes
No
Yes
Yes
Yes
Unclear
No
Unclear
Baseline
Outcomes
Similar?
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Unclear
Yes
Unclear
Unclear
Unclear
Unclear
Unclear
Baseline
Characteristics
Similar?
No
Unclear
Yes
Yes
Unclear
No
No
Yes
No
Yes
Yes
Yes
Yes
Unclear
Unclear
Unclear
Yes
Incomplete
Data
Addressed?
Unclear
Unclear
Yes
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Unclear
Yes
No
Unclear
Unclear
Unclear
Yes
Outcomes
Assessed Blind
to Intervention?
Unclear
Unclear
Unclear
Yes
Unclear
Unclear
Yes
Yes
Yes
Yes
Unclear
Yes
Yes
Yes
Unclear
Yes
Unclear
No
Contamination?
Unclear
Unclear
Unclear
Unclear
Yes
Unclear
Yes
Unclear
Yes
Yes
Unclear
Yes
Unclear
Unclear
Unclear
Yes
Unclear
Free of
Selective
Outcome
Reporting
Risk?
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Yes
Unclear
Yes
Yes
Yes
Yes
Unclear
Yes
No
Yes
Yes
CCT
RCT
RCT
RCT
RCT
No
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Unclear
Unclear
Unclear
Unclear
Yes
Unclear
Unclear
Yes
Yes
Yes
Yes
Unclear
Unclear
Yes
Yes
Unclear
Yes
Unclear
Yes
Yes
Unclear
Unclear
Yes
Unclear
Yes
Yes
Unclear
Unclear
Yes
Yes
No
Unclear
Unclear
Yes
Yes
CCT
RCT
CCT
RCT
RCT
RCT
RCT
CCT
RCT
RCT
RCT
RCT
No
Yes
No
Yes
Yes
Yes
Yes
No
Yes
No
Yes
Yes
No
Unclear
No
Yes
Unclear
Unclear
Yes
No
No
No
Yes
Yes
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
No
Unclear
Yes
Unclear
Unclear
Unclear
Unclear
Yes
No
Yes
Yes
Unclear
Yes
Unclear
Yes
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Yes
Unclear
Yes
Yes
Unclear
Yes
No
Yes
Yes
Yes
Yes
Unclear
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Unclear
Yes
Unclear
Unclear
Yes
Unclear
Yes
Yes
Yes
Yes
Unclear
Yes
Yes
Yes
Yes
Unclear
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
Yes
D-277
Free from
other
Bias?
No
Yes
Yes
Yes
Unclear
Unclear
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Unclear
Unclear
Study, Year
35
Scullin et al, 2011
36
Siu et al, 1996
37
Steeman et al, 2006
38
Stewart et al, 1998
39
Styrborn et al, 1995
40
Thomas et al, 1993
41
Voss et al, 2011
42
Walker et al, 2009
Weinberger et al,
43
1996
Study
Design
CCT
RCT
CCT
RCT
CCT
RCT
CCT
CCT
Allocation
Sequence
Random?
No
Yes
Unclear
Yes
No
Yes
No
No
Allocation
Concealed?
Unclear
Unclear
Unclear
Unclear
No
Unclear
No
No
Baseline
Outcomes
Similar?
Unclear
Unclear
Unclear
Yes
No
Unclear
Yes
Unclear
Baseline
Characteristics
Similar?
Unclear
Unclear
Yes
Unclear
No
Yes
No
Unclear
Incomplete
Data
Addressed?
Unclear
Unclear
Unclear
Unclear
No
Yes
Unclear
Unclear
Outcomes
Assessed Blind
to Intervention?
Yes
Yes
Yes
Unclear
Unclear
Yes
Yes
Unclear
No
Contamination?
Unclear
No
Unclear
Yes
Yes
Unclear
Yes
Yes
Free of
Selective
Outcome
Reporting
Risk?
Yes
Yes
Unclear
Unclear
Unclear
Yes
Unclear
Yes
RCT
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Unclear
D-278
Free from
other
Bias?
Unclear
No
Unclear
Yes
No
No
Yes
Yes
Yes
Table 3, Chapter 37. Included studies and outcomes

Pre-discharge Interventions
Source
Adler et al,
1
2009
Al-Rashed et
2
al, 2002
Balaban et al,
3
2008
Bolas et al,
4
2004
Brand et al,
5
2004
Coleman et
6
al, 2004
Coleman et
7
al, 2006
Courtney et
8
al, 2009
Cowan et al,
9
2006
Dellasega et
10
al, 2000
Einstadler et
11
al, 1996
Forster et al,
12
2005
Gallagher et
13
al, 2011
Gillespie et al,
14
2009
Gow et al,
15
1999
Graumlich et
16
al, 2009
Hellstrom et
17
al, 2011
Hogan and
18
Fox, 1990
Risk
Stratification
Individualized Patient
Record
MultiPatient
Facilitation of disciplinary
Engagement Communication
Team
Dedicated
Discharge
Advocate
Medication
Reconciliation
Post-discharge Interventions
Facilitated
Clinical
Outreach
Medication
Follow-up to Patients Reconciliation
D-279
Source
Jack et al,
19
2009
Koehler et al,
20
2009
Lim et al,
21
2003
Lipton and
22
Bird, 1994
Makowsky et
23
al, 2009
Martin et al,
24
1994
Mudge et al,
25
2006
Naylor et al,
26
1999
Naylor,
27
1990
Nazareth et
28
al, 2001
Nikolaus et al,
29
1999
Palmer et al,
30
2002
Parry et al,
31
2009
Schillig et al,
32
2011
Schnipper et
33
al, 2006
Scullin et al,
34
2007
Scullin et al,
35
2011
Siu et al,
36
1996
Steeman et
37
al, 2006
Stewart et al,
38
1998
Risk
Stratification
Record
MultiPatient
Team
Dedicated
Discharge
Advocate
Medication
Reconciliation
Facilitated
Clinical
Outreach
Medication
D-280
Source
Styrborn et al,
39
1995
Thomas et al,
40
1993
Voss et al,
41
2011
Walker et al,
42
2009
Weinberger et
43
al, 1996
Risk
Stratification
Record
MultiPatient
Team
Dedicated
Discharge
Advocate
Medication
Reconciliation
Facilitated
Clinical
Outreach
Medication
D-281
Table 4, Chapter 37. Studies reporting 30-day readmissions

Author, Year
Study
Design
Bridging
Intervent
ion (Y/N)
Total
number of
intervention
s
30-day
readmiss
ion rate control
group
30-day
readmissi
on rate interventi
on group
ARR - 30-day
readmissions
Al-Rashed et
2
al, 2002
CCT
29.5%*
11.1%*
18.4%
Balaban et al,
3
2008
Coleman et al,
6
2004
RCT
8.2%
8.5%
-0.3%
2.0%
2.1%
-0.1%
CCT
13.8%
8.9%
4.9%
14.2%
11.0%
3.2%
Coleman et al,
7
2006
RCT
11.9%
8.3%
3.6%
Courtney et al,
8
2009
RCT
~15%,
only in
graphics
~5%, only
in
graphics
Dellasega and
10
Zerbe, 2000
Einstadler et
11
al, 1996
Forster et al,
12
2005
Gow et al,
15
1999
Jack et al,
19
2009
RCT
NR
NR
CCT
14.4%
19.8%
-5.4%
19.2%
21.8%
-2.6%
RCT
13.0%
19.4%
-6.4%
7.8%
9.6%
-1.8%
CCT
15.2%
16.7%
-1.5%
RCT
20.7%
14.9%
5.8%
D-282
30-day
ED visits
- control
group
30-day ED
visits intervention
group
ARR - 30day ED
visits
Statistically
significant
ARR (for
either
readmits or
ED visits)
Y
Comment
*Outcomes
measured at 15-22
days post-discharge
Significant
difference for
readmissions, but
not ED visits
te
kkappears only in
graphical format;
p<0.05 for
comparison with
control for
Data presented only
in graphical format;
significant
improvement for ED
visits
Exact rates not
supplied
N
24.5%
16.5%
8.0%
Significant
difference for ED
visits only
Author, Year
Study
Design
Bridging
Intervent
ion (Y/N)
Total
number of
intervention
s
30-day
readmiss
ion rate control
group
30-day
readmissi
on rate interventi
on group
Koehler et al,
20
2009
RCT
38.1%
10.0%
-28.1%
11.2%
15.4%
-4.2%
5.8%
5.5%
0.3%
Lipton and
21
Bird, 1994
Palmer et al,
30
2002
Parry et al,
31
2009
Schnipper et
33
al, 2006
Steeman et al,
37
2006
41
Voss, 2011
Walker et al,
42
2009
ARR - 30-day
readmissions
30-day
ED visits
- control
group
30-day ED
visits intervention
group
ARR - 30day ED
visits
Statistically
significant
ARR (for
either
readmits or
ED visits)
Y
Comment
Composite
outcome of ED
visits and
readmissions
i.e. 284 CCT.284
Control 284
Intervention 284
General Medical
284 Mixed 284
RCT
CCT
Y
16.7%
6.8%
9.9%
RCT
RCT
30%
30%
0%
CCT
5.1%
2.8%
2.3%
CCT
CCT
Y
Y
6
7
18.6%
18.0%
12.8%
22.1%
5.8%
-4.1%
D-283
12.3%
9.5%
2.8%
Y
N
Composite
outcome of total ED
visits and
readmissions
Outcomes
measured at 14
days post discharge
Table 5, Chapter 37. Studies reporting adverse events (including adverse drug events)
Study and
Country
Study
Design
Population
Sample
Size
Intervention
Control
Balaban et al.,
3
2008
US
RCT
General
medical
(Safety net)
96
Discharge-transfer
intervention
Forster et al.,
12
2005
Canada
Gallagher et al.,
13
2011
Ireland
RCT
General
medical
361
RCT
Geriatric
400
Gillespie et al.,
14
2009
Sweden
Graumlich et al.,
16
2009
US
Hellstrom et al.,
17
2011 *
Sweden
RCT
Geriatric/
General
Medical
Mixed Patient
Population
400
CCT
Geriatric
210
RCT
Geriatric/
General
medical
General
Medical/
Mixed (on
warfarin
therapy)
40
Naylor, 1990
27
Schillig et al.,
32
2011
US
US
RCT
RCT
631
500
ARR (95%
Cl)
Usual care
Adverse Drug Events

(Control vs.
Intervention rates, pvalue
-
Discharge coordination
led by nurse specialist
Usual care
Inpatient medication
screening using
validated criteria
(STOPP/START)
Pharmacist discharge
counseling and postdischarge follow-up
Software-assisted
discharge
Usual care
Usual care
12 months
1 month
Usual care
All post-discharge
adverse events (7.3
vs. 7.3%, N/A)
-
Pharmacist-led
systematic medication
reconciliation and
review
Comprehensive
discharge planning led
by nurse specialist
Pharmacist-directed
anticoagulation service
Medication-related
readmissions (24.2 vs.
4.8%, NR)
Probable adverse drug
event (5.4 vs. 5.4%,
N/A)
Composite of
medication-related
admissions & ED visits
(12.0 vs. 5.6%, 0.138)
-
3 months
Post-discharge
infection rates (50 vs.
33.3%, NR)
-
12 weeks
Inpatient &
30 days
(combined)
Usual care
Usual care
Usual care
Composite endpoint
(14.8 vs. 10.0%, 0.104)
INR >5 (14.8 vs. 9.6%,
0.076)
Major bleeding (0.8 vs.
0.4%, 0.563)
Thrombosis (0 vs. 0%,
N/A)
D-284
6.4% (-1.214.8)
Other Adverse
ARR (95%
Events (Control vs.
Cl)
intervention rates, pvalue)
Failure to complete
recommended
outpatient work-up
(31.3 vs. 11.5%, 0.11)
Post-discharge
adverse events (22.8
vs. 23.6%, 0.87)
Falls (8.4 vs. 5.8%,
0.332)
Timing
NR
30 days
6 months
Study and
Country
Schnipper et al.,
33
2006
US
Study
Design
Population
Sample
Size
Intervention
Control
RCT
General
medical
176
Pharmacist discharge
counseling and postdischarge follow-up
Usual care
*This was a nonrandomized controlled trial. All other studies are randomized controlled trials.
D-285
Adverse Drug Events

(Control vs.
Intervention rates, pvalue
Preventable
medication-related ED
visit or admission (8
vs. 1%, 0.03)
Preventable ADEs (11
vs. 1%, 0.01)
All ADEs (16 vs. 18%,
>0.99)
ARR (95%
Cl)
Other Adverse
ARR (95%
Events (Control vs.
Cl)
intervention rates, pvalue)
-
Timing
30 days
References for Tables 1-5

1.
2.
3.
4.
Adler A, Lipkin C, Cooper L, et al. Effect of

social work intervention on hospital
discharge transition planning in a special
needs population. Manag Care.
2009;18(11):50-3.19999254
discharge. Br J Clin Pharmacol.
2002;54(6):657-64.12492615
Balaban RB, Weissman JS, Samuel PA, et
al. Redefining and redesigning hospital
discharge to enhance patient care: a
randomized controlled study. J Gen Intern
Med. 2008;23(8):1228-33.18452048
Bolas H, Brookes K, Scott M, et al.
Evaluation of a hospital-based community
liaison pharmacy service in Northern
Ireland. Pharm World Sci. 2004;26(2):11420.15085948
5.
Brand CA, Jones CT, Lowe AJ, et al. A

transitional care service for elderly chronic
disease patients at risk of readmission. Aust
Health Rev. 2004;28(3):275-84.15595909
6.
Coleman EA, Parry C, Chalmers S, et al.

The care transitions intervention: results of a
randomized controlled trial. Arch Intern
Med. 2006;166(17):1822-8.17000937
7.
Coleman EA, Smith JD, Frank JC, et al.

Preparing patients and caregivers to
participate in care delivered across settings:
the care transitions intervention. Journal of
the American Geriatrics Society.
2004;52(11):1817-25.2005044404
8.
Courtney M, Edwards H, Chang A, et al.

Fewer emergency readmissions and better
quality of life for older adults at risk of
hospital readmission: a randomized
controlled trial to determine the
effectiveness of a 24-week exercise and
telephone follow-up program. J Am Geriatr
Soc. 2009;57(3):395-402.19245413
9.
Cowan MJ, Shapiro M, Hays RD, et al. The

effect of a multidisciplinary
hospitalist/physician and advanced practice
nurse collaboration on hospital costs. J Nurs
Adm. 2006;36(2):79-85.16528149
D-286
10.
Dellasega CA, Zerbe TM. A multimethod

study of advanced practice nurse
postdischarge care. Clin Excell Nurse Pract.
2000;4(5):286-93.11858450
11.
Einstadter D, Cebul RD, Franta PR. Effect

of a nurse case manager on postdischarge
follow-up. J Gen Intern Med.
1996;11(11):684-8.9120655
12.

Effect of a nurse team coordinator on
outcomes for hospitalized medicine patients.
Am J Med. 2005;118(10):114853.16194647
13.
Gallagher PF, OConnor MN, OMahony D.

Prevention of potentially inappropriate
prescribing for elderly patients: a
randomized controlled trial using
STOPP/START criteria. Clin Pharmacol
Ther. 2011;89(6):845-54.21508941
14.

Intern Med. 2009;169(9):894-900.19433702
15.
Gow P, Berg S, Smith D, et al. Care coordination improves quality-of-care at South

Auckland Health. J Qual Clin Pract.
1999;19(2):107-10.10408752
16.
Graumlich JF, Novotny NL, Stephen Nace

G, et al. Patient readmissions, emergency
visits, and adverse events after softwareassisted discharge from hospital: cluster
randomized trial. J Hosp Med.
2009;4(7):E11-9.19479782
17.
Hellstrom LM, Bondesson A, Hoglund P, et

al. Impact of the Lund Integrated Medicines
Management (LIMM) model on medication
appropriateness and drug-related hospital
revisits. Eur J Clin Pharmacol.
2011;67(7):741-52.21318595
18.
Hogan DB, Fox RA. A prospective

controlled trial of a geriatric consultation
team in an acute-care hospital. Age Ageing.
1990;19(2):107-13.2337005
19.

trial. Ann Intern Med. 2009;150(3):17887.19189907
28.

trial. Age Ageing. 2001;30(1):3340.11322670
20.

a targeted care bundle. J Hosp Med.
2009;4(4):211-8.19388074
29.
Nikolaus T, Specht-Leible N, Bach M, et al.

A randomized trial of comprehensive
geriatric assessment and home intervention
in the care of hospitalized patients. Age
Ageing. 1999;28(6):543-50.10604506
30.
Palmer HC, Jr., Halperin A, Elnicki M, et al.

Effect of a patient care partnership project
on cost and quality of care at an academic
teaching hospital. South Med J.
2002;95(11):1318-25.12540000
31.
Parry C, Min SJ, Chugh A, et al. Further

application of the care transitions
intervention: results of a randomized
controlled trial conducted in a fee-forservice setting. Home Health Care Serv Q.
2009;28(2-3):84-99.20182958
32.
Schillig J, Kaatz S, Hudson M, et al. Clinical

and safety impact of an inpatient
Pharmacist-Directed anticoagulation service.
J Hosp Med. 2011;6(6):322-8.21834113
33.

hospitalization. Arch Intern Med.
2006;166(5):565-71.16534045
34.
Scullin C, Hogg A, Luo R, et al. Integrated

medicines management - can routine
implementation improve quality? J Eval
Clin Pract. 2011.21504517
35.
Scullin C, Scott MG, Hogg A, et al. An

innovative approach to integrated medicines
management. J Eval Clin Pract.
2007;13(5):781-8.17824872
36.
Siu AL, Kravitz RL, Keeler E, et al.

Postdischarge geriatric assessment of
hospitalized frail elderly patients. Arch
Intern Med. 1996;156(1):76-81.8526700
37.
Steeman E, Moons P, Milisen K, et al.

Implementation of discharge management
for geriatric patients at risk of readmission
or institutionalization. Int J Qual Health
Care. 2006;18(5):352-8.16861721
21.
Lim WK, Lambert SF, Gray LC.

Effectiveness of case management and postacute services in older people after hospital
discharge. Med J Aust. 2003;178(6):2626.12633482
22.
Lipton HL, Bird JA. The impact of clinical

pharmacists consultations on geriatric
patients compliance and medical care use: a
randomized controlled trial. Gerontologist.
1994;34(3):307-15.8076871
23.
Makowsky MJ, Koshman SL, Midodzi WK,

et al. Capturing outcomes of clinical
activities performed by a rounding
pharmacist practicing in a team
environment: the COLLABORATE study
[NCT00351676]. Med Care.
2009;47(6):642-50.19433997
24.
25.
Martin F, Oyewole A, Moloney A. A

randomized controlled trial of a high support
hospital discharge team for elderly people.
Age Ageing. 1994;23(3):228-34.8085509
Mudge A, Laracy S, Richter K, et al.
Controlled trial of multidisciplinary care
teams for acutely ill medical inpatients:
enhanced multidisciplinary care. Intern Med
J. 2006;36(9):558-63.16911547
26.
Naylor MD. Comprehensive discharge

planning for hospitalized elderly: a pilot
study. Nurs Res. 1990;39(3):15661.2188217
27.
Naylor MD, Brooten D, Campbell R, et al.

Comprehensive discharge planning and
home follow-up of hospitalized elders: a
randomized clinical trial. JAMA.
1999;281(7):613-20.10029122
D-287
38.
Stewart S, Pearson S, Luke CG, et al.

Effects of home-based intervention on
unplanned readmissions and out-of-hospital
deaths. J Am Geriatr Soc. 1998;46(2):17480.9475445
39.
Styrborn K. Early discharge planning for

elderly patients in acute hospitals--an
intervention study. Scand J Soc Med.
1995;23(4):273-85.8919370
40.
Thomas DR, Brahan R, Haywood BP.

Inpatient community-based geriatric
assessment reduces subsequent mortality. J
Am Geriatr Soc. 1993;41(2):101-4.8426028
41.
Voss R, Gardner R, Baier R, et al. The care

efficacy to effectiveness. Arch Intern Med.
2011;171(14):1232-7.21788540
42.

study. Arch Intern Med. 2009;169(21):200310.19933963
43.
Weinberger M, Oddone EZ, Henderson WG.

Does increased access to primary care
reduce hospital readmissions? Veterans
Affairs Cooperative Study Group on
Primary Care and Hospital Readmission. N
Engl J Med. 1996;334(22):1441-7.8618584
D-288
Evidence Tables for Chapter 38. Use of Simulation Exercises

in Patient Safety Efforts
Evidence Tables for Chapter 39. Obtaining Informed Consent

From Patients: Brief Update Review
Evidence Tables for Chapter 40. Team-Training in Health

Care: Brief Update Review
Evidence Tables for Chapter 41. Computerized Provider

Order Entry With Clinical Decision Support Systems: Brief
Update Review
Evidence Tables for Chapter 42. Tubing Misconnections:

Brief Review (NEW)
Evidence Tables for Chapter 43. Limiting Individual

Providers Hours of Service: Brief Update Review
D-289

Bookshelf NBK133363 PDF

Uploaded by

Copyright:

Available Formats

Bookshelf NBK133363 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bookshelf NBK133363 PDF

Uploaded by

Copyright:

Available Formats

What is the purpose of the document?

What is the purpose of the document?

What organizations were involved in preparing the document?

What organizations were involved in preparing the document?

Evidence Report/Technology Assessment

Making Health Care

Evidence Report/Technology Assessment

Making Health Care Safer II: An Updated Critical

AHRQ Publication No. 13-E001-EF

This report is based on research conducted by the Southern California-RAND Evidence-based

Jean Slutsky, P.A., M.S.P.H.

Stephanie Chang, M.D., M.P.H.

James Battles, Ph.D.

Authors and Affiliations

Fang Sun, M.D., Ph.D.

Johns Hopkins University:

Kathryn Martinez, M.P.H.; Ph.D.

Aneesa Motala, B.A.

University of California, San Francisco/Stanford:

Peter B. Angood, M.D., FACS, FCCM

Richard Lilford Ph.D., FRCP, FRCOG, FFPHM

Rainu Kaushal, M.D., M.P.H.

Allison Gross, M.S., M.L.S.

Additional Chapter Contributors

Despina Contopoulos-Ioannidis, M.D.

Sean M. Berenholtz, M.D., M.H.S.

Kristina M. Cordasco, M.D., M.P.H.,

Vineet Chopra, M.D., M.Sc.

Paul Dallas, M.D.

Elliott R. Haut, M.D., FACS

Cyrus Engineer, Dr.P.H.

Susanne Hempel, Ph.D.

Mohamad G. Fakih, M.D., Ph.D.

Lawrence A. Ho, M.D.

Olavo A. Fernandes, B.Sc.Phm., Pharm.D.

John Ioannidis, M.D.

Tabor Flickinger, M.D., M.S.

Devan Kansagara, M.D., M.C.R., FACP

David A. Ganz, M.D., Ph.D.

Sarah Kianfar, M.S.

Melinda Maggard-Gibbons, M.D., M.S.H.S.

Sarah L. Krein, Ph.D., R.N.

Peter Glassman, M.B.B.S., M.Sc.

Janice Kwan, M.D.

Sara N. Goldhaber-Fiebert, M.D.

Brandyn D. Lau, M.P.H.

Oanh K. Nguyen, M.D.

Lisha Lo, M.Sc.

Russell N. Olmsted, M.P.H., C.I.C.

Julia Lonhart, B.S., B.A.

Noelle Pineda, B.A.

Lisa Lubomski, Ph.D.

Michael A. Rosen, Ph.D.

Yimdriuska Magan, B.S.

Nasia Safdar, M.D., Ph.D.

Brian Matesic, B.S.

Marin Schweizer, Ph.D.

Jennifer Meddings, M.D., MS.c.

Marwa H. Shoeb, M.D., M.S.