Includes Flash Drive!

Root Cause Analysis in

Health Care:
Tools and Techniques
F I F T H EDI T I O N
Executive Editor: James Parker
Project Manager: Lisa King
Associate Director, Publications: Helen M. Fry, MA
Associate Director, Production and Meeting Support: Johanna Harris
Executive Director, Publications and Education: Catherine Chopp Hinckley, MA, PhD
Joint Commission/JCR Reviewers for the fifth edition:
Lisa Buczkowski; Maureen Carr; Gerard Castro; Paul vanOstenberg; Ronald Wyatt, MD
Joint Commission Resources Mission
The mission of Joint Commission Resources (JCR) is to continuously improve the safety and quality of health care in the
United States and in the international community through the provision of education, publications, consultation, and
evaluation services.
Joint Commission Resources educational programs and publications support, but are separate from, the accreditation
activities of The Joint Commission. Attendees at Joint Commission Resources educational programs and purchasers of Joint
Commission Resources publications receive no special consideration or treatment in, or confidential information about, the
accreditation process.
The inclusion of an organization name, product, or service in a Joint Commission Resources publication should not be
construed as an endorsement of such organization, product, or service, nor is failure to include an organization name,
product, or service to be ­construed as disapproval.
This publication is designed to provide accurate and authoritative information in regard to the subject matter covered.
Every attempt has been made to ensure accuracy at the time of publication; however, please note that laws, regulations,
and standards are subject to change. Please also note that some of the examples in this publication are specific to the
laws and regulations of the locality of the facility. The information and examples in this publication are provided with
the understanding that the publisher is not engaged in providing medical, legal, or other professional advice. If any such
assistance is desired, the services of a competent professional person should be sought.
© 2015 The Joint Commission
Joint Commission Resources, Inc. (JCR), a not-for-profit affiliate of The Joint Commission, has been designated by The
Joint Commission to publish publications and multimedia products. JCR reproduces and distributes these materials under
license from The Joint Commission.
All rights reserved. No part of this publication may be reproduced in any form or by any means without written permission
from the publisher.
Printed in the USA 5 4 3 2 1
Requests for permission to make copies of any part of this work should be mailed to
Permissions Editor
Department of Publications and Education
Joint Commission Resources
1515 W. 22nd Street, Suite 1300W
Oak Brook, Illinois 60523 USA
[email protected]
ISBN: 978-1-59940-898-9
Library of Congress Control Number: 2009931192
For more information about Joint Commission Resources, please visit http://www.jcrinc.com.
 Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

CHAPTER 1: Root Cause Analysis: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Investigating Patient Safety Events: The Need for Comprehensive Systematic Analysis . . . . . . . . . . . 1
When Can a Root Cause Analysis Be Performed? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Variation and the Difference Between Proximate and Root Causes . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Benefits of Root Cause Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Maximizing the Value of Root Cause Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
The Root Cause Analysis and Action Plan: Doing It Right . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

CHAPTER 2: Addressing Sentinel Events in Policy and Strategy . . . . . . . . . . . . . . . . . 17

The Range of Adverse Events in Health Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Signals of Risk: Near Misses and No‑Harm Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
The Joint Commission’s Sentinel Event Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Reasons for Reporting a Sentinel Event to The Joint Commission . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Required Response to a Reviewable Sentinel Event . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Joint Commission International’s Sentinel Event Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Related Joint Commission International Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Developing Your Own Sentinel Event Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Leadership, Culture, and Sentinel Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Early Response Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Event Investigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Onward with Root Cause Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

CHAPTER 3: Preparing for Root Cause Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Step 1  Organize a Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Step 2  Define the Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Step 3  Study the Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

CHAPTER 4: Determining Proximate Causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Step 4  Determine What Happened . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Step 5 Identify Contributing Process Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Step 6 Identify Other Contributing Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Step 7 Measure—Collect and Assess Data on Proximate and Underlying Causes . . . . . . . . . . . . . 71
Step 8 Design and Implement Immediate Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

iii
ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

CHAPTER 5: Identifying Root Causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Step 9  Identify Which Systems Are Involved—The Root Causes . . . . . . . . . . . . . . . . . . . . . . . . 81
Step 10 Prune the List of Root Causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Step 11 Confirm Root Causes and Consider Their Interrelationships . . . . . . . . . . . . . . . . . . . . . . 86

CHAPTER 6: Designing and Implementing an Action Plan for Improvement . . . . . . . 91

Step 12 Explore and Identify Risk Reduction Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Step 13  Formulate Improvement Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Step 14 Evaluate Proposed Improvement Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Step 15  Design Improvements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Step 16 Ensure Acceptability of the Action Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Step 17 Implement the Improvement Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Step 18 Develop Measures of Effectiveness and Ensure Their Success . . . . . . . . . . . . . . . . . . . . . . 113
Step 19 Evaluate Implementation of Improvement Efforts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Step 20  Take Additional Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Step 21  Communicate the Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

CHAPTER 7: Tools and Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

What Is Lean Six Sigma? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Affinity Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Brainstorming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Capability Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Change Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Change Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Check Sheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Control Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Failure Mode and Effects Analysis (FMEA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Fishbone Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Flowchart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Gantt Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Histogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Kaizen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Multivoting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Operational Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Pareto Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Relations Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Run Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Scatter Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
SIPOC Process Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
Stakeholder Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
Standard Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
Value Stream Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

iv
 Contents

CHAPTER 8: Root Cause Analysis Case Studies from the Field . . . . . . . . . . . . . . . . . 175

Root Cause Analysis of Serious Adverse Events Among Older Patients in the
Veterans Health Administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Using Root Cause Analysis and Form Redesign to Reduce Incorrect Ordering of HIV Tests . . . . 186
Counting Matters: Lessons from the Root Cause Analysis of a Retained Surgical Item . . . . . . . . . 194

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

v
ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

vi
 Introduction
Patient safety events can cause serious harm or death. adverse events in acute care hospital admissions range from
They affect anyone. To address and prevent these threats, 2.9% in the United States to 5.0%–10.0% in the United
health care organizations must dig deep to unearth the root Kingdom, 7.5% in Canada, 12.9% in New Zealand, and
cause(s) and develop solutions that address the problems 16.6% in Australia.2
from a systems perspective.
Although these reports and chart reviews illuminate the
Indeed, the very presence of patient safety events indicates problem, it is virtually impossible to know how many
a continuing paradox in contemporary health care. Despite patients suffer as a result of health care system failures;
remarkable advances in almost every field of health care, however, any single patient safety event is a cause for
the occurrence of errors, or failures—the term used increas- concern. These events can result in tragedy for individuals
ingly instead of errors—persists. When such failures harm served and their families, add costs to an already over-
patients, the results can be heartbreaking. Most failures and burdened health care system, adversely affect the public’s
sentinel events—that is, a patient safety event (not primarily perception of an organization, and lead to litigation. They
related to the natural course of the patient’s illness or can also deeply affect health care professionals who are
under­­lying condition) that reaches a patient and results in dedicated to the well-being of their patients.
death, permanent harm, or severe temporary harm—are the
result of system and process flaws. These flaws are often not Health care organizations, then, have no choice but to
immediately apparent and require investigation answer one key question: Why do these errors or failures
continue to occur?
The prevalence of patient safety events had been thrust into
the limelight with the watershed report To Err Is Human: To answer this, a comprehensive systematic analysis must
Building a Safer Health System, published in 2000 by the be done. The most commonly used form of comprehensive
Institute of Medicine (IOM). The IOM report, however, systematic analysis among Joint Commission–accredited
was just the tip of the iceberg. More reports followed, organizations is root cause analysis—a process for identi-
illustrating the need to improve the quality of care being fying the basic or causal factor(s) underlying variation in
delivered in the United States. For example, researchers performance, including the occurrence or possible occur-
at Johns Hopkins Children’s Center and the US Agency rence of a sentinel event—and all of its related tools. Root
for Healthcare Research and Quality reviewed 5.7 million cause analysis can be used to uncover the factors that lead
records of patients younger than 19 years of age from 27 to patient safety events and move organizations to deliver
states who were hospitalized in 2000. Of the 52,000 chil- safer care.
dren identified by the researchers as being harmed by unsafe
medical care during their hospital stays, 4,483 suffered a Although health care organizations in the United States
fatal injury.1 often use root cause analysis to help improve quality en
route to accreditation, such analysis has many broader
Quality-of-care issues such as these are a problem for applications around the world. High-quality care is high-
­hospitals around the world. According to a 2007 report, quality care, whether it is delivered in New York City or
hospital chart reviews in various countries indicate that Dubai or Singapore. Organizations worldwide should

vii
 ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

consider how root cause analysis can be used to help procedure—in this case a plastic surgical clip that was
improve quality. left inside a patient’s skull.7
• In May 2015, the United Kingdom’s National Health
The Current Health Care Environment Service paid £15,000 in damages to the mother of an
Health care continues to experience dramatic change. infant who died in utero due to a medical error.8
Health care organizations are evolving constantly because
of changes in reimbursement, new technology, regulatory The above examples are only a few of the serious patient
requirements, and staffing levels. These modifications cause safety events that have attracted media attention in recent
policies and procedures to change often and, in most cases, years. These events cast a shadow on the public’s trust of
quickly. As health care organizations become more complex, health care. Stakeholders, including patients, justifiably
their systems and processes are increasingly interdependent. ask, “What’s going on?” Failure detection, reduction, and
This interdependence increases the risk of failures and can prevention strategies are receiving new impetus as the
make the recovery from failure more difficult. Clinical and health care community recognizes the value of a proactive
support staff workloads are growing heavier, resulting in approach to reducing risk.
greater stress and fatigue for many health care professionals.
Caregivers are working in new settings and performing new Root cause analysis is one such approach. Historically used
functions, sometimes with minimal training. Consequently, to investigate sentinel events, root cause analysis shows great
maintaining consistency in processes and systems is chal- promise as a proactive tool. Increasingly, health care organi-
lenging, leading to variation. Often, this variation results in zations are using this methodology to investigate close calls
increased risk to patients. (or near misses), no-harm patient safety events, and other
signals of risk. Health care organizations no longer have to
Media reports about patient safety events are occurring with wait until after a sentinel event occurs to perform a root
increasing regularity, including the following examples: cause analysis.
• In September 2013, researchers estimated that the
number of premature deaths associated with preventable When an adverse outcome, a sentinel event, or a cluster
harm to patients in US hospitals was more than 400,000 of less serious incidents or near misses occurs, organiza-
per year. This makes patient safety events the third tions must develop an understanding of the contributing
leading cause of death in the United States. Incidents factors and the interrelationship of those factors. Next, the
resulting in serious patient harm were estimated to be organization must implement an action plan to fortify its
10- to 20-times more common than lethal harm.3 systems against vulnerabilities with the potential to impact
• In November 2014, the journal Pediatrics reported that patients. Resilience is the degree to which a system contin-
an annual average of 63,358 medication errors occur uously prevents, detects, mitigates, or ameliorates hazards
in children younger than age 6 in the United States in or incidents.9
nonhospital settings and that 25% of those errors are in
infants, younger than 12 months old. This means that Purpose of This Book
a medication error affecting a child in the United States Root Cause Analysis in Health Care: Tools and Techniques,
occurs every eight minutes.4 Fifth Edition, is intended to help health care organizations
• In February 2015, the state of Minnesota reported that prevent systems failures by using root cause analysis to do
98 patients in that state were seriously injured, and the following:
another 13 patients died, as a result of patient safety • Identify causes and contributing factors of a sentinel
events during 2014.5 event or a cluster of incidents
• In May 2015, the Jordanian Ministry of Health began • Identify system vulnerabilities that could lead to
investigating an alleged medical error that resulted in a patient harm
Saudi patient becoming comatose.6 • Implement risk reduction strategies that decrease the
• In May 2015, the State of California fined a hospital likelihood of a recurrence of the event or incidents
$100,000 after the unintended retention of a foreign • Determine effective and efficient ways of measuring and
object in a patient’s body following an invasive improving performance

viii
 Introduction

Root cause analysis is an effective technique most commonly to meet the unique needs of the team and organization.
used after an error has occurred to identify underlying causes. Appropriate tools for use in each stage of root cause anal-
Failure mode and effects analysis (FMEA) is a proactive ysis are identified in each chapter. A chapter-by-chapter
technique used to prevent process and product problems ­description of the contents follows.
before they occur.10 Health care ­organizations should learn
both techniques to reduce the likelihood of adverse events. Chapter 1, “Root Cause Analysis: An Overview,” takes a
holistic look at root cause analysis. It describes variation,
Root Cause Analysis in Health Care: Tools and Techniques, how proximate and root causes differ, when root cause anal-
Fifth Edition, provides health care organizations worldwide ysis can be conducted, and the benefits of root cause anal-
with up-to-date information on The Joint Commission’s ysis. One of the benefits involves effectively meeting Joint
Sentinel Event Policy and safety-related requirements. Commission and Joint Commission International require-
It also describes the Sentinel Event Policy of Joint ments that relate to the management of sentinel events. The
Commission International. The book includes examples chapter also provides guidelines on the characteristics of a
that guide the reader through application of root cause thorough and credible root cause analysis and action plan.
analysis to the investigation of specific types of sentinel
events, such as medication errors, suicide, treatment delay, Chapter 2, “Addressing Sentinel Events in Policy and Strategy,”
and elopement. For ease of access and use by root cause describes the types of adverse events occurring in health
analysis teams, practical checklists and worksheets are care. The Joint Commission’s Sentinel Event Policy and
offered at the end of each chapter. requirements are listed in full, including a description
of reportable and reviewable events. Joint Commission
This publication provides and explains The Joint International’s Sentinel Event Policy also is discussed. The
Commission’s framework for conducting a root cause chapter provides practical guidelines on how an organiza-
­analysis. It also helps organizations do the following: tion can develop its own sentinel event policy, including
• Identify the processes that could benefit from root cause the role that an organization’s culture and leadership play in
analysis risk reduction and prevention. It describes the need for root
• Conduct a thorough and credible root cause analysis cause analysis and provides practical guidance on the early
• Interpret analysis results steps involved in responding to an adverse or sentinel event.
• Develop and implement an action plan for improvement
• Assess the effectiveness of risk reduction efforts Chapter 3, “Preparing for Root Cause Analysis,” covers the
• Integrate root cause analysis with other programs early steps involved in performing a root cause analysis. The
first of four hands-on workbook chapters, it describes how
Even without the occurrence of an adverse event, health to organize a root cause analysis team, define the problem,
care organizations should embrace the use of root cause and gather the information and measurement data to study
analysis to minimize the possibility of patient safety events the problem. Details are provided about team composi-
and thereby to improve the care, treatment, and services tion and ground rules. The chapter also covers how to use
provided at their facilities. information gleaned from The Joint Commission’s Sentinel
Event Database and accreditation requirements to identify
Overview of Contents problem areas in need of root cause analysis. The chapter
Root Cause Analysis in Health Care: Tools and Techniques, provides guidance on recording information obtained
Fifth Edition, provides health care organizations with during a root cause analysis, conducting interviews, and
­practical, how-to information on conducting a root cause gathering physical and documentary evidence.
analysis. Twenty-one steps are described (in Chapters 3
through 6). Teams conducting a root cause analysis Chapter 4, “Determining Proximate Causes,” provides
might not follow these steps in a sequential order. Often, practical guidance on the next stage of root cause analysis—
numerous steps will occur simultaneously, or the team will determining what happened and the reasons it happened.
return to earlier steps before proceeding to the next step. Organized in a workbook format, the chapter describe s
It is crucial for teams to customize or adapt the process how to further define the event, identify process problems,

ix
 ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

determine which care processes are involved with the Chapter 8, “Root Cause Analysis Case Studies from the Field,”
problem, and pinpoint the human, process, equipment, presents root cause analyses that resulted from real-life inci-
environmental, and other factors closest to the problem. dents at health care organizations. In these studies, the tools
The chapter also addresses how to collect and assess data on and techniques used to dig down to the root causes of the
proximate and underlying causes. In addition, the chapter events are identified and explained.
describes the process of designing and implementing
interim changes. Finally, the Glossary provides definitions of key terms used
throughout the book.
Chapter 5, “Identifying Root Causes,” provides practical
guidance, through workbook questions, on identifying A Word About Terminology
or uncovering the root causes—the systems that underlie The terms patient, individual served, and care recipient all
sentinel events—and the interrelationship of the root causes describe the individual, client, consumer, or resident who
to one another and to other health care processes. Systems actually receives health care, treatment, and/or services. The
are explored and described, including human resources, term care includes care, treatment, services, rehabilitation,
information management, environment of care, leader- habilitation, or other programs instituted by an organiza-
ship, communication, and uncontrollable factors. The tion for individuals served.
chapter also addresses how to differentiate root causes and
­contributing factors. Acknowledgments
Joint Commission Resources is grateful to the many
Chapter 6, “Designing and Implementing an Action Plan content experts who have contributed to this publication,
for Improvement,” includes practical guidelines on how to including the following:
design and implement an action plan—the improvement • Lisa Buczkowski, RN, MS, associate director, Office of
portion of a root cause analysis. During this stage, an orga- Quality and Patient Safety, The Joint Commission
nization identifies risk reduction strategies and designs and • Maureen Carr, RN, project director, Division of
implements improvement strategies to address underlying Standards and Survey Methods, The Joint Commission
systems problems. Characteristics of an acceptable action • Gerard Castro, PhD, MPH, project director, Office of
plan are provided, as is information on how to assess the Patient Safety, The Joint Commission
effectiveness of improvement efforts. The chapter concludes • Doreen Finn, RN, associate director, Standards
with information on how to effectively communicate the Interpretation Group, The Joint Commission
results in improvement initiatives. • Paul van Ostenberg, senior executive director,
International Accreditation and Standards, Joint
Chapter 7, “Tools and Techniques,” presents the tools and Commission International
techniques used during root cause analysis. Each tool • Ronald Wyatt, MD, medical director, Office of Quality
profile addresses the purpose of the tool, the appropriate and Patient Safety, The Joint Commisison
stage(s) of root cause analysis for the tool’s use, simple steps
for success, and tips for effective use. Twenty-three tools In addition, JCR would like to thank writer James Foster
are profiled: affinity diagrams, brainstorming, capability for his help in revising this book.
charts, change analysis, change management, check sheets,
control charts, failure mode and effects analysis, fishbone
diagrams, flowcharts, Gantt charts, histograms, run charts,
scatter diagrams, SIPOC process maps, stakeholder analysis,
and other tools. Preceding the tool descriptions is a discus-
sion of a performance improvement methodology, Lean Six
Sigma, that incorporates many of these tools.

x
 Introduction

7. San Jose Mercury News. State Fines Marin General Hospital

References $100,000 for Leaving Object in Patient’s Skull. Halstead R.
1. Miller MR, Zhan C. Pediatric patient safety in hospitals: A
May 8, 2015. Accessed Oct 14, 2015. http://www.mercurynews
national picture in 2000. Pediatrics.2004 Jun;113(6):1741–1746.
.com/crime-courts/ci_28163048state-fines-marin-general
2. Peters GA, Peters BJ. Medical Error and Patient Safety: Human -hospital-100-000-leaving.
Factors in Medicine. Boca Raton, FL: CRC Press, 2007.
8. STV Group. Health Board Apologises After Unborn Baby Dies
3. James JT. A new evidence-based estimate of patient harms Following Doctor Error. May 19, 2015. Accessed Oct 14, 2015.
associated with hospital care. J Patient Saf. 2013 Sep;9(3):122–128. http://news.stv.tv/stirling-central/1320942-forth-valley-hospital
4. Smith MD, et al. Out-of-hospital medication errors among -apology-after-sarah-mackinlay-of-falkirk-lost-baby/.
young children in the United States, 2002–2012. Pediatrics. 2014 9. World Health Organization (WHO). More Than Words:
Nov;134(11):867–876. Conceptual Framework for the International Classification for
5. Minnesota Department of Health (MDH). Adverse Health Events Patient Safety: Final Technical Report, ver. 1.1. Geneva: WHO,
in Minnesota: 11th Annual Public Report. St. Paul: MDH: 2015. 2009. Accessed Oct 14, 2015. http://www.who.int/patientsafety
Accessed Oct 14, 2015. http://www.health.state.mn.us /taxonomy/icps_full_report.pdf.
/patientsafety/ae/2015ahereport.pdf. 10. The Joint Commission. Failure Mode and Effects Analysis in
6. Arab News. Saudi Brain Dead After ‘Medical Error’ in Jordan Health Care: Proactive Risk Reduction, 3rd ed. Oak Brook, IL:
Hospital. May 8, 2015. Accessed Oct 14, 2015. http://www Joint Commission Resources, 2010.
.arabnews.com/saudi-arabia/news/743946.

xi
ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

xii
 CHAPTER

1
Root Cause Analysis: An Overview

Learning Objectives
• Understand the need for comprehensive systematic analysis of sentinel events an other adverse outcomes

• Learn the basics of root cause analysis (RCA), the most common method of comprehensive systematic analysis

• Know how RCA and action plans relate to The Joint Commission’s Sentinel Event Policy

Investigating Patient Safety Events: sentinel event as an unanticipated occurrence involving

The Need for Comprehensive death or major permanent loss of function unrelated to the
Systematic Analysis natural course of the patient’s illness or underlying condi-
The Joint Commission’s Sentinel Event Policy requires tion. A root cause analysis focuses primarily on systems and
accredited health care organizations to conduct a compre- processes, not individual performance. To be successful,
hensive systematic analysis in the wake of a sentinel event. the objective of an RCA must not be to assign individual
Comprehensive systematic analysis seeks to go beyond blame. Rather, through the RCA determine process, a team
individual performance issues to determine how gaps works to understand a process or processes, the causes or
in policies and safety systems may have contributed to potential causes of variation that can lead to error, and
an adverse event and to identify changes to policies and identify process changes that would make variation less
procedures that may prevent similar events from occurring likely to recur.
in the future. The Joint Commission reviews methods of
comprehensive systematic analysis on a case-by-case basis to A root cause is the most fundamental reason (or one of
determine their credibility, thoroughness, and acceptability. several fundamental reasons) a failure, or a situation
The Joint Commission also provides advice and resources in which performance does not meet expectations, has
to institutions to assist them in assessing analytical tools. occurred. In common usage, the word cause suggests
However, RCA is by far the most common method and is responsibility or a factor to blame for a problem. In the
the method preferred by The Joint Commission. context of an RCA, however, the use of the word cause does
not imply an assignment of blame. Instead, the cause refers
What Is Root Cause Analysis? to a relationship or potential relationship between certain
Root cause analysis is a process for identifying the basic factors that enable a sentinel event to occur. The focus in
or causal factor(s) underlying variation in performance.
* Severe temporary harm is critical, potentially life-threatening harm
Variation in performance can (and often does) produce lasting for a limited time with no permanent residual, but requires
unexpected and undesired adverse outcomes, including the transfer to a higher level of care/monitoring for a prolonged period of
occurrence or risk of a sentinel event. The Joint Commission time, transfer to a higher level of care for a life-threatening condition,
or additional major surgery, procedure, or treatment to resolve the
defines sentinel event as a patient safety event (not condition. Adapted from Healthcare Performance Improvement, LLC.
primarily related to the natural course of the patient’s illness The HPI SEC & SSER Patient Safety Measurement System for Healthcare,
or underlying condition) that reaches a patient and results rev. 2. Throop C, Stockmeier C. May 2011. Accessed Oct 15, 2015.
http://hpiresults.com/publications/HPI%20White%20Paper%20-%20
in death, permanent harm, or severe temporary harm.* SEC%20&%20SSER%20Measurement%20System%20REV%202%20
Similarly, Joint Commission International (JCI) defines MAY%202011.pdf.

1
 ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

an RCA is on a positive, preventive approach to system Sidebar 1-1.

and process changes following a sentinel event, a near-miss
Root Cause Analysis Case Example
sentinel event, or a cluster of less serious yet potentially
harmful incidents. A 16-year-old patient came to the hospital to deliver
her baby. During the process of her care, an infu-
As shown in Sidebar 1-1, right, root cause analysis can do sion intended exclusively for the epidural route was
more than discover that “A caused B.” The process also connected to the patient’s peripheral intravenous line
can help an organization determine that “if we change A and infused by pump. The patient experienced cardio-
because we had a problem with it, we can reduce the possi- vascular collapse. A cesarean section resulted in the
bility of B recurring or in fact prevent B from occurring in delivery of a healthy infant, but the medical team was
unable to resuscitate the mother.
the first place.”
The media attention surrounding the error accelerated
RCA is a powerful and useful tool that can help health care through the national provider and safety community
organizations around the world reduce errors and move when the nurse was charged with a criminal offense.
These events set in motion intense internal and
quality efforts forward. However, health care organization
external scrutiny of the hospital’s medication and safety
leaders must realize that RCA is not a panacea but one procedures.
tool that should be used in conjunction with others to
improve care. To further understanding about latent systems gaps
and process failure modes, a root cause analysis of the
event was conducted. A team conducted a one-week
When Can a Root Cause Analysis evaluation of the medication use system and the orga-
Be Performed? nization’s current environment, systems and processes,
Historically, root cause analysis has most commonly been staffing patterns, leadership, and culture to help shape
used retrospectively—to probe the reasons for a bad outcome the recommended improvements. For each of the four
or for failures that have already occurred. Root cause anal- proximate causes of the event, performance-shaping
ysis can also be used to probe a near-miss event or pattern factors were identified.
of events or as part of other performance improvement Although the hospital’s organizational learning was
redesign initiatives, such as gaining an understanding of painful, this event offered an opportunity for increasing
variations observed in systematically collected data. The organizational competency and capacity for designing
best RCAs look at the entire process and all support systems and implementing patient safety. Structures and
involved in a specific event to minimize overall risk associ- processes, including safety nets and fail-safe mecha-
nisms, were implemented to promote safer behavioral
ated with that process, as well as the recurrence of the event
choices for providers.
that prompted the root cause analysis.1 The goal of the root
cause analysis is to produce an action plan that identifies the The hospital took a number of clinical steps to improve
strategies the organization intends to implement to reduce the safety of medication administration, including
removing the barriers to scanning medication bar
the risk of similar events occurring in the future.
codes, implementing consistent scanning-compliance
tracking, and providing teamwork training for all nursing
Root cause analysis is also used increasingly by organi- and physician staff practicing in the birth suites.
zations as one step of a proactive risk reduction effort
Source: Smetzer J, et al. Shaping systems for better behavioral
using failure mode and effects analysis (FMEA). FMEA is a choices: Lessons learned from a fatal medication error. Jt Comm J Qual
proactive, prospective approach used to prevent process and Patient Saf. 2010 Apr;36(4):152–163.

product problems before they occur. It provides a look not

only at what problems could occur—the failure modes— Figure 1-1, page 3, shows a frequency/severity matrix one
but also at how severe the effects of those problems could organization uses to help decide when to conduct a root
be. The goal is to prevent poor results, which in health care cause analysis as part of an FMEA. In this matrix, if a
means harm to patients. One step of FMEA involves iden- process failure gets a high score both for severity (or poten-
tifying the root causes of the failure modes.2 At this point in tial severity) of outcome and for frequency, then a root cause
the process, the FMEA team can use the RCA approach. analysis should be done.

2
 CHAPTER 1  |  Root Cause Analysis: An Overview

RCA in High Reliability Industries and workers are trained accordingly to anticipate, recognize,
Industries that are regarded as highly reliable, such as and either avoid or quickly recover from errors. In contrast,
nuclear power or the aerospace industry, also employ RCA sentinel events in the health care environment occur with
methodology to investigate adverse events. In the nuclear relative frequency and tend to be handled reactively.
power and aerospace industries, sentinel events are rare
because they have been anticipated. These high reliability Domestic and International Requirements
industries have adopted a systems approach, in which errors Both The Joint Commission, which accredits health care
are viewed as an expected part of the workplace, the result organizations in the United States, and Joint Commission
of a chance misalignment of weaknesses in the underlying International (JCI), which accredits health care organi-
system.3 (Imagine a stack of slices of Swiss cheese. Each slice zations in countries other than the United States, have
has holes in different places, thus only if a hole in each slice a Sentinel Event Policy and standards related to sentinel
aligned perfectly with a hole in all the other slices could an events. For example, JCI Quality Improvement and Patient
object pass through the entire stack. The Swiss cheese model Safety (QPS) standards require each accredited organization
shown in Figure 1-2, page 4, represents how an error could to establish which unanticipated events are significant and
possibly penetrate multiple layers of barriers, defenses, and the process for their intense analysis.
safeguards in a system.)
While the determination of what constitutes a signifi-
Consequently, systems, often with significant redundancies, cant event must be consistent with the general definition
have been built to protect against the occurrence of errors, of sentinel event as described in JCI’s policy, accredited

Figure 1-1. Frequency/Severity Matrix for Prioritizing Safety-Related Problems

This matrix helps the organization apply its resources (such as time) to areas where the opportunity to improve
safety is greatest. The larger the number, the more urgent the problem and the more useful a root cause analysis
would be.
Source: US Department of Veterans Affairs National Center for Patient Safety. VHA National Patient Safety Improvement Handbook. Mar 4, 2011, page
B-2. Accessed Oct 15, 2015. http://www.va.gov/vhapublications/ViewPublication.asp?pub_ID=2389.

3
 ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

organizations have some latitude in setting more-specific Both JCI– and Joint Commission–accredited organizations
parameters to define unanticipated and major permanent loss are expected to identify and respond appropriately to all
of ­function. At a minimum, an organization must include sentinel events that occur in the organization or that are
those events that are subject to review under JCI standards, associated with services that the organization provides or
such as unanticipated death related to the natural course of provides for. Appropriate response includes conducting a
the patient’s illness or underlying condition; major perma- timely, thorough, and credible comprehensive systematic
nent loss of function unrelated to the natural course of the analysis; developing an action plan designed to implement
patient’s illness or underlying condition; or wrong-site, improvements to reduce risk; implementing the improve-
wrong-procedure, wrong-patient surgery. ments; and monitoring the effectiveness of those improve-
ments. (See Sidebar 1-2, page 5, for a discussion of
For organizations based in the United States, standards multiple events.)
require the organization to have an organizationwide,
integrated patient safety program within its performance Variation and the Difference Between
improvement activities. Proximate and Root Causes
Whether addressing a sentinel event or a cluster of less
Particularly, thorough and credible comprehensive serious low-harm or near-miss events, root cause analysis in
systematic analyses (for example, root cause analyses) in all environments provides two challenges:
response to sentinel events they are required to conduct. 1. To understand why the event occurred
RCA is the most commonly used form of comprehensive 2. To prevent the same or a similar event from occurring
systematic analysis used by Joint Commission–accredited in the future through prospective process design
­organizations to comply with this requirement. or redesign

Figure 1-2. Swiss Cheese Model

The Swiss cheese model shows how an error could penetrate multiple layers of defenses, barriers, and safeguards
in a system.
Source: Reason J.: The Human Contribution: Unsafe Acts, Accidents and Heroic Recoveries. Farnham, England: Ashgate Publishing, 2008.
© Managing the Risks of Organizational Accidents, by James Reason, 1997, Ashgate. Reproduced by permission.

4
 CHAPTER 1  |  Root Cause Analysis: An Overview

Sidebar 1-2. those events or possible events with a significant negative

or potentially negative impact on the patient. The criterion
Investigating Multiple Patient
for a sentinel event is generally death, permanent harm, or
Safety Events severe temporary harm to the patient.
Although root cause analysis (RCA) is associated
frequently with the investigation of a single sentinel
Adverse or sentinel events involve unexpected variation in
event, the methodology also can be used to determine a process. When this variation occurs, the probability of
the cause of multiple occurrences of low-harm events. a serious adverse outcome increases. As mentioned previ-
For example, 40 Danish community pharmacies worked ously, RCA is a process for identifying the basic or causal
together, using RCA, to gain insight into medication factor(s) underlying variation in performance. Variation is
errors, a problem that can result in serious conse- a change in the form, position, state, or qualities of a thing.
quences for patients. Although a sentinel event is the result of an unexpected
The root cause analysis included investigation of variation in a process, variation is inherent in every process.
401 errors, many of which had potential clinical signif- To reduce variation, it is necessary to determine its cause.
icance, even though each error did not necessarily What’s more, variation can be classified by what caused it.
result in harm. Analyzed as a cluster, however, the RCA
resulted in the identification of four common medication
error causes1:

1. Illegible handwritten prescriptions Figure 1-3. The Relationships Between a

2. Similarities in packaging or names, or strength and Sentinel Event and Its Causes
dosage stated in misleading ways
3. Lack of effective control of prescription label and
medicine
4. Lack of concentration caused by interruptions.
Reference:
1. Knudsen P, et al. Preventing medication errors in community
pharmacy: Root-cause analysis of transcription errors. Qual Saf
Health Care. 2007 Aug;16(4):285–290.

To meet these challenges, organizations must under-

stand not only the proximate causes, or active failures (the
apparent, seemingly obvious reasons an error occurred) but
also the underlying causes, or latent conditions (the aspects
of a process than can allow an error to occur), and the
interrelationship of the two. As shown in Figure 1-3, right,
active failures are only the “tip of the iceberg”—that which
is visible or proximal to the patient—while latent condi-
tions lurk unseen “underwater,” posing a hidden poten-
tial danger. Root cause analysis helps organizations delve
beneath the proximate causes to find the underlying causes
of a sentinel event.

Conducting an RCA has significant resource implications.

Active failures, shown as the “tip of the iceberg,”
A team approach, involving a full range of disciplines and
reveal only the proximate causes of a sentinel
departments in the process being studied, is mandatory, as event; latent conditions loom below the waterline,
will be described in Chapter 3. Organizations therefore will representing the underlying causes.
want to conduct root cause analysis principally to explore

5
 ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

Understand Common-Cause Variation Special causes in one process are usually the result of
Common-cause variation, although inherent in every common causes in a larger system of which the process is
process, is a consequence of the way a process is designed to a part. For example, mechanical breakdown of a piece of
work. For example, an organization is examining the length equipment used during surgery may indicate a problem
of time required by the emergency department to obtain a with an organization’s preventive maintenance activities.
routine radiology report. The time may vary depending on
how busy the radiology service is or by when the report is Understand the Relationships Between
requested. On a particular day, the radiology department Common and Special Causes
may have received many concurrent requests for reports, In health care, all the clinical and organizational processes
making it difficult for the department to fill one specific and subprocesses associated with an event under review
request. Or the report may have been requested between need to be delineated and evaluated to identify the degree of
midnight and 6:00 a.m. when fewer radiology technologists common-cause and/or special-cause variation. This process
are on duty. Variation in the process of providing radiology will help organizations identify whether variation is due to
reports is inherent, resulting from common causes such as clinical processes or organizational processes or both.
staffing levels and emergency department census.
Any variation in performance, including a sentinel event,
A process that varies only because of common causes is said may be the result of a common cause, a special cause, or
to be stable. The level of performance of a stable process both. In the case of a sentinel event, the direct or proximate
or the range of the common-cause variation in the process special cause could be uncontrollable factors. For example,
can be changed only by redesigning the process. Common- a patient death results from a hospital’s total loss of elec-
cause variation is systemic and endogenous (that is, produced trical power during a storm. This adverse outcome is clearly
from within). The organization needs to determine whether the result of a special cause in the operating room that is
the amount of common-cause variation will be tolerated. uncontrollable by the operating room staff. Staff members
may be able to do little to prevent a future power outage
Special-Cause Variation and more deaths. However, the power outage and resulting
Special-cause variation arises from unusual circumstances or death can also be viewed as the result of a common cause in
events that may be difficult to anticipate and may result in the organization’s system for preparing for and responding
marked variation and an unstable, intermittent, and unpre- to a utility failure and other emergencies. Perhaps the
dictable process. Special-cause variation is not inherently backup generator that failed was located in the basement,
present in systems. It is exogenous (that is, produced from which flooded during the storm, and the organization had
the outside), resulting from factors that are not part of the no contingency plan for such a situation.
system as designed. Mechanical malfunctions, fatigued
employees, and natural disasters such as floods, hurri- When looking at the chain of causation, proximate or
canes, and earthquakes are examples of special causes that direct causes tend to be nearest to the origin of the event.
result in variation. Organizations should strive to identify, For example, proximate causes of a medication error may
mitigate, and/or eliminate special causes wherever possible. include an outdated drug, product mislabeling or misiden-
However, removing a special cause eliminates only that tification, or an improper administration technique. By
current abnormal performance in the process. It does not contrast, root causes are systemic and appear far from the
prevent the same special cause from recurring. For example, origin of the event, often at the foundation of the processes
firing an overly fatigued employee who was involved in a involved in the event. For example, root causes of a medi-
medication error does little to prevent the recurrence of cation error might include manufacturer’s production or
the same error. Instead, organizations should investigate, labeling of two different types or strengths of drugs so that
understand, and address underlying common causes within one looks too much like the other, storage setup that places
their systems and processes such as staffing arrangements, different dosages of the same medication too close together,
employee education, complacency, information manage- an inadequate medication procurement process, communi-
ment, and communication. cation problems, or any number of system, issues that set
people up to make a mistake.

6
 CHAPTER 1  |  Root Cause Analysis: An Overview

Most root causes alone are not sufficient to cause a failure; Maximizing the Value of Root Cause Analysis
rather, the combination of root cause(s) and other contrib- Root cause analysis is designed to answer the following
uting factors sets the stage. For example, flaws in the process three questions:
for communicating changes in the condition of a patient, 1. What happened?
a poorly designed emergency call system, and an inadequate 2. Why did it happen?
assessment process can be root causes of a patient’s fall from 3. What can be done to prevent it from happening again?
bed. Organizations that are successful in effectively identi-
fying all the root causes and their interactions can eliminate The problem, however, is that health care organizations
a plethora of risks when redesigning processes.4 Elimination frequently use root cause analysis to answer these questions
of one root cause reduces the likelihood of that one specific but never determine whether the risk of recurrence of an
adverse outcome occurring again. However, if the organi- adverse event has actually been reduced. Therefore, some
zation misses two or three other root causes, it is possible health care organizations may dedicate resources to root
that they could interact to cause a different but equally cause analysis without knowing whether the investment
adverse outcome. has any payoff. To make root cause analysis more useful,
follow-up activities that measure the implementation of
process changes and improvements in patient outcomes
TIP should become a standardized component of the process.5
Organizations must steer clear of the “myth of one
root cause” and expect to find several possibilities. To ensure that the RCA yields improvement, health care
leaders must address fundamental challenges that can limit
the value of the incident investigation. Consider the strate-
Benefits of Root Cause Analysis gies listed in Sidebar 1-3, page 8.
All health care organizations experience problems of
varying persistence and magnitude. Organizations can The outcome of the root cause analysis is an action plan
improve the efficiency and effectiveness of their operations that the organization intends to implement in order to
and the quality and safety of care through addressing the reduce the risk of similar events occurring in the future.
roots of such problems. Individual accountability for faulty The plan should address responsibility for implementa-
performance should not be the focus of a root cause anal- tion, oversight, pilot testing as appropriate, time lines, and
ysis. (See Chapter 3 for additional discussion of individual ­strategies for measuring the effectiveness of actions.
accountability.) If a question arises regarding whether
an individual acted appropriately, it should be addressed The Root Cause Analysis and Action Plan:
through the organization’s employee or physician perfor- Doing It Right
mance management system. For the purpose of an RCA, How can an organization ensure that its RCA and action
the focus should be on systems—how to improve systems to plan represent an appropriate response to a particular
prevent the occurrence of sentinel events or problems. This sentinel event? The Joint Commission and JCI provide
approach involves digging into the organization’s systems to guidance to organizations conducting RCAs in their respec-
find new ways to do things. Root cause analysis helps orga- tive sentinel event policies. These policies provide criteria
nizations identify risk or weak points in processes, under- that organizations can use to evaluate their RCA for accept-
lying or systemic causes, and corrective actions. Moreover, ability, thoroughness, and credibility. Organizations can use
information from RCAs shared between and among organi- the tool in Figure 1-4, pages 9–10, to review their RCAs
zations can help prevent future sentinel events. Knowledge based on these criteria.
shared in the health care field can contribute to proactive
improvement efforts and yield results across the health care Crafting an Acceptable Action Plan
delivery system. The Joint Commission and JCI also provide criteria for an
acceptable action plan within their sentinel event policies.

7
 ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

Sidebar 1-3. reduction activities. This information may benefit organiza-

tions that are developing their own action plans. Data gath-
Strategies for an Effective RCA
ered by The Joint Commission between January 1995 and
To ensure the effectiveness of their root cause analyses
December 2014 from review of more than 8,876 sentinel
(RCAs), organizations and their leaders should imple- events indicate that nearly 90% of these events fall into the
ment the following strategies: following categories:
• Anesthesia-Related Event
►► Find and resolve latent conditions as well as root
causes. Finding the cause of a problem is a good
• Criminal Event
start, but to be effective, identify the latent condi- • Delay in Treatment
tions that allowed the problem to materialize in the • Dialysis-Related Event
first place. • Elopement
►► Treat the cause rather than try to change people.
• Fall
Many RCA actions are aimed at changing the • Fire
behavior of staff members. Such interventions are • Infant Abduction
rarely effective. Although one person might change • Infant Discharge to Wrong Family
his or her behavior, another person might still • Infection-Related Event
do the same thing as before because the under- • Inpatient Drug Overdose
lying problem with the system has not changed. • Maternal Death
Leaders must understand that processes need to be
• Medical Equipment–Related
changed—and, therefore, more effective systems-
based solutions should be developed.
• Medication Error
• Op/Post-Op Complication
►► Follow through to ensure change. Often health
• Perinatal Death/Injury
care organization leaders initiate an improvement
• Radiation Overdose
program based on the findings of a root cause anal-
ysis, but the initiative loses steam. It is important for
• Restraint-Related Event
leaders to make sure that improvement programs • Self-Inflicted Injury
are fully implemented. • Severe Neonatal Hyperbilirubinemia
• Suicide
Source: The Joint Commission. Patient safety systems. In 2015 • Transfer-Related Event
Compre­hen­sive Accreditation Manual for Hospitals. Oak Brook, IL: 2015.
• Transfusion Error
• Unassigned
According to these policies, an action plan will be consid- • Unintended Retention of a Foreign Body
ered acceptable if it does the following: • Utility System Failure
• Identifies changes that can be implemented to reduce • Ventilator Death
risk or formulates a rationale for not undertaking • Wrong-Patient, Wrong-Site, Wrong-Procedure
such  changes
• Identifies, in situations in which improvement actions An organization experiencing a sentinel event in one of
are planned, including the following: these categories is expected to conduct a thorough and
–– Who is responsible for implementation ­credible root cause analysis, which, at a minimum, investi-
–– When the action will be implemented gates each of the areas identified for that category of event.
(including any pilot testing) This inquiry should determine that there is, or is not,
–– How the effectiveness of the actions will be evaluated opportunity with the associated system(s), process(es), or
–– The point at which alternative actions will be function(s) to redesign or otherwise take action to reduce
considered if improvement targets are not met risk. A root cause analysis submitted in response to a
sentinel event in one of the listed categories is considered
Review of the root cause analyses of sentinel events has unacceptable if it does not, at a minimum, address each of
allowed The Joint Commission to identify patterns for risk the areas specified for that type of event.

8
 CHAPTER 1  |  Root Cause Analysis: An Overview

Figure 1-4. Root Cause Analysis Evaluation Checklist

Root Cause Analysis Evaluation Checklist

Date of Evaluation: Date of Event:

Completed by:
Brief Description of Incident:

Root Cause Analysis Team Participants:

Evaluation Level 1: Acceptability

Follow-Up Action Follow-Up Action
Criteria Criteria Met? (Y/N) Required? Completed (Date)
Progresses from special causes in clinical
processes to common causes in organiza-
tional processes.
Repeatedly digs deeper by asking “Why?”
and then, when answered, asks “Why?”
again, until it no longer makes sense to
ask again.
Identifies changes that could be made in
systems and processes (either through
redesign or development of new systems or
processes) that would reduce the risk of such
events occurring in the future.

Evaluation Level 2: Thoroughness

Follow-Up Action Follow-Up Action
Criteria Criteria Met? (Y/N) Required? Completed (Date)
Determines the factors most directly
­associated with the sentinel event and
the process(es) and systems related to its
­occurrence, including human factors
Analyzes the underlying systems and
pro­cesses through a series of “Why?”
­questions to determine where redesign
might reduce risk
Inquires into all areas appropriate to the
specific type of event
Identifies the risk points and their potential
contributions to this type of event

(continued)

9
 ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

Figure 1-4. Root Cause Analysis Evaluation Checklist (continued)

Evaluation Level 2: Thoroughness (continued)

Determines potential improvement in
processes or systems that would tend to
decrease the likelihood of such events in the
future, or a determination, after analysis, that
no such improvement opportunities exist

Evaluation Level 3: Credibility

Follow-Up Action Follow-Up Action
Criteria Criteria Met? (Y/N) Required? Completed (Date)
Includes participation by the patient safety
director and by individuals most closely
involved in the processes and systems
under review.
Is internally consistent (that is, the analysis
does not contradict itself or leave obvious
questions unanswered).
Provides an explanation for all findings of
“not applicable” or “no problem.”
Includes consideration of any relevant liter-
ature, guidelines, or evidence-based best
practices.

When conducting a root cause analysis, investigators must Orientation and Training of Staff
ask questions that meet the minimum scope of root cause • Does the hospital offer medication safety training?
analysis for specific types of sentinel events. The following • Did the nurse involved in the error participate in
are sample questions that might be used when investigating medication safety training?
a medication error.
Competency Assessment/Training
Patient Identification Process • Are nurses at the hospital required to demonstrate
• Are specific patient identification processes and protocols competency in medication administration?
in place? • Did the nurse who was involved in the error demonstrate
• Did the nurse verify the patient’s identity? medication administration competency?
• Was the patient identified by a bar-coded wristband or
any other means? Supervision of Staff
• Who was supervising the nurse who was involved
Staffing Levels in the error?
• What are the typical staffing levels on the unit? • How many other nurses was the supervisor
• How many staff members were working on the unit responsible for?
where the error occurred? • Does the supervisor specifically oversee the medication
• How many patients were assigned to the nurse who was administration process?
involved in the error?

10
 CHAPTER 1  |  Root Cause Analysis: An Overview

TIP Sidebar 1-4. Get Support

Use a WWW (Who, What, When): A WWW is a simple
An organization can seek clarification of any questions
tool that helps RCA teams clarify the actions that
about The Joint Commission’s Sentinel Event Policy
need to be taken, who is responsible for completing
and the requirements for a comprehensive systematic
or facilitating those actions, the time frames for when
analysis by visiting http://www.jointcommission.org,
those actions will be completed. The tool consists
which includes detailed information on the Sentinel
of a simple grid, such as in the example below. The
Event Policy and root cause analysis (see Figure 1-5,
first column lists the individual responsible for the
pages 13–16). Note that The Joint Commission will not
action, the second lists the action itself, and the third
give a determination of reviewability at this point but
indicates a due date.
can answer questions and provide support.

WHO WHAT WHEN The Joint Commission established the Sentinel Event
Hotline to respond to inquiries about the Sentinel Event
Policy. For more information visit http://www.joint
commission.org/SentinelEvents/PolicyandProcedures/.

Physical Environment
• Did any environmental factors make it difficult for the
nurse to properly carry out medication administration
duties?
Communication Among Staff Members • What environmental factors (for example, lighting,
• Are there established processes and protocols in place for space considerations) would make it easier for staff
nurses to communicate with physicians and pharmacists members to properly carry out the medication
about medication orders? administration process?
• Did all staff members involved properly follow the
communication protocols? Control of Medications: Storage/Access
• Were the medications in question stored in the accepted
Availability of Information manner?
• Does the hospital routinely supply information about • Were the medications accessed in the accepted manner?
medications?
• Did the staff members involved review all information Labeling of Medications
available to them? • Were the medications in question properly labeled?
• What processes or protocols are in place to verify that
Adequacy of Technological Support the label matched the medication?
• Are there any technologies in place to support the • Are there any protocols in place to ensure that “look-
medication administration process? alike” prescriptions are properly identified on the label?
• If “yes,” were these technologies properly used?
• If “no,” are there technologies available that would Sidebar 1-5, page 12, outlines the high-level key tasks
enhance the medication administration process? involved in performing a thorough and credible root cause
analysis and action plan. Overall, a thorough and credible
Equipment Maintenance/Management root cause analysis should do the following:
• Were all medication distribution systems (for example, • Be clear (understandable information)
medication cabinets) in working order? • Be accurate (validated information and data)
• How often are these systems maintained? • Be precise (objective information and data)

11
ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

Sidebar 1-5.
• Be relevant (focus on issues related or potentially related
Conducting a Root Cause Analysis and
to the sentinel event)
Implementing an Action Plan • Be complete (cover all causes and potential causes)
• Be systematic (methodically conducted)
1. Assign an interdisciplinary team to assess the
sentinel event.
• Possess depth (ask and answer all of the relevant
“Why” questions)
2. Establish a way to communicate progress to senior
leadership. • Possess breadth of scope (cover all possible systemic
3. Create a high-level work plan with target dates, factors wherever they occur)
responsibilities, and measurement strategies.
4. Define all the issues clearly. The Joint Commission’s framework and JCI’s framework for
5. Brainstorm all possible or potential contributing a root cause analysis and action plan are similar and appear
causes and their interrelationships. as Figure 1-5 at the end of this chapter. This framework,
6. Sort and analyze the cause list. to be used extensively in Chapters 3 through 6, provides a
7. For each cause, determine which process(es) and solid foundation for root cause analyses and action plans.
system(s) it is a part of and the interrelationship of The tool selection matrix, found in Chapter 7 as Table 7-1
causes. (page 135), can also be used as a guide to ensure that an
8. Determine whether the causes are special causes, organization considers and selects the most appropriate
common causes, or both. tools and techniques for root cause analysis.
9. Begin designing and implementing changes while
finishing the root cause analysis. References
10. Assess the progress periodically. 1. Croteau RJ, Schyve PM. Proactively error-proofing health care
11. Repeat activities as needed (for example, processes. In Spath PL, editor: Error Reduction in Health Care: A
brainstorming). Systems Approach to Improving Patient Safety. San Francisco: Jossey-
12. Be thorough and credible. Bass, 2000, 179–198.

13. Focus improvements on the larger system(s). 2. The Joint Commission: Failure Mode and Effects Analysis in Health
Care: Proactive Risk Reduction, 3rd ed. Oak Brook, IL: Joint
14. Redesign to eliminate the root cause(s) and the
Commission Resources, 2010.
interrelationship of root causes that can create an
adverse outcome. 3. Reason J. Human error: Models and management. BMJ. 2000
Mar 18;320(7237):768–770.
15. Measure and assess the new design.
4. Knudsen P, et al. Preventing medication errors in community
pharmacy: Root-cause analysis of transcription errors. Qual Saf
Health Care. 2007 Aug;16(4):285–290.
5. Wu AW, Lipshutz AK, Pronovost PJ. Effectiveness and
efficiency of root cause analysis in medicine. JAMA. 2008 Feb
13;299(6):685–687.

12
 Figure 1-5. A Framework for a Root Cause Analysis and Action Plan in Response to a Sentinel Event

A Framework for a Root Cause Analysis and Action Plan

in Response to a Sentinel Event
Page 1 of 4
Level of Analysis Questions Findings Root Ask Take
Cause? “Why?” Action
What happened? Sentinel Event What are the details of the
event? (Brief description)
When did the event occur?
(Date, day of week, time)
What area/service was
impacted?
Why did it The process or What are the steps in the
happen? activity in which process, as designed? (A
the event occurred flow diagram may be
helpful here.)
What were the What steps were involved
most proximate in (contributed to) the
factors? event?

13
(Typically “special Human factors What human factors were
cause” variation)
relevant to the outcome?

Equipment factors How did the equipment

performance affect the
outcome?
Controllable What factors directly
environmental affected the outcome?
factors
Uncontrollable Are they truly beyond the
external factors organization’s control?

Other Are there any other factors

that have directly
influenced this outcome?
What other areas or
services are impacted?
CHAPTER 1  |  Root Cause Analysis: An Overview

(continued)
 Figure 1-5. A Framework for a Root Cause Analysis and Action Plan in Response to a Sentinel Event (continued)

A Framework for a Root Cause Analysis and Action Plan

in Response to a Sentinel Event (continued)
Page 2 of 4
This template is provided as an aid in organizing the steps in a root cause analysis. Not all possibilities and questions will apply in every
case, and there may be others that will emerge in the course of the analysis. However, all possibilities and questions should be fully
considered in your quest for “root cause” and risk reduction.
As an aid to avoiding “loose ends,” the three columns on the right are provided to be checked off for later reference:
• Root Cause? should be answered “yes” or “no” for each finding. A root cause is typically a finding related to a process or system that has a
potential for redesign to reduce risk. If a particular finding that is relevant to the event is not a root cause, be sure that it is addressed later in
the analysis with a “Why?” question. Each finding that is identified as a root cause should be considered for an action and addressed in the
action plan.
• Ask “Why?” should be checked off whenever it is reasonable to ask why the particular finding occurred (or didn’t occur when it should have) –
in other words, to drill down further. Each item checked in this column should be addressed later in the analysis with a “Why?” question. It is
expected that any significant findings that are not identified as root causes themselves have “roots.”
• Take action? should be checked for any finding that can reasonably be considered for a risk reduction strategy. Each item checked in this
column should be addressed later in the action plan. It will be helpful to write the number of the associated Action Item on page 4 of this tool
in the Take Action? column for each of the findings that requires an action.

14
Level of Analysis Questions Findings Root Ask Take
Cause? “Why?” Action
Why did that Human Resources To what degree are staff
happen? What issues properly qualified and
systems and currently competent for
processes their responsibilities?
underlie those
proximate
factors?
(Common-cause How did actual staffing
variation here may
lead to special-cause
compare with ideal levels?
ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

variation in dependent
processes)
What are the plans for
dealing with contingencies
that would tend to reduce
effective staffing levels?
To what degree is staff
performance in the operant
process(es) addressed?

(continued)
 Figure 1-5. A Framework for a Root Cause Analysis and Action Plan in Response to a Sentinel Event (continued)

A Framework for a Root Cause Analysis and Action Plan

in Response to a Sentinel Event (continued)
Page 3 of 4
Level of Analysis Questions Findings Root Ask Take
Cause? “Why?” Action
How can orientation and
in-service training be
improved?
Information To what degree is all
management issues necessary information
available when needed?
Accurate? Complete?
Unambiguous?
To what degree is
communication among
participants adequate?
Environmental To what degree was the
management issues physical environment
appropriate for the pro-

15
cesses being carried out?
What systems are in place
to identify environmental
risks?
What emergency and
failure-mode responses have
been planned and tested?
Leadership issues: To what degree is the
- Corporate culture
culture conducive to risk
identification or reduction?
- Encouragement of What are the barriers to
communication
communication of
potential risk factors?
- Clear communication To what degree is the
of priorities
prevention of adverse
outcomes communicated
as a high priority? How?
Uncontrollable What can be done to protect
factors against the effects of these
uncontrollable factors?
CHAPTER 1  |  Root Cause Analysis: An Overview

(continued)
 Figure 1-5. A Framework for a Root Cause Analysis and Action Plan in Response to a Sentinel Event (continued)

A Framework for a Root Cause Analysis and Action Plan

in Response to a Sentinel Event (continued)
Page 4 of 4
Action Plan Risk Reduction Strategies Measures of Effectiveness
For each of the findings identified in the analysis Action Item #1:
as needing an action, indicate the planned action
expected, implementation date, and associated
measure of effectiveness. OR . . .

If after consideration of such a finding, a decision Action Item #2:

is made not to implement an associated risk
reduction strategy, indicate the rationale for not
taking action at this time.

Check to be sure that the selected measure will Action Item #3:
provide data that will permit assessment of the
effectiveness of the action.

Consider whether pilot testing of a planned Action Item #4:

improvement should be conducted.

16
Improvements to reduce risk should ultimately be Action Item #5:
implemented in all areas where applicable, not just
where the event occurred. Identify where the
improvements will be implemented.

Action Item #6:

Action Item #7:

ROOT CAUSE ANALYSIS IN HEALTH CARE: TOOLS AND TECHNIQUES, Fifth Edition

Action Item #8:

Cite any books or journal articles that were considered in developing this analysis and action plan:

This framework outlines several questions that may be used to probe for systems problems underlying problematic processes. In each area,
consider whether and how the factors can be improved, as well as the pros and cons of expending resources to make improvements.