Full Guideline PDF 183081997

The management of hip fracture in adults
Update information
January 2023: NICE's original guidance on hip fracture was published in 2011
and updated in 2017. The section on surgical procedures (see section 10.3) has
been updated by the 2023 update.
See the NICE website for the guideline recommendations and the evidence
reviews for the 2023 update. This document preserves evidence reviews and
committee discussions for areas of the guideline that were not updated in 2023 or
2017.
May 2017: NICE has made new recommendations on hip replacements for
patients with a displaced intracapsular fracture. In addition, a footnote has been
added to recommendation 4.2.6 on cemented implants to highlight
safety guidance. The recommendations in this guideline on pages 34, 37,
107, 108 and 109 that are marked with grey shading have been replaced.
These changes can be seen in the short version of the

guideline at https://www.nice.org.uk/guidance/cg124.
METHODS, EVIDENCE & GUIDANCE
Produced by the National Clinical Guideline Centre

3
Published by the National Clinical Guideline Centre at The Royal College of Physicians, 11 St Andrews
Place, Regent’s Park, London, NW11 4LE
First published 2011
© National Clinical Guideline Centre 2011
Apart from any fair dealing for the purposes of research or private study, criticism or review, as
permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be
reproduced, stored or transmitted in any form or by any means, without the prior written permission
of the publisher or, in the case of reprographic reproduction, in accordance with the terms of
licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction
outside the terms stated here should be sent to the publisher at the UK address printed on this page.
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence
of a specific statement, that such names are exempt from the relevant laws and regulations and
therefore for general use.
The rights of the National Clinical Guideline Centre to be identified as Author of this work have been
asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.
Citation: National Clinical Guideline Centre, (2011) [The Management of Hip Fracture in Adults].
London: National Clinical Guideline Centre. Available from: www.ncgc.ac.uk
Contents
CONTENTS ..................................................................................................................................................... 3
GUIDELINE DEVELOPMENT GROUP MEMBERS ......................................................................................................... 1
NCGC STAFF MEMBERS OF THE GUIDELINE DEVELOPMENT GROUP .............................................................................. 2
EXPERT ADVISORS ........................................................................................................................................... 2
ACKNOWLEDGEMENTS ..................................................................................................................................... 3
ACRONYMS AND ABBREVIATIONS ........................................................................................................................ 4
1 INTRODUCTION ...................................................................................................................................... 6
2 DEVELOPMENT OF THE GUIDELINE ............................................................................................................ 10
2.1 WHAT IS A NICE CLINICAL GUIDELINE? ....................................................................................................... 10
4 HIP FRACTURE
2.2 REMIT .................................................................................................................................................. 11

2.3 WHO DEVELOPED THIS GUIDELINE? ............................................................................................................ 11
2.4 WHAT THIS GUIDELINE COVERS.................................................................................................................. 12
2.5 WHAT THIS GUIDELINE DOES NOT COVER ..................................................................................................... 13
2.6 RELATIONSHIPS BETWEEN THE GUIDELINE AND OTHER NICE GUIDANCE ............................................................. 13
3 METHODS .......................................................................................................................................... 15
3.1 DEVELOPING THE REVIEW QUESTIONS AND OUTCOMES .................................................................................. 15
3.2 SEARCHING FOR EVIDENCE........................................................................................................................ 20
3.3 EVIDENCE OF EFFECTIVENESS..................................................................................................................... 21
3.4 EVIDENCE OF COST-EFFECTIVENESS ............................................................................................................ 27
3.5 DEVELOPING RECOMMENDATIONS ............................................................................................................. 30
3.6 VALIDATION PROCESS .............................................................................................................................. 31
3.7 UPDATING THE GUIDELINE ........................................................................................................................ 31
3.8 DISCLAIMER .......................................................................................................................................... 31
3.9 FUNDING .............................................................................................................................................. 31
4 GUIDELINE SUMMARY ........................................................................................................................... 32
4.1 MAP OF RECOMMENDATIONS ................................................................................................................... 32
AN ALGORITHM WILL BE ADDED BEFORE PUBLICATION.KEY PRIORITIES FOR IMPLEMENTATION ......................................... 32
KEY PRIORITIES FOR IMPLEMENTATION ................................................................................................................. 33
4.2 FULL LIST OF RECOMMENDATIONS.............................................................................................................. 35
4.3 RESEARCH RECOMMENDATIONS ................................................................................................................ 39
5 IMAGING OPTIONS IN OCCULT HIP FRACTURE............................................................................................... 45
5.1 INTRODUCTION ...................................................................................................................................... 45
5.2 REVIEW QUESTION .................................................................................................................................. 46
5.3 RADIOGRAPHS ....................................................................................................................................... 46
5.4 RADIONUCLIDE BONE SCAN (RNS) ............................................................................................................. 47
5.5 ULTRASOUND (US) ................................................................................................................................. 48
5.6 COMPUTED TOMOGRAPHY (CT) ................................................................................................................ 50
5.7 RESEARCH RECOMMENDATION ON IMAGING OPTIONS IN OCCULT HIP FRACTURE ................................................. 53
6 TIMING OF SURGERY ............................................................................................................................. 53
6.1 INTRODUCTION ...................................................................................................................................... 53
6.2 RESEARCH RECOMMENDATIONS ON TIMING OF SURGERY ................................................................................ 68
7 ANALGESIA ......................................................................................................................................... 68
7.1 INTRODUCTION ...................................................................................................................................... 68
7.2 SYSTEMIC ANALGESIA .............................................................................................................................. 69
7.3 NERVE BLOCKS COMPARED TO SYSTEMIC ANALGESIA ...................................................................................... 70
7.4 RESEARCH RECOMMENDATIONS ON ANALGESIA ............................................................................................ 81
8 REGIONAL (SPINAL OR EPIDURAL) VERSUS GENERAL ANAESTHESIA.................................................................... 82
8.1 INTRODUCTION ...................................................................................................................................... 82
8.2 REGIONAL VERSUS GENERAL ANAESTHESIA ................................................................................................... 82
8.3 RESEARCH RECOMMENDATION ON ANAESTHESIA .......................................................................................... 87
9 SURGEON SENIORITY ............................................................................................................................. 88
9.1 INTRODUCTION ...................................................................................................................................... 88
9.2 SURGEON SENIORITY ............................................................................................................................... 89
9.3 RECOMMENDATIONS AND LINK TO EVIDENCE ............................................................................................... 91
10 SURGICAL PROCEDURES ......................................................................................................................... 94
10.1 INTRODUCTION ................................................................................................................................. 94
10.2 SURGERY WITH REGARD TO EARLY MOBILISATION ..................................................................................... 94
10.3 DISPLACED INTRACAPSULAR FRACTURES ................................................................................................. 95
CONTENTS 5
10.4 USE OF CEMENT IN ARTHROPLASTY ...................................................................................................... 112

10.5 SURGICAL APPROACH TO HEMIARTHROPLASTY ....................................................................................... 121
10.6 EXTRACAPSULAR FRACTURE FIXATION................................................................................................... 124
11 MOBILISATION STRATEGIES .................................................................................................................. 133
11.1 INTRODUCTION ............................................................................................................................... 133
11.2 EARLY VS. DELAYED MOBILISATION ...................................................................................................... 134
11.3 INTENSITY OF PHYSIOTHERAPY ............................................................................................................ 137
12 MULTIDISCIPLINARY MANAGEMENT........................................................................................................ 144
12.1 INTRODUCTION ............................................................................................................................... 144
12.2 HOSPITAL-BASED MULTIDISCIPLINARY REHABILITATION VERSUS USUAL CARE ................................................ 146
12.3 RESEARCH RECOMMENDATIONS ON HOSPITAL MULTIDISCIPLINARY REHABILITATION ...................................... 163
12.4 COMMUNITY-BASED MULTIDISCIPLINARY REHABILITATION VERSUS USUAL CARE ............................................ 165
12.5 RESEARCH RECOMMENDATIONS ON COMMUNITY MULTIDISCIPLINARY REHABILITATION .................................. 175
13 PATIENT AND CARER VIEWS AND INFORMATION ........................................................................................ 176
13.1 INTRODUCTION ............................................................................................................................... 176
13.2 PATIENT AND CARER VIEWS ................................................................................................................ 176
13.3 INFORMATION FOR PATIENTS.............................................................................................................. 183
13.4 CARER INVOLVMENT ......................................................................................................................... 185
13.5 RESEARCH RECOMMENDATIONS.......................................................................................................... 185
GLOSSARY ................................................................................................................................................. 186
APPENDICES ............................................................................................................................................... 199
APPENDIX A: SCOPE .................................................................................................................................... 206
13.6 GUIDELINE TITLE .............................................................................................................................. 206
13.7 THE REMIT...................................................................................................................................... 206
13.8 CLINICAL NEED FOR THE GUIDELINE ...................................................................................................... 206
13.9 THE GUIDELINE ................................................................................................................................ 209
13.10 RELATED NICE GUIDANCE ................................................................................................................. 212
13.11 FURTHER INFORMATION .................................................................................................................... 213
14 APPENDIX B: DECLARATIONS OF INTEREST ............................................................................................... 214
14.1 INTRODUCTION ............................................................................................................................... 214
14.2 DECLARATIONS OF INTERESTS OF THE GDG MEMBERS ............................................................................. 215
14.3 DECLARATIONS OF INTERESTS OF THE NCGC MEMBERS ........................................................................... 228
14.4 DECLARATIONS OF INTERESTS OF THE EXPERT ADVISORS .......................................................................... 229
15 APPENDIX C: REVIEW PROTOCOLS.......................................................................................................... 229
15.1 REVIEW PROTOCOL – IMAGING IN OCCULT HIP FRACTURE......................................................................... 229
15.2 REVIEW PROTOCOL – TIMING OF SURGERY ............................................................................................ 231
15.3 REVIEW PROTOCOL – ANALGESIA- SYSTEMIC MEDICATIONS ...................................................................... 233
15.4 REVIEW PROTOCOL – ANALGESIA- NERVE BLOCKS COMAPARED TO SYSTEMIC ANALGESICS .............................. 235
15.5 REVIEW PROTOCOL - ANAESTHESIA ..................................................................................................... 237
15.6 REVIEW PROTOCOL – SURGEON SENIORITY ............................................................................................ 239
15.7 REVIEW PROTOCOL – CEMENT............................................................................................................ 240
15.8 REVIEW PROTOCOL – INTRACAPSULAR FRACTURES.................................................................................. 242
15.9 REVIEW PROTOCOL – SURGICAL APPROACH ........................................................................................... 244
15.10 REVIEW PROTOCOL – HEMIARTHROPLASTY STEM DESIGN ......................................................................... 246
15.11 REVIEW PROTOCOL – EXTRACAPSULAR FRACTURES ................................................................................. 248
15.12 REVIEW PROTOCOL – MOBILISATION STRATEGIES ................................................................................... 250
15.13 REVIEW PROTOCOL – MULTIDISCIPLINARY REHABILITATION ...................................................................... 252
15.14 REVIEW PROTOCOL – CARER INVOLVEMENT .......................................................................................... 258
15.15 REVIEW PROTOCOL – HEALTH ECONOMICS ........................................................................................... 260
6 HIP FRACTURE
16 APPENDIX D: LITERATURE SEARCH STRATEGIES .......................................................................................... 262

16.1 SEARCH STRATEGIES ......................................................................................................................... 262
16.2 SEARCH TERMS ................................................................................................................................ 264
17 APPENDIX E: EVIDENCE TABLES - CLINICAL STUDIES .................................................................................... 284
17.1 EVIDENCE TABLE 1: IMAGING OPTIONS IN OCCULT HIP FRACTURE ............................................................. 286
17.2 EVIDENCE TABLE 2: TIMING OF SURGERY ............................................................................................. 289
17.3 EVIDENCE TABLE 3: OPTIMAL ANALGESIA ............................................................................................ 307
17.4 EVIDENCE TABLE 4: ANAESTHESIA ...................................................................................................... 310
17.5 EVIDENCE TABLE 5: SURGEON SENIORITY ............................................................................................. 313
17.6 EVIDENCE TABLE 6: DISPLACED INTRACAPSULAR FRACTURES .................................................................... 317
17.7 EVIDENCE TABLE 7: SURGERY – CEMENT VERSUS NO CEMENT.................................................................. 326
17.8 EVIDENCE TABLE 8: EXTRACAPSULAR FIXATION ..................................................................................... 328
17.9 EVIDENCE TABLE 9: SURGICAL APPROACH TO HEMIARTHROPLASTY............................................................ 376
17.10 EVIDENCE TABLE 10: MOBILISATION STRATEGIES .................................................................................. 378
17.11 EVIDENCE TABLE 11: MULTIDISCIPLINARY REHABILITATION ..................................................................... 387
17.12 EVIDENCE TABLE 12: PATIENT VIEWS .................................................................................................. 411
18 APPENDIX F: EVIDENCE TABLES - ECONOMIC STUDIES ................................................................................. 427
18.1 EVIDENCE TABLE 13: GENERAL VERSUS REGIONAL ANAESTHESIA .............................................................. 428
18.2 EVIDENCE TABLE 14: DISPLACED INTRACAPSULAR FRACTURES.................................................................. 430
18.3 EVIDENCE TABLE 15: CEMENTED ARTHROPLASTIES ................................................................................ 434
19 APPENDIX G: FOREST PLOTS ................................................................................................................. 449
19.1 RADIOLOGY .................................................................................................................................... 452
19.2 TIMING OF SURGERY......................................................................................................................... 453
19.3 ANALGESIA ..................................................................................................................................... 459
19.4 ANAESTHESIA.................................................................................................................................. 465
19.5 SURGICAL INTERVENTIONS ................................................................................................................. 471
19.6 MOBILISATION STRATEGIES ................................................................................................................ 510
19.7 INTENSIVE EXERCISE OR PHYSIOTHERAPY VS. USUAL CARE ......................................................................... 512
19.8 MULTIDISCIPLINARY REHABILITATION ................................................................................................... 517
19.9 HOME-BASED MDR VERSUS USUAL INPATIENT REHABILITATION ................................................................ 522
20 APPENDIX H: HEALTH ECONOMIC ANALYSIS.............................................................................................. 524
20.1 COST ANALYSIS OF NERVE BLOCKS, NON-OPIOID ANALGESICS AND OPIOID ANALGESICS ................................... 524
20.2 HOURLY WAGE COSTS FOR A PLANNED TRAUMA LIST ............................................................................... 527
20.3 PRICES FOR SLIDING HIP SCREWS AND SHORT AND LONG INTRAMEDULLARY NAILS ......................................... 530
20.4 COST ANALYSIS OF THE INTERVENTIONS FOR INTENSIVE MOBILISATION STRATEGIES. ...................................... 532
20.5 COST-EFFECTIVENESS ANALYSIS OF HOSPITAL INVESTMENT VERSUS NO HOSPITAL INVESTMENT FOR EARLY SURGERY
535
20.6 COST-EFFECTIVENESS ANALYSIS OF HOSPITAL MDR VS USUAL CARE .......................................................... 552
20.7 COST-EFFECTIVENESS ANALYSIS OF COMMUNITY MDR VS USUAL CARE ...................................................... 586
20.8 COST ANALYSIS OF CEMENTED VS. UNCEMENTED IMPLANTS (NEWER DESIGNS OF ARTHROPLASTY) ......................... 598
21 APPENDIX I: HIGH PRIORITY RESEARCH RECOMMENDATIONS ....................................................................... 603
21.1 IMAGING OPTIONS IN OCCULT HIP FRACTURE ......................................................................................... 603
21.2 ANAESTHESIA.................................................................................................................................. 606
21.3 DISPLACED INTRACAPSULAR HIP FRACTURES .......................................................................................... 609
21.4 INTENSIVE REHABILITATION THERAPIES AFTER HIP FRACTURE ..................................................................... 613
21.5 EARLY SUPPORTED DISCHARGE IN CARE HOME PATIENTS ........................................................................ 616
22 APPENDIX J: EXCLUDED STUDIES ............................................................................................................ 619
BIBLIOGRAPHY ............................................................................................................................................ 628
GUIDELINE DEVELOPMENT GROUP MEMBERS 1
Guideline development group members
Professor Cameron Swift (Chair) Emeritus Professor of Health Care of the

Elderly
Mr Tim Chesser Consultant Trauma and Orthopaedic Surgeon
Mr Anthony Field Patient Member
Dr Richard Griffiths Consultant Anaesthetist
Mr Robert Handley Consultant Trauma and Orthopaedic Surgeon
Mrs Karen Hertz Advanced Nurse Practitioner
Dr Sally Hope General Practitioner
Dr Antony Johansen Consultant Orthogeriatrician
Professor Sarah (Sallie) Lamb Professor of Rehabilitation, Director of

Warwick Clinical Trials Unit, Professor of
Trauma Rehabilitation
Professor Opinder Sahota Consultant Physician
Mrs Tessa Somerville Patient Member
Mrs Heather Towndrow Clinical Manager, Day Rehabilitation and Falls

Prevention
Mr Martin Wiese Consultant in Emergency Medicine

2 HIP FRACTURE
NCGC staff members of the guideline

development group
Dr Saoussen Ftouh Senior Research Fellow / Project Manager
Ms Joanna Ashe Information Scientist
Miss Elisabetta Fenu Senior Health Economist
Dr Jennifer Hill Operations Director
Dr Antonia Morga Health Economist
Dr Sarah Riley Research Fellow
Mr Carlos Sharpin Senior Information Scientist / Research Fellow
Expert advisors
Professor Judy Adams Consultant and Honorary Professor of

Diagnostic Radiology
Mrs Pamela Holmes Practice Development Manager (SCIE)
Mr Martyn Parker Consultant Orthopaedic Surgeon
Dr Luigi Siciliani Senior Lecturer in Health Economics

ACKNOWLDEGEMENTS 3
Acknowledgements
The development of this guideline was greatly assisted by the following people
NCGC
Kate Lovibond, David Wonderling, Abigail Jones, Maggie Westby, Laura Sawyer,
Clare Jones, Dr Lee-Yee Chong, Tamara Diaz
External
Dr Stuart White, Professor Maria Crotty, Dr Michelle Miller, The members of the
“Hip Fracture Anaesthesia” NHS Network, especially Dr Michael McBrien, Mr.
David Chittenden, Mr. John Ellington, Mr. David Drust, Ms Liz Hedinger and Ms
Sally Lambert
4 HIP FRACTURE
Acronyms and abbreviations
ADL Activities of Daily Living

ANOVA Analysis of variance
AO Arbeitsgemeinschaft für Osteosynthesefragen
BNF British National Formulary
CCA Cost-consequences analysis
CEA Cost-effectiveness analysis
c.f. Confer (refer to)
CI / 95% CI Confidence interval / 95% confidence interval
CT Computed tomography
CUA Cost-utility analysis
DH Department of Health
DSA Deterministic Sensitivity Analysis
ED Emergency Department
ESD Early Supported Discharge
EQ-5D EuroQol-5D
GA General anaesthesia
GORU Geriatric Orthopaedic Rehabilitation Unit
GDG Guideline Development Group
GP General Practitioner
GRADE Grading of Recommendations Assessment, Development and Evaluation
HES Hospital Episode Statistics
HFP Hip fracture programme
HR Hazard Ratio
HRQoL Health-related quality of life
HTA Health technology assessment
IC Intermediate care
ICD-10 International Classification of Diseases, 10th edition
ICER Incremental cost-effectiveness ratio
ACRONYMS AND ABBREVIATIONS 5
IQR Interquartile range

INMB Incremental Net Monetary Benefit
IRR Inter-rater reliability
ITT Intention to treat
LOS Length of Stay
+
LR Positive likelihood ratio
LR- Negative likelihood ratio
LY Life-Year
MD Mean Difference
MDR Multi-Disciplinary Rehabilitation
MARU Mixed Assessment and Rehabilitation Unit
MRI Magnetic resonnance imaging
NCGC National Clinical Guideline Centre
N/A Not applicable
NHS National Health Service
NHSEED The NHS Economic Evaluation Database
NICE National Institute for Health and Clinical Excellence
NNT Number needed to treat
NPV Negative predictive value
NSAID Non steroidal anti-inflammatory drugs
OR Odds ratio
PICO Framework incorporating patients, interventions, comparison and outcome
POMA Tinetti’s performance oriented mobility assessment
PFF Proximal femoral fracture
PPP Purchasing Power Parity
PPV Positive predictive value
p.r.n Pro re nata
PSA Probabilistic sensitivity analysis
QALY Quality-adjusted life year
QUADAS Quality assessment tool for diagnostic accuracy studies
RA Regional anaesthesia
RCT Randomised controlled trial
RNS Radionuclide scan
ROC Receiver operating characteristic
RR Relative risk
6 HIP FRACTURE
SCIE Social Care Institute for Excellence

SD Standard deviation
SE Standard error
SPC Summary of product characteristics
SR Systematic review
US Ultrasound
WTP Willingness to pay
1 Introduction
Hip fracture is the plain English term for a proximal femoral fracture or PFF. It refers to a
fracture occurring in the area between the edge of the femoral head and 5 centimetres
below the lesser trochanter (Figure 1). These fractures are generally divided into two main
groups depending on their relationship to the capsule of the hip joint. Those above the
insertion of the capsule are termed intracapsular, subcapital or femoral neck fractures.
Those below the insertion are extracapsular. The extracapsular group is split further into
trochanteric (inter- or pertrochanteric and reverse oblique) and subtrochanteric as shown.
The division into intra and extracapsular fractures relates to both the blood supply of the
femoral head and the mechanics of fixation.
Hip fracture is a major public health issue due to an ever increasing ageing population.
About 70,000 to 75,000 hip fractures (proximal femoral fractures) occur annually in the
UK39, with a cost (including medical and social care) amounting to about £2 billion a year.
Demographic projections indicate that the UK annual incidence will rise to 91,500 by 2015
and 101,000 in 202039, with an associated increase in annual expenditure. The majority of
this expenditure will be accounted for by hospital bed days and a further substantial
contribution will come from health and social aftercare. At present about a quarter of
patients with hip fracture are admitted from institutional care, and about 10–20% of those
admitted from home ultimately move to institutional care.
Hip fracture is the commonest reason for admission to an orthopaedic trauma ward and is
usually a ‘fragility’ fracture1 caused by a fall affecting an older person with osteoporosis or
1
The strict definition of a fragility fracture is one caused by a fall from standing height or less. For the
purposes of this guidance, the definition is slightly more flexible to encompass all hip fractures judged to
have an osteoporotic or osteopaenic basis
INTRODUCTION 7
osteopaenia (a condition in which bones lose calcium and become thinner, but not as much
as in osteoporosis). The National Hip Fracture Database reports the average age of a person
with hip fracture as 84 years for men and 83 for women, 76% of fracture occur in women.
Mortality is high – about 10% of people with a hip fracture die within 1 month and about
one third within 12 months. Most of the deaths are due to associated co morbidities and
not just to the fracture itself reflecting the high prevalence of comorbidity in people with
hip fracture. It is often the occurrence of a fall and fracture that signals underlying ill health.
Thus, hip fracture is by no means an exclusively surgical concern. Its effective management
requires the co-ordinated application of medical, surgical, anaesthetic and multidisciplinary
rehabilitation skills and a comprehensive approach covering the full time course of the
condition from presentation to subsequent follow-up, including the transition from hospital
to community.
Although hip fracture is predominantly a phenomenon of later life, it may occur at any age
in people with osteoporosis or osteopenia, and this guidance is applicable to adults across
the age spectrum. Skills in its management have, however been accrued, researched and
reported especially by collaborative teams specialising in the care of older people (using the
general designation ‘orthogeriatrics’). These skills are applicable in hip fracture irrespective
of age, and the guidance includes recommendations that cover the needs of younger
patients by drawing on such skills in an organised manner.
This guidance covers the management of hip fracture from the point of admission to
secondary care through to final return to the community and discharge from specific
follow-up. It assumes that anyone clinically suspected of having a hip fracture will be
referred for immediate hospital assessment other than in exceptional circumstances. It
excludes (other than by cross-reference) aspects covered by parallel NICE guidance, most
notably primary and secondary prevention of fragility fractures, but recognises the
importance of effective linkage to these closely related elements of comprehensive care.
The diagnosis of hip fracture is easily missed and in a small minority of patients the fracture
may not be apparent on a plain X-ray. In view of the serious nature of hip fracture the
guidance has sought to identify the most cost-effective imaging strategies to ensure this
does not happen.
Although not a structured service delivery evaluation, the Guideline Group was required to
extend its remit to cover essential implications for service organisation within the NHS
where these are fundamental to hip fracture management, and this has been done. In
general it is the case that suboptimal care and/or fragmentation of care result in longer
periods of dependency and/or hospitalisation leading to greater cost as well as inferior
outcome. There is substantial variation and lack of clarity in the UK in the extent, timing,
manner and organisation of the necessary collaborative and multidisciplinary elements of
effective management, including the timely achievement of rehabilitation after surgery
according to individual need. A further concern is the occurrence of delay before necessary
surgery is carried out. Prompt surgery has been generally recognised to be important, but
surgery is sometimes delayed for administrative or clinical reasons. Emerging evidence from
the National Hip Fracture database indicates substantial variation across centres in England
and Wales in this and other indicators of clinical and service quality. Such variation has
potentially profound economic implications, and priority has been given where appropriate
to underpinning recommendations with any available evidence of cost-effectiveness in the
NHS. Since work began on the guideline the Department of Health in England has launched
a high priority Best Practice Tariff initiative targeting a range of performance variables for
8 HIP FRACTURE
hip fracture, and the GDG have been aware of this contextual change as well as of
humanitarian issues in evaluating the evidence and formulating recommendations.
At all stages of hip fracture management, the importance of optimal communication with,
and support for, patients themselves and those who provide or will provide care –
including unpaid care family members or others – has been a fundamental tenet of
guidance development.
The view of the GDG is that an exceptional contemporary window of opportunity exists in
the NHS to achieve major improvements in the delivery of hip fracture care, to the benefit
not only of patients but of the system as a whole in terms of efficiency and cost. It is hoped
that implementation of this guidance will be instrumental to that end.
INTRODUCTION 9
Figure 1: Types of hip fracture (Parker M & Johansen A, 2006)259,270
Reproduced from BMJ, Parker, M., Johansen, A., 333(7557), 27-30, 2006 with
permission from BMJ Publishing Group Ltd
10 HIP FRACTURE
2 Development of the guideline
2.1 What is a NICE clinical guideline?

NICE clinical guidelines are recommendations for the care of individuals in specific clinical
conditions or circumstances within the NHS – from prevention and self-care through
primary and secondary care to more specialised services. We base our clinical guidelines on
the best available research evidence, with the aim of improving the quality of health care.
We use predetermined and systematic methods to identify and evaluate the evidence
relating to specific review questions.
NICE clinical guidelines can:
• provide recommendations for the treatment and care of people by health

professionals
• be used to develop standards to assess the clinical practice of individual health

professionals
• be used in the education and training of health professionals
• help patients to make informed decisions
• improve communication between patient and health professional
While guidelines assist the practice of healthcare professionals, they do not replace their
knowledge and skills.
We produce our guidelines using the following steps:
• Guideline topic is referred to NICE from the Department of Health
• Stakeholders register an interest in the guideline and are consulted throughout the
development process.
• The scope is prepared by the National Clinical Guideline Centre (NCGC)
• The NCGC establishes a guideline development group
• A draft guideline is produced after the group assesses the available evidence and
makes recommendations
DEVELOPMENT OF GUIDELINE 11
• There is a consultation on the draft guideline.
• The final guideline is produced.
The NCGC and NICE produce a number of versions of this guideline:
• the full guideline contains all the recommendations, plus details of the methods
used and the underpinning evidence
• the NICE guideline lists the recommendations
• the quick reference guide (QRG) presents recommendations in a suitable format

for health professionals
• information for the public (‘understanding NICE guidance’ or UNG) is written using
suitable language for people without specialist medical knowledge.
This version is the full version. The other versions can be downloaded from NICE
www.NICE.org.uk and the NCGC website www.ncgc.ac.uk.
2.2 Remit
NICE received the remit for this guideline from the Department of Health. They
commissioned the NCGC to produce the guideline.
The remit for this guideline is:
To prepare a clinical guideline on the management of fractured neck of femur.
2.3 Who developed this guideline?

A multidisciplinary Guideline Development Group (GDG) comprising professional group
members and consumer representatives of the main stakeholders developed this guideline
(see section on Guideline Development Group Membership and acknowledgements).
The National Institute for Health and Clinical Excellence funds the National Clinical
Guideline Centre (NCGC) and thus supported the development of this guideline. The GDG
was convened by the NCGC and chaired by Professor Cameron Swift in accordance with
guidance from the National Institute for Health and Clinical Excellence (NICE).
The group met every 6-8 weeks during the development of the guideline. At the start of the
guideline development process all GDG members declared interests including
consultancies, fee-paid work, share-holdings, fellowships and support from the healthcare
industry. At all subsequent GDG meetings, members declared arising conflicts of interest,
which were also recorded.
Members were either required to withdraw completely or for part of the discussion if their
declared interest made it appropriate. The details of declared interests and the actions
taken are shown in Appendix B.
12 HIP FRACTURE
Staff from the NCGC provided methodological support and guidance for the development
process. The team working on the guideline included a project manager, systematic
reviewers, health economists and information scientists. They undertook systematic
searches of the literature, appraised the evidence, conducted meta analysis and cost
effectiveness analysis where appropriate and drafted the guideline in collaboration with the
GDG.
2.4 What this guideline covers

The population of this guideline covers:
a) Adults aged 18 years and older presenting to the health service with a clinical
diagnosis (firm or provisional) of fragility fracture of the hip.
b) People with the following types of hip fracture:
• intracapsular (undisplaced and displaced)
• extracapsular (trochanteric and subtrochanteric).
c) Those with comorbidity strongly predictive of outcome, and those without such
comorbidity. The influence (if any) of advanced age or gender on clinical decision-
making, management and outcome will be specifically evaluated.
For further details please refer to the scope in Appendix A and review protocols in
Appendix C.
Key clinical areas in this guideline are:
a) Using alternative radiological imaging to confirm or exclude a suspected hip

fracture in patients with a normal X-ray.
b) Involving a physician or orthogeriatrician in the care of patients presenting with hip

fracture.
c) Early surgery (within 48 hours).
d) Optimal preoperative and postoperative analgesia (pain relief), including the use of
nerve blockade.
e) Regional (spinal – also known as ‘epidural’) versus general anaesthesia in patients

undergoing surgery for hip fracture.
f) Surgeon experience and seniority
g) For displaced intracapsular fracture:
• Internal fixation versus arthroplasty (hip replacement surgery)
• Total hip replacement versus hemiarthroplasty (replacing the head of the

femur only).
DEVELOPMENT OF GUIDELINE 13
h) Choice of surgical implants - Sliding hip screw versus intramedullary nail for
trochanteric extracapsular fracture.
i) Choice of surgical implants - Sliding hip screw versus intramedullary nail for
subtrochanteric extracapsular fracture.
j) Cemented versus non-cemented arthroplasty implants.
k) Hospital-based multidisciplinary rehabilitation for patients who have undergone hip

fracture surgery.
l) Early transfer to community-based multidisciplinary rehabilitation for patients who

have undergone hip fracture surgery.
2.5 What this guideline does not cover

The population of this guideline does not cover:
a) People younger than 18 years.
b) People with fractures caused by specific pathologies other than osteoporosis or

osteopaenia (because these would require more condition-specific guidance).
Clinical areas not included in this guideline are:
a) Primary and secondary prevention of fragility fracture.
b) Prevention and management of pressure sores.
c) Prophylaxis for venous thromboembolism.
d) Prevention and management of infection at the surgical site.
e) Nutritional support.
f) Selection of prostheses for hip replacement.
g) Complementary and alternative therapies.
2.6 Relationships between the guideline and other NICE guidance

Related NICE Health Technology Appraisals:
Alendronate, etidronate, risedronate, raloxifene, strontium ranelate and teriparatide for

the secondary prevention of osteoporotic fragility fractures in postmenopausal women
(amended). NICE technology appraisal guidance TA161 (2011). Available from
www.nice.org.uk/TA161
Alendronate, etidronate, risedronate, raloxifene and strontium ranelate for the primary
prevention of osteoporotic fragility fractures in postmenopausal women (amended). NICE
technology appraisal guidance TA160 (2011). Available from www.nice.org.uk/TA160
14 HIP FRACTURE
Denosumab for the prevention of osteoporotic fractures in postmenopausal women. NICE

technology appraisal guidance TA204 (2010). Available from www.nice.org.uk/TA204
Guidance on the use of metal on metal hip resurfacing arthroplasty. NICE technology
appraisal guidance 44 (2002). Available from www.nice.org.uk/guidance/TA44
The selection of prostheses for primary total hip replacement. NICE technology appraisal
guidance TA2 (2000). Available from www.nice.org.uk/TA2
Related NICE Interventional Procedures Guidance:
Minimally invasive hip replacement. NICE interventional procedure guidance (2010).

Available from www.nice.org.uk/guidance/IPG363
Related NICE Clinical Guidelines:
Delirium: diagnosis, prevention and management of delirium. NICE clinical guideline CG103
(2010). Available from http://guidance.nice.org.uk/CG103
Venous thromboembolism – reducing the risk. NICE clinical guideline CG92 (2010). Available
from http://guidance.nice.org.uk/CG92
Surgical site infection. NICE clinical guideline CG74 (2008). Available from
www.nice.org.uk/CG74
Dementia: supporting people with dementia and their carers in health and social care. Nice
clinical guideline CG42 (2006). Available from www.nice.org.uk/CG42
Nutrition support in adults. NICE clinical guideline CG32 (2006).Available from

The management of pressure ulcers in primary and secondary care. NICE clinical guideline
CG29 (2005). Available from www.nice.org.uk/CG29
Falls. NICE clinical guideline CG21 (2004). Available from www.nice.org.uk/CG21
Preoperative tests. NICE clinical guideline CG3 (2003). Available from www.nice.org.uk/CG3
NICE Related Guidance currently in development:
Osteoporosis: risk assessment of people with osteoporosis. NICE clinical guideline.

Publication date to be confirmed.
METHODS 15
3 Methods
This guidance was developed in accordance with the methods outlined in the NICE
Guidelines Manual 2009 233
3.1 Developing the review questions and outcomes

Review questions were developed in a PICO framework (patient, intervention, comparison
and outcome) for intervention reviews, and with a framework of population, index tests,
reference standard and target condition for reviews of diagnostic test accuracy. This was to
guide the literature searching process and to facilitate the development of
recommendations by the guideline development group (GDG). They were drafted by the
NCGC technical team and refined and validated by the GDG. The questions were based on
the key clinical areas identified in the scope (Appendix A). Further information on the
outcome measures examined follows this section.
16 HIP FRACTURE
Chap Review question Outcomes

ter
Radiology In patients with a continuing clinical ▪ Sensitivity
suspicion of hip fracture, despite negative ▪ Specificity
radiographic findings, what is the clinical ▪ Positive and negative
and cost-effectiveness of additional predictive values
imaging (radiography after at least 48 ▪ Positive and negative
hours), Radionuclide scanning (RNS), likelihood ratios
ultrasound (US) and computed
tomography (CT), compared to magnetic
resonance imaging (MRI), in confirming,
or excluding, a hip fracture?
Timing of In patients with hip fractures what is the ▪ Mortality (30 days, 3 months,
surgery clinical and cost effectiveness of early 1 year)
surgery (within 24, 36 or 48 hours) on the ▪ Length of stay in secondary
incidence of complications such as care
mortality, pneumonia, pressure sores, ▪ Length of time before
cognitive dysfunction and increased community
length of hospital stay? resettlement/discharge
▪ Place of residence (compared
with baseline) 12 months
after fracture
▪ Functional status (30 days, 3
months, 1 year)
▪ Quality of life (30 days, 3
months, 1 year)
▪ Complications (including
pressure ulcers)
Analgesia In patients who have or are suspected of ▪ Pain Need for ‘breakthrough’
having a hip fracture, what is the clinical analgesia
and cost effectiveness of nerve blocks ▪ Mortality
compared to systemic analgesia in ▪ Adverse effects
providing adequate pain relief and
reducing side effects and mortality?
Anaesthesia In patients undergoing surgical repair for ▪ Patient preference

hip fractures, what is the clinical and cost ▪ Early mortality up to 1 month
effectiveness of regional (spinal/epidural) ▪ Functional status up to 1 year
anaesthesia compared to general ▪ Pain Adverse effects
anaesthesia in reducing complications
such as mortality, cognitive dysfunction
thromboembolic events, postoperative
respiratory morbidity, renal failure and
length of stay in hospital?
METHODS 17
Surgeon Does surgeon seniority (consultant or ▪ Mortality (30 days, 3 months,

seniority equivalent) reduce the incidence of 1 year)
mortality, operative revision and poor ▪ Length of stay in secondary
functional outcome? care
▪ Reoperation rate
▪ Dislocations
▪ Wound infection
Cement In hip fracture patients undergoing total ▪ Perioperative mortality

hip replacement what is the clinical and ▪ Mortality at 30 days, 3
cost effectiveness of cemented total hip months & 1 year or longer
replacement versus uncemented total hip ▪ Functional status up to 1 year
replacement on mortality, surgical ▪ Pain (generally measured by
revision, functional status, length of stay, visual analogue scale or verbal
quality of life, pain and place of residence rating)
after hip fracture? ▪ Quality of life
▪ Requirement for reoperation
▪ Length of stay in
hospital/acute care
▪ Length of stay in to
community or resettlement
(i.e. superspell)
▪ Place of residence 12 months
after fracture
▪ Wound healing complications
Intracapsular In patients undergoing repair for ▪ Mortality at 30 days, 3

fractures intracapsular hip fractures what is the months & 1 year or longer
clinical and cost effectiveness of internal ▪ Functional status up to 1 year
fixation compared to hemiarthroplasty ▪ Pain (generally measured by
compared to total hip replacement on visual analogue scale or verbal
mortality, surgical revision, functional rating)
status, length of stay, quality of life, pain ▪ Quality of life
and place of residence after hip fracture? ▪ Requirement for reoperation
hospital/acute care
(i.e. superspell)
after fracture
Surgical In patients having surgical treatment for ▪ Mortality (30 days, 3 months,
approach intracapsular hip fracture with 1 year)
hemiarthroplasty what is the clinical and ▪ Length of hospital stay
cost effectiveness of anterolateral ▪ Reoperation rate
compared to posterior surgical approach ▪ Dislocations
on mortality, number of reoperations, ▪ Functional status
dislocation, functional status, length of ▪ Quality of life
hospital stay, quality of life and pain? ▪ Pain
18 HIP FRACTURE
Hemiarthrop In patients undergoing surgery for hip ▪ Mortality at 30 days, 3

lasty stem fracture what is the clinical and cost months & 1 year or longer
design effectiveness of ‘OEDP 10A rating’ ▪ Functional status up to 1 year
designs of stems in preference to Austin ▪ Pain (generally measured by
Moore or Thompson stems when visual analogue scale or verbal
inserting a hemiarthroplasty on mortality, rating)
surgical revision, functional status, length ▪ Quality of life
of stay, quality of life, pain and place of ▪ Requirement for reoperation
residence after hip fracture? ▪ Length of stay in
hospital/acute care
(i.e. superspell)
after fracture
Extracapsula In patients undergoing repair for ▪ Mortality at 30 days, 3
r fractures trochanteric extracapsular hip fractures months & 1 year or longer
what is the clinical and cost effectiveness ▪ Functional status up to 1 year
of extramedullary sliding hip screws ▪ Pain (generally measured by
compared to intramedullary nails on visual analogue scale or verbal
(operative or postoperative
fracture of the femur, cut-out
and non-union)
hospital/acute care
(i.e. superspell)
Extracapsula In patients undergoing repair for ▪ Mortality at 30 days, 3

r fractures subtrochanteric extracapsular hip months & 1 year or longer
fractures, what is the effectiveness of ▪ Functional status up to 1 year
extramedullary sliding hip screws ▪ Pain (generally measured by
compared to intramedullary nails on visual analogue scale or verbal
(operative or postoperative
fracture of the femur, cut-out
and non-union)
hospital/acute care
(i.e. superspell)
METHODS 19
Mobilisation In patients who have undergone surgery ▪ Mortality at 30 days, 3

strategies for hip fracture, what is the clinical and months & 1 year or longer
cost effectiveness of early mobilisation ▪ Functional status up to 1 year
(<48 hours after surgery) compared to ▪ Pain (generally measured by
late mobilisation on functional status, visual analogue scale or verbal
mortality, place of residence/discharge, rating)
pain and quality of life? ▪ Quality of life
▪ Discharge destination
Mobilisation In patients who have undergone surgery ▪ Mortality at 30 days, 3
strategies for hip fracture, what is the clinical and months & 1 year or longer
cost effectiveness of intensive ▪ Functional status up to 1 year
physiotherapy compared to non intensive ▪ Pain (generally measured by
physiotherapy on functional status, visual analogue scale or verbal
mortality, place of residence/discharge, rating)
pain and quality of life? ▪ Quality of life
▪ Mobility
Multidiscipli In patients with hip fracture what is the ▪ Mortality (30 days, 3 months,
nary clinical and cost effectiveness of 1 year)
rehabilitatio 'orthogeriatrician' involvement in the ▪ Length of stay in secondary
n whole pathway of assessment, peri- care
operative care and rehabilitation on ▪ Length of time before
functional status, length of stay in community
secondary care, mortality, place of resettlement/discharge
residence/discharge, hospital ▪ Place of residence (compared
readmission and quality of life? with baseline) 12 months
after fracture
months, 1 year)
▪ Hospital readmission
months, 1 year)
nary clinical and cost effectiveness of hospital- 1 year)
rehabilitatio based multidisciplinary rehabilitation on ▪ Length of stay in secondary
n functional status, length of stay in care
secondary care, mortality, place of ▪ Length of time before
residence/discharge, hospital community
readmission and quality of life? resettlement/discharge
after fracture
months, 1 year)
months, 1 year)
20 HIP FRACTURE
nary clinical and cost effectiveness of 1 year)
rehabilitatio community-based multidisciplinary ▪ Length of stay in secondary
n rehabilitation on functional status, length care
of stay in secondary care, mortality, place ▪ Length of time before
of residence/discharge, hospital community
readmission and quality of life? resettlement/discharge
after fracture
months, 1 year)
months, 1 year)
Carer In patients who have been discharged ▪ Mortality (30 days, 3 months,
involvement after hip fracture repair, what is the 1 year)
clinical and cost effectiveness of having a ▪ Length of stay in secondary
non paid carer (e.g. spouse, relative, care
friends) on mortality, length of stay, place ▪ Length of time before
of residence/discharge, functional status, community
hospital readmission and quality of life? resettlement/discharge
after fracture
months, 1 year)
months, 1 year)
3.2 Searching for evidence
3.2.1 Clinical literature search
Systematic literature searches were undertaken to identify evidence within published

literature in order to answer the review questions as per The Guidelines Manual 233.
Clinical databases were searched using relevant medical subject headings, free-text
terms and study type filters where appropriate. Studies published in languages other
than English were not reviewed. Where possible, searches were restricted to articles
published in English language. All searches were conducted on core databases,
MEDLINE, Embase and The Cochrane Library. Additional subject specific databases were
used for some questions: PsycInfo for patient views and patient education questions;
Cinahl for every question except those on anaesthesia, analgesia and the surgical
procedures. All searches were updated on the 31st August 2010. No papers after this
date were considered.
METHODS 21
Search strategies were checked by looking at reference lists of relevant key papers,
checking search strategies in other systematic reviews and asking the GDG for known
studies. The questions, the study types applied, the databases searched and the years
covered can be found in Appendix D.
During the scoping stage, a search was conducted for guidelines and reports on the
websites listed below and on organisations relevant to the topic. Searching for grey
literature or unpublished literature was not undertaken. All references sent by
stakeholders were considered.
• Guidelines International Network database (www.g-i-n.net)
• National Guideline Clearing House (www.guideline.gov/)
• National Institute for Health and Clinical Excellence (NICE) (www.nice.org.uk)
• National Institutes of Health Consensus Development Program (consensus.nih.gov/)
• NHS Evidence (www.evidence.nhs.uk/)
3.2.2 Health economic literature search
Systematic literature searches were also undertaken to identify health economic evidence
within published literature relevant to the review questions. The evidence was identified by
conducting a broad search relating to the guideline population in the NHS economic
evaluation database (NHS EED) and health technology assessment (HTA) database with no
date restrictions. Additionally, the search was run on MEDLINE and Embase, with a specific
economic filter, to ensure recent publications that had not yet been indexed by these
databases were identified. This was supplemented by additional searches that looked for
economic papers specifically relating to the radiological imaging question on MEDLINE,
Embase, NHS EED and HTA databases, and the Health Economic Evaluations Database
(HEED) as it became apparent that some papers in this area were not being identified
through the first search. Studies published in languages other than English were not
reviewed. Where possible, searches were restricted to articles published in English
language.
The search strategies for health economics are included in Appendix D. All searches were
updated on the 31st August 2010. No papers published after this date were considered.
3.3 Evidence of effectiveness

The Research Fellow
• Identified potentially relevant studies for each review question from the relevant search
results by reviewing titles and abstracts – full papers were then obtained.
• Reviewed full papers against pre-specified inclusion / exclusion criteria to identify studies
that addressed the review question in the appropriate population and reported on
outcomes of interest (review protocols are included in Appendix C).
22 HIP FRACTURE
• Critically appraised relevant studies using the appropriate checklist as specified in The
Guidelines Manual233.
• Extracted key information about the study’s methods and results into evidence tables
(evidence tables are included in Appendix E).
• Generated summaries of the evidence by outcome (included in the relevant chapter

write-ups):
o Randomised studies: meta analysed, where appropriate and reported in GRADE

profiles (for clinical studies) – see below for details
o Observational studies: data presented as a range of values in GRADE profiles
o Diagnostic studies: data presented as a range of values in adapted GRADE profiles
o Qualitative studies: each study summarised in a table where possible, otherwise

presented in a narrative.
3.3.1 Inclusion/exclusion
See the review protocols in Appendix C for full details.
3.3.2 Methods of combining clinical studies
Data synthesis for intervention reviews
Where possible, meta-analyses were conducted to combine the results of studies for each
review question using Cochrane Review Manager (RevMan5) software. Fixed-effects
(Mantel-Haenszel) techniques were selected to calculate risk ratios (relative risk) for the
binary outcomes. The continuous outcomes were analysed using an inverse variance
method for pooling weighted mean differences and where the studies had different scales,
standardised mean differences were used.
Statistical heterogeneity was assessed by considering the chi-squared test for significance
at p<0.05 or an I-squared inconsistency statistic of >50% to indicate significant
heterogeneity. Where significant heterogeneity was present, we carried out predefined
subgroup analyses as defined in the protocol for each question (Appendix C). Sensitivity
analysis based on the quality of studies was also carried out if there were differences, with
particular attention paid to allocation concealment, blinding and loss to follow-up (missing
data).
Assessments of potential differences in effect between subgroups were based on the chi-
squared tests for heterogeneity statistics between subgroups. If no sensitivity analysis was
found to completely resolve statistical heterogeneity then a random effects (DerSimonian
and Laird) model was employed to provide a more conservative estimate of the effect.
For binary outcomes, absolute event rates were also calculated using the GRADEpro
software using event rate in the control arm of the pooled results.
METHODS 23
Data synthesis for diagnostic test accuracy review
For diagnostic test accuracy studies, the following outcomes were reported: sensitivity,
specificity, positive predictive value, negative predictive value and positive and negative
likelihood ratios.In cases where the outcomes were not reported, 2 by 2 tables were
constructed from raw data to allow calculation of these accuracy measures. Summary
receiver operative characteristic (ROC) curves were not generated as we did not explore the
effect of different cut-off thresholds on sensitivity and specificity for the imaging questions.
3.3.3 Appraising the quality of evidence by outcomes

The evidence for outcomes from the included RCT and observational studies were
evaluated and presented using an adaptation of the ‘Grading of Recommendations
Assessment, Development and Evaluation (GRADE) toolbox’ developed by the international
GRADE working group (http://www.gradeworkinggroup.org/). The software (GRADEpro)
developed by the GRADE working group was used to assess the quality of each outcome,
taking into account individual study quality and the meta-analysis results.The summary of
findings was presented as two separate tables in this guideline. The “Clinical/Economic
Study Characteristics” table includes details of the quality assessment while the “Clinical
/Economic Summary of Findings” table includes pooled outcome data, where appropriate,
an absolute measure of intervention effect and the summary of quality of evidence for that
outcome. In this table, the columns for intervention and control indicate the sum of the
sample size for continuous outcomes. For binary outcomes such as number of patients with
an adverse event, the event rates (n/N: number of patients with events divided by sum of
number of patients) are shown with percentages. Reporting or publication bias was only
taken into consideration in the quality assessment and included in the Clinical Study
Characteristics table if it was apparent. Each outcome was examined separately for the
quality elements listed and defined in Table 3-1 and each graded using the quality levels
listed in
Table 3-2. The main criteria considered in the rating of these elements are discussed below
(see section 3.3.4 Grading of Evidence). Footnotes were used to describe reasons for
grading a quality element as having serious or very serious problems. The ratings for each
component were summed to obtain an overall assessment for each outcome.
Table 3-3. The GRADE toolbox is currently designed only for randomised trials and
observational studies but we adapted the quality assessment elements and outcome
presentation for diagnostic accuracy studies.
24 HIP FRACTURE
Table 3-1: Descriptions of quality elements in GRADE for intervention studies

Quality element Description
Limitations Limitations in the study design and implementation may bias the
estimates of the treatment effect. Major limitations in studies
decrease the confidence in the estimate of the effect
Inconsistency Inconsistency refers to an unexplained heterogeneity of results
Indirectness Indirectness refers to differences in study population, intervention,
comparator and outcomes between the available evidence and the
review question, or recommendation made
Imprecision Results are imprecise when studies include relatively few patients
and few events and thus have wide confidence intervals around
the estimate of the effect relative to the clinically important
threshold
Publication bias Publication bias is a systematic underestimate or an overestimate
of the underlying beneficial or harmful effect due to the selective
publication of studies
Table 3-2: Levels for quality elements in GRADE

Level Description
None There are no serious issues with the evidence
Serious The issues are serious enough to downgrade the outcome
evidence by one level
Very serious The issues are serious enough to downgrade the outcome
evidence by two levels
Table 3-3: Overall quality of outcome evidence in GRADE

Level Description
High Further research is very unlikely to change our confidence in the
estimate of effect
Moderate Further research is likely to have an important impact on our
confidence in the estimate of effect and may change the estimate
Low Further research is very likely to have an important impact on our
confidence in the estimate of effect and is likely to change the
estimate
Very low Any estimate of effect is very uncertain
3.3.4 Grading the quality of clinical evidence
After results were pooled, the overall quality of evidence for each outcome was considered.
The following procedure was adopted when using GRADE:
1. A quality rating was assigned, based on the study design. RCTs start HIGH and
observational studies as LOW, uncontrolled case series as LOW or VERY LOW
2. The rating was then downgraded for the specified criteria: Study limitations,
inconsistency, indirectness, imprecision and reporting bias. These criteria are detailed
below. Observational studies were upgraded if there was: a large magnitude of effect,
dose-response gradient, and if all plausible confounding would reduce a
demonstrated effect or suggest a spurious effect when results showed no effect. Each
METHODS 25
quality element considered to have “serious” or “very serious” risk of bias were rated
down -1 or -2 points respectively.
3. The downgraded/upgraded marks were then summed and the overall quality rating
was revised. For example, all RCTs started as HIGH and the overall quality became
MODERATE, LOW or VERY LOW if 1, 2 or 3 points were deducted respectively.
4. The reasons or criteria used for downgrading were specified in the footnotes.
The details of criteria used for each of the main quality element are discussed further in the
following sections 4.3.5 to 4.3.8.
3.3.5 Study limitations
The main limitations for randomised controlled trials are listed in Table 3-4.
The GDG accepted that investigator blinding in surgical intervention studies was impossible
and participant blinding was also impossible to achieve in most situations. Therefore, open-
label studies for surgery were not downgraded in the quality rating across the guideline.
Studies were downgraded for unclear or inadequate allocation concealment. .
Table 3-4 lists the limitations considered for randomised controlled trials.
Table 3-4: Study limitations of randomised controlled trials

Limitation Explanation
Allocation Those enrolling patients are aware of the group to which the
concealment next enrolled patient will be allocated (major problem in
“pseudo” or “quasi” randomised trials with allocation by day
of week, birth date, chart number etc.)
Lack of blinding Patient, caregivers, those recording outcomes, those
adjudicating outcomes, or data analysts are aware of the arm
to which patients are allocated
Incomplete Loss to follow-up not accounted and failure to adhere to the
accounting of intention to treat principle when indicated
patients and
outcome events
Selective outcome Reporting of some outcomes and not others on the basis of
reporting the results
Other limitations For example:
• stopping early for benefit observed in randomised
trials, in particular in the absence of adequate
stopping rules
• use of unvalidated patient-reported outcomes
• carry-over effects in cross-over trials
• recruitment bias in cluster-randomised trials
3.3.6 Inconsistency
Inconsistency refers to an unexplained heterogeneity of results. When estimates of the

treatment effect across studies differ widely (i.e. heterogeneity or variability in results), this
suggests true differences in underlying treatment effect. When heterogeneity was
26 HIP FRACTURE
measured at either Chi square p<0.05 or I- squared inconsistency statistic of >50%, but no
plausible explanation can be found, the quality of evidence was downgraded by one or two
levels, depending on the extent of uncertainty to the results contributed by the
inconsistency in the results. In addition to the I- square and Chi square values, the decision
for downgrading was also dependent on factors such as whether the intervention is
associated with benefit in all other outcomes or whether the uncertainty about the
magnitude of benefit (or harm) of the outcome showing heterogeneity would influence the
overall judgment about net benefit or harm (across all outcomes).
If inconsistency could be explained based on prespecified subgroup analysis, the GDG took
this into account and considered whether to make separate recommendations based on
the identified explanatory factors, i.e. population and intervention.Where subgroup
analysis gives a plausible explanation of heterogeneity, the quality of evidence would not
be downgraded.
3.3.7 Indirectness
Directness refers to the extent to which the populations, intervention, comparisons and
outcome measures are similar to those defined in the inclusion criteria for the reviews.
Indirectness is important when these differences are expected to contribute to a difference
in effect size, or may affect the balance of harms and benefits considered for an
intervention.
3.3.8 Imprecision
The sample size, event rates and the resulting width of confidence intervals were the main
criteria considered. Where the minimal important difference (MID) of an outcome is
known, the optimal information size (OIS), i.e. the sample size required to detect the
difference with 80% power and p≤0.05 was calculated and used as the criteria. The criteria
applied for imprecision are based on the confidence intervals for pooled or the best
estimate of effect as illustrated in Figure 3-1.
Figure 3-1: Illustration of precise and imprecise outcomes based on the confidence interval of
outcomes in a forest plot
METHODS 27
MID = minimal important difference determined for each outcome. The MIDs are the threshold for
appreciable benefits and harms. The confidence intervals of the top three points of the diagram were
considered precise because the upper and lower limits did not cross the MID. Conversely, the bottom three
points of the diagram were considered imprecise because all of them crossed the MID and reduced our
certainty of the results. Figure adapted from GRADEPro software.
The following are the MID for the outcomes and the methods used to calculate the OIS in
this guideline:
• Any statistically significant difference in mortality
• The default confidence intervals in GRADE for relative risk of 0.75 and 1.25 for all
other outcomes.
3.4 Evidence of cost-effectiveness

Evidence on cost-effectiveness related to the key clinical issues being addressed in the guideline
was sought. The health economist:
• Undertook a systematic review of the economic literature
• Undertook new cost-effectiveness analysis in priority areas
3.4.1 Literature review

The Health Economist:
• Identified potentially relevant studies for each review question from the economic search
results by reviewing titles and abstracts – full papers were then obtained.
• Reviewed full papers against pre-specified inclusion / exclusion criteria to identify

relevant studies (see below for details).
• Critically appraised relevant studies using the economic evaluations checklist as specified
in The Guidelines Manual233.
• Extracted key information about the study’s methods and results into evidence tables
(evidence tables are included in Appendix F).
• Generated summaries of the evidence in NICE economic evidence profiles – see below for
details.
3.4.1.1 Inclusion/exclusion
28 HIP FRACTURE
Full economic evaluations (cost-effectiveness, cost–utility, cost-benefit and cost-

consequence analyses) and comparative costing studies that addressed the review
question in the relevant population were considered potentially applicable as economic
evidence.
Studies that only reported cost per hospital (not per patient), or only reported average
cost effectiveness without disaggregated costs and effects, were excluded. However,
studies reporting the cost per hospital were included when it was possible to ascertain the
cost per patient of each intervention. Abstracts, posters, reviews, letters/editorials,
foreign language publications and unpublished studies were excluded. Studies judged to
have had an applicability rating of ‘not applicable’ were excluded (this included studies
that took the perspective of a non-OECD country).
Remaining studies were prioritised for inclusion based on their relative applicability to the
development of this guideline and the study limitations. For example, if a high quality,
directly applicable UK analysis was available other less relevant studies may not have
been included. Where exclusions occurred on this basis, this is noted in the relevant
section.
For more details about the assessment of applicability and methodological quality see the
economic evaluation checklist (The Guidelines Manual, Appendix H233 and the health
economics research protocol in Appendix C.
When no relevant economic analysis was found from the economic literature review,
relevant UK NHS unit costs related to the compared interventions were presented to the
GDG to inform the possible economic implication of the recommendation to make.
3.4.2 NICE economic evidence profiles

The NICE economic profile has been used to summarise cost and cost-effectiveness
estimates. The economic evidence profile shows, for each economic study, an assessment
of applicability and methodological quality, with footnotes indicating the reasons for the
assessment. These assessments were made by the health economist using the economic
evaluation checklist from The Guidelines Manual, Appendix H233. It also shows
incremental costs, incremental outcomes (e.g. QALYs) and the incremental cost-
effectiveness ratio from the primary analysis, as well as information about the assessment
of uncertainty in the analysis. See Table 3-5 for more details.
If a non-UK study was included in the profile, the results were converted into pounds
sterling using the appropriate purchasing power parity246.
Table 3-5: Content of NICE economic profile

Item Description
Study First author name, reference, date of study publication and country perspective.
Limitations An assessment of methodological quality of the study*:
• Minor limitations – the study meets all quality criteria, or the study fails to
meet one or more quality criteria, but this is unlikely to change the
conclusions about cost effectiveness.
• Potentially serious limitations – the study fails to meet one or more
quality criteria, and this could change the conclusion about cost
METHODS 29
effectiveness
• Very serious limitations – the study fails to meet one or more quality
criteria and this is very likely to change the conclusions about cost
effectiveness. Studies with very serious limitations would usually be
excluded from the economic profile table.
Applicability An assessment of applicability of the study to the clinical guideline, the current
NHS situation and NICE decision-making*:
• Directly applicable – the applicability criteria are met, or one or more
criteria are not met but this is not likely to change the conclusions about
cost effectiveness.
• Partially applicable – one or more of the applicability criteria are not met,
and this might possibly change the conclusions about cost effectiveness.
• Not applicable – one or more of the applicability criteria are not met, and
this is likely to change the conclusions about cost effectiveness.
Other Particular issues that should be considered when interpreting the study.
comments
Incremental cost The mean cost associated with one strategy minus the mean cost of a comparator
strategy.
Incremental The mean QALYs (or other selected measure of health outcome) associated with
effects one strategy minus the mean QALYs of a comparator strategy.
ICER Incremental cost-effectiveness ratio: the incremental cost divided by the
respective QALYs gained
Uncertainty A summary of the extent of uncertainty about the ICER reflecting the results of
deterministic or probabilistic sensitivity analyses, or stochastic analyses of trial
data, as appropriate.
*Limitations and applicability were assessed using the economic evaluation checklist from The Guidelines
Manual, Appendix H 233
When no cost-effectiveness evidence was available, the cost of the interventions being
evaluated has in some cases been determined by conducing original cost analyses there
were reported in Appendix H. Alternatively, the GDG was presented with the cost figures
from relevant sources, such as the NHS reference cost for England and Wales.
3.4.3 Undertaking new health economic analysis

As well as reviewing the published economic literature for each review question, as
described above, new economic analyses were undertaken by the Health Economist in
priority areas. Priority areas for new health economic analysis were agreed by the GDG
after formation of the review questions and consideration of the available health
economic evidence.
Additional data for the analysis was identified as required through additional literature
searches undertaken by the Health Economist, and discussion with the GDG. Model
structure, inputs and assumptions were explained to and agreed by the GDG members
during meetings, and they commented on subsequent revisions.
See Appendix H for details of the health economic analyses undertaken for the guideline.
30 HIP FRACTURE
3.4.4 Cost-effectiveness criteria
NICE’s report ‘Social value judgements: principles for the development of NICE
guidance’232 sets out the principles that GDGs should consider when judging whether an
intervention offers good value for money. In general, an intervention was considered to
be cost effective if either of the following criteria applied (given that the estimate was
considered plausible):
a) The intervention dominated other relevant strategies (that is, it was both less costly
in terms of resource use and more clinically effective compared with all the other
relevant alternative strategies), or
b) The intervention cost less than £20,000 per quality-adjusted life-year (QALY) gained
compared with the next best strategy.
If the GDG recommended an intervention that was estimated to cost more than £20,000
per QALY gained, or did not recommend one that was estimated to cost less than £20,000
per QALY gained, the reasons for this decision are discussed explicitly in the ‘from
evidence to recommendations’ section of the relevant chapter. This is written with
reference to the issues regarding the plausibility of the estimate or to the factors set out
in the Social value judgements report 232.
3.5 Developing recommendations

Over the course of the guideline development process, the GDG was presented with:
• Evidence tables of the clinical evidence (Appendix E) and economic evidence

(Appendix F) reviewed from the literature.
• Summary of clinical and economic evidence and quality (as presented in chapters 5
to 13).
• Forest plots (Appendix G)
• A description of the methods and results of the cost-effectiveness analysis

undertaken for the guideline (Appendix H)
Recommendations were drafted on the basis of the GDG interpretation of the available
evidence, taking into account the balance of benefits, harms and costs. When clinical and
economic evidence was of poor quality, conflicting or absent, the GDG drafted
recommendations based on their expert opinion. The considerations for making consensus
based recommendations include the balance between potential harms and benefits,
economic or implications compared to the benefits, current practices, recommendations
made in other relevant guidelines, patient preferences and equality issues. The consensus
recommendations were done through discussions in the GDG.
METHODS 31
3.5.1 Research recommendations
When areas were identified for which good evidence was lacking, the guideline development
group considered making recommendations for future research. Decisions about inclusion
were based on factors such as:
• the importance to patients or the population
• national priorities
• potential impact on the NHS and future NICE guidance
• ethical and technical feasibility
3.6 Validation process

The guidance is subject to an eight week public consultation and feedback as part of the
quality assurance and peer review of the document. All comments received from registered
stakeholders are responded to in turn and posted on the NICE website when the pre-
publication check of the full guideline occurs.
3.7 Updating the guideline

Following publication, and in accordance with the NICE guidelines manual, NICE will ask a
National Collaborating Centre or the National Clinical Guideline Centre to advise NICE’s
Guidance executive whether the evidence base has progressed significantly to alter the
guideline recommendations and warrant an update.
3.8 Disclaimer
Health care providers need to use clinical judgement, knowledge and expertise when deciding whether
it is appropriate to apply guidelines. The recommendations cited here are a guide and may not be
appropriate for use in all situations. The decision to adopt any of the recommendations cited here
must be made by the practitioners in light of individual patient circumstances, the wishes of the
patient, clinical expertise and resources.
The National Clinical Guideline Centre disclaim any responsibility for damages arising out of the use or
non-use of these guidelines and the literature used in support of these guidelines.
3.9 Funding
The National Clinical Guideline Centre was commissioned by the National Institute for Health
and Clinical Excellence to undertake the work on this guideline.
32 HIP FRACTURE
4 Guideline summary
4.1 Map of recommendations

The algorithms can be found in the quick reference guide, available from
www.nice.org.uk/guidance/CG124/QuickRefGuide. In addition, NICE is developing a pathway,
which will be published on the NICE website.
GUIDELINE SUMMARY 33
Key priorities for implementation

The GDG identified ten key priorities for implementation. The decision was made after
discussion and voting by the GDG. They selected recommendations that would:
• Have a high impact on outcomes that are important to patients (A)
• Have a high impact on reducing variation in care and outcomes (B)
• Lead to a more efficient use of NHS resources (C)
• Promote patient choice (D)
• Promote equalities (E)
• Mean patients reach critical points in the care pathway more quickly (F).
In doing this the GDG also considered which recommendations were particularly likely to
benefit from implementation support. They considered whether a recommendation:
• Requires changes in service delivery (W)
• Requires retraining of professionals or the development of new skills and

competencies (X)
• Affects and needs to be implemented across various agencies or settings (complex

interactions) (Y)
• May be viewed as potentially contentious, or difficult to implement for other

reasons (Z)
For each key recommendation listed below, the selection criteria and implementation
support points are indicated by the use of the letters shown in brackets above and are
shown in the linking evidence to recommendations sections in the relevant chapters.
➢ Perform surgery on the day of, or the day after, admission. (A, B, C, F, W, Y and Z).
➢ Identify and treat correctable comorbidities immediately so that surgery is not

delayed by:
• anaemia
• anticoagulation
• volume depletion
• electrolyte imbalance
• uncontrolled diabetes
• uncontrolled heart failure
• correctable cardiac arrhythmia or ischaemia

34 HIP FRACTURE
• acute chest infection
• exacerbation of chronic chest conditions (A, B, C, F, Y and Z).
➢ Schedule hip fracture surgery on a planned trauma list (A, B, C, F, W, and Z).
➢ Perform replacement arthroplasty (hemiarthroplasty or total hip replacement) in

patients with a displaced intracapsular fracture (A, B, C, F and Z).
➢ Offer total hip replacements to patients with a displaced intracapsular fracture who:
• were able to walk independently out of doors with no more than the use of a
stick and
• are not cognitively impaired and
• are medically fit for anaesthesia and the procedure (A, B, C, X, and Z).
➢ Use extramedullary implants such as a sliding hip screw in preference to an

intramedullary nail in patients with trochanteric fractures above and including the
lesser trochanter (AO classification types A1 and A2) (A, B, C, and Z).
➢ Offer patients a physiotherapy assessment and, unless medically or surgically

contraindicated, mobilisation on the day after surgery (A, B, C, D, E, F, W, X, Y and Z).
➢ Offer patients mobilisation at least once a day and ensure regular physiotherapy
review (A, B, F, and W).
➢ From admission, offer patients a formal, acute, orthogeriatric or orthopaedic ward-

based Hip Fracture Programme that includes all of the following:
• orthogeriatric assessment
• rapid optimisation of fitness for surgery
• early identification of individual goals for multidisciplinary rehabilitation to

recover mobility and independence, and to facilitate return to prefracture
residence and long-term well-being.
• continued co-ordinated orthogeriatric and multidisciplinary review
• liaison or integration with related services, particularly mental health, falls

prevention, bone health, primary care and social services.
• clinical and service governance responsibility for all stages of the pathway of
care and rehabilitation, including those delivered in the community.
(A,B,C,D,E,F,W,X,Y and Z).
➢ Consider early supported discharge as part of the Hip Fracture Programme, provided
the Hip Fracture Programme multidisciplinary team remains involved, and the
patient:
• is medically stable and

• has the mental ability to participate in continued rehabilitation and
• is able to transfer and mobilise short distances and
• has not yet achieved their full rehabilitation potential, as discussed with the
patient, carer and family (A,B,C, E,F,W, and Z).
4.2 Full list of recommendations

Some aspects of hip fracture management are already covered by NICE guidance
and are therefore outside the scope of this guideline. In order to ensure
comprehensive management and continuity, the following NICE guidance should be
referred to when developing a complete programme of care for each patient::
osteoporotic fragility fracture prevention (TA 160, 161 & 204)234-236, falls (CG21)227, pressure
ulcers (CG29)228, nutrition support (CG32)229, dementia (CG42)230, surgical site infection
(CG74)231, venous thromboembolism (CG92)237 and delirium (CG103)230.
4.2.1 Imaging options in occult hip fracture
➢ Offer magnetic resonance imaging (MRI) if hip fracture is suspected despite negative
X-rays of the hip of an adequate standard. If MRI is not available within 24 hours or is
contraindicated, consider computed tomography (CT).
4.2.2 Timing of surgery
➢ Perform surgery on the day of, or the day after, admission.
➢ Identify and treat correctable comorbidities immediately so that surgery is not

delayed by:
• anaemia
• anticoagulation
• exacerbation of chronic chest conditions.
4.2.3 Analgesia
➢ Assess the patient’s pain:
36 HIP FRACTURE
• immediately upon presentation at hospital and
• within 30 minutes of administering initial analgesia and
• hourly until settled on the ward and
• regularly as part of routine nursing observations throughout admission.
➢ Offer immediate analgesia to patients presenting at hospital with suspected hip

fracture, including people with cognitive impairment.
➢ Ensure analgesia is sufficient to allow movements necessary for investigations (as

indicated by the ability to tolerate passive external rotation of the leg), and for
nursing care and rehabilitation.
➢ Offer paracetamol every 6 hours preoperatively unless contraindicated.
➢ Offer additional opioids if paracetamol alone does not provide sufficient preoperative
pain relief.
➢ Consider adding nerve blocks if paracetamol and opioids do not provide sufficient
preoperative pain relief, or to limit opioid dosage. Nerve blocks should be
administered by trained personnel. Do not use nerve blocks as a substitute for early
surgery.
➢ Offer paracetamol every 6 hours postoperatively unless contraindicated.
➢ Offer additional opioids if paracetamol alone does not provide sufficient

postoperative pain relief.
➢ Non steroidal anti-inflammatory drugs (NSAIDs) are not recommended.
4.2.4 Anaesthesia
➢ Offer patients a choice of spinal or general anaesthesia after discussing the risks and
benefits.
➢ Consider intraoperative nerve blocks for all patients undergoing surgery.
4.2.5 Planning the theatre team
➢ Schedule hip fracture surgery on a planned trauma list.
➢ Consultants or senior staff should supervise trainee and junior members of the
anaesthesia, surgical and theatre teams when they carry out hip fracture procedures.
4.2.6 Surgical procedures
➢ Operate on patients with the aim to allow them to fully weight bear (without
restriction) in the immediate postoperative period.
➢ Perform replacement arthroplasty (hemiarthroplasty or total hip replacement) in

patients with a displaced intracapsular fracture.
➢ Offer total hip replacement to patients with a displaced intracapsular fracture who:
• were able to walk independently out of doors with no more than the use of a
stick and
• are medically fit for anaesthesia and the procedure
➢ Use a proven femoral stem design rather than Austin Moore or Thompson stems for
arthroplasties. Suitable designs include those with an Orthopaedic Data Evaluation
Panel rating of 10A, 10B, 10C, 7A, 7B, 5A, 5B, 3A or 3B.
➢ Use cemented implants in patients undergoing surgery with arthroplasty.
➢ Consider an anterolateral approach in favour of a posterior approach when inserting a

hemiarthroplasty.
➢ Use extramedullary implants such as a sliding hip screw in preference to an

intramedullary nail in patients with trochanteric fractures above and including the
lesser trochanter (AO classification types A1 and A2).
➢ Use an intramedullary nail to treat patients with a subtrochanteric fracture.
4.2.7 Mobilisation strategies
➢ Offer patients a physiotherapy assessment and, unless medically or surgically

contraindicated, mobilisation on the day after surgery.
➢ Offer patients mobilisation at least once a day and ensure regular physiotherapy
review.
4.2.8 Multidisciplinary management

➢ From admission, offer patients a formal, acute orthogeriatric or orthopaedic ward-
based Hip Fracture Programme that includes all of the following:
• early identification of individual goals for multidisciplinary rehabilitation to

recover mobility and independence, and to facilitate return to prefracture
residence and long-term wellbeing.
• continued, coordinated, orthogeriatric and multidisciplinary review

• liaison or integration with related services, particularly mental health, falls
prevention, bone health, primary care and social services.
• clinical and service governance responsibility for all stages of the pathway of
care and rehabilitation, including those delivered in the community.
38 HIP FRACTURE
➢ If a hip fracture complicates or precipitates a terminal illness, the multidisciplinary

team should still consider the role of surgery, as part of a palliative care approach
that:
• minimises pain and other symptoms and
• establishes patients' own priorities for rehabilitation and
• considers patients' wishes about their end-of-life care.
➢ Healthcare professionals should deliver care that minimises the patient’s risk of
delirium and maximises their independence, by:
• actively looking for cognitive impairment when patients first present with hip
fracture
• reassessing patients to identify delirium that may arise during their admission
• offering individualised care in line with ‘Delirium’ (NICE clinical guideline 103).
➢ Consider early supported discharge as part of the Hip Fracture Programme, provided
the Hip Fracture Programme multidisciplinary team remains involved, and the
patient:
• has the mental ability to participate in continued rehabilitation and
• has not yet achieved their full rehabilitation potential, as discussed with the
patient, carer and family.
➢ Only consider intermediate care (continued rehabilitation in a community hospital or

residential care unit) if all of the following criteria are met:
• intermediate care is included in the Hip Fracture Programme and
• the Hip Fracture Programme team retains the clinical lead, including patient
selection, agreement of length of stay and ongoing objectives for
intermediate care and
• the Hip Fracture Programme team retains the managerial lead, ensuring that
intermediate care is not resourced as a substitute for an effective acute
hospital Programme.
➢ Patients admitted from care or nursing homes should not be excluded from
rehabilitation programmes in the community or hospital, or as part of an early
supported discharge programme.
4.2.9 Patient and carer information
➢ Offer patients (or, as appropriate, their carer and/or family) verbal and printed
information about treatment and care including:
• diagnosis
• choice of anaesthesia
• choice of analgesia and other medications
• surgical procedures
• possible complications
• postoperative care
• rehabilitation programme
• long-term outcomes
• healthcare professionals involved.
4.3 Research recommendations
The GDG identified the following priority areas for research:
• Imaging options in occult hip fracture
• Anaesthesia
• Displaced intracapsular hip fracture
• Early supported discharge
• Physiotherapy
4.3.1 Research recommendation on imaging options in occult hip fracture
➢ In patients with a continuing suspicion of a hip fracture but whose radiographs are
normal, what is the clinical and cost effectiveness of computed tomography (CT)
compared to magnetic resonance imaging (MRI), in confirming or excluding the
fracture?
Why this is important
The GDG’s consensus decision to recommend CT over a radionuclide bone scan as an

alternative to MRI to detect occult hip fractures reflects current NHS practice but assumes
that advances in technology have made the reliability of CT comparable with that of MRI. If
modern CT can be shown to have similar reliability and accuracy to MRI, then this has
considerable implications because of its widespread availability out of hours and lower cost.
40 HIP FRACTURE
It is therefore a high priority to confirm or refute this assumption by direct randomised

comparison. The study design would need to retain MRI as the ‘gold standard’ for cases of
uncertainty and to standardise the criteria, expertise and procedures for radiological
assessment. Numbers required would depend on the degree of sensitivity and specificity
(the key outcome criteria) set as target requirement for comparability, but need not
necessarily be very large.
4.3.2 Research recommendation on anaesthesia
➢ What is the clinical and cost effectiveness of regional versus general anaesthesia on
postoperative morbidity in patients with hip fracture?
No recent randomised controlled trials were identified that fully address this question. The
evidence is old and does not reflect current practice. In addition, in most of the studies the
patients are sedated before regional anaesthesia is administered, and this is not taken into
account when analysing the results. The study design for the proposed research would be
best addressed by a randomised controlled trial. This would ideally be a multi-centre trial
including 3000 participants in each arm. This is achievable given that there are about
70,000 to 75,000 hip fractures a year in the UK39. The study should have three arms that
look at spinal anaesthesia versus spinal anaesthesia plus sedation versus general
anaesthesia; this would separate those with regional anaesthesia from those with regional
anaesthesia plus sedation. The study would also need to control for surgery, especially type
of fracture, prosthesis and grade of surgeon.
A qualitative research component would also be helpful to study patient preference for
type of anaesthesia.
4.3.3 Research recommendation on displaced intracapsular hip fracture
➢ What is the clinical and cost effectiveness of large-head total hip replacement versus
hemiarthroplasty on functional status, reoperations and quality of life in patients with
displaced intracapsular hip fracture?
Large-head total hip replacement is a development of traditional total hip replacement,

where a larger head makes the joint more stable and hence reduces the risks of dislocation.
Three small trials have shown traditional small-head total hip replacement to have better
outcomes and function, albeit with an increased dislocation rate in selected groups of
patients. The drawback with large-head arthroplasty is the additional implant cost and
theatre time. This cost can account for up to 20% of current NHS tariff (up to £2000) and
the study aims to address whether this translates to improved patient outcome. The study
design for the proposed research would be best addressed by a randomised controlled trial.
This would have two arms to compare current standard care (using hemiarthroplasty) with
using large-head total hip replacement for patients sustaining displaced intracapsular hip
fractures. The primary outcome would be patient mobility at 1 year and secondary
outcomes would include functional outcomes, quality of life and cost effectiveness of the
intervention.
It would be expected that a sample size of approximately 500 patients would be required to
show a significant difference in the mobility, hip function and quality of life (assuming 80%
power, p < 0.05). By recruiting through a trauma research network it is estimated that 10
centres would be able to recruit 20 patients per month (from 45 eligible patients) giving a
recruitment period of 25 months.
4.3.4 Research recommendation on early supported discharge
➢ What is the clinical and cost effectiveness of early supported discharge on mortality,
quality of life and functional status in patients with hip fracture who are admitted
from a care home?
Residents of care and nursing homes account for about 30% of all patients with hip fracture
admitted to hospital. Two-thirds of these come from care homes and the remainder from
nursing homes. These patients are frailer, more functionally dependent and have a higher
prevalence of cognitive impairment than patients admitted from their own homes. One-
third of those admitted from a care home are discharged to a nursing home and one-fifth
are readmitted to hospital within 3 months. There are no clinical trials to define the optimal
rehabilitation pathway following hip fracture for these patients and therefore represent a
discrete cohort where the existing meta-analyses do not apply. As a consequence, many
patients are denied structured rehabilitation and are discharged back to their care home or
nursing home with very little or no rehabilitation input.
Given the patient frailty and comorbidities, rehabilitation may have a limited effect on
clinical outcomes for this group. The fact that they already live in a home where they are
supported by trained care staff, however, clearly provides an opportunity for a systematic
approach to rehabilitation. Early multidisciplinary rehabilitation based in care homes
ornursing homes would take advantage of the day-to-day care arrangements already in
place and provide additional NHS support to deliver naturalistic rehabilitation, where
problems are tackled in the patient’s residential setting.
Early supported multidisciplinary rehabilitation could reduce hospital stay, improve early
return to function, and affect both readmission rates and the level of NHS-funded nursing
care required.
The research would follow a two-stage design: (1) an initial feasibility study to refine the
selection criteria and process for reliable identification and characterisation of those
considered most likely to benefit, together with the intervention package and measures for
collaboration between the Hip Fracture Programme team, care-home staff and other
community-based professionals, and (2) a cluster randomized controlled comparison (with
two or more intervention units and matched control units) set against agreed outcome
criteria. The latter should include those specified above, together with measures of the
impact on care-home staff activity and cost, as well as qualitative data from patients on
relevant quality-of-life variables.
4.3.5 Research recommendation on physiotherapy
➢ What is the clinical and cost effectiveness of additional intensive physiotherapy

and/or occupational therapy (for example progressive, resistance training) after hip
fracture?

42 HIP FRACTURE
The rapid restoration of physical and self care functions is a critical to recovery from hip
fracture, particularly where the goal is to return to the patient to preoperative levels of
function and residence. Approaches that are worthy of future development and
investigation include progressive resistance training, progressive balance and gait training,
supported treadmill gait re-training, dual task training, and activities of daily living training.
The optimal time point at which these interventions should be started requires clarification.
The ideal study design is a randomised controlled trial. Initial studies may have to focus on
proof of concept and be mindful of costs. A phase III randomised controlled trial is required
to determine clinical effectiveness and cost-effectiveness. The ideal sample size will be
around 400 to 500 patients, and the primary outcome should be physical function and
health related quality of life. Outcomes should also include falls. A formal sample size
calculation will need to be undertaken. Outcomes should be followed over a minimum of 1
year, and compare if possible, either the recovery curve for restoration of function or time
to attainment of functional goals.
4.3.6 Additional research recommendations
The following research questions were selected by the GDG but were not prioritised in the
top five recommendations for research.
4.3.6.1 Analgesia
The GDG recommended the following research question:
➢ What is the clinical and cost effectiveness of preoperative and postoperative nerve
blocks in reducing pain and achieving mobilisation and physiotherapy goals sooner in
patients with hip fracture?
Nerve blocks may potentially find an important role in the management of hip fracture
pain, both pre- and postoperatively, because of their potential to reduce the requirement
for opioids and their associated unwanted effects. Economically there are considerations
for staff training, but also for the potential benefits in terms of duration of stay and early
mobilisation. It is not possible from the existing literature to determine this with any
confidence and there is a pressing need for a definitive trial comparing these outcomes
with nerve blocks against a defined protocol of systemic opioid use.
4.3.6.2 Timing of surgery

➢ What is the clinical and cost effectiveness of surgery within 36 hours of admission
compared to surgery later than 36 hours from admission in mortality, morbidity
and quality of life in patients with hip fracture?
Early and appropriate surgery for hip fractures is the most effective form of pain relief,
potentially quickening the rehabilitation and reducing complications. Within the current
literature no specific time interval threshold has been identified (up to 24hr) below which a
reduction in delay has shown no benefit. In addition to the evidence of the cost
effectiveness below 48hr, pragmatic, organisational and humanitarian considerations have

been utilised to arrive at the recommendation to operate not later than the day after
admission. A formal study within the NHS based on an arbitrary but realistic 36hr threshold
would provide additional important data to that already available, in order to inform more
precisely the forward clinical and cost-effectiveness of the strategy. For ethical reasons, the
research design would be an observational cohort study, correcting for confounding
variables, possibly set in the context of the National Hip Fracture Database and examining
the effect of the time to surgery and its cost on key outcomes, including mortality,
complications, length of stay, time taken to rehabilitate and qualitative aspects of the
experiences of patients.
4.3.6.3 Reverse oblique trochanteric

fractures
➢ What is the clinical and cost effectiveness of intramedullary versus extramedullary

total hip replacement on mortality, functional status and quality of life in patients
with reverse oblique trochanteric hip fracture?
Reverse oblique trochanteric fractures account for approximately 5 % of all trochanteric hip
fractures. This means it affects approximately over 1000 patients per year in the UK.
Presently there is little evidence as to which is the preferable implant (which can be either
extramedullary – outside the bone, or intramedullary - inside the bone). The potential
biomechanical advantage of intramedullary advantage may be offset by increased cost
(which can be over £1000 more expensive). A randomised trial comparing the two implants
using patient mobility, function and re-operation would allow a more informed choice of
treatment for this injury.
4.3.6.4 Designated hip fracture units

➢ What is the clinical and cost effectiveness of a designated hip fracture unit within
the trauma ward compared to units integrated into acute trusts on mortality,
quality of life and functional status in patients with hip fracture?
The increasingly structured approach to hip fracture care has led to a number of UK units
considering or establishing a specific ‘hip fracture ward’ as a specialist part of their acute
orthopaedic service.
Designated hip fracture wards may prove an effective means of delivering the whole
programme of coordinated perioperative care and multidisciplinary rehabilitation which
this NICE Guidance has proposed, but at present there is no high quality evidence of their
clinical effectiveness when compared to such care within general orthopaedic or trauma
beds.
It may not be practical to run an RCT within a trauma unit, but there is certainly potential
for cohort studies to explore the effect of such units on individual patients' mobility,
44 HIP FRACTURE
discharge residence, mortality and length of stay. Units considering the establishment of
hip fracture wards should be encouraged to consider performing such trials.
4.3.6.5 Care/nursing home residents

➢ Do patients admitted to hospital with a fractured hip who live permanently in a

care/nursing home have equal access to multidisciplinary rehabilitation as patients
admitted from home?
The existing literature on the effectiveness of multidisciplinary rehabilitation typically

excludes patients who live in care/nursing homes. From an equality perspective it
hypothesised that this group of people do not have access to the same multidisciplinary
rehabilitation as patients who are returning home as it is assumed patients returning to
care/nursing homes will have their care needs met by the home. The research design would
be a prospective observational cohort study to determine the extent and quality of
rehabilitation services available to this group in comparison to patients returning to their
own homes.
4.3.6.6 Patient and carer quality of life
➢ What quality of life value do individual patients and their carers place on different
mobility, independence and residence states following rehabilitation?
It is important in evaluating future priorities for intervention to determine whether the

perceived clinical and health economic benefits of rehabilitation outcomes in the research
literature are matched over the same time-frame by the quality of life judgements,
aspirations and expectations of patients themselves and their carers. There is currently no
evidence.
4.3.6.7 Patient experience

➢ What is the patient’s experience of being admitted to hospital with a hip fracture in
relation to surgery, pain management, timeliness of information given, and
rehabilitation?
No studies from NHS populations were identified where patients commented specifically on
their surgery, their pain management and rehabilitation programme. There were comments
in the patient views studies about not being kept informed about the management of their
condition, however there was no information identified about the appropriate time to be
told. It may be that different patients want the information at different times. The studies
suggest that patients suffer from fear, pain and delirium until after surgery and it is
important to learn what (if anything) can be done to alleviate this which for many will be
considered the worst stage in their treatment.
5 Imaging options in occult hip fracture
5.1 Introduction
The occult, or ‘hidden’, hip fracture is one in which the clinical findings are suggestive of a
fracture but this is not confirmed by radiographs.
Most hip fractures can be readily diagnosed using radiographs, consisting of an antero-
posterior (AP) and a lateral projection of the hip, whenever the clinical suspicion of a
fracture first arises. Importantly, no clinical decision rule has yet become available that
would allow clinicians to exclude a hip fracture without imaging. To avoid misdiagnosis
with hip pain being attributed erroneously to soft tissue injury and the patient being
discharged, a high index of clinical suspicion of hip fracture is required. This applies in all
patients presenting with a typical history - usually hip pain following trauma, e.g. a fall – as
certain typical features, such as the inability to bear weight or a shortened, abducted and
externally rotated leg, may be absent.
Achieving an accurate diagnosis as soon as possible is advantageous for a variety of

reasons. The primary reason is that without an accurate diagnosis it is not possible to
formulate a proper management plan. A fracture which is not obviously evident on
radiographs is likely to be undisplaced. Once the hip fracture is demonstrated early
diagnosis may allow for a simple procedure to fix the fracture in situ. Should it be confirmed
that no hip fracture is present then other diagnoses may be sought, there is less chance of
the patient being kept unnecessarily immobile and the patient may not need to stay in
hospital.
Hip radiographs have an estimated sensitivity of between 90% and 98%, and the initial films
will therefore miss only a small proportion of hip fractures. It is, however, essential to
ensure that the radiographs are of satisfactory quality. In particular, if the initial AP film of
the entire pelvis together with the lateral hip projection (taken in the position of comfort)
show no fracture, a third film is sometimes taken centred on the hip with the hip in 10
degrees of internal rotation to position the femoral neck at 90 degrees to the x-ray beam
and ensure an optimum view of this area. All subsequent discussion and recommendations
assume that clinicians suspect a fracture despite two or three radiographs of adequate
quality as detailed above.
The prevalence of occult hip fractures is estimated to be around 3 – 4%; up to 9% in some

series (though a proportion of this may reflect radiographs of inadequate standard as
46 HIP FRACTURE
discussed above). Bone resorption around the fracture site, or cortical displacement, will
render most occult hip fractures visible if radiographs are repeated after a few days. This is
due to bone resorption occurring along the fracture line making it radiographically more
obvious, but displacement or impaction may occur during this interval due to the patient
having walked with the fracture. Delays in surgery due to late diagnosis are associated with
prolonged suffering and poorer health outcomes for patients, and expose clinicians to the
risk of litigation.
Optimal strategy for patient selection and timing of secondary imaging strategies to ensure
early diagnosis of occult hip fractures, while avoiding over investigation of patients with
soft tissue injury only, is yet to be determined. However, the inability to weight bear on the
day following the injury, in spite of adequate analgesia, should prompt clinicians to re-
evaluate the patient and have a high index of suspicion of hip fracture.
Imaging modalities used to assist in the early detection of occult hip fractures include
computed tomography (CT), radionuclide scan (RNS), magnetic resonance imaging (MRI)
and, rarely, ultrasound scanning (US). The type of secondary imaging modalities used locally
is often determined by considerations of access, particularly outside normal working hours,
and radiological expertise available. MRI is usually considered to be the reference standard,
as numerous studies have found MRI to have the highest accuracy (100% sensitivity and
between 93% and 100% specificity, depending on experience and skill of radiologist
interpreting the images).
In this chapter we consider the clinical and cost-effectiveness of a number of alternative

imaging modalities that can be used to detect an occult hip fracture when MRI is
unavailable or precluded for safety or technical reasons.
5.2 Review question

In patients with a continuing clinical suspicion of hip fracture, despite negative radiographic
findings, what is the clinical and cost-effectiveness of additional imaging (radiographs after
at least 48 hours, RNS, US and CT, compared to MRI, in confirming, or excluding, a hip
fracture?
5.3 Radiographs
5.3.1 What is the diagnostic accuracy of additional radiographs (X-Rays) after 48 hours
compared to MRI in the diagnosis of occult hip fractures
Radiographs are the most widely available imaging technique (in- and out-of hours) utilised
for diagnosis of hip fracture. They can be acquired quickly (5 minutes) and experience in
image interpretation is widespread.
A hip fracture not visible on the original radiographs may become evident on films taken a
few days later because of bone resorption (reduced bone density) along the fracture line,
impaction (fracture line becomes more dense) or displacement.
IMAGING 47
5.3.1.1 Clinical evidence

No studies were identified.
5.3.1.2 Economic evidence.
5.3.1.3 Recommendations and link to

evidence
See Section 5.6.2
5.4 Radionuclide bone scan (RNS)

For a RNS of the skeleton a short-life radio-isotope (technetium 99m) is linked to methylene
diphosphonate (MDP) which is taken up in areas of bone formation (osteoblastic activity)
resulting in ‘hot spots’. The isotope is injected intravenously and then there has to be a
delay of three hours before scanning, using a gamma camera and which takes 30 minutes,
will detect increased uptake in the skeleton. Other causes of high bone turnover such as
arthritis, synovitis and tumor may lead to false positive results and these are more frequent
in patients over the age of 70. It is common practice to defer RNS until 72 hours after injury
to avoid false negative scans but some authors suggest that the modern three-phase
technique may give accurate results after only 24 hours.
5.4.1 What is the diagnostic accuracy of RNS compared to MRI in the diagnosis of occult
hip fractures
Two RCTs with a total of 99 particpants were identified. See Evidence Table 1, Appendix E.

Table 5-6: Bone scanning – Quality assessment
Outcome Number Design Limitations Inconsistency Indirectness Other
of considerations
studies
Diagnostic 2 Cross Serious No serious No serious Bone scanning
accuracy 88,286 sectional limitations inconsistency indirectness was carried out
(a), (b)
study up to 72 after
admission
(a) Evans 199488 study did not clearly report patient demographics
(b) Not clear who interpreted the results and whether they were blind to the results of the reference
standard test
Table 5-7: - Clinical summary of findings

48 HIP FRACTURE
Outcome Sensitivity Specificity NPV PPV Likelihood Likelihood Ratio Quality

(%) (%) (%) (%) Ratio (-ve)
(+ve)
Diagnostic 75-98 100 93-96 100 0 0.02-0.25 Low
accuracy
5.4.2 Economic evidence

No studies were identified. The cost of the procedures in England and Wales were
presented to the GDG: a category 3 RNS costs £205, and an MRI (one area, no contrast)
costs £206 (source: National schedule of reference costs 2008-09; NHS trusts and PCTs
combined).
5.4.2.1 Evidence statement(s)
Clinical The sensitivity of bone RNS compared to MRI ranged from 75% to 98% and
specificity was 100%. This means that between 2% and 25% of those who
have a fracture, the fracture will have been missed. However, all patients who
tested positively do actually have a fracture. (LOW QUALITY)
Economic No studies were identified on the cost-effectiveness of the diagnostic

accuracy of RNS compared to MRI in the diagnosis of occult hip fractures.
5.4.3 Recommendations and link to evidence
See section 5.6.2
5.5 Ultrasound (US)

In ultrasound (US) imaging a probe emits ultrasound waves which are reflected off surfaces
and recored to form the image. Good contact is required between skin and probe
(coupling), generally achieved with gel, but may be problematic if there is pain or soft tissue
swelling in the site being scanned, which may be the case in hip fracture. US is widely
available, both in- and out-of-hours, does not use ionising radiation and is relatively
inexpensive. However, it takes considerable skill and expertise to acquire optimum images
and for interpretation of the appearances. Currently this kind of US scanning is performed
by a minority of specialised musculo-skeletal radiologists in the UK.
Ultrasound scanning of the hip may detect bone surface changes, effusions or haemorrhage
in patients with fractures but the results are non-specific and usually require confirmation
by MRI or CT. The technique is highly operator-dependent.
IMAGING 49
5.5.1 Diagnostic accuracy of ultrasound (US) compared to MRI in the diagnosis of occult hip
fractures
One study with 30 participants was identified. See Evidence Table 1, Appendix E and forest plot
G2 in Appendix G
Table 5-8: Ultrasound (US) – Quality assessment

Outcome Number Design Limitations Inconsistency Indirectness Other
of considerations
studies
Diagnostic 1 Cross No serious No serious No serious Sonographic
accuracy297 sectional limitations inconsistency indirectness examinations
were performed
by highly
experienced
muskuloskeletal
radiologists
Table 5-9: Ultrasound (US) - Clinical summary of findings

Outcome Sensitivity Specificity NPV PPV Likelihood Likelihood Ratio Quality
(%) (%) (%) (%) Ratio (-ve)
(+ve)
Diagnostic 100 65 100 59 2.85 0 Moderate
Accuracy
5.5.1.2 Economic evidence

No studies were identified. The costs of the procedures in England and Wales were
presented to the GDG: ultrasound (US) costs £48 for a procedure lasting less than 20
minutes, and £62 for a procedure lasting more than 20 minutes. The cost of an MRI (one
area, no contrast) is £206 (source: National schedule of reference costs 2008-09; NHS trusts
and PCTs combined)
Clinical The sensitivity of ultrasound (US) compared to MRI was 100% and specificity
was 65%. This means that none of the patients who had a fracture have been
missed. However, of those who tested positive 35% do not actually have a
fracture – i.e. there is a high percentage of false positives (sonographic
abnormalities indistinguishable from those attributable to conditions other
than fracture) (LOW QUALITY)
accuracy of ultrasound (US) compared to MRI in the diagnosis of occult hip
fractures.
See section 5.6.2

50 HIP FRACTURE
5.6 Computed tomography (CT)

CT uses rings of sensitive detectors and an X-ray tube which rotates around the patient to
acquire transverse axial images through the body. CT is a readily available imaging
modality but its value for the detection of occult hip fractures has not been extensively
evaluated. There is evidence that undisplaced fractures running parallel to the axial plane
can be missed and limited resolution of osteoporotic trabecular bone may make the
technique less reliable for the detection of fractures of the hip than of other areas of the
body. However, technical developments in CT (spiral, multi-detector referred to as MDCT)
have enabled thin 2 dimensional (2D) sections to be acquired very rapidly and from which
3D volumetric reconstructions can be acquired and displayed at bone, or a variety of soft
tissue, settings. This has greatly enhanced the potential application of CT to imaging occult
hip fractures. The scan is rapid (2minutes)(slice thickness 1.25mm; MAs between 100 to
355 depending on patient size/weight; field of view 36cm) and from which coronal, sagittal
and other planar/3D reformations can be generated. CT is particularly good for imaging
bone, but does not show the marrow changes (oedema) which occur in hip fracture
adjacent to the fracture line.
No studies that meet our inclusion criteria were identified.

No studies were identified. The costs of the procedures in England and Wales were
presented to the GDG: the cost for a CT scan (one area, no contrast) is £101. The cost of an
MRI (one area, no contrast) is £206 (source: National schedule of reference costs 2008-09;
NHS trusts and PCTs combined)
Clinical No studies were identified directly comparing the diagnostic accuracy of CT

with MRI and that meet our inclusion criteria.
accuracy of CT compared to MRI in the diagnosis of occult hip fracture.
Recommendation Offer magnetic resonance imaging (MRI) if hip fracture is

suspected despite negative X-rays of the hip of an adequate
standard. If MRI is not available within 24 hours or is
contraindicated, consider computed tomography (CT).
Relative values of different Reliability (in terms of diagnostic accuracy) was considered the
outcomes primary outcome of interest. A false positive diagnosis carries the
risks either of unnecessary surgery or of delay and increased cost
caused by the need for additional radiographic investigation; a
false negative result carries the risks associated with subsequent
fracture displacement and its consequences as well as avoidable
IMAGING 51
prolonged immobility and pain. It is therefore important for the

selected method to minimise both false positives and false
negatives.
Trade off between clinical MRI cannot be used in patients with certain types of metallic
benefits and harms implants but does not otherwise have known harmful effects other
than the potential to cause claustrophobia due to the need for
patients to remain in a confined space for a considerable length of
time. MRI was considered to be the first choice option in view of
its superior diagnostic accuracy (up to 100% specificity and
sensitivity).
If limitations in the local availability of MRI lead to unacceptably
prolonged delay to diagnosis offering an RNS or CT may have a net
benefit to the patient even though both carry the risks of exposure
to ionising radiation. A delay of several days may, however, be
required for RNS to achieve the required sensitivity, it is also
generally unavailable out-of-hours (a further cause of delay), and
may provide less precise information for surgical planning.
Repeat radiographs after 48 hours have limited sensitivity and
carry the risks of displacement during the intervening period, as
well as those of delay to surgery.
Ultrasound (US) has no known harms but it’s low specificity means
that further imaging confirmation (with resulting delay) is required
to determine whether a positive US represents a fracture, thus
limiting its use. Conversely, a negative US reliably excludes
fracture and could in theory enable immediate discharge of this
small subset of patients from Emergency departments.
The advent of MRI has enabled the accurate early identification of
occult hip fractures that would previously have been missed. The
precise natural history of such occult fractures (and therefore the
precise place of surgical intervention) has therefore only begun to
be fully clarified. It is at least theoretically possible that a
proportion of occult fractures might not require surgery. At the
same time techniques of fracture fixation have also become less
traumatic and invasive. Unless and until these issues of
benefit/harm are fully resolved, precise and reliable early diagnosis
as a basis for surgical decision making remains a clinical priority.
Economic considerations In England and Wales, the cost of a radionuclide scan (RNS) and of
an MRI is very similar: a category 3 RNS costs £205, and an MRI
(one area, no contrast) costs £206. However, an MRI is cost saving
compared to an RNS, as the latter may result in a longer length of
hospital stay (and the possible consequences of delay to surgery)
before the fracture is diagnosed.
The GDG also considered MRI to be cost-effective compared to US,
since in the case of a positive US, its low specificity would still
necessitate additional imaging (notably MRI or CT) to confirm the
diagnosis. The possible consequences of delay to surgery would
need to be added to those of additional imaging.
Quality of evidence Two cross sectional studies comparing RNS to MRI were identified.
52 HIP FRACTURE
These studies had serious methodological limitations due to the

limited reporting of patient demographics and lack of clarity as to
whether the assessors were blinded to the results of the index test
when interpreting the results of the reference standard and vice
versa.
One cross sectional study comparing ultrasound (US) to MRI was
identified. This study was of moderate quality. The GDG considered
that the reproducibility was a potential limitation as the
sonographic readings were performed by highly experienced
muskuloskeletal radiologists.There were no serious inconsistencies
or indirectness in any of the identified studies.
The assumption that MRI is the gold standard for detecting occult
hip fracture and the recommendation advising use of CT as an
alternative to MRI were based on unanimous GDG consensus.
Other considerations The diagnosis and management of occult hip fracture is still very
much an evolving area of practice. In the absence of an evidence-
based clinical decision rule clinicians must exert clinical judgement
to decide when suspicion of hip fracture after normal plain
radiographs is great enough to warrant additional imaging.
Before radiographs are regarded as excluding a hip fracture one
should ensure that radiographic quality is optimized. When AP
pelvic or hip radiographs are performed the leg should be a little
internally rotated with the great toes of the feet overlapping so as
to bring the anteverted femoral neck parallel to the X-ray table. In
this position little of the lesser trochanter should be visible medial
to the femoral cortex (the more externally rotated is the leg the
more obvious is the lesser trochanter). Optimising the positioning
enables the greater trochanter to be better visualized and not
obscured behind the femur. When a hip fracture is present it may
prove impossible to position the leg in this optimum position
because of pain, but this may be compensated for by appropriate
X-ray tube angulation. It should also be ensured that the X-ray
exposure factors are optimum to demonstrate both the entire
pelvis, to check that fractures are not present in sites additional to
the hip, and also for the hip suspected of fracture. To attain this
separate exposures and radiographs may be required.
Whilst the GDG considered that MRI was the best test to use to
detect occult hip fracture and that this should be the first choice,
they noted that there may be occasions where MRI is not available
and thought it was important to give guidance as to which test to
use in these circumstances. The GDG’s consensus decision to
recommend CT over RNS is based on greater availability, especially
outside the working week, and shorter delay to diagnosis. It also
reflects current NHS practice.
In addition, the technical aspects of RNS of bone (a 3 hour delay
after radionuclide is given until gamma emission can be recorded;
also increased uptake of radionuclide depends on increased
osteoblastic activity which may take several days to occur following
fracture; lack of availability out of hours) makes this the least
IMAGING 53
appropriate now for imaging occult hip fractures and is now not
often used in this scenario, since the advent of CT and MRI.
The GDG were also aware that rapid advances in CT technology,
such as 64-slice scanners and sophisticated 3 dimensional
reconstruction algorithms, may well overcome the limitations of CT
reported in the published literature about its value for detection of
occult hip fractures.
5.7 Research recommendation on imaging options in occult hip

fracture
In patients with a continuing suspicion of a hip fracture but whose radiographs are normal,
what is the clinical and cost effectiveness of computed tomography compared to magnetic
resonance imaging, in confirming or excluding the fracture?

that advances in technology have made the reliability of CT comparable to that of MRI. If
modern CT indeed can be shown to have similar reliability and accuracy to MRI, then this
has considerable implications because of its widespread availability out of hours and lower
cost. It is a high priority, therefore, to confirm or refute this assumption by direct
randomised comparison. The study design would need to retain MRI as “gold standard” for
cases of uncertainty and would clearly need to standardise the criteria, expertise and
procedures for radiological assessment. Numbers required would depend on the degree of
sensitivity/specificity (the key outcome criteria) set as target requirement for comparability,
but need not necessarily be very large.
6 Timing of surgery
6.1 Introduction
The timing of treatment for patients sustaining fractures of the proximal femur remains one
of the biggest challenges to a health care system. It involves multidisciplinary co-ordination
between accident and emergency departments, acute orthopaedic trauma services,
orthogeriatricians, anaesthetists, as well as the availability of appropriate theatre space
54 HIP FRACTURE
with trained staff and relevant equipment. In the past these patients were given low
priority in the hospital system, which led to many delays and repeated periods of
starvation. It is recognised that it is not only the time a patient takes to get to surgery that
is important, but that the patient has to be medically optimised, with the anaesthetic,
surgical and theatre team being appropriately experienced. When planning any emergency
care it is not always possible to predict the number of cases which can present, so any
system which is set up must have the flexibility to adapt to the peaks and troughs of
admissions. This can lead to potential free theatre capacity in quieter periods.
As it would be unethical to enforce an unnecessary delay for patients sustaining fractures of

the proximal femur, all studies reported are retrospective cohort studies. As such the level
and quality of the evidence is poor.
The timing of surgery is an early marker of a patient's progress following a hip fracture. The
surgery does not stand alone. The pathway to safe, timely surgery includes proper
organisation and expertise in diagnosis, medical optimisation and anaesthesia. In the last
decade many orthopaedic trauma emergencies are now treated on dedicated planned
trauma lists. A planned trauma list is one with a rostered senior anaesthetist, senior
surgeon and dedicated theatre time. These by their nature usually concentrate the
expertise required.
There are sometimes legitimate reasons for delay and it is important to look at the
excluded patients in these studies. In a few patients delay to surgery is unavoidable.
However, it should be anticipated that many patients with hip fractures will be frail and
have comorbidities. The following would be common findings in patients presenting with
hip fractures:
• Anaemia
• Anticoagulation
• Volume depletion
• Electrolyte imbalance
• Uncontrolled diabetes
• Uncontrolled heart failure
• Correctable cardiac arrhythmia or ischaemia
• Acute chest infection
• Exacerbation of chronic chest conditions
Provided these problems are sought and measures initiated to correct them are taken
promptly the majority can be optimised within 24 hours.
When looking at the timings measured it is generally accepted the time of diagnosis should
be the initial time recorded and the time to the start of the anaesthetic procedure be the
index time measured. Objective outcomes used to compare timing of surgery include early
and late mortality, length of hospital stay, return to mobility, complications including chest
infections and pressure sores, change of residence and other surgical complications. What
has not been measured in the past is the pain and suffering experienced with prolonged
TIMING OF SURGERY 55
delay and what is the ethical time period the elderly, who are often very frail, should wait
for treatment.
6.1.1 Review question
In patients with hip fractures what is the clinical and cost effectiveness of early surgery
(within 24, 36 or 48 hours) on the incidence of complications such as mortality, pneumonia,
pressure sores, cognitive dysfunction and increased length of hospital stay?
10 studies met the inclusion criteria for this question, with a total of 193,793 participants.
Data are given for studies where outcomes have been adjusted for confounding factors
such as comorbidity and age using logistic regression (7 studies). A separate subgroup is
given which excludes patients who are unfit for surgery i.e. reason for delay is due to
unavailability of staff, theatres or equipment (3 studies). Delay to surgery in the identified
studies was from time to admission. All studies report surgical delay versus early surgery to
investigate the harm of delaying surgery.
The cut-off for delay to surgery in this analysis is 24, 36 and 48 hours.
See evidence table 2, Appendix E and forest plots G2 to G22 in Appendix G.

56 HIP FRACTURE
Table 6-10: Late (>24h) versus early surgery for hip fracture – Clinical study characteristics
Outcome Numbe Desig Limitations Inconsistency Indirectness Other
r of n considerations/
studies imprecision
Mortality – In 2 Obser No serious No serious No serious Serious
hospital19,351 vation limitations inconsistency indirectness imprecision (e)
(b, d)
al
Mortality – 30 2 Obser Serious No serious No serious Serious
days30 vation limitations inconsistency indirectness imprecision (e)
(a) (a, b, d)
al
Mortality – 3 1 Obser No serious No serious No serious Serious
months351 vation limitations inconsistency indirectness (b) imprecision (e)
al
months4 vation limitations inconsistency indirectness imprecision (e)
al
year351 vation limitations inconsistency indirectness (b) imprecision (e)
al
Return to 1 Obser No serious No serious No serious Serious
independent vation limitations inconsistency indirectness imprecision (e)
4
living al
4
Pressure ulcers 1 Obser No serious No serious No serious No serious
vation limitations inconsistency indirectness imprecision
al
Major 1 Obser No serious No serious No serious No serious
complications (c) vation limitations inconsistency indirectness (d) imprecision
19
al
(a) In Bottle and Aylin, 2006 30 baseline data, such as age is given for the entire cohort and also
stratified by type of surgery e.g. fixation, replacement, other procedure. No baseline data stratified
by delay to surgery. Patients were all admitted from their own home.
(b) In Weller et al., 2005351 baseline data, such as age is stratified per hospital.No baseline data
stratified by delay to surgery.
(c) Severe complications were defined as cerebrovascular accident, cardiorespiratory complications,
digestive complications except unspecific paralytic ileus, and dialysis.
(d) The comparison is 24-48h vs. 0-24 h time to surgery for Bergeron 200619
(e) The wide confidence intervals around the estimate make it difficult to determine and effect size for
this outcome.
Table 6-11: Late (>24 hours) versus early surgery for hip fracture - Clinical summary of findings
Late Adjusted Odds
Outcome surgery(a) Early surgery (a)
Ratio Absolute effect Quality
Mortality – in hospital
325 523 0.88 (0.55 - 1.41) N/A Very low
25320 20303 1.17 (1.08 - 1.26) N/A Low
Mortality – 30 days
45862 69080 1.25 (1.19 - 1.31) N/A Very low
Mortality – 3 months
25320 20303 1.11 (1.05 - 1.17) N/A Very low
225 209 1.07 (0.67 - 1.70) N/A Very low
Mortality – 1 year
25320 20303 1.13 (1.05 - 1.22) N/A Very low
Return to
independent living 225 209 0.86 (0.45 - 1.65) N/A Very low
Pressure ulcers
225 209 2.19 (1.21 - 3.96) N/A Low
Major complications 325 523 0.87 (0.58 - 1.29) N/A Low
(a) Numbers of patients in each study arm. No event data is given as the data provided is odds ratios
adjusted using logistic regression for confounding factors.
studies imprecision
Mortality – in 1 Obser No serious No serious No serious Serious
hospital vation limitations inconsistency indirectness imprecision (a)
189
al (a)
Minor 1 Obser No serious No serious No serious Serious
complications vation limitations inconsistency indirectness imprecision (a)
189
al (a)
Major 1 Obser No serious No serious No serious Serious
complications vation limitations inconsistency indirectness imprecision (a)
189
al (a)
Pressure ulcers 1 Obser No serious No serious No serious Serious
189
vation limitations inconsistency indirectness imprecision (a)
al (a)
months vation limitations inconsistency indirectness imprecision (a)
4
al
Pressure ulcers 1 Obser No serious No serious No serious No serious
4
vation limitations inconsistency indirectness imprecision
al
independent vation limitations inconsistency indirectness imprecision (a)
living al
4
(a) Baseline data given for entire cohort not by time to surgery.
(b) Late surgery is between 24-48h with early surgery defined as <24h.
(a) The wide confidence intervals around the estimate make it difficult to determine and effect size for
this outcome.
58 HIP FRACTURE
Late Early Adjusted Odds
Outcome surgery(a) surgery(a) Ratio Absolute effect Quality
264 245 0.82 (0.42 - 1.62) N/A Very low
Minor complications
264 245 1.53 (1.05 - 2.22) N/A Very low
Major complications
264 245 0.96 (0.52 - 1.75) N/A Very low
Pressure ulcers
264 245 1.23 (0.71 - 2.12) N/A Very low
194 550 1.5 (0.63 – 1.74) N/A Very low
Pressure ulcers
194 550 3.42 (1.94 – 6.03) N/A Low
Return to
independent living 194 550 0.44 (0.21 – 0.91) N/A Very low
studies imprecision
Mortality – In 3 Obser No serious No serious No serious Serious
hospital19,189,351 vation limitations inconsistency indirectness imprecision (e)
(b,d)
al
Mortality – 30 2 Obser Serious No serious No serious Serious
days 30,125 vation limitations inconsistency indirectness imprecision (e)
(a)
al
Mortality – 3 1 Obser No serious No serious No serious No serious
months vation limitations inconsistency indirectness (b) imprecision
351
al
months vation limitations inconsistency indirectness imprecision (e)
4
al
Mortality – 1 year 1 Obser No serious No serious No serious No serious
351
vation limitations inconsistency indirectness (b) imprecision
al
independent vation limitations inconsistency indirectness imprecision (e)
living al
4
Pressure 3 Obser No serious No serious No serious No serious

ulcers4,125,189 vation limitations inconsistency indirectness imprecision
al
Major 2 Obser No serious No serious No serious Serious
complications vation limitations inconsistency indirectness (d) imprecision (e)
(c)19,189
al
Minor 1 Obser No serious No serious No serious No serious
complications 189 vation limitations inconsistency indirectness (d) imprecision
al
(a) In Bottle and Aylin, 2006 30 baseline data, such as age is given for the entire cohort and also
stratified by type of surgery e.g. fixation, replacement, other procedure. No baseline data stratified
by delay to surgery.Patients were all admitted from their own home.
(b) In Weller et al., 2005351 baseline data, such as age is stratified per hospital.No baseline data
stratified by delay to surgery.
(c) In Bergeron 200619, severe complications were defined as cerebrovascular accident,
cardiorespiratory complications, digestive complications except unspecific paralytic ileus, and
dialysis.
(d) The comparison is >48h vs. 0-24 h time to surgery
(e) The wide confidence intervals around the estimate make it difficult to determine the effect size for
this outcome.
Late Early Adjusted Odds
Outcome surgery(a) surgery(a) Ratio Absolute effect Quality
Mortality – In hospital
19 129 848 1.16 (0.64 - 2.13) N/A Very low
189 98 509 0.93 (0.38 - 2.33) N/A Very low
Mortality – In hospital
351 7314 20303 1.60 (1.42 - 1.80) N/A Low
Mortality – 30 days 30 24391 90551 1.36 (1.29 - 1.43) N/A Very low
125
Mortality – 30 days 3805 4578 0.71 (0.45 - 1.10) N/A Very low
Mortality – 3 months 7314 20303 1.40 (1.28 - 1.54) N/A Low
Mortality – 4 months 98 646 0.86 (0.44 - 1.69) N/A Very low
Mortality – 1 year 7314 20303 1.58 (1.26 - 1.99) N/A Low
Return to
98 646 0.33 (0.14 - 0.78) N/A Very low
independent living
Pressure ulcers 4 98 646 4.34 (2.34 - 8.04) N/A Low
Pressure ulcers 125 3805 4578 1.20 (0.9 - 1.6) N/A Very low
Pressure ulcers 189 98 509 2.29 (1.19 - 4.40) N/A Low
Major complications
19 129 848 1.32 (0.79 - 2.20) N/A Very low
Major complications
189 98 509 2.21 (1.01 - 4.34) N/A Very low
Minor complications 98 509 2.27 (1.38 - 3.72) N/A Low
Table 6-16: Late (>48h) versus early surgery for hip fracture (length of hospital stay outcomes)–
Clinical study characteristics
studies imprecision
Postoperative 1 Obser No serious No serious No serious No serious
length of hospital vation limitations inconsistency indirectness imprecision
stay19 al (a)

(a)
stay; without al
comorbidity19
(a)
stay (including al
60 HIP FRACTURE

studies imprecision
rehab)308
(a) Mean and standard deviations are not provided, only median or mean and 95% confidence interval.
Table 6-17: Late (>48h) versus early surgery for hip fracture - Clinical summary of findings;
length of hospital stay
Late Early Median (days) Median (days)
Outcome surgery(c) surgery(c) Late surgery Early surgery Quality
Postoperative length
129 848 28 18 Low
of hospital stay (a)
of hospital stay; 30 248 20 16 Low
without comorbidity
of hospital stay 174 3454 36.5 (b) 21.6 (b) Low
(including rehab)
(a) Data is unadjusted for co-morbidity, which is more frequent in the delayed surgery study arm.
(b) Mean number of days given, 95% confidence interval = 5.7 to 16.0, p < 0.0001.
(c) Numbers of patients in each study arm. No event data is given as the data provided is odds ratios
Table 6-18: Late (>24h) versus early surgery for hip fracture (exclusion of patients unfit for
surgery) – Clinical study characteristics
studies imprecision
Mortality 30 days 1 Obser Serious No serious No serious Serious
215
vation limitations inconsistency indirectness imprecision (c)
(a, b)
al
Mortality and 1 Obser Serious No serious No serious Serious
needing total vation limitations inconsistency indirectness imprecision (c)
(a)
assistance in al
locomotion at 6
months
250
Major 1 Obser Serious No serious No serious Serious

postoperative vation limitations inconsistency indirectness imprecision (c)
(a)
complications al
250
(a) Baseline data not reported separately for the restricted cohort.
(b) No protocol for determining which patients were unfit for surgery and anaesthesia, therefore
variation between clinicians.
(c) The wide confidence intervals around the estimate make it difficult to determine and effect size for
this outcome.
Table 6-19: Late (>24 hours) versus early surgery for hip fracture (exclusion of patients unfit for
surgery) - Clinical summary of findings
Early
Outcome Late surgery surgery Risk Ratio Absolute effect Quality
Mortality 30 days 85/1166 85/982 0.84 (0.63 - 1.12) N/A Very low
Mortality and needing
total assistance in
509 0.62 (0.35 -1.08) (a) N/A Very low
locomotion at 6
months
Major postoperative 0.26 (0.07 – 0.95)
273 (a) N/A Very low
complications
(a) Adjusted odds ratio
62 HIP FRACTURE
Table 6-20: Late (>48h) versus early surgery for hip fracture (exclusion of patients unfit for
surgery) – Clinical study characteristics
studies imprecision
Mortality 30 1 Obser Serious No serious No serious Serious
days215 vation limitations inconsistency indirectness imprecision (c)
(a, b)
al
Mortality at 1 1 Obser Serious No serious No serious No serious
year308 vation limitations inconsistency indirectness imprecision
(a)
al
Change in 1 Obser Serious No serious No serious Serious
residence (more vation limitations inconsistency indirectness imprecision (c)
308 (a)
dependent) al
Return to original 1 Obser Serious No serious No serious No serious
308
residence vation limitations inconsistency indirectness imprecision
(a)
al
(a) Baseline data not reported separately for the restricted cohort.
(b) No protocol for determining which patients were unfit for surgery and anaesthesia, therefore
variation between clinician decisions.
(c) The wide confidence intervals around the estimate make it difficult to determine and effect size for
this outcome.
Table 6-21: Late (>48 hours) versus early surgery for hip fracture (exclusion of patients unfit for
surgery) - Clinical summary of findings
Early
Outcome Late surgery surgery Risk Ratio Absolute effect Quality
Mortality 30 days 36/497 134/1651 0.89 (0.63 – 1.27) N/A Very low
Mortality at 1 year 24/174 238/3454 0.5 (0.34 – 0.74) N/A Very low
Change in residence
22/174 240/3454 0.55 (0.37 – 0.83) N/A Very low
(more dependent)
Return to original
128/174 2974/3454 1.17 (1.07 – 1.28) N/A Very low
residence
One study304,304 was found which calculated the mean hospital costs for hip fracture
patients who had received surgery at different points in time from admission. This study
was excluded because of serious methodological limitations, as no reason was given as to
why patients had faced delays before receiving surgery (whether it was because of medical
or administrative reasons)
An original decision analytical model was developed to compare the cost-effectiveness of a

strategy consisting in adding extra half-day operating lists to increase the proportion of
patients operated within 48 hours from admission against a non-investment strategy.
Please see Appendix H, section 20.5 for further details.
Table 6-22: Early versus late (>48h) surgery for hip fracture - Economic study characteristics
Study Limitations Applicability Other Comments
NCGC decision model Minor limitations (a) Partial applicability (b)
(a) Cost-effectiveness analysis based on a Markov model.
(b) The findings of the model may not be generalized to the whole UK NHS because its treatment
effects and cost data are based on evidence from two specific hospital settings. The addition of
extra operating lists may not be feasible for those providers where no spare theatre capacity is
available.
Table 6-23: Early versus late (>48h) surgery for hip fracture - Economic summary of findings
Incremental effects
Study Incremental cost (£) (QALYs) ICER Uncertainty
NCGC 1) £1,000 for the first 1) 0.0425 for the first 1) £22,542/QALY 95% CI: cost saving –
decision year of implementation year of for the first year of dominated (both in
model of extra operating lists (a) implementation of implementation of the first and in the
extra operating lists extra operating lists second year of
implementation of
2) £ 800 for the second extra operating lists
(d)
year of implementation 2) 0.094 for the second 2) £8,933/QALY for
of extra operating lists (b) year of the second year of
implementation of implementation of
extra operating lists (c) extra operating lists
(a) In the first year of implementation of extra operating lists, the mean costs for investment in extra
operating lists early surgery were £47.4, and for the non-investment strategy £46.4.
(b) For the second year, the mean costs associated with the strategy of investment for early surgery were
£47.3, and for the non-investment strategy £46.4.
(c) In the first year of implementation of extra operating lists, the mean effectiveness for the strategy of
investment for early surgery was 2.3637, and for the non-investment strategy 2.3212. In the second
year, they corresponded to 2.415 and 2.321 respectively.
(d) 95% CI of ICERs calculated from the 10,000 Monte Carlo simulations. The high uncertainty of the
model is due to all the types of variables, including the effectiveness of interventions. We have tested
the uncertainty of all categories of inputs in the model (costs, utilities, relative risks), by making
probabilistic one category at a time while keeping the others deterministic, and under all scenarios
the findings showed great uncertainty, with a 95% CI cost saving – dominated”.
6.1.1.3 Evidence statement (s)
Clinical All patients

Early surgery (<24h) shows a statistically significant and clinically significant
reduction in mortality (in 4 out of 7 studies) (VERY LOW QUALITY) and
reduction in pressure ulcers (LOW QUALITY) with early surgery compared to
late surgery. No statistically significant difference shown for return to
independent living or major complications (LOW QUALITY).
Early surgery (<36h) – statistically significant and clinically significant
reduction in pressure ulcers with early surgery compared to late surgery
(LOW QUALITY). Statistically significant, but not clinically significant increased
return to independent living (VERY LOW QUALITY). No statistically significant
difference in mortality at 4 months (VERY LOW QUALITY).
Early surgery (<48h) shows a statistically significant and clinically significant
reduction in mortality (in 4 out of 8 studies) (VERY LOW QUALITY), increased
return to independent living (VERY LOW QUALITY), reduced pressure ulcers
(LOW QUALITY), reduced major and minor complications with early surgery
compared to late surgery (VERY LOW QUALITY).
Exclusion of patients unfit for surgery
64 HIP FRACTURE
Early surgery (<24h) – Statistically significant, but not clinically significant

reduction in major postoperative complications with early surgery compared
to late surgery. No statistically significant difference in mortality, with early
surgery compared to late surgery. (VERY LOW QUALITY)
Early surgery (<48h) – Statistically significant, and clinically significant
reduction in mortality at 1 year and patients changing residence (more
dependent) and increased return to original residence (VERY LOW QUALITY).
No statistically significant difference in mortality at 30 days with early surgery
compared to late surgery. (VERY LOW QUALITY).
Economic Investing in adding extra operating lists as a way to increase the proportion of
patients operated within 48 hours from admission is only marginally above
the £20k/QALYs threshold in the first year of implementation, but becomes
clearly cost-effective from the second year onwards.
This evidence has minor limitations and partial applicability.
Recommendation Perform surgery on the day of, or the day after, admission.
Relative values of different The GDG recognised that hip fracture surgery was often
outcomes disproportionately delayed in comparison with other operations,
and that this in part reflected a lack of sufficient priority afforded
to this group of patients.
On humanitarian criteria alone, initiatives to avoid delay were
considered to be of high priority in developing the guidance. It was
considered that surgery was the best form of pain relief, and that
to spend more than one night in hospital without operation was
generally unacceptable.
Postponement of surgery carries increased risk of complications, as
well as prolongation of pain, and the need for repeated
preoperative fasting.
Of the outcomes derived from the literature, mortality, return to
independent living, occurrence of specific complications (notably
pressure ulcers) and duration of hospital stay were all considered
of parallel and inter-related importance as indicators of care
standard and efficacy.
Trade off between clinical There was no instance in the literature of any advantage in
benefits and harms delaying surgery, nor of disadvantage in reducing delay.
Although the range of studies utilised a range of arbitrary or
pragmatic time thresholds (governed to some degree by service
context and organisation), there was no definitive cut-off point (up
to and including 24 hours) beyond which further reduction of delay
ceased to confer measurable benefit in one or more outcomes.
Therefore the GDG considered it could not be prescriptive about
the precise time threshold from the literature alone.

The trade off between early surgery and harms relate to the
difficulties and infrastructure required to treat this population who
present as emergencies. It is recognized surgery is the best form of
analgesia and as over 30% present with cognitive impairment, it
can be otherwise difficult to assess patients suffering. It is also
considered humane not to leave this frail patient group waiting
treatment (often being repeatedly starved). The potential harm of
earlier surgery include the risks of not medically resuscitating and
optimizing the patients health prior to a further surgical insult and
ensuring the surgical team is experienced and available. A delay up
to 36 hours allows for appropriate assessment and planning. It
allows patients to be operated on in planned trauma lists and
should allow most hospitals to cope with peaks in emergency
admissions.
Only one study 4 looked at complications, return to independent
living and pressure sores. Whilst this study did report a small
benefit in protecting against pressure sores it did not demonstrate
any additional benefits. Regarding mortality one study 351 showed a
small difference in mortality at one year, though again the
difference and numbers were small.
Alani et al., 20084 is the only study which looked at the 36 hour
time frame. It failed to show improvement in mortality at four
months yet showed a slight benefit in return to independent living
and avoidance of pressure ulcers.
When comparing surgery at 48 hours, again the data is limited. The
overall number of patients included is small and there is a reported
decrease in mortality in two out of the five studies included30,351.
Apart from the benefits already reported in Alani’s study, other
outcomes were either not reported or did not show any difference.
Economic considerations To be able to offer surgery for hip fracture patients by an
experienced surgical team, within the recommended time period,
it is recognized there may have to be an investment in
infrastructure, specifically planned trauma operating lists with
experienced surgical, anaesthetic and theatre teams. Generally
these should occur in the normal working day. As admission
numbers, including peaks and troughs, cannot be always predicted
then this capacity may not always be utilised.
The potential costs of reducing delay to surgery were recognised-
such as additional theatre time, out-of-hours staffing (including
senior staff), out of-hours lists and planned trauma lists.
These costs will be at least partially offset by potential savings from
reduced length of stay, reduced complications and enhanced
return to independent living.
There was no definitive health economic study for any time
threshold in the literature. The guideline group therefore
considered that an original decision model was crucial to inform
the broad economic feasibility of any recommendation on reducing
66 HIP FRACTURE
surgical delay. As discussed in Appendix H, the GDG agreed that,

out of the evidence included in the clinical review, the outcome
data to undertake this analysis were adequate only to provide a
model based on a 48hr threshold, and as a consequence this
specific cut-off point was selected for the economic analysis.
The economic model demonstrates that investing to add extra
operating lists in order to undertake surgery within 48 hours from
admission is only marginally above the £20k/QALYs threshold in
the first year of implementation, but becomes clearly cost-effective
in the following years.
Furthermore, the implementation of extra operating lists will also
achieve a more equitable distribution of health care resources in
favour of patients that had previously been made to wait for
surgery as other cases were given higher priority.
However, the model does not capture the possibility that the extra
operating lists could potentially be used to treat cases in addition
to hip fracture patients (thus resulting in an increase of activity for
the hospital trust and subsequent QALYs gains for the patients
treated).
In addition, our cost-effectiveness estimates are also conservative
in that we do not look at the impact that early surgery has on the
pain relief of our population.
Quality of evidence The available clinical evidence covering this issue is of low quality,
but in aggregate supports the avoidance of surgical delay.
For this reason there is an element of consensus in the wording of
the recommendation which, in addition to the evidence of clinical
benefit and NHS economic feasibility, also reflects a strong
humanitarian case. The consensus was unanimous within the GDG.
The health economic analysis reported in Appendix H showed that
surgery performed with 48 hours was cost effective.
Although the evidence base for this question is predominantly
retrospective, cohort studies of low quality (all low or very low) it is
not considered ethical to conduct an RCT to answer this question.
The main studies included were cohort studies that adjusted for
confounding factors by logistic regression, which although were
low quality were considered higher quality than cohort studies
without any adjustment. The subgroup studies did not adjust for
confounding factors, but were considered as similar quality to
those studies using logistic regression as the population excluded
those unfit for surgery.
Other considerations The context of implementation has changed during guideline
development in such a way as to highlight the relevance and
feasibility of the recommendation, in that the Department of
Health has introduced a Best Practice Tariff initiative to achieve hip
fracture surgery within 36 hours of admission.
Recommendation Identify and treat correctable comorbidities immediately so that

surgery is not delayed by:
• anaemia
• anticoagulation
• exacerbation of chronic chest conditions.
Relative values of different The most important outcomes considered here were mortality,
outcomes length of stay in hospital and postoperative complications.
Trade off between clinical Patients should not be delayed for routine tests which will not
benefits and harms affect the surgical or anaesthetic procedure. It has been shown in
the majority of patients that longer delay leads to an increase in
complications and length of stay in those medically fit.
A number of medical conditions that might pose a concern to the
surgeon or the anaesthetist are so commonly encountered among
patients presenting with hip fracture that their occurrence should
be anticipated, and admission assessment and management
protocols designed that will expedite their management and so
prevent their delaying surgery. The process of pro-actively seeking
to identify such conditions will also help in identifying other less
common potential concerns that might need more individual
assessment - by experienced physicians (often orthogeriatricians)
or anaesthetists - when a medical delay may be required.
Economic considerations The early identification and treatment of patients’ comorbidities
may require additional resources in terms of personnel’s rounds
and ad-hoc tests. These costs would be at least partially off-set by
savings linked with a lower length of hospital stay associated with
the possibility of performing surgery at an earlier stage.
Quality of evidence The evidence included in this chapter did not cover treatment of
comorbidities. The main studies adjusted for these factors and the
subgroup excluded patients unfit for surgery.
Other considerations There should be the availability of experienced orthogeriatricians /
physicians and anaesthetists to assess patients who may require
further optimization. Regular review and communication with the
surgical team is essential.
68 HIP FRACTURE
6.2 Research recommendations on timing of surgery
6.2.1 Surgery within 36 hours
➢ What is the clinical and cost effectiveness of surgery within 36 hours of admission
compared to surgery later than 36 hours from admission in mortality, morbidity
and quality of life in patients with hip fracture?
Early and appropriate surgery for hip fractures is the most effective form of pain relief,
potentially quickening the rehabilitation and reducing complications. Within the current
literature no specific time interval threshold has been identified (up to 24hr) below which a
reduction in delay has shown no benefit. In addition to the evidence of the cost
effectiveness below 48hr, pragmatic, organisational and humanitarian considerations have
been utilised to arrive at the recommendation to operate not later than the day after
admission. A formal study within the NHS based on an arbitrary but realistic 36hr threshold
would provide additional important data to that already available, in order to inform more
precisely the forward clinical and cost-effectiveness of the strategy. For ethical reasons, the
research design would be an observational cohort study, correcting for confounding
variables, possibly set in the context of the National Hip Fracture Database and examining
the effect of the time to surgery and its cost on key outcomes, including mortality,
complications, length of stay, time taken to rehabilitate and qualitative aspects of the
experiences of patients.
7 Analgesia
7.1 Introduction
Pain is a major component of the patient experience following a hip fracture. Fracture and
postoperative pain, along with fracture and surgical site blood loss, constitute the major
physiological stresses facing these patients. Fear of pain is a major concern to them and
their relatives. The best form of analgesia is surgical repair, but there will usually be a
period when assessment is taking place when some analgesia is needed. Prompt and
adequate relief of pain has long been identified as a major priority in the management of
hip fracture, and one that has not always historically been achieved.
ANALSGESIA 69
Pain relief is obviously important for simple humanitarian reasons and for acute nursing
care, but also improves patients' wellbeing, reduces the risk of delirium, and facilitates the
return to mobility and independence.
It is often difficult to assess the need for analgesia when the patients are lying still. They
may require more pain relief when moved passively for investigations, such as radiological
procedures and subsequently for the active mobilisation essential to their successful
recovery. Many patients with hip fracture may be unable to express their pain, either
because of cognitive impairment, acute delirium or an underlying expressive dysphasia.
Systemic analgesics act through the bloodstream on the whole body rather than on a
localised area or region. They are still the most widely used drugs for providing pain relief in
acute painful situations. Systemic analgesics used for pain relief in hip fracture include
simple analgesics such as paracetamol, and a wide range of opioids. Non-steroidal anti-
inflammatory drugs are usually avoided or used with caution because of their side effects.
These include upper gastrointestinal bleeding, nephrotoxicity and fluid retention – to all of
which the older population and are well known to exhibit increased susceptibility.
The nerves supplying the proximal femur may also be blocked by injecting local anaesthetic
around the femoral nerve. These injections are referred to as nerve blocks and are
sometimes administered to patients to reduce pain if simple analgesics and opioids have
not proven to be sufficient. They are also thought to improve pain scores and mobility and
to help avoid excessive opioid usage.
The aim of this chapter is to identify optimal preoperative and postoperative analgesia
including the use of nerve blocks as adjuncts or alternatives to simple analgesics such as
paracetamol and opioids.
The use of nerve blocks as with anaesthesia is covered in Chapter 8 on regional compared
to general anaesthesia.
7.2 Systemic analgesia
In patients who have or are suspected of having a hip fracture, what is the comparative
effectiveness and cost effectiveness of systemic analgesics in providing adequate pain relief
and reducing side effects and mortality?
No studies on the effectiveness of these drugs in hip fracture patients were identified.
No relevant studies were identified. We conducted a cost analysis of a nerve block, non-
opioids and other analgesics. We found that a nerve block would cost approximately
£54.66. The average cost for opioids controlled drugs is £11.84 (where £1.34 is the average
cost per dose of the drugs and £10.50 the personnel cost of two trained nurses required for
the administration of the drugs). The price of opioids non-controlled drugs is estimated at
£1.96 per doses. The cost of non-opioids analgesics is less than £0.1p per dose. Please see
Appendix H section 20.1 for further details.
70 HIP FRACTURE
In order to present the recommendations in a logical manner and retain their sequential
order, the recommendations for this section are presented below in section 7.3.2
7.3 Nerve blocks compared to systemic analgesia
In patients who have or are suspected of having a hip fracture, what is the clinical and cost
effectiveness of nerve blocks compared to systemic analgesia in providing adequate pain
relief and reducing side effects and mortality?
One systematic review 262 was identified including 17 RCTs with a total of 888 participants.
See evidence table 3, Appendix E and forest plots G23 to G37 in Appendix G.
The review considered any nerve block that affects the nerves supplying the proximal
femur. These include the subcostal nerve, the lateral cutaneous nerve of the thigh, the
femoral nerve, psoas (lumbar plexus), fascia iliaca compartment block (FICB) and triple
(femoral, obturator and sciatic) nerve.
The literature search retrieved one Cochrane review (Parker et al 2002)262. A further update
search was then conducted to look for any papers that may have been published since the
publication of this review. No additional studies were retrieved and therefore the clinical
evidence presented in this chapter is based on the Parker et al results with the addition of
the GRADE analysis.
Table 7-24: Nerve blocks versus systemic analgesia – Clinical study characteristics
Other
Number Design( considerations/
p)
Outcome of studies Limitations Inconsistency Indirectness imprecision
116,182,220
Pain 3 RCT Serious No serious No serious Serious
limitations(a) inconsistency indirectness imprecision(o)
Unsatisfactory 5 RCT Serious No serious No serious No serious
pain control limitations(b) inconsistency indirectness imprecision
preoperatively or
need for
‘breakthrough’
analgesia51,98,116,18
2,220
Unsatisfactory 2 RCT Serious No serious No serious Serious

pain control limitations(c) inconsistency indirectness imprecision(o)
postoperatively51,
62
ANALSGESIA 71
Other
Number Design( considerations/
p)
Outcome of studies Limitations Inconsistency Indirectness imprecision
Nausea and/or 6 RCT Serious No serious No serious Serious
vomiting62,98,116,22 limitations(d) inconsistency indirectness imprecision(o)
0,318,331
Need for anti- 1 RCT Serious No serious No serious Serious

emetics331 limitations(e) inconsistency indirectness imprecision(o)
Wound 1 RCT No serious No serious No serious Serious
infection99 limitations inconsistency indirectness imprecision(o)
Pneumonia95,99,129 5 RCT No serious No serious No serious Serious
,207,352
limitations inconsistency indirectness imprecision(o)
Any cardiac 2 RCT Serious No serious No serious Serious
complication99,207 limitations(f) inconsistency indirectness imprecision(o)
Myocardial 1 RCT Serious No serious No serious Serious
infarction207 limitations(g) inconsistency indirectness imprecision(o)
Puritis331 1 RCT Serious No serious No serious Serious
limitations(h) inconsistency indirectness imprecision(o)
Pulmonary 2 RCT No serious No serious No serious Serious
embolism99,129 limitations inconsistency(m) indirectness imprecision(o)
Deep vein 5 RCT Serious No serious No serious Serious
thrombosis62,95,99, limitations(i) inconsistency indirectness imprecision(o)
129,352
Mortality62,95,99,129 8 RCT Serious No serious No serious Serious

,153,165,207,352
limitations(j) inconsistency indirectness imprecision(o)
Pressure 3 RCT Serious No serious No serious Serious
sores62,129,182 limitations(k) inconsistency(n) indirectness imprecision(o)
Confusional 3 RCT Serious No serious No serious Serious
state62,182,352 limitations(l) inconsistency indirectness imprecision(o)
116
(a) One study (Gille 2006) did not state the method of randomisation. All 3 studies were not adequately
blinded.
(b) High risk of bias due to lack of allocation concealment. 2 (GIlle 2006 and Chudinov 1999)51,116 out of the
5 studies did not specify their method of randomisation.
(c) One study (Chudinov 1999)51 did not clearly report its randomisation method and did not report any
allocation concealment.
(d) Low risk of bias. 2 out of the 6 studies did not clearly report randomisation method and allocation
concealment.
(e) High risk of bias due to unclear reporting of the method of randomisation
(f) One of the 2 studies (Matot 2003)207 has a high risk of selection bias due to unclear methods of
concealment and randomisation.
(g) This study has a high risk of selection bias due to unclear methods of concealment and randomisation
(h) This study has a high risk of selection bias due to unclear methods of concealment and randomisation. It
also had a very short follow up (24 hours).
(i) One of the 5 studies (White 1980)352 has a high risk of selection bias due to unclear methods of
concealment and randomisation.
(j) Two of the studies (white and Hood)153,352 had a high risk of selection bias due to unclear methods of
concealment and randomisation. One study also had a high number of drop outs in one the trial arms.
(k) studies has a high risk of selection bias due to unclear methods of concealment and randomisation
(l) One of the studies (white 1980)352 had a high risk of selection bias due to unclear methods of
concealment and randomisation. One study also had a high number of drop outs in one the trial arms.
(m) There was some non statistically significant heterogeneity I2 =31% p=0.23.
(n) There was some non statistically significant heterogeneity I 2 =30% p=0.23.
(o) The wide confidence intervals around the estimate make the result imprecise. Consequently, it is
difficult to determine the true effect size for this outcome.
72 HIP FRACTURE
(p) The following studies included nerve blocks in conjunction with general anaesthesia: Foss et al (2005) 99,
Tuncer et al (2003)331, Spansberg et al (1996)318, Hood et al (1991)153, Jones et al (1985)165, White at al
(1980)352.
Table 7-25: Nerve blocks versus systemic analgesia - Clinical summary of findings
Relative risk (95%
Outcome Intervention Control confidence interval) Absolute effect Quality
Pain 106 104 N/A SMD -0.52 (-0.8 Low
to -0.25)
Unsatisfactory pain 18/150 47/148 RR 0.37 200 fewer per Low
control preoperatively (12%) (31.8%) (0.23-0.61) 1000 (from 124
or need for fewer to 245
‘breakthrough’ fewer)
analgesia
Unsatisfactory pain 1/20 10/20 (50%) RR 0.1 Low
control (5%) (0.01-0.71) 549 fewer per
postoperatively 15/21 15/21 RR 1 1000 (from 177
(71.5%) (71.5%) (0.68-1.47) fewer to 604
fewer)
Nausea and/or 18/141 25/159 RR 1.05 8 more per 1000 Moderate
vomiting (12.8%) (15.7%) (0.63-1.75) (from 58 fewer
to 118 more)
Need for anti-emetics 0/20 5/20 RR 0.09 227 fewer per Low
(0%) (25%) (0.01-1.54) 1000 (from 248
fewer to 135
more)
Wound infection 0/28 2/27 RR 0.019 60 fewer per Moderate
(0%) (7.4%) (0.01-3.85) 1000 (from 73
fewer to 164
more)
Pneumonia 12/129 25/130 RR 0.49 98 fewer per Moderate
(9.3%) (19.2%) (0.26-0.94) 1000 (12 fewer
to 142 fewer)
Any cardiac 3/62 12/62 RR 0.25 145 fewer per Low
complication (4.8%) (19.4%) (0.07-0.84) 1000 (from 31
fewer to 180
fewer)
Myocardial infarction 1/34 4/34 RR 0.25 88 fewer per Low
(3%) (12%) (0.03-2.12) 1000 (from 114
fewer to 132
more)
Pruritis 0/20 5/20 RR 0.09 227 fewer per Low
(0%) (25%) (0.01-1.54) 1000 (from 248
fewer to 135
more)
Pulmonary embolism 1/53 (1.9%) 2/52 (3.8%) RR 0.66 13 fewer per Low
(0.11-3.86) 1000 (31 fewer
to 110 more)
Deep vein thrombosis 7/116 (6%) 7/137 RR 1.12 6 more per 1000 Low
(5.1%) (0.43-2.93) (29 fewer to 99
more)
Mortality 9/189 19/205 RR 0.59 38 fewer per Low
(4.8%) (9.3%) (0.29-1.21) 1000 (66 fewer
to 99 more)
ANALSGESIA 73
Pressure sores 3/86 9/106 RR 0.51 42 fewer per Low

(3.5%) (8.5%) (0.11-2.39) 1000 (76 fewer
to 118 more)
Confusional state 15/77 34/101 RR 0.63 125 fewer per Low
(19.5%) (33.7%) (0.37-1.06) 1000 (212 fewer
to 20 more)
No relevant studies were identified. We conducted a cost analysis of a nerve block, non-
opioids and other analgesics. We found that a nerve block would cost approximately
£54.66. The average cost for opioids controlled drugs is £11.84 (where £1.34 is the average
cost per dose of the drugs and £10.50 the personnel cost of two trained nurses required for
the administration of the drugs). The price of opioids non-controlled drugs is estimated at
£1.96 per doses. The cost of non-opioids analgesics is less than £0.1p per dose. Please see
Appendix H section 20.1 for further details.
Clinical There is a statistically significant but not clinically significant reduction in pain
when using nerve blocks compared to systemic analgesia. (LOW QUALITY).
There is a statistically significant but not clinically significant reduction in
pneumonia when using nerve blocks compared to systemic analgesia
(MODERATE QUALITY).
There is no statistically significant difference between nerve blocks and
systemic analgesia in all other outcomes (LOW QUALITY).
No studies on the cost-effectiveness of nerve blocks for hip fracture patients

Economic
were identified.
Recommendation Assess the patient’s pain:

• immediately upon presentation at hospital and
• within 30 minutes of administering initial analgesia and
• hourly until settled on the ward and
• regularly as part of routine nursing observations throughout
admission.
Relative values of different This group of patients is most commonly elderly and frail and pain
outcomes is one of the main physiological and psychological stresses they
face. Therefore, the GDG considered pain relief (for example as
indicated by the need for ‘breakthrough analgesia’) to be the most
important outcome. The GDG also considered adverse events
74 HIP FRACTURE
outcomes to be important.
Trade off between clinical Regular assessments mean that the patients benefit from analgesia
benefits and harms that is tailored to their needs and ensure that the analgesic agents
have taken effect. There are no identifiable harms associated with
this.
Economic considerations The GDG agrees that the additional costs linked with the staff time
required for regular pain assessment are likely to be offset by the
beneficial outcomes of ensuring adequate analgesia.
Quality of evidence There have been no studies of this approach to achieving adequate
analgesia. The recommendation is based on GDG consensus.
Other considerations Satisfactory and timely pain relief can only be ensured by regular
re-assessment.
To maintain an adequate level of pain relief, analgesia should be
administered routinely and not ‘on demand’. It is good practice to
re-assess a patient in severe pain after 30 minutes, as analgesia will
have taken effect in this time and the need (or not) for additional
analgesia can be determined. The 30-minute interval also reflects
the pharmacokinetic/pharmacodynamic profiles of morphine and
its active metabolite morphine-6-glucuronide. Adequate analgesic
response is usual by 15 minutes after administration and should
invariably be achieved by 30 minutes. Upward dose titration is
otherwise required. The duration of effect varies, ranging from 2 to
24 hours reflecting inter-individual variability in morphine-6-
glucuronide clearance and response. If further analgesia is
required, the need for subsequent hourly reassessment is justified
not only by the need to ensure a satisfactory response, but also to
assess any unwanted effects. This hourly interval is also partly
pragmatic, consistent with safe, common good clinical practice,
and in line with CEM recommendations. For these reasons, the
GDG felt that the recommended 30-minute check to ascertain and
achieve initial response, and hourly observation thereafter to
determine its duration, together with any adverse effects, are
appropriate. The same intervals apply to dosage switches.
Some patients may be unable to express their need for pain relief
to health care professionals. Regular assessment of pain and
tailoring of medication accordingly will reduce the risk of these
patients suffering because of inadequate pain control.
The GDG also considered evidence on patient views. Two studies in
which patients mentioned pain management were identified
(Section 13.2). In one, pain management did not seem to be a
problem314. However, in the other the patient had to keep asking
for pain relief after surgery274. This highlights the importance of
regular assessment.
Additional broad guidance on the assessment of pain in general in
older people is given in a joint British Pain Society and British
Geriatrics Society document to be found at:
http://www.bgs.org.uk/Publications/Publication%20Downloads/Se
p2007PainAssessment.pdf
ANALSGESIA 75
Recommendation Offer immediate analgesia to patients presenting at hospital with

suspected hip fracture, including people with cognitive
impairment.
Trade off between clinical Immediate pain control not only improves patients' wellbeing but
benefits and harms may reduce the risk of delirium, and facilitate rehabilitation and a
return to mobility and independence. The risks of pain relief are
the side effects of the individual agents used to achieve it (see
below).
Economic considerations The GDG agrees that the costs of providing immediate and
adequate analgesia are likely to be offset by the improvement in
patients’ wellbeing.
Quality of evidence There have been no studies on the timing of analgesia on patient
outcome. The evidence for efficacy is that of each agent. The
recommendation is based on GDG consensus.
Other considerations It is a humanitarian necessity that these patients receive adequate
analgesia, even if cognitively impaired, or limited in their ability to
express pain.
Particular skill and sensitivity may be required in the management
of pain in those who also show signs of delirium (see NICE delirium
Guideline224)
It must be remembered that patients may require more analgesia
for investigations such as X Rays.
Recommendation Ensure analgesia is sufficient to allow movements necessary for

investigations (as indicated by the ability to tolerate passive
external rotation of the leg), and for nursing care and
rehabilitation.
Trade off between clinical Providing adequate levels of analgesia is essential in improving the
benefits and harms patients’ wellbeing and minimising their discomfort whilst clinical
76 HIP FRACTURE
investigations are being carried out. Gentle rotation of the leg may
be associated with some degree of pain but would not otherwise
cause any additional harm to the patient. There are no other
identifiable harms from carrying out this assessment.
Economic considerations The beneficial outcomes of ensuring that adequate analgesia is
provided to allow patients’ movements are likely to offset the staff
time required).
Quality of evidence There have been no studies of this approach to achieving adequate
analgesia. The recommendation is based on GDG consensus.
Other considerations In both the pre and postoperative periods if the patient can
tolerate passive rotation of the leg then this gives an indication
they will be comfortable for preoperative radiographs as well as
initial postoperative mobilisation. This procedure should
adequately predict the adequacy of analgesia when patients
subsequently have to be moved (e.g. on and off examination
surfaces) for investigational procedures, such as X-rays.
Recommendation Offer paracetamol every 6 hours preoperatively unless

contraindicated.
Trade off between clinical Simple regular prescribed analgesia such as paracetamol is not
benefits and harms associated with any significant harm or side effects. However, it
should be avoided or used with caution in patients with known
hypersensitivity to paracetamol and in liver and renal disease.
Economic considerations The cost of paracetamol is minimal (Appendix H, section 8.1). The
administration of paracetamol would be part of routine drug
rounds, and therefore it will not involve additional staff or
administrative costs.
Quality of evidence There are no placebo-controlled trials of the efficacy of
preoperative administration of paracetamol in hip fracture patients
as these are unethical. In a randomised controlled trial, Cuvillion et
al 200762 have shown that 2g of intravenous propacetamol
(equivalent to 1g intravenous paracetamol ) can be as effective as
nerve blocks or morphine in the postoperative phase. There were
no studies comparing paracetamol administered via the oral or
rectal routes (which are associated with greater variation in
bioavailability than than the intravenous route). Therefore, the
recommendation for the use of paracetamol is supported by
evidence of low to moderate quality with respect to intravenous
use, but made on the basis of consensus with respect to oral or
ANALSGESIA 77
rectal administration.
Other considerations Complications are especially more likely to develop when stronger
analgesia is administered in the elderly. Regular paracetamol is
first-line unless contra-indicated.
This and subsequent recommendations follow a logical hierarchy
for the use of analgesic agents as indicated in the World Health
Organisation pain relief ladder.
Recommendation Offer additional opioids if paracetamol alone does not provide

sufficient preoperative pain relief.
indicated or by the need for ‘breakthrough analgesia’) to be the
most important outcome. The GDG also considered adverse events
Trade off between clinical Repeated use of opioids may cause dependence and tolerance.
benefits and harms While this should be borne in mind, it should not deter the
achievement of effective pain relief in the acute situation of hip
fracture. In those for whom the fracture is an incident within the
pathway of a terminal illness, the palliative context of that illness
should also be an important consideration. In particular, if there is
a history of previous opioid use, the existence of acquired
tolerance may necessitate the use of higher doses to relieve hip
fracture pain. Many older patients may have impaired respiratory
function and opioids should be used with caution in these patients.
Smaller doses may be required in older patients.
Harm may come from excessive opioid administration:
• Some patients may develop nausea and constipation from
stronger opioids and codeine. Regular laxatives may need to
be administered.
• Severe constipation may exacerbate other chronic conditions
like diverticulitis.
• The significant sedation from even mild opioids in this
vulnerable group may slow down their postoperative
mobilisation, and upset their balance.
There is a trade off between using stronger analgesia with more
side effects and the benefit of better pain relief. Elderly patients
are more susceptible to the harmful effects of opioid analgesics.
Opioids and NSAIDs can both cause harm in elderly patients with
comorbidities. Most elderly hip fracture patients do have multiple
chronic conditions such as decreased renal function , hiatus hernia
78 HIP FRACTURE
or previous gastric or duodenal erosions, vertigo, diverticulitis , or

mild cognitive problems that may be exacerbated by these forms
of analgesia.
Economic considerations The administration of some opioids requires two trained nurses for
approximately 15 minutes. Please see Appendix H section 20.1 for
further details. The GDG agrees that the additional costs are likely
to be offset by the beneficial outcomes of ensuring adequate
analgesia (see Recommendation 1).
Quality of evidence No studies on the effectiveness of opioids compared to placebo or
to other drugs in hip fracture patients were identified.
Other considerations None
Recommendation Consider adding nerve blocks if paracetamol and opioids do not

provide sufficient preoperative pain relief, or to limit opioid
dosage. Nerve blocks should be administered by trained
personnel. Do not use nerve blocks as a substitute for early
surgery.
outcomes to be important. Adequate pain relief is beneficial.
Reduction in the required administration of opioids and the
associated side effects may also be an important outcome.
Trade off between clinical Local nerve blocks are effective and may serve as a means of
benefits and harms reducing the need for, and side effects of, opioids and other
analgesia. However, as there they are associated with a very rare
incidence of nerve damage, administering them in a busy casualty
department may require a rolling programme of training junior
doctors or nurses to be competent with this technique.
Economic considerations The additional cost of nerve blocks versus the cost of opioid drugs
may be offset by savings in the resources that would be required to
treat the side effects of opioids. The GDG agrees that the
additional costs are likely to be offset by the beneficial outcomes
of ensuring adequate analgesia.
Quality of evidence There are a limited number of clinical trials that have examined the
effectiveness of nerve blocks in conjunction with general
anaesthesia. Some studies have looked at the impact of inserting
nerve blocks before the surgical procedure, to see if this may
reduce analgesic requirements and improve pain management.
These studies show that nerve blocks reduce the degree of pain
compared to systemic analgesia alone and that they may have
fewer side effects compared to systemic analgesia.
ANALSGESIA 79
Other considerations Although studies have shown that nerve blocks are better than
systemic analgesia at relieving pain, the GDG considered that this
should not be the be first line treatment. The GDG wished to
ensure that the administration of analgesics is done in a step wise
approach as some patients may benefit from simple analgesics
such as paracetamol and therefore avoid the more serious side
effects of stronger analgesics.
Recommendation Offer paracetamol every 6 hours postoperatively unless

contraindicated.
face. It is also of central importance in achieving early mobilisation
postoperatively. Therefore, the GDG considered pain relief (for
example as indicated by the need for ‘breakthrough analgesia’) to
be the most important outcome. The GDG also considered adverse
events outcomes to be important.
Trade off between clinical Paracetamol administered first-line and regularly in standard
benefits and harms dosage at this frequency is commonly effective and lacks the
unwanted effects of second-line systemic agents (see below). It
should be avoided or used with caution in patients with known
hypersensitivity to paracetamol and in liver and renal disease.
Economic considerations The cost of paracetamol is minimal. The administration of
paracetamol would be part of routine drug rounds, and therefore it
will not involve additional staff or administrative costs. (Appendix
H, section 8.1.
Quality of evidence Cuvillion et al have shown that 2g intravenous propacetamol
(equivalent to 1g paracetamol) is as effective as nerve blocks or
morphine in the postoperative phase.
Other considerations Paracetamol should be the first option as opioids often sedate
patients when they need to be alert to understand and remember
important instructions from the physiotherapist on early effective
mobilisation. Also opioids may make patients feel dizzy and
unconfident about their balance.
Postoperatively active mobilisation may require additional pain
relief. Pain may be a critical barrier to be overcome for effective
early mobilisation.
80 HIP FRACTURE
Recommendation Offer additional opioids if paracetamol alone does not provide

sufficient postoperative pain relief.
example as indicated by Visual Analogue Scales or by the need for
‘breakthrough analgesia’) to be the most important outcome. The
GDG also considered adverse events outcomes to be important.
Trade off between clinical Opioids do have significant side effects of sedation, nausea,
benefits and harms dizziness and constipation. However, pain is also a significant
barrier to early mobilisation. Getting the analgesia right at each
step of the hip fracture pathway is a skilled judgement for each
individual patient until they are discharged.
Often opioids sedate patients when they need to be alert to
understand and remember important instructions from the
physiotherapist on early effective mobilisation. Also opioids may
make patients feel dizzy and unconfident about their balance.
Economic considerations The GDG believe that the side-effects of opioids and additional
costs are likely to be offset by the benefits of pain relief.
Quality of evidence No studies on the effectiveness of opioids compared to placebo or
to other drugs in hip fracture patients were identified. This
Other considerations None.
Recommendation Non steroidal anti-inflammatory drugs (NSAIDs) are not

recommended.
example as indicated by the need for ‘breakthrough analgesia’) to
be the most important outcome. The GDG also considered adverse
events outcomes to be important.
Trade off between clinical The benefits of pain relief are outweighed by the potential side
benefits and harms effects of these drugs particularly (but not exclusively) in the
elderly population. There is a known age-related increase in
susceptibility to the harmful effects of NSAIDs including upper
gastrointestinal bleeding, nephrotoxicity and fluid retention.
Economic considerations The use of NSAIDs is expected to result in a QALY loss, mainly
associated with the side effects and adverse events of NSAIDs in
ANALSGESIA 81
our population. The incremental cost savings would have to be

considerably high to outweigh these negative benefits, and given
the recommended interventions this is highly unlikely.
Quality of evidence No RCTs on the effectiveness of NSAIDs compared to placebo or to
other drugs in hip fracture patients were identified. This
Other considerations The side effects of these drugs are too great in the elderly.
Therefore, the GDG decided that they should be avoided as there
are other safer alternatives are available such as paracetamol and
opioids.
As discussed, many of these patients have comorbidities of hiatus
hernia, gastric or duodenal erosions, or chronic renal impairment,
which can all be made worse by regular use of NSAIDs.
7.4 Research recommendations on analgesia

➢ What is the clinical and cost effectiveness of preoperative and postoperative nerve
blocks in reducing pain and achieving mobilisation and physiotherapy goals sooner
in patients with hip fracture?
Nerve blocks may potentially find an important role in the management of hip fracture
pain, both pre- and postoperatively, because of their potential to reduce the requirement
for opioids and their associated unwanted effects. Economically there are considerations
for staff training, but also for the potential benefits in terms of duration of stay and early
mobilisation. It is not possible from the existing literature to determine this with any
confidence and there is a pressing need for a definitive trial comparing these outcomes
with nerve blocks against a defined protocol of systemic opioid use.
82 HIP FRACTURE
8 Regional (spinal or epidural) versus general

anaesthesia
8.1 Introduction
Patients who have a proximal femoral fracture are usually offered surgery to treat the
injury. The vast majority of these operations will require some type of anaesthesia.
Anaesthesia may be general anaesthesia or regional anaesthesia.
General anaesthesia involves complete loss of consciousness. This may be achieved by

either inhalational agents or intravenous anaesthetic agents. Regional anaesthesia is
conducted by numbing the nerves that supply sensation to the lower limbs, with the
injection of local anaesthetic solution into the fluid surrounding the spinal cord. There are
two types of regional anaesthesia, spinal and epidural. During a spinal, local anaesthetic
drugs, sometimes in combination with opioid painkillers are injected directly into the
cerebro-spinal fluid of the spinal cord. The majority regional anaesthesia administered to
hip fracture patients is spinal anaesthesia rather than epidural.
Hip fracture patients are generally elderly and have significant comorbidities. This increases
the risks from all types of anaesthesia. At present both regional and general anaesthesia are
administered but the eventual choice is the preference and experience of the anaesthetist
in discussion with the patient and their carers.
The aim of this review is to identify whether regional anaesthesia confers any benefit
compared to general anaesthesia with regards to reducing complications and improving
patient outcomes after surgery.
8.2 Regional versus general anaesthesia
In patients undergoing surgical repair or replacement for hip fractures, what is the clinical
and cost-effectiveness of regional (spinal/epidural) anaesthesia compared to general
anaesthesia in reducing complications such as mortality, cognitive dysfunction,
thromboembolic events, postoperative respiratory morbidity, renal failure and length of
stay in hospital?

ANAESTHESIA 83
The literature search retrieved one Cochrane review (Parker et al 2004)266 including 22 RCTs
with a total of 2567 participants. A further update search was then conducted to search for
any papers that may have been published since the publication of this review. No additional
studies were retrieved and therefore the clinical evidence presented in this chapter is based
on the Parker et al results with the addition of the GRADE analysis.
In addition, we conducted a systematic review on patient views to look for evidence on

patient preference as this was one of the main outcomes.
See evidence table4, Appendix E, forest plots G38 to G49.
Table 8-26: General vs. regional anaesthesia – Clinical study characteristics

Other
Number considerations/
Outcome of studies Design Limitations Inconsistency Indirectness imprecision
Mortality (early 11 RCT Serious No serious No serious Serious
up to 1 limitations inconsistency indirectness Imprecision (c)
month)1,20,23,65,66,1 (a), (b)
67,210,211,277,334,339
Mortality at 1 8 RCT Serious No serious No serious Serious

month20,65,66,167,210 limitations inconsistency indirectness Imprecision (c)
,211,277,339 (a), (b) (d)
Length of stay in 2 RCT Serious No serious No serious Serious

hospital210,277 limitations inconsistency indirectness imprecision
(a), (b)
23,211
Vomiting 2 RCT Serious No serious No serious Serious
limitations inconsistency indirectness Imprecision (c)
(a), (b)
Acute 5 RCT Serious No serious No serious Serious

confusional limitations inconsistency indirectness imprecision (c)
state20,23,46,169,277 (a), (b)
Pneumonia1,20,23,6 9 RCT Serious No serious No serious Serious

5,66,167,210,211,277
limitations inconsistency indirectness Imprecision (c)
(a), (b)
Myocardial 6 RCT Serious No serious No serious Serious

infarction65,66,167,2 limitations inconsistency indirectness Imprecision (c)
10,211,277 (a), (b)
Pulmonary 9 RCT Serious No serious No serious Serious

embolism1,20,23,36,6 limitations inconsistency indirectness Imprecision (c)
5,66,210,211,277 (a), (b) (e)
Deep vein 4 RCT Serious No serious No serious Serious

thrombosis36,65,210 limitations inconsistency indirectness Imprecision (c)
,211 (a), (b)
(a) Some of the studies did not report definite allocation concealment
(b) None of the trials clearly stated whether it was an intention to treat analysis
(c) The relatively few events and few patients give wide confidence intervals around the estimate of effect.
This makes it difficult to know the true effect size for this outcome.
(d) Pooling of the results showed some but not statistically significant heterogeneity: I2 = 31% (p= 0.18)
(e) The results of pooling all pulmonary embolism events showed statistical heterogeneity I2 = 47% (p=
0.06). The authors suggest this is mainly due to the significantly different in trials presenting results for
fatal and non fatal pulmonary embolism. These were subsequently analysed in separate meta-analyses.
84 HIP FRACTURE
Table 8-27: General vs. regional anaesthesia - Clinical summary of findings

Outcome Intervention Control Relative risk (95% CI) Absolute effect Quality
Mortality (early up to 64/912 (7%) 93/966 RR 0.73 26 fewer per Low
1 month) (9.6%) (0.54-0.99) 1000 (from 1
fewer to 44
fewer)
Mortality at 1 month 56/811 86/857 RR 0.69 31 fewer per Low
(6.9%) (10%) (0.50, 0.95) 1000 (from 5
fewer to 50
fewer)
Length of stay in 108 110 N/A Mean Difference Low
hospital 0.21 (-5.21-4.78)
Vomiting 2/46 (4.3%) 3/49 (6.1%) RR 0.7 18 fewer per Low
(0.12-3.94) 1000 (from 54
fewer to 179
more)
Acute confusional 11/117 23/120 RR 0.5 96 fewer per Low
state (9.4%) (19.2%) (0.26-0.95) 1000 (from 10
fewer to 142
fewer)
Pneumonia 21/574 29/612 RR 0.76 11 fewer per Low

(3.7%) (4.7%) (0.44-1.3) 1000 (from 26
fewer to 14
more)
Myocardial infarction 5/502 (1%) 11/531 RR 0.55 9 fewer per 1000 Low
(2.1%) (0.22-1.37) (from 16 fewer
to 8 more)
Pulmonary embolism 9/605 13/640 (2%) RR 0.88 2 fewer per 1000 Low
(1.5%) (0.32-2.39) (from 14 fewer
to 28 more)
Deep vein thrombosis 39/129 61/130 RR 0.64 169 fewer per Low
(30.2%) (36.9%) (0.48-0.86) 1000 (from 66
fewer to 244
fewer)
One study was identified. Chakladar 201048 is a cost study of general vs. spinal anaesthesia based
on a survey. Please see Economic Evidence table 13 in Appendix F for further details.
Table 8-28: General anaesthesia vs regional anaesthesia- Economic study characteristics

Chakladar 201048 Potentially serious limitations Partially applicable (b) Cost analysis of general
(a)
anaesthesia vs. spinal
anaesthesia.
(a) Not a full economic evaluation – costs but not health effects. Cost analysis based on responses to a
questionnaire, not on a direct audit of equipment usage. Overhead costs and cost of treating side
effects were not included. No sensitivity analysis.
(b) UK study but does not estimate QALYs.
ANAESTHESIA 85
Table 8-29: General anaesthesia vs regional anaesthesia - Economic summary of findings

Incremental cost Incremental
Study (£) effects ICER Uncertainty
(a)
Chakladar 76.77 NA NA NR
201048
(a) General anaesthesia more costly than regional anaesthesia (SD):£270.58 (44.68) vs 193.81
(44.68); p<0.0001
Clinical There is a statistically and clinically significant reduction in early mortality (up
to 1 month) in patients having regional anaesthesia compared to those having
general anaesthesia (LOW QUALITY).
There is a statistically significant but not clinically significant improvement in
postoperative confusion and reduction in incidence of deep vein thrombosis
in patients receiving regional compared to general anaesthesia (LOW
QUALITY).
There were no statistically significant differences in length of stay in hospital,
vomiting, pneumonia, myocardial infarction and pulmonary embolism (LOW
QUALITY).
Economic One study found general anaesthesia to be more costly than spinal
anaesthesia. This evidence has very serious limitations since it did not
evaluate effectiveness and may not have included all important cost
differences.
Recommendation Offer patients a choice of spinal or general anaesthesia after

discussing the risks and benefits.
Relative values of different The GDG considered early mortality (up to 1 month) and patient
outcomes preference to be the most important outcomes.
Trade off between clinical Most clinical benefit was seen in patients undergoing regional
benefits and harms anaesthesia. However, there is a small chance of nerve damage
following regional anaesthesia.
Potential benefits with regional also include, reduction in venous
thromboembolic (VTE) complications but studies were conducted
in patients not receiving VTE prophylaxis which may lead to some
false positive results. However, this finding is supported by a more
comprehensive review of DVT and PE across all surgical patients in
the NICE guideline on venous thromboembolism prophylaxis225.
A potential benefit of general anaesthesia includes lack of
awareness throughout the surgical procedure. Indeed some
patients perceive unconsciousness during general anaesthesia as a
benefit. However, others fear the loss of control. A potential
disadvantage of general anaesthesia is that recovery on the first
postoperative day may be slower.
86 HIP FRACTURE
Economic considerations The GDG felt that because of the potentially serious limitations of
the study included as economic evidence there were insufficient
data to claim that the overall costs of the general and regional
anaesthesia are substantially different.
However, there was agreement in acknowledging that spinal
anaesthesia usually involves lower costs for drugs, anaesthesia
equipment and airway equipment than general anaesthesia.
Nevertheless, these lower costs of regional anaesthesia could be
offset by its longer administration time. The GDG debated at length
whether regional anaesthesia required more time to be
administered compared to general anaesthesia, but no agreement
was reached.
Quality of evidence The studies comparing the two types of anaesthesia were mainly of
low methodological quality. They included small numbers of
participants and only reported a few outcome measures. These
varied between studies making pooling of the data difficult. The
studies lacked methodological rigour in particular regarding
allocation concealment, assessor blinding and intention to treat
analysis. The studies are now considered to be out of date and no
longer relevant to current anaesthesia and perioperative care. In
addition, they do not account for the advances in safety in the field
of anaesthesia. For example in some of the studies patients
allocated to general anaesthesia did not receive thrombo-
prophylaxis as part of routine care.
The economic evidence has very serious limitations, as it is based
on responses to a questionnaire on a hypothetical anaesthetic
technique, and not a direct audit of actual equipment usage.
Moreover, the analysis did not look at whether there are any
potential savings linked to a reduction in the cases of confusion
when regional anaesthesia is used.
Other considerations The GDG also considered the evidence for other outcomes such as
length of stay in hospital and adverse events including vomiting,
acute confusional state and respiratory and cardiac complications.
In the absence of any strong evidence favouring one method over
the other, the GDG decided that the choice of anaesthesia should
be based on the patient preference after being given sufficient
information about the options available and the expertise of the
anaesthetist.
Recommendation Consider intraoperative nerve blocks for all patients undergoing

surgery.
Relative values of different The GDG considered pain relief, postoperative mobility and
outcomes reduction in opioid usage to be the main outcomes.
ANAESTHESIA 87
Trade off between clinical As discussed in chapter 7 on using nerve blocks for hip fracture
benefits and harms analgesia, local nerve blocks may serve as a means of reducing the
need for, and side effects of, opioids and other analgesia.
However, they are associated with a very rare incidence of nerve
damage and must be admisitered by trained health care
professionals.
Economic considerations The GDG agreed this likely to be cost-effective because the
administration of nerve blocks avoids the complications and side
effects of opioids, and therefore might result in a reduced length of
hospital stay. Please see the analgesia chapter for evidence on the
cost-effectiveness of nerve blocks in general.
Quality of evidence The evidence that nerve blocks reduce the degree of pain and the
requirement for opioid analgesics compared to other forms of
analgesia alone, and that they may have fewer side effects
compared to systemic analgesia, is presented under Analgesia
(Chapter 7). This includes several studies studies that have
investigated the effectiveness of nerve blocks in conjunction with
general anaesthesia to determine if this reduces the requirements
for opioid analgesics and improve pain management. These studies
show that nerve blocks reduce the degree of pain compared to
systemic analgesia alone and that they may have fewer side effects
compared to systemic analgesia. However, these studies could not
be subgrouped in a meaningful way as they looked at different
outcomes and differed in the way they reported them. Therefore,
this recommendation was partly based on consensus.
Other considerations Nerve blocks are often administered before a spinal anaesthetic, in
order to position the patient. They are usually administered before
a general anaesthetic and many are now conducted using
ultrasound guidance. This reduces the chance of complications,
such as an intraneural injection and also enables the dose of local
anaesthetic administered to be lower. The use of nerve blocks in
surgery has increased in recent years and has almost become
routine practice. Therefore, studies to show any benefit may now
be difficult to conduct, as withholding analgesia from such patients
may be unethical. Administration of nerve blocks should not delay
surgery.
8.3 Research recommendation on anaesthesia

➢ What is the clinical and cost effectiveness of regional versus general anaesthesia on
postoperative morbidity in patients with hip fracture?

88 HIP FRACTURE
patients are sedated before regional anaesthesia is administered and this is not taken into
best addressed by a randomised controlled trial. This would ideally be a multi-centred trial
including 3,000 participants in each arm. This is achievable if one considers that there are
70, 000 hip fractures a year in the UK39. The study should have three arms which look at
spinal anaesthesia versus spinal anaesthesia plus sedation versus general anaesthesia, this
would separate those with regional anaesthesia from those with regional anaesthesia plus
sedation. The study would also need to control for surgery, especially type of fracture,
prosthesis and grade of surgeon.
A qualitative research component would also be helpful to study patient preference for
9 Surgeon seniority
9.1 Introduction
As a general observation of life one would conclude that to have a job completed
thoroughly, effectively and efficiently it would be appropriate to give the task to somebody
with adequate training and experience. Whether this can be extrapolated to the
relationship of the management of hip fractures to the seniority of the surgeon involved is
the purpose of this chapter.
The historical background of this question has to be considered in relation to the

environment in which hip fracture patients were treated. In the United Kingdom hip
fractures were commonly regarded as the surgical material for trainee surgeons to gain
their experience. In the past much of this work would have been unsupervised, and in the
main the trainees would have enjoyed the challenge and responsibility this gave them.
The operations were often performed outside of scheduled list times as extra or emergency
cases. Under these circumstances it was more likely that the anaesthetist involved in the
procedure would be more junior and the nursing scrub team not specifically from a trauma
theatre.
Any variations in outcome which may be simply labelled as related to surgeon seniority may
in fact have multiple underlying causes. A more senior surgeon is more likely to be
SURGEON SENIORITY 89
operating on a scheduled list, with more senior anaesthetists and a regular nursing scrub
team.
9.2 Surgeon seniority
What is the clinical and cost effectiveness of surgeon seniority (consultant or equivalent) in
reducing the incidence of mortality, the number of patients requiring reoperation, and poor
outcome in terms of mobility, length of stay, wound infection and dislocation? (See
evidence table 5, Appendix E and forest plots G50 and G51 in Appendix G).
9.2.2 Clinical evidence
No randomised evidence was identified. Three prospective cohorts including 2018

participants that adjusted for some confounding factors were identified.
Table 9-30: Junior/less senior surgeon vs. senior surgeon – Clinical study characteristics
Numbe Other
r of considerations/
Outcome studies Design Limitations Inconsistency Indirectness imprecision
Reoperations 1 Cohort serious no serious serious serious
(follow up 6 limitations inconsistency indirectness(c,d,
imprecision (h)
256 (a,b) e)
months)
Dislocation in 1 Cohort serious no serious serious serious
hemiarthroplasty limitations inconsistency indirectness imprecision (h)
(b) (f,g)
(follow up 0 to 10
years)85
Dislocation in 1 Cohort serious no serious serious serious
total hip limitations inconsistency indirectness imprecision (h)
(c) (f,g)
replacement
(follow up 0 to 11
years)85
(a) Senior surgeons operated on significantly more patients with a poor pre-fracture mobility score and
performed significantly more arthroplasties and significantly fewer osteosyntheses.
(b) Only a limited number of confounders were included in the analysis. No adjustment or mention of
the anaesthetists experience or grade.
(c) Surgeon seniority measured by years experience rather than the grade of surgeon. Experienced
surgeons with more than 3 years orthopaedic surgical experience either performing surgery or
supervising junior registrars were compared unsupervised orthopaedic junior registrars with less
than 3 years orthopaedic surgical experience.
(d) Only the technically demanding fractures were included in the analysis, not all surgery for hip
fractures.
(e) Reoperation rate only measured at 6 months, not longer.
(f) The focus of the study is on surgical approach therefore baseline data by surgeon seniority is not
reported.
(g) Dislocation is not a primary outcome.
(h) The wide confidence intervals make the estimate of effect imprecise.
Table 9-31: Junior/less senior surgeon vs senior surgeon – Clinical summary of findings
Outcome Intervention Control Relative risk Absolute effect Quality
90 HIP FRACTURE
Reoperations (follow 289 more per

multivariate odds
up 6 months) 16/56 47/309 1000 (from 3
ratio 2.01 (1.01 to Very low
(28.6%) (15.2%) more to 864
4.02)
more)
Dislocation in 18 more per
hemiarthroplasty 37/404 8/135 multivariate odds 1000 (from 24
Very low
(median follow up 4.3 (9.2%) (5.9%) ratio 1.3 (0.6 to 3) fewer to 118
(0 to 10) years) more)
Dislocation in total hip 10 fewer per
replacement (median 37/636 8/77 multivariate odds 1000 (from 73
Very low
follow up 2.3 (0 to 11) (5.8%) (10.4%) ratio 0.9 (0.3 to 2.8) fewer to 187
years) more)
No studies were identified on the cost-effectiveness of junior/less senior surgeon vs.

senior surgeon. However, we conducted a cost-analysis around the personnel cost of
a planned trauma list compared to the personnel cost of a general emergency
theatre. We found that a planned trauma list involves an additional cost per hour of
£94, See Appendix H section 20.2for further details.
Clinical There is a statistically significant, but not clinically significant increased

reoperation rate at 6 months with unsupervised junior orthopaedic registrars
with less than 3 years experience than with experienced surgeons with more
than 3 years experience. (VERY LOW QUALITY).
There is no statistically significant difference between Swedish post registrars
and registrars in dislocation rate at a median follow up of 2.3 years after
hemiarthroplasty in patients with hip fracture. (VERY LOW QUALITY).
There is no statistically significant difference between Swedish post registrars
and registrars in dislocation rate at a median follow up of 2.3 years after total
hip replacement in patients with hip fracture. (VERY LOW QUALITY).
There was no evidence identified for mortality, mobility, length of stay or
wound infection.
Economic No studies were identified on the cost-effectiveness of junior/less senior
surgeon vs. senior surgeon. However, we conducted a cost-analysis around
the personnel cost of a planned trauma list compared to the personnel cost of
a general emergency theatre. We found that a planned trauma list involves an
additional cost per hour of £94, See Appendix H section 20.2for further
details.
9.3 Recommendations and link to evidence
Recommendation Schedule hip fracture surgery on a planned trauma list
Relative values of different Mortality, reoperation rate, dislocations, length of stay in

outcomes secondary care and wound infection were considered the main
outcomes. Complications, pain and functional status were also
considered.
Trade off between clinical No RCTs were identified evaluating a planned trauma list. Evidence
benefits and harms is extrapolated from the surgeon seniority data. This shows a
significantly higher reoperation rate with unsupervised/junior
orthopaedic surgeons with less than 3 years experience than senior
more experienced surgeons. There was no statistically significant
difference in dislocation rates. No other outcomes were reported.
Economic considerations A planned trauma list consists of a period of time allocated to the
surgical management of patients with unplanned admissions
following musculoskeletal injury. For this period there will be an
adequate operating theatre, with supporting equipment including
an image intensifier. The responsible senior surgical, anaesthetic
and theatre staff will have work plan allocating time to the list to
carry out procedures or supervise their junior staff. Thus, a planned
trauma list implies allocation and involvement of senior staff, who
will either carry out the necessary procedures in the operating
theatre or will adequately supervise the junior staff.
The GDG suggested that a possible comparator for a planned
trauma list could be a general emergency theatre, shared by many
different specialities, often occurring outside of normal working
hours and staffed by trainees.
If we consider the case of a planned trauma list where operations
are performed by a consultant surgeon and a consultant
anaesthetist and if we take as comparator a general emergency
theatre where both surgeon and anaesthetist are registrars, and
we assume no other difference in the professional grade of the
remaining staff involved in the operation, then the planned trauma
list would result in an additional personnel cost per hour of £94
over the general emergency theatre. In particular, the personnel
cost per hour for a planned trauma list with a consultant surgeon
and consultant anaesthetist correspond to £337, and for a general
emergency list with a registrar surgeon and a registrar anaesthetist
(and with a consultant surgeon and consultant anaesthetist on
call), to £243 (please see Appendix H section 20.2for further
details). However, there is great uncertainty as to whether there
are other differences in other categories of costs (e.g. overheads,
diagnostic devices, etc) between a planned trauma list and a
general emergency theatre,and therefore our estimate should be
considered only as an approximation of the overall cost difference
between a planned trauma list and a general emergency theatre.
Furthermore, there is uncertainty around the right baseline
92 HIP FRACTURE
intervention as after the introduction of the BPT for hip fracture,

senior staff should be performing the surgery. In particular, the
GDG noted that it is not clear as to what we should consider as the
usual alternative to a planned trauma list, as it is quite uncertain
what could represent the “baseline” case for a hospital, and this
can change depending on the type of hospital. It was also pointed
out that since the introduction of the Best Practice Tariff (BpT) for
hip fracture in April 2010 the hospitals that do not have planned
trauma list in place on a daily basis would however have employed
relevant senior staff (consultant surgeons and anaesthetist) to
meet the tariff’s requirements, and therefore senior staff are
already part of the comparator.
Nevertheless, the GDG thinks that these potential additional
personnel costs of a planned trauma list would be at least partially
off-set by savings due to lower re-operation rates and by a higher
number of patients operated per hour.
Quality of evidence No RCTs were identified evaluating a planned trauma list. There is
extrapolated evidence from surgeon seniority showed no evidence
for the majority of the outcomes and only very low quality
evidence from non-randomised studies for two outcomes:
reoperation rate and dislocations. The recommendation is based
on a consensus agreement within the GDG.
Other considerations We have specified in the recommendation that surgery for hip
fractures should occur on a planned trauma list. To establish a
scheduled trauma list management and clinicians are required to
provide adequate facilities and staff for it to run. For a planned list
it is necessary to have a chain of responsibility to a consultant
surgeon and consultant anaesthetist who have time in their
programs to execute that responsibility. To run a planned trauma
list requires ready access to an image intensifier and radiographer.
The nursing team would need to be appropriate to the work
planned for that theatre. The recommendation therefore
recognises the need for adequate seniority of the surgeon but
makes what we believe to be a reasonable assumption that this
recognition should also apply to the rest of the operating theatre
team caring for the hip fracture patient.
The GDG noted that there is high uncertainty regarding the
implementation costs linked with this recommendation, as these
costs will vary depending on the current set up and infrastructure
of each hospital . For example, the GDG recognised that smaller
hospitals may not currently provide this service at weekends.
This recommendation is in line with the British Orthopaedic
Association’s Advisory book on consultant trauma and orthopaedic
services 38. The GDG consider this recommendation a key priority
for implementation.
Recommendation Consultants or senior staff should supervise trainee and junior

members of the anaesthesia, surgical and theatre teams when
they carry out hip fracture procedures.
Relative values of different Mortality, reoperation rate, dislocations, length of stay in
outcomes secondary care and wound infection were considered the main
outcomes. Complications, pain and functional status were also
considered.
Trade off between clinical There is a significantly higher reoperation rate with
benefits and harms unsupervised/junior orthopaedic surgeons with less than 3 years
experience than senior more experienced surgeons. There was no
statistically significant difference in dislocation rates. No other
outcomes were reported.
Economic considerations Higher grade surgeons or those with more experience are likely to
be entitled to a higher wage than junior surgeons. However, as
their rate of re-operations is statistically significantly lower, having
hip fracture patients operated on by experienced surgeons will
plausibly result in cost savings and improved health outcomes. In
addition, the GDG believe experienced surgeons use theatre time
more efficiently allowing greater throughput of cases.
Quality of evidence There is no evidence for the majority of the outcomes and only
very low quality evidence from non-randomised studies for two
outcomes: reoperation rate and dislocations.
Other considerations The level of supervision required for a trainee or junior staff
member for a particular case depends on two main factors: the
junior’s ability and the complexity of the case. It is therefore
implicit that the senior staff responsible for the trauma list must
have knowledge of both of these factors before determining the
level of supervision required. Potential surgical, anaesthetic or
nursing problems may be evident to an experienced surgeon,
anaesthetist or nurse preoperatively. This gives the opportunity to
both avoid the problem occurring and to enhance the training
opportunity. An unsupervised list would therefore be one in which
those responsible did not have adequate prior knowledge of the
capabilities of the more junior members of the team and the
specific problems they may encounter, or when they did not use
this knowledge to provide adequate supervision.
94 HIP FRACTURE
10 Surgical procedures
10.1 Introduction
The options for hip fracture surgery depend on the type of fractures. They can be divided
into two main groups according to their relationship to the capsular attachment of the hip
joint. Those above the insertion of the capsule are termed intracapsular and those below
are termed extracapsular. Extracapsular fractures can be further divided into three types:
pertrochanteric (also called intertrochanteric), reverse oblique or subtrochanteric.
Broadly speaking there are two surgical options for treating hip fractures, replacement
arthroplasty or internal fixation. Replacement arthroplasty involves removing part or all of
the damaged bone and replacing it with a prosthesis which then functions in place of the
removed bone. It may describe a hemiarthroplasty or a total hip arthroplasty. Both involve
replacement of the femoral head with a metal implant, the stem of which is secured in the
femoral shaft. A total hip arthroplasty involves, in addition, replacement of the socket. Both
implants can be inserted with or without the use of cement. Internal fixation involves
returning the bone fragments to an acceptable position and then holding that position with
screws, plates or nails. This should allow healing of the facture fragments in an acceptable
position for long term function and maintenance of patient function whilst that healing
occurs.
10.2 Surgery with regard to early mobilisation

This section relates to the section on early mobilisation (chapter 11) as well as surgery.
When embarking on any surgical procedure there should be a clear objective. In
orthopaedic and trauma surgery it is easy to attach a rather bland aim of "safe restoration
of function". Prior to any surgery commencing the surgeon should already know what his
planned postoperative care of that patient is to be. Given the poor reserve functional
capacity of many hip fracture patients any prescribed limits on mobility and weight-bearing
may significantly alter and restrict their postoperative care. In particular unnecessary
restriction of weight-bearing has the potential to compromise independence, discharge
destination, general health and final level of function. As a consequence of these
considerations, and as a result of the recommendation for early mobilisation (section
11.2.2) the GDG felt it appropriate to make a recommendation on postoperative weight-
bearing status.
SURGICAL PROCEDURES 95
Recommendation Operate on patients with the aim to allow them to fully weight
bear (without restriction) in the immediate postoperative period.
Relative values of different The aim of surgery and rehabilitation is for patients to regain their
outcomes prefracture functional status. Early mobilisation with a
physiotherapist appears safe and is effective in promoting early
recovery. The most important outcomes considered by the GDG
were functional status, mobility, pain and quality of life.
Trade off between clinical The evidence from the early mobilisation question shows that the
benefits and harms only outcome relating to harm or safety was mortality, which
showed no statistically significant difference. If safety issues were a
concern it is likely that they would be reflected in the overall
functional outcomes, all of which improved or had no significant
effect, therefore we don't believe that harm is caused harm from
this evidence.
Economic considerations See also early mobilisation section 8.2. One of the main aims of
surgery is for patients to regain their pre-fracture functional status.
As the GDG has agreed to consider early mobilisation strategy as a
cost-effective intervention for our population, this
recommendation is unlikely to result in extra costs.
Quality of evidence There is no direct evidence relating to this recommendation, but
the evidence from the early mobilisation review question is
indirectly applicable, see Chapter 8.
Other considerations Elderly patients may be physically frail, suffering from cognitive
impairment or delirium and so cannot be expected to mobilise
non-weight-bearing or partially weight-bearing. Postoperative
instructions requesting non-or partial weight-bearing will
frequently result in the patient not mobilising at all.
10.3 Displaced intracapsular fractures

In an intracapsular fracture the proximal fragment includes the femoral head alone or the
femoral head with a small portion of neck. The size and shape of this fragment combined
with the often soft nature cancellous bone of which it is constituted makes secure fixation
difficult. This can potentially compromise early function. In addition, the blood supply of
the femoral head may be disrupted, leading to poor healing or bone death.
The displacement of an intracapsular fracture is determined on the anteroposterior and

lateral radiographs of the area. An undisplaced fracture may as its name suggests
demonstrate no change in position from that it would have occupied prior to the injury.
However it is also customary to include in the undisplaced group valgus impacted fractures.
In this impacted group the harder bone of the femoral neck has been driven into the softer
bone of the femoral head. In both of these these undisplaced fracture types there is
96 HIP FRACTURE
generally already inherent stability and little likelihood of damage to the blood supply.
Fixation in situ is generally accepted
In practice a displaced fracture is one in which the preoperative radiographs demonstrate

the fragments have moved in relation to each other to an unacceptable position for fixation
in situ. The implication of this is that the fragments have moved in relation to each other to
a greater extent. The particular anatomy of the region means that the blood supply to the
femoral head is at risk. There will also be less inherent stability either as a consequence of
fragmentation along the fracture line or difficulties in obtaining precise reduction.
In patients with these displaced intracapsular fractures a decision initially needs to be made
as to whether to reduce the fracture and internally fix it or to carry out some form of
replacement arthroplasty. Each has potential advantages and disadvantages. Internal
fixation retains the patient's own tissues and is often a smaller procedure. However, it may
require a more prescriptive postoperative regime to protect the healing bone. Should
replacement arthroplasty be appropriate it is necessary to determine the indications for a
hemiarthroplasty in which only the damaged bone of the proximal femur is replaced or a
total hip replacement when both the femoral head and the hip socket are replaced.
10.3.1 Internal fixation versus hemiarthroplasty
10.3.1.1 Review question
In patients having treatment for displaced intracapsular hip fracture what is the clinical and
cost effectiveness of internal fixation compared to hemiarthroplasty on mortality, number
of reoperations, functional status, length of stay in hospital, total time to resettlement in
the community, quality of life, pain and place of residence after hip fracture.
One systematic review264 was identified and one additional RCT102. Overall, there were 13
RCTs with 2195 participants. See evidence table 7, Appendix E and forest plots G74 to G82
in Appendix G.
Table 10-32: Internal fixation vs hemiarthroplasty – Clinical study characteristics

Numbe Other
r of Desig considerations/
Outcome studies n Limitations Inconsistency Indirectness imprecision
Mortality at 1 1 RCT serious no serious no serious serious
month102 limitations (a) inconsistency indirectness imprecision (b)
Mortality at 3 to 10 RCT serious no serious no serious serious
6 limitations (a) inconsistency indirectness imprecision (b)
months27,102,161,174
,267,276,317,324,341,343
Mortality at 1 9 RCT serious no serious no serious no serious

year27,102,161,174,267, limitations (a) inconsistency indirectness imprecision
317,324,341,343
Mortality at 2 to 10 RCT serious no serious no serious no serious

3 limitations (a) inconsistency indirectness imprecision
years27,102,161,174,26
7,276,317,324,341,343
Numbe Other
Total no. of 13 RCT serious serious no serious no serious
reoperations limitations (a) inconsistency (c) indirectness imprecision
(follow-up 1 to 5
years)27,69,102,161,17
4,267,276,288,313,317,324,
341,343
Failure to return 2 RCT serious no serious no serious serious

to same limitations (a) inconsistency indirectness imprecision (b)
residence
(follow-up 1 to 3
years)161,267
Failure to regain 6 RCT serious serious no serious serious
mobility (follow- limitations (a) inconsistency (f) indirectness imprecision (b)
up 1 to 5
years)27,161,267,288,3
17,341
No. of patients 3 RCT serious serious no serious serious

reporting pain at limitations (a) inconsistency (d) indirectness imprecision (b)
27,174,267
1 year
Harris Hip Score 1 RCT no serious no serious no serious serious
(follow-up 1 limitations inconsistency indirectness imprecision (e)
102
year)
Harris Hip Score 1 RCT no serious no serious no serious serious
(follow-up 2 limitations inconsistency indirectness imprecision (e)
102
years)
Number of 1 RCT no serious no serious no serious serious
patients with limitations inconsistency indirectness imprecision (e)
Barthel Index
Score of 95 or
100 (follow-up 1
year)102
patients with limitations inconsistency indirectness imprecision (e)
Barthel Index
Score of 95 or
100 (follow-up 2
years)102
Eq-5d (Euroqol) 1 RCT no serious no serious no serious serious
Index Score limitations inconsistency indirectness imprecision (e)
(follow-up 1
year)102
Eq-5d (Euroqol) 1 RCT no serious no serious no serious serious
Index Score limitations inconsistency indirectness imprecision (e)
(follow-up 2
years)102
Length of 4 RCT no serious no serious no serious serious
hospital limitations inconsistency indirectness imprecision (e)
102,174,267,341
stay
(a) The studies with the most weight in the meta-analysis have inadequate or unclear allocation
concealment.
(b) The relatively few events and few patients give wide confidence intervals around the estimate of
effect. This makes it difficult to know the true effect size for this outcome.
98 HIP FRACTURE
(c) There is significant unexplained statistical heterogeneity between the studies. This could be due to
the different types of implant or arthroplasty and different follow up periods.
(d) There is significant statistical heterogeneity between the studies. This could be due to the different
types of implant or arthroplasty.
(e) The wide confidence intervals around the estimate make the result imprecise. Consequently, it is
difficult to determine the true effect size for this outcome.
(f) There is significant statistical heterogeneity between the studies. This Cochrane review reports this
is likely to be due to the variation in the definition for this outcome.
Table 10-33: Internal fixation vs hemiarthroplasty - Clinical summary of findings

Mortality at 1 month 28 fewer per
7/112 10/110 RR 0.69 1,000 (from 66
Low
(6.3%) (9.1%) (0.27 to 1.74) fewer to 67
more)
Mortality at 3 to 6 32 fewer per
107/765 122/709 RR 0.81
months 1,000 (from 60 Low
(14%) (16.7%) (0.64 to 1.03)
fewer to 5 more)
Mortality at 1 year 17 fewer per
148/636 143/584 RR 0.93 1,000 (from 52
Moderate
(23.3%) (23.6%) (0.78 to 1.12) fewer to 28
more)
Mortality at 2 to 3 15 fewer per
years 265/750 254/683 RR 0.96 1,000 (from 60
Moderate
(35.3%) (37.8%) (0.84 to 1.09) fewer to 34
more)
Total no. of 243 more per
reoperations (follow- 355/1001 99/1033 RR 3.59 1,000 (from 181
Low
up 1 to 5 years) (35.5%) (9.4%) (2.93 to 4.39) more to 319
more)
Failure to return to 38 fewer per
same residence 29/187 34/185 RR 0.84 1,000 (from 109
Low
(follow-up 1 to 3 (15.5%) (23.6%) (0.54 to 1.33) fewer to 78
years) more)
Failure to regain 9 more per 1,000
155/287 165/306 RR 1.02
mobility (follow-up 1 (from 119 fewer Very low
(54%) (45.7%) (0.74 to 1.39)
to 5 years) to 178 more)
No. of patients 13 fewer per
reporting pain at 1 126/280 127/281 RR 0.97 1,000 (from 150
Very low
year (45%) (44.2%) (0.66 to 1.44) fewer to 194
more)
Harris Hip Score MD -6.8
87 74 N/A Moderate
(follow-up 1 year) (-12 to -1.6)
Harris Hip Score MD -3.3
71 68 N/A Moderate
(follow-up 2 years) (-9.1 to 2.5)
Number of patients 176 fewer per
with Barthel Index 31/87 39/73 RR 0.67 1,000 (from 27
Moderate
Score of 95 or 100 (35.6%) (53.4%) (0.47 to 0.95) fewer to 283
(follow-up 1 year) more)
Number of patients 34 fewer per
with Barthel Index 24/69 26/68 RR 0.91 1,000 (from 160
Moderate
Score of 95 or 100 (34.8%) (38.2%) (0.58 to 1.42) fewer to 160
(follow-up 2 years) more)
Eq-5d (Euroqol) Index
MD -0.09
Score (follow-up 1 70 62 N/A Moderate
(-0.2 to 0.02)
year)
Eq-5d (Euroqol) Index

MD -0.11
Score (follow-up 2 52 52 N/A Moderate
(-0.21 to -0.01)
years)
Length of hospital stay MD -0.6
486 478 N/A Moderate
(-2.04 to 0.83)
Two economic studies were identified 173,291. Rogmark et al (2003)291 is a cost-consequence

analysis based on a RCT but it was excluded because it does not distinguish patients on the
basis of whether they received hemiarthroplasty or total hip replacement. Keating et al
(2005)173 compare internal fixation vs. hemiarthroplasty in a cost-consequence analysis
based on a RCT. Please see Economic Evidence Table 14 in Appendix Ffor further details
Table 10-34: Internal Fixation vs Hemiarthroplasty - Economic study characteristics

Keating 2005 173 Minor limitations (a) Partially applicable (b) Costs not discounted
because mainly incurred
within 1 year of injury
(a) Small number of patients.
(b) UK study, but does a CUA.
Table 10-35: Internal Fixation vs Hemiarthroplasty - Economic summary of findings

Study per patient (£) effects ICER Uncertainty
Keating 2005 173 £2726(a) Various (b) N/A Two-way sensitivity
analysis showed that the
direction of change in cost
did not change when cost
of prostheses and cost of
readmission were varied
over a range from -50% to
+100% around the baseline
values.
(a) The mean cost per patient for internal fixation was £12,623 (95% CI: 10,768 – 14,478) and for £9,897
(95% CI: 8,062 – 11,732) for hemiarthroplasty (2001 GBP)
(b) Several outcomes were reported. Internal fixation entailed lower mortality at 4 and 12 months from the
operation than hemiarthroplasty (3% vs. 5%; 8% vs. 10%) and slightly higher EQ-5D scores at 24 months
(0.55 vs 0.53); (all effects were not statistically significant). Hemiarthroplasty involved a significantly
lower number of patients needing further surgery at 12 and 24 months (31% vs. 5% and 39% vs. 5%),
and higher EQ-5D scores at 4 and 12 months (0.56 vs. 0.61 and 0.58 vs.0.64; difference not statistically
significant).
Clinical There is a statistically and clinically significant decrease in patients who

require reoperations with hemiarthroplasty than with internal fixation. The
follow up varied between 1 and 5 years. (LOW QUALITY)
There is a statistically significant, but not clinically significant, increase in
patients who have a Barthel Index Score of 95 or 100 at 1 year with
100 HIP FRACTURE
hemiarthroplasty compared to internal fixation but there is no statistically

significant difference at 2 years (MODERATE QUALITY)
patients who have a higher Harris Hip Score at 1 year with hemiarthroplasty
compared to internal fixation but there is no statistically significant difference
at 2 years (MODERATE QUALITY)
patients who have a higher Eq-5d (Euroqol) score at 2 years with
hemiarthroplasty compared to internal fixation but there is no statistically
significant difference at 1 year (MODERATE QUALITY)
There is no statistically significant difference between internal fixation and
hemiarthroplasty in mortality at 1 months (LOW QUALITY), 3 to 6 months
(LOW QUALITY) or 1 to 2 years (MODERATE QUALITY), the number of patients
reporting pain at 1 year (VERY LOW QUALITY), the number of patients failing
to return to the same residence at 1 to 3 years (LOW QUALITY), failure to
regain mobility at 1 to 5 years and length of hospital stay (MODERATE
QUALITY).
No RCT evidince was identified reporting on total time to resettlement in the
community.
Economic Hemiarthroplasty is cost saving with respect to internal fixation. This evidence
has minor limitations and partial applicability.
10.3.2 Internal fixation versus total hip replacement
In patients having treatment for intracapsular hip fracture what is the clinical and cost
effectiveness of internal fixation compared to total hip replacement on mortality, number
of reoperations, functional status, length of stay in hospital, total time to resettlement in
the community, quality of life, pain and place of residence after hip fracture.
One systematic review264 was identified. Overall, there were 6 RCTs with 888 participants
were included. See evidence table 7, Appendix E and forest plots G83 to 86 in Appendix G.
Table 10-36: Internal fixation vs. total hip replacement – Clinical study characteristics
Numbe Other
months162,174,239,32
7

months)162,174,239
Mortality at 2 4 RCT serious no serious no serious serious
years162,166,174,327 limitations (a) inconsistency indirectness imprecision (b)
Numbe Other
Reoperations – 6 RCT serious serious no serious no serious
any (follow-up 1 limitations (a) inconsistency (c) indirectness imprecision
to 13
years)162,166,174,239,
313,327

patients limitations inconsistency indirectness imprecision (b)
reporting pain at
1 year166,174
hospital stay174 limitations inconsistency indirectness imprecision (d)
concealment.
(c) There is significant statistical heterogeneity between the studies. This could be due to the different
types of implant or arthroplasty and different follow up periods. One study had a 13 year follow up
whereas the others varied between 1 and 4 years.
(d) The wide confidence intervals around the estimate make it difficult to determine and effect size for
this outcome.
Table 10-37: Internal fixation vs total hip replacement - Clinical summary of findings
Mortality at 2 to 4 45 more per
months 15/210 6/196 RR 2.21 1,000 (from 3
Low
(7.1%) (3.7%) (0.91 to 5.4) fewer to 163
more)
Mortality at 12 to 18 8 more per 1,000
25/157 21/147 RR 1.08
months) (from 36 fewer Low
(15.9%) (10%) (0.64 to 1.82)
to 82 more)
Mortality at 2 years 21 more per
44/224 34/209 RR 1.18 1,000 (from 24
Low
(19.6%) (11.6%) (0.79 to 1.75) fewer to 87
more)
Reoperations – any 160 more per
(follow-up 1 to 13 126/325 44/308 RR 2.70 1,000 (from 93
Low
years) (38.8%) (9.4%) (1.99 to 3.67) more to 251
more)
Number of patients 150 more per
reporting pain at 1 47/78 34/79 RR 1.4 1,000 (from 8
Moderate
year (60.3%) (37.7%) (1.02 to 1.9) more to 339
more)
Length of hospital stay MD -1.7 (-4.45 to
69 69 - Moderate
1.05)
Three economic studies were identified 163,173,291. Rogmark et al (2003)291 is a cost-

consequence analysis based on a RCT which was excluded because it does not distinguish
patients on the basis of whether they received hemiarthroplasty or total hip replacement.
102 HIP FRACTURE
Keating et al (2005)173 compare Internal Fixation vs Total Hip Replacement in a cost-

consequences analysis included in a Health Technology Assessment based on a RCT.
Johansson et al (2006)163 is a cost-consequence analysis based on a RCT. Please see
Economic Evidence Tables 14 in Appendix F for further details.
Table 10-38: Internal fixation vs total hip replacement - Economic study characteristics
Keating 2005 173 Minor limitations (a) Partial applicability (b) Costs not discounted
Johansson 2006 Potentially serious limitations Partial applicability (d)
163 (c)

(b) Study set in the UK, but not a CUA.
(c) Costs were derived from just one hospital. No sensitivity analysis was conducted.
(d) Study set in Sweden.
Table 10-39: Internal fixation vs total hip replacement - Economic summary of findings
Keating 2005 173 £3224 (a) THR has higher THR dominant Two-way sensitivity
EQ-5D scores at 4, analysis showed that the
12 and 24 months direction of change in cost
by 0.08; 0.12 and did not change when cost
0.14 respectively (b) of prostheses and cost of
values.
Johansson 2006 £265 More patients with THR dominant NR
163
good/fair Harris
hip score at 1 and
2 years in THR
group (c)
(a) The mean cost per patient included cost of hospital admission (inpatient and day case), theatre
costs, prosthesis and profile of hardware. The mean cost per patient for internal fixation was
£12,623 (95% CI: 10,768 – 14,478) and £9,399 (95% CI: 8,265-10,532) for THR.
(b) THR had better outcomes than internal fixation: lower number of deaths within 4, 12 and 24
months from operation: (3% vs. 4%; 8% vs. 6% and 15% vs. 9%; p value not significant). Lower
number of patients requiring further surgery within 4, 12 and 24 months from operation: 22% vs.
7%; 31% vs. 9% and 39% vs. 9%; p value not reported). Higher mean EQ-5D scores at 4, 12 and 24
months from operation: 0.56 vs 0.68 (p value not significant); 0.58 vs 0.70 (p = 0.04); 0.55 vs 0.69 (p
value not significant).
(c) Percentage of patients with a Harris hip score excellent or good/fair or poor at 1 year: 12.5% vs.
100% (p value: <0.0001); at 2 years: 14.29% vs.95.23% (p value: <0.001)
Clinical There is a statistically and clinically significant decrease in patients who

require reoperations with total hip replacement than with internal fixation.
The follow up varied between 1 and 13 years. (LOW QUALITY)

patients who reported pain at 1 year with internal fixation compared to total
hip replacement (MODERATE QUALITY).
There is no statistically significant difference in mortality at 2 to 4 months, 12
to 18 months or 2 years (LOW QUALITY) and length of hospital stay
(MODERATE QUALITY) between internal fixation and total hip replacement.
No RCT evidence was identified reporting functional status, quality of life,
total time to resettlement in the community and place of residence after hip
fracture.
Economic THR is the dominant strategy with respect to internal fixation (less costly and
more effective). This evidence has minor limitations and partial applicability.
10.3.3 Hemiarthroplasty versus total hip replacement
In patients having treatment for intracapsular hip fracture what is the clinical and cost
effectiveness hemiarthroplasty versus total hip replacement on mortality, number of
reoperations, functional status, length of stay in hospital, total time to resettlement in the
community, quality of life, pain and place of residence after hip fracture.
One systematic review265 was identified. From this, 7 RCTs with 734 participants met the
inclusion criteria. See evidence table 7, Appendix E and forest plots G87 to G95 in Appendix
G.
Table 10-40: Hemiarthoplasty vs total hip replacement – Clinical study characteristics

Numbe Other
Mortality (follow 3 RCT serious no serious no serious serious
up 3-6 limitations (a) inconsistency indirectness imprecision (b)
months)174,197,313
up 1 year)26,218,313 limitations (a) inconsistency indirectness imprecision (b)
up 2-4 limitations (a) inconsistency indirectness imprecision (b)
years)11,174,197,218
Total no. of 6 RCT serious no serious no serious serious
reoperations limitations (a) inconsistency indirectness imprecision (b)
(follow-up 8 to 48
months)11,26,73,174,
218,313
No. of patients 2 RCT no serious serious no serious serious

reporting pain at limitations inconsistency (d) indirectness imprecision (b)
1 years174,313
104 HIP FRACTURE
Numbe Other
Harris Hip Score 1 RCT serious no serious no serious serious
for pain - 12 limitations (a) inconsistency indirectness imprecision (c)
months26
Failure to regain 2 RCT serious no serious no serious serious
mobility (follow- limitations (a) inconsistency indirectness imprecision (b)
up 1 to 4
years)73,313
Oxford Hip Score 1 RCT no serious no serious no serious serious
- mean of 40 limitations inconsistency indirectness imprecision (c)
11
months
Barthel score - 1 RCT serious no serious no serious serious
one year218 limitations (a) inconsistency indirectness imprecision (c)
Barthel score - 1 RCT serious no serious no serious serious
four years218 limitations (a) inconsistency indirectness imprecision (c)
Hip rating 1 RCT no serious no serious no serious serious
questionnaire - limitations inconsistency indirectness imprecision (c)
174
24 months
Harris Hip Score - 2 RCT serious no serious no serious serious
total score - 12 limitations (a) inconsistency indirectness imprecision (c)
26,218
months
Harris Hip Score - 1 RCT serious no serious no serious serious
total score - four limitations (a) inconsistency indirectness imprecision (c)
years218
Harris Hip Score 1 RCT serious no serious no serious serious
for function - 12 limitations (a) inconsistency indirectness imprecision (c)
26
months
Short form 36 1 RCT no serious no serious no serious serious
physical score - limitations inconsistency indirectness imprecision (c)
mean of 40
months11
Self reported 1 RCT no serious no serious no serious serious
walking distance limitations inconsistency indirectness imprecision (c)
(kilometres) -
mean of 40
months11
EuroQol (EQ-5d) 1 RCT no serious no serious no serious serious
questionnaire - limitations inconsistency indirectness imprecision (c)
174
24 months
hospital stay174 limitations inconsistency indirectness imprecision (c)
concealment.
(c) The wide confidence intervals around the measurement make the result imprecise. This makes it
difficult to know the true effect size for this outcome.
(d) There is significant heterogeneity between the studies which maybe due to the types of
arthroplasty used.
Table 10-41: Hemiarthroplasty vs total hip replacement - Clinical summary of findings

25/192 11/166 RR 1.88 57 more per

Mortality (follow up 3- (13%) (6.6%) (0.96 to 3.68) 1,000 (from 3
Low
6 months) fewer to 174
more)
15 more per
Mortality (follow up 1 42/272 32/252 RR 1.15 1,000 (from 25
Low
year) (15.4%) (10.3%) (0.76 to 1.74) fewer to 76
more)
44 more per
Mortality (follow up 2- 38/176 29/169 RR 1.23 1,000 (from 38
Low
4 years) (21.6%) (19.1%) (0.8 to 1.87) fewer to 166
more)
Total no. or 6 more per 1,000
42/350 36/331 RR 1.06
reoperations (follow- (from 32 fewer Low
(12%) (10.6%) (0.7 to 1.6)
up 8 to 48 months) to 64 more)
50/133 29/123 RR 1.68 161 more per
No. of patients
(37.6%) (23.8%) (1.16 to 2.42) 1,000 (from 38
reporting pain (follow- Low
more to 338
up 1 years)
more)
Harris Hip Score for MD -4
55 56 N/A Low
pain - 12 months (-6.33 to -1.67)
43 fewer per
Failure to regain
17/110 20/101 RR 0.78 1,000 (from 111
mobility (follow-up 1 Low
(15.5%) (19.5%) (0.43 to 1.4) fewer to 78
to 4 years)
more)
Oxford Hip Score - MD 3.50
33 36 N/A Moderate
mean of 40 months (0.34 to 6.66)
Barthel score - one MD -8
30 33 N/A Low
year (-13.61 to -2.39)
Barthel score - four MD -5.7
20 23 N/A Low
years (-11.19 to -0.21)
Hip rating
MD -6.1
questionnaire – 2 50 56 N/A Moderate
(-12.38 to 0.18)
years
Harris Hip Score - total MD -5.47
85 89 N/A Low
score at 1 year (-8.39 to -2.55)
Harris Hip Score - total MD -4.2
20 23 N/A Low
score at 4 years (-7.66 to -0.74)
Harris Hip Score for MD -3.7
55 56 N/A Low
function - 12 months (-7.13 to -0.27)
Short form 36 physical
MD -2.43
score - mean of 40 33 36 N/A Moderate
(-7.56 to 2.7)
months
Self reported walking
MD -1.7
distance (kilometres) - 33 36 N/A Moderate
(-3.28 to -0.12)
mean of 40 months
EuroQol (EQ-5d)
MD -0.16
questionnaire at 2 65 66 N/A Moderate
(-0.28 to -0.04)
years
MD -0.80
Length of hospital stay 69 69 N/A Moderate
(-3.82 to 2.22)

106 HIP FRACTURE
Two studies were identified. Rogmark et al (2003)291 is a cost-consequence analysis based

on a RCT which was excluded because it does not distinguish patients on the basis of
whether they received hemiarthroplasty or total hip replacement. A cost-consequence
analysis comparing internal fixation vs. total hip replacement by Keating et al (2005)173 was
included. (Economic Evidence Table 14 in Appendix F)
Table 10-42: Hemiarthroplasty vs total hip replacement - Economic study characteristics

Keating 2005 173 Minor limitations (a) Partially applicability (b) Costs not discounted

(b) UK study but did not a CUA.
Table 10-43: Hemiarthroplasty vs total hip replacement - Economic summary of findings

Keating 2005 173 £498 (b) Hemiarthroplasty NA Two-way sensitivity
has lower EQ-5D analysis showed that the
scores at 4, 12 and direction of change in cost
24 months (b) did not change when cost
of prostheses and cost of
values.
(a) The mean cost per patient for hemiarthroplasty was 9,897 (95% CI: 8,062 – 11,732) and £9,399 (95% CI:
8,265-10,532) for THR.
(b) Hemiarthroplast had higher number of deaths within 4, 12 and 24 months from operation than THR: 5%
vs.4%; 10% vs. 6% and 16% vs. 9%; (p values not significant), but lower reoperation rates at 4, 12 and 24
months: 5% vs. 7%; 5% vs 9%; and 5% vs. 9% (p value NR). THR had higher mean EQ-5D scores at 4, 12
and 24 months: 0.61 vs. 0.68 (not significant); 0.64 vs. 0.70 (not significant); 0.53 vs 0.69 (p=0.008).
Clinical There is a statistically significant, but not clinically significant, decrease in

patients who reported pain and had a lower Harris Hip score for pain
(indicating better function) at 1 year with total hip replacement compared to
hemiarthroplasty (LOW QUALITY).
patients who have a lower Oxford Hip Score at 40 months (indicating better
function) with total hip replacement compared to hemiarthroplasty
(MODERATE QUALITY).
patients who have a higher Barthel Score (indicating better function) at 1 and
4 years (LOW QUALITY), a higher total Harris Hip Score at 1 and 4 years (LOW
QUALITY), a higher Harris Hip Score for function at 1 year (LOW QUALITY)and
a longer self reported walking distance at 40 months (MODERATE QUALITY)
with total hip replacement compared to hemiarthroplasty.

patients who have a higher Eq-5d (Euroqol) score at 2 years with total hip
replacement compared to hemiarthroplasty (MODERATE QUALITY).
There is no statistically significant difference in mortality at 2 to 4 months
(LOW QUALITY), 6 months (MODERATE QUALITY), 1 year (LOW QUALITY) or 2
to 4 years (LOW QUALITY), the number of reoperation at 8 to 48 months
(LOW QUALITY), the number of patients who fail to regain mobility at 1 to 4
years (LOW QUALITY), the Hip Rating Questionnaire Score at 2 years
(MODERATE QUALITY), the Short Form 36 (SF 36) score (MODERATE QUALITY)
and length of hospital stay (MODERATE QUALITY) between hemiarthroplasty
and total hip replacement.
No RCT evidence was identified reporting total time to resettlement or place
of residence after hip fracture for studies comparing total hip replacement
and hemiarthroplasty.
Economic THR is dominant compared to hemiarthroplasty. This evidence has minor
limitations and partial applicability.
Recommendation Perform replacement arthroplasty (hemiarthroplasty or total hip

replacement) in patients with a displaced intracapsular fracture.
Relative values of different The number of reoperations, functional status, pain and quality of
outcomes life were considered the important outcomes with the number of
reoperations being the most important. The interventions were
not anticipated to have a significant impact on mortality so this
was considered to be less important. Place of residence after hip
fracture was also considered to be less important as it is a
surrogate measurement for functional status.
Trade off between clinical Compared to internal fixation there was a significantly lower
benefits and harms reoperation rate with both hemiarthroplasty and total hip
replacement, less patient reported pain with total hip replacement
and better functional or quality of life scores with
hemiarthroplasty. There was no significant difference for mortality,
length of stay, failure to return to the same place of residence and
failure to regain mobility. None of the reported outcomes showed
any advantage of internal fixation over arthroplasty.
Economic considerations Evidence partially applicable to the UK with only minor limitations
was available on the cost-effectiveness of internal fixation vs.
hemiarthroplasty and internal fixation vs. total hip replacement.
The evidence shows that hemiarthroplasty is cost saving
compared to internal fixation. In particular, hemiarthroplasty
involved a significantly lower number of patients needing further
surgery at 12 and 24 months compared to internal fixation.
Similarly, THR required a lower rate of re-operation than internal
108 HIP FRACTURE
fixation, albeit not statistically significant.

Quality of evidence The evidence was of low or moderate quality. Most outcomes
were downgraded due to poor or uncertain allocation
concealment. Several results were imprecise as the confidence
intervals were near to one, making it difficult to determine the true
effect size. Some studies were also heterogenous that could be due
to the different types of arthroplasty.
Overall, the GDG felt that despite some of the results being of low
quality and data not being available for some outcomes where
there is a difference it shows arthroplasty being better than
internal fixation. Consequently arthroplasty is recommended.
Other considerations There maybe rare circumstances where reduction and internal
fixation is appropriate for displaced intracapsular fragility fractures.
People with cognitive impairment were excluded from a lot of the
studies. However, the GDG felt there is no reason for this group of
patients should be excluded from equal treatment to others.
All patients should be allowed to be mobilised full weight bearing
after hip fracture surgery (see section 10.2). All modern implants
are designed to be load sharing devices to facilitate this.
The GDG consider this recommendation a key priority for
implementation.
Recommendation Offer total hip replacement to patients with a displaced

intracapsular fracture who:
• were able to walk independently out of doors
with no more than the use of a stick and
• are medically fit for anaesthesia and the

procedure.
outcomes life were considered the important outcomes with the number of
reoperations being the most important. The interventions were
not anticipated to have a significant impact on mortality so this
fracture was also considered to be less important as it is a
surrogate measurement for functional status.
Trade off between clinical There was a significantly less patient reported pain and a better
benefits and harms Oxford Hip Score, Barthel Score, Harris Hip Score, self reported
walking distance and quality of life score (Eq-5d) with total hip
replacement compared to hemiarthroplasty. There was no
significant difference for mortality, length of stay, failure to return
to the same place of residence and failure to regain mobility. None
of the reported outcomes showed any advantage of
hemiarthroplasty over total hip replacement in the selected

patient group.
Economic considerations The cost-effectiveness evidence shows that THR replacement was
cost-saving compared to both hemiarthroplasty and internal
fixation.
Quality of evidence The evidence was of low or moderate quality. Most outcomes
were downgraded due to poor or uncertain allocation
concealment. Several results were imprecise as the confidence
intervals were near to one making it difficult to determine the true
effect size. Some studies were also heterogenous that could be due
to the different types of arthroplasty.
Overall, the GDG felt that despite some of the results being of low
quality and data not being available for some outcomes where
there is a difference it all shows total hip replacement being better
than hemiarthroplasty in the selected patient group. Consequently
total hip replacement is recommended for that group.
Other considerations All but one of the studies excluded patients who were not
medically fit, were not independently mobile before the fracture
and were cognitively impaired. Consequently this recommendation
does not include these groups. All the studies included in this
review used a small head size for total hip replacement. Modern
total hip replacements use a larger head which can reduce the risk
of dislocation.
The GDG consider this recommendation a key priortiy for
implementation.
Recommendation Use a proven femoral stem design rather than Austin Moore or
Thompson stems for arthroplasties. Suitable designs include
those with an Orthopaedic Data Evaluation Panel rating of 10A,
10B, 10C, 7A, 7B, 5A, 5B, 3A or 3B.
outcomes life were considered the important outcomes. The interventions
were not believed to have a significant impact on mortality so this
fracture was also considered to be less important.
Trade off between clinical Stem designs recommended here have a revision rate less than
benefits and harms other stem designs. A higher failure rate would lead to a lower
quality of life for patients.
Economic considerations No economic evidence was found. Stems with a higher failure rate
would require more reoperations and consequently, increased
110 HIP FRACTURE
costs and a lower quality of life for patients. Data supplied by an

expert advisor reported the cost of an Exeter Trauma stem (ETS)
monoblock as an example of a proven femoral stem design as £249
at 2008 prices.
Quality of evidence No randomised evidence comparing modern stems with older
stems was found.
Other considerations There is a move towards modern style cemented stems. The
Orthopaedic Data Evaluation Panel (OEDP) was set up in response
to the NICE guidance on selection of prosthesis for primary total
hip replacement226. The ratings used relate to the revision rate of
stems and cups in arthroplasty. The results are available via the
NHS Supply Chain website
(http://www.supplychain.nhs.uk/portal/page/portal/Communities/
Orthopaedics/ODEP%20database). A rating of 10A, 10B or 10C
relates to devices with a failure rate of arthroplasty of 10% or less
at 10 years. A rating of 7A and 7B relate to a failure rate of 7% or
less at 7 years. A rating of 5A and 5B relate to a failure rate of 5%
or less at 5 years. A rating of 3A and 3B relate to a failure rate of
3% or less at 3 years.
This recommendation was based on NICE guidance on selection of
prosthesis for pirmary total hip replacement and expert opinion. In
the light of such good evidence being available for the adequacy of
femoral stem designs for patients with degenerative change it was
thought that specific research in the fracture group would not be
appropriate.
Patients with hip fracture, particularly older patients have been
treated by methods which have evolved very little over the last 50
years. This has led to a perception that they may be receiving
second-class treatment. An example is the difference in the design
of hip replacement implants used in patients with fractures
compared with those used in patients with degenerative change.
Many of those used in the fracture patients now appear archaic
and their equivalents in the elective orthopaedic patients were
superseded many years ago.
Long-term follow-up studies to identify function and durability of a
replacement component in a fracture patient are difficult to carry
out as so many of the patients are frail and their life expectancy is
limited. However such studies are easier in patients with
degenerative change and there is a well recognised system of
assessing the adequacy of the design of a femoral stem for these
patients.
10.3.5 Research recommendations on displaced intracapsular fractures
10.3.5.1 Large head total hip replacement versus hemiarthroplasty
➢ What is the clinical and cost effectiveness of large-head total hip replacement
versus hemiarthroplasty on functional status, reoperations and quality of life in
patients with displaced intracapsular hip fracture?

intervention.
112 HIP FRACTURE
10.4 Use of cement in arthroplasty

The cement used in securing a hip replacement is not an adhesive but a grout, that is it is
used to fill the gaps between the metal prosthesis and the bone. Thus, a component fixed
with cement may be more secure resulting in less pain after surgery and decreased need
for surgical revision due to loosening of the prosthesis. However, it has been suggested that
cementing may induce side effects including cardiac arrhythmias and cardiorespiratory
collapse, both of which may be fatal. NPSA data reports 26 deaths and six cases of severe
harm when bone cement was used during hip surgery between October 2003 and October
2008. Data from the MHRA reports 20 deaths and four cases of severe harm with bone
cement between 2000 and 2008. The NPSA published advice on cementing techniques to
reduce such risk. However, patients undergoing surgery for proximal femoral fractures are
often elderly and frequently have multiple comorbidities, often severe. Therefore some
intraoperative deaths may occur and be unrelated to the use of cement.
10.4.1 Use of cement in original Thompson and Austin Moore designs of

arthroplasty
In patients having replacement arthroplasty for hip fracture what is the clinical and cost
effectiveness of a cemented stem versus an uncemented stem on mortality, number of
reoperations, wound healing complications, functional status, length of stay in hospital and
total time to resettlement in the community, quality of life, pain and place of residence
after hip fracture?
One systematic review265 including 6 RCTs with 899 participants was identified. See
Evidence Table 7 and forest plots G52 to G66 in Appendix G.
Table 10-44: Cemented vs. uncemented stem (original Thompson and Austin Moore designs of
arthroplasty) – Clinical study characteristics
Numbe Other
Perioperative 2 RCT no serious no serious serious serious
mortality136,260 limitations inconsistency indirectness (b) imprecision (b)
(b)
Mortality (follow 2 RCT no serious no serious serious serious

up <1 month)81,260 limitations inconsistency indirectness (b) imprecision (b)
up 3 limitations inconsistency indirectness (b) imprecision (b)
months)81,136,260,31
6
Mortality (follow 5 RCT no serious no serious no serious serious

up 1 limitations inconsistency indirectness imprecision (g)
year)34,81,136,260,298
Numbe Other
Failure to regain 4 RCT no serious serious no serious serious
mobility (follow- limitations inconsistency indirectness imprecision (k)
(i,j)
up 12 to 17
months)81,260,316
Change in 1 RCT no serious no serious serious serious
mobility score limitations inconsistency indirectness imprecision (l)
(a,l)
(follow-up 12
months; better
indicated by
less)260
Length of 4 RCT serious no serious no serious serious
hospital limitations inconsistency indirectness imprecision (b)
stay81,136,260,298 (d,e)
Failure to return 2 RCT no serious no serious no serious serious

home (follow up limitations inconsistency indirectness imprecision (b)
1.5 to 5
years)81,260
Pain (follow up 3 2 RCT no serious no serious serious serious
months)260,316 limitations inconsistency indirectness imprecision (b)
(a,m)
Pain (follow up 1- 3 RCT no serious no serious serious no serious

2 years)81,260,316 limitations inconsistency indirectness (b) imprecision
Pain score (follow 1 RCT no serious no serious serious serious
up 6 months)260 limitations inconsistency indirectness imprecision (b)
(a,m)
Reoperations 2 RCT no serious no serious serious serious

(follow-up 8 to 20 limitations inconsistency indirectness (a) imprecision (b)
months)34,260
Deep Sepsis 4 RCT no serious no serious no serious serious
(follow-up 1 to 5 limitations inconsistency indirectness imprecision (b)
136,260,298,316
years)
Wound 1 RCT no serious no serious no serious serious
haematoma limitations inconsistency indirectness imprecision (b)
(follow-up 1 to 5
years)260
(a) Data only available for unipolar hemiarthroplasty
(c) The result is calculated using only one of the two studies with no allocation concealment or
blinding of the intervention. However, the effect size is similar between the two studies, the second
study is larger and does not having any serious limitations in design. There fore the evidence has
not been downgraded on the basis of study quality.
(d) Unclear or no allocation concealment in 2 of the 4 studies which account for over 75% of the weight
of the result.
(e) Randomisation method by odd or even hospital number in 1 of the 4 studies, and by alternate days
in another of the 4 studies. These 2 of the 4 studies account for over 75% of the weight of the result.
(f) The estimate of effect is derived from the data relating to unipolar hemiarthroplasty. There is a
small study relating to bipolar arthroplasty, this has little impact on the overall result.
(g) The confidence intervals around the estimate of effect are wide enough to suggest some
uncertainty in the estimate of the effect. A larger number of patients may show a statistically
significant difference in the outcome.
(h) The estimate of effect is calculated with the better quality studies having more weight than the
lower quality studies. Consequently, the result has not been downgraded for quality.
114 HIP FRACTURE
(i) There is significant statistical heterogeneity in the results: there is no statistical for unipolar
hemiarthroplasty; Significantly more patients failed to regain mobility with uncemented bipolar
hemiarthroplasty than cemented bipolar hemiarthroplasty.
(j) The definition for failure to regain mobility is different in the studies. The two studies, one showing
no statistical difference the other favouring cement, measure the number of people with a change
in their walking status. The third study showing no statistical difference measures the number of
people unable to walk properly (this includes walking without a limp) .
(k) The confidence intervals around the estimate of effect are wide enough to suggest some
uncertainty in the estimate of the effect.
(l) Definition of mobility score not given. Unable to determine if it is a valid measurement for mobility
or if the estimate of effect is clinically significant.
(m) How pain was measured is not reported for the study with the most weight in the meta-analysis.
Unable to determine if it is a valid measurement or if the estimate of effect is clinically significant.
Table 10-45: Cemented vs uncemented stem (original Thompson and Austin Moore designs of
arthroplasty) - Clinical summary of findings
0 fewer per
Perioperative 1/277 0/266 RR 2.58 (0.11 to
1,000 (from 0 Low
mortality (0.4%) (0%) 62.21)
fewer to 0 more)
10 fewer per
Mortality (follow up 11/227 13/226 1,000 (from 28
RR 0.84 (0.38 to 1.84) Low
<1 month) (4.8%) (6.6%) fewer to 54
more)
3 fewer per 1000
Mortality (follow up 3 49/359 49/349
RR 0.98 (0.68 to 1.41) (from 45 fewer Low
months) (13.6%) (13%)
to 57 more)
28 fewer per
Mortality (follow up 1 101/395 113/398 1000 (from 82
RR 0.9 (0.71 to 1.13) Moderate
year) (25.6%) (26.4%) fewer to 37
more)
109 fewer per
Failure to regain
117/196 124/182 1000 (from 245
mobility (follow-up 12 RR 0.84 (0.64 to 1.11) Low
(59.7%) (68.1%) fewer to 75
to 17 months)
more)
Change in mobility
score (follow-up 12 MD -0.8 (-1.23 to
150 144 N/A Low
months; better -0.37)
indicated by less)
MD -1.42 (-3.15
Length of hospital stay 354 342 N/A Low
to 0.32)
45 fewer per
Failure to return home
16/219 26/220 1000 (from 78
(follow up 1.5 to 5 RR 0.62 (0.34 to 1.12) Moderate
(7.3%) (11.8%) fewer to 14
years)
more)
106 fewer per
Pain (follow up 3 67/192 84/183 1000 (from 9
RR 0.77 (0.6 to 0.98) Low
months) (34.9%) (45.9%) fewer to 184
fewer)
187 fewer per
Pain (follow up 1-2 44/193 73/176 1000 (from 104
years) (22.8%) (41.5%) fewer to 249
fewer)
Pain score (follow up 6 MD -0.6 (-0.9 to -
147 142 - Low
months) 0.3)
34 fewer per
Reoperations (follow- 10/238 19/253 1000 (from 55
RR 0.55 (0.27 to 1.14) Low
up 8 to 60 months) (4.2%) (7.5%) fewer to 11
more)
4 more per 1000
Deep sepsis (follow up 8/385 6/376
RR 1.25 (0.48 to 3.24) (from 8 fewer to Moderate
1 to 5 years) (2.1%) (1.6%)
36 more)
Wound Haematoma 5 more per 1000
1/200 RR 2.01 (0.18 to
(follow up 2 to 5 2/200 (1%) (from 4 fewer to Moderate
(0.5%) 22.35)
years) 107 more)
Two economic studies were identified. Santini (2005)298 is a cost-consequence analysis

based on a RCT included in our clinical review (see 10.3.3.2). See evidence table 15 in
Appendix F for additional details. Marinelli (2008) 206 was excluded because of poor
methodology.
Table 10-46: Cemented vs. uncemented hemiarthroplasty - Economic study characteristics

Santini 2005298 Potentially serious limitations Partially applicable (b)Based on RCT included in
(a)
our clinical review (see
10.3.3.2).
(a) Surgical time not included in cost calculation although it was significantly different (patients in the
uncemented hemiarthroplasty group had shorter operating time). The only difference considered was
the cost of prostheses.
(b) Not a cost-utility analysis. Study conducted in Italy.
Table 10-47: Cemented vs. uncemented hemiarthroplasty - Economic summary of findings

Santini 2005298 Cost saving (-£710) (b)
N/R N/R
(b)
(a) Cost of medical and nursing staff, drugs, diagnostic procedures, prostheses, blood transfusion and
hospital sta. Converted into GBP from 2001 euro using the Purchasing Power Parities.
(b) Different outcomes were reported but none of them were significantly different.
Clinical There is a statistically significant, but not clinically significant, increase in

patients who have a lower reduction in mobility score (less loss of mobility) at
12 months (LOW QUALITY).
There is a statistically significant, but not clinically significant, decrease in
patients who reported pain at 3 months (LOW QUALITY) and 1 to 2 years
(MODERATE QUALITY). However, there was no significant difference in a pain
score at 6 months (LOW QUALITY).
There is no statistically significant difference in perioperative mortality (LOW
QUALITY), mortality at 3 months (LOW QUALITY) or 1 year (MODERATE
QUALITY), failure to return home (MODERATE QUALITY), length of hospital
stay (LOW QUALITY), number of patients requiring reoperations (LOW
116 HIP FRACTURE
QUALITY), number of patients failing to regain mobility (LOW QUALITY), deep

sepsis (MODERATE QUALITY), wound haematoma (MODERATE QUALITY) and
all medical complications combined (VERY LOW QUALITY).
No RCT evidence was identified reporting quality of life, total length of stay to
community resettlement or place of residence after hip fracture
No RCT evidence was identified to suggest there is a safety issue with using
cement.
Economic Cemented hemiarthroplasty is cost saving compared to uncemented

hemiarthroplasty. This evidence has potentially serious limitations and partial
applicability.
10.4.2 Use of cement in newer designs of arthroplasty
In patients having replacement arthroplasty for hip fracture what is the clinical and cost
effectiveness of a cemented stem versus an uncemented stem on mortality, number of
reoperations, wound healing complications, functional status, length of stay in hospital and
total time to resettlement in the community, quality of life, pain and place of residence
after hip fracture.
One RCT94 including 220 participants was identified. See Evidence Table 7 and forest plots
G67 to G73 in Appendix G.
Table 10-48: Cemented vs. uncemented stem (newer designs of arthroplasty) – Clinical study
characteristics
Numbe Other
up 30 days)94 limitations inconsistency indirectness (b) imprecision (b)
up 90 days)94 limitations inconsistency indirectness (b) imprecision (b)
up 1 year)94 limitations inconsistency indirectness (b) imprecision (b)
up 2 years)94 limitations inconsistency indirectness (a) imprecision (b)
Total number of 1 RCT no serious no serious serious serious
reoperations limitations inconsistency indirectness (a) imprecision (b)
(follow up 12
months)94
Need for pain 1 RCT no serious no serious serious serious
medication limitations inconsistency indirectness (a) imprecision (b)
(follow up 12
months)94
Numbe Other
Unable to walk 1 RCT no serious no serious serious serious
without aids limitations inconsistency indirectness (a) imprecision (b)
(follow up 12
months)94
Barthel score of 1 RCT no serious no serious serious serious
less than 19 limitations inconsistency indirectness (a) imprecision (b)
(follow up 12
months)94
Harris Hip Score 1 RCT no serious no serious serious serious
(a)
(follow up 12 limitations inconsistency indirectness imprecision (b)
94
months)
Eq-5d index score 1 RCT no serious no serious serious serious
(follow up 12 limitations inconsistency indirectness (a) imprecision (b)
94
months)
Eq-5d visual 1 RCT no serious no serious serious serious
analogue score limitations inconsistency indirectness (a) imprecision (b)
(follow up 12
months)94
Length of 1 RCT no serious no serious serious serious
hospital stay94 limitations inconsistency indirectness (a) imprecision (c)
(a) Data only available for bipolar hemiarthroplasty
(c) The effect size is uncertain as the confidence intervals suggest the length of stay could be over 2
days shorter or over 1 day longer with cemented hemiarthroplasty.
Table 10-49: Cemented vs uncemented stem - Clinical summary of findings

Mortality (follow up 35 fewer per 1000
8/142 10/153 RR 0.47 (0.15 Low
30 days) (from 56 fewer to 37
(5.6%) (6.5%) to 1.57) quality
more)
Mortality (follow up 23 fewer per 1000
13/108 15/105 RR 0.84 (0.42 Low
90 days) (from 83 fewer to 97
(12%) (14.3%) to 1.68) quality
more)
Mortality (follow up 1 105 fewer per 1000
34/142 46/153 RR 0.65 (0.39 Low
year) (from 183 fewer to 21
(23.9%) (30.1%) to 1.07) quality
more)
Mortality (follow up 2 48 fewer per 1000
32/108 36/105 RR 0.86 (0.58 Low
years) (from 144 fewer to 96
(29.6%) (34.3%) to 1.28) quality
more)
Total number of 12 fewer per 1000
7/112 8/108 RR 0.84 (0.32 Low
reoperations (follow (from 50 fewer to 93
(6.3%) (7.4%) to 2.25) quality
up 12 months) more)
Need for pain 71 more per 1000
23/91 14/77 RR 1.39 (0.77 Low
medication (follow up (from 42 fewer to 275
(25.3%) (18.2%) to 2.51) quality
12 months) more)
Unable to walk 34 fewer per 1000
RR 0.56 (0.17 Low
without aids (follow 4/91 (4.4%) 6/77 (7.8%) (from 65 fewer to 72
to 1.93) quality
up 12 months) more)
Barthel score of less 128 more per 1000
46/91 29/77 RR 1.34 (0.94 Low
than 19 (follow up 12 (from 23 fewer to 343
(50.5%) (37.7%) to 1.91) quality
months) more)
118 HIP FRACTURE
Harris Hip Score MD 0.9 lower (6 lower Low

90 77 N/A
(follow up 12 months) to 4.2 higher) quality
Eq-5d index score MD 0.07 higher (0.03 Low
56 57 N/A
(follow up 12 months) lower to 0.17 higher) quality
Eq-5d visual analogue
MD 4 lower (10.75 Low
score (follow up 12 61 60 N/A
lower to 2.75 higher) quality
months)
Length of hospital stay MD 0.6 lower (2.48
109 106 N/A
lower to 1.28 higher)
No cost-effectiveness evidence was identified. A cost analysis was conducted based on

the resources used in the Figved study94 and on GDG expert opinion. Please see section
20.8 of Appendix H of this guideline for further details.
Table 10-50: Cemented stems versus uncemented stems (newer designs of arthroplasty) –
Economic study characteristics
NCGC cost analysis Minor limitations (a) Partially applicable (b) Cost analysis based on
resources reported in
Figved (2009)94 and on
GDG’s expert opinion
(a) No sensitivity analysis.

(b) Cost analysis based on one study alone by Figved94 and on GDG’s expert opinon. The study by Figved94 is
not UK based and therefore may not completely reflect current NHS practice.
Table 10-51: Cemented stems versus uncemented stems (newer designs of arthroplasty) –
Economic summary of findings
Incremental
Study Incremental cost (£) effects ICER Uncertainty
NCGC cost analysis £171.79(a) N/A N/A N/R
(cost saving)
(a) The following cost categories were considered in the cost analysis: cost of implants;
length of hospital stay; cost of cement accessorises; theatre time costs; re-operation
costs. The costs of length of stay and re-operation were considered in the analysis even
if in the RCT by Figved94 there was not statistically significant difference between the
two groups for these outcomes. The total cost for the new design cemented stems was
estimated to correspond to £2751.64 and that for the new design uncemented stems
to £2923.43. The estimate for the total cost for the cemented stems could increase up
to £2859.75 when a more thorough set of accessories are assumed to be used in the
operation, in which case the incremental savings associated with using cemented
stems would amount to £63.68. See Appendix H section 20.8 for further details.
Clinical There is no statistically significant difference in mortality at 30 days, 90 days,

1 year or 2 years (LOW QUALITY).
There is no statistically significant difference at 1 year in the number of

patients requiring reoperations, number of patients pain requiring
medication, number of patients unable to walk without aids, Barthel Score of
less than 19, Harris Hip Score, Eq-5d index score and visual analogue score,
deep wound sepsis, any wound infection, length of hospital stay (LOW
QUALITY).
No RCT evidence was identified reporting total time to resettlement in the
community and place of residence after hip fracture
No RCT evidence was identified to suggest there is a safety issue with using
cement.
Economic No studies were identified on the cost-effectiveness of cemented vs.
uncemented stem (newer designs of arthroplasty). An NCGC cost analysis
found that cemented stems are £171.79 cheaper than the newer design
uncemented stems. This evidence has minor limitation and partial
applicability.
Recommendation Use cemented implants in patients undergoing surgery with

arthroplasty
Relative values of different The outcomes considered were mortality, functional status, quality
outcomes of life, pain, requirement for reoperation, non-healing and
requirement for surgical revision, total length of stay (i.e. the time
in hospital plus any time spent in rehabilitation). Mortality was of
particular importance because of reported deaths by the NPSA.
Trade off between clinical There is no significant difference in mortality. There is evidence of
benefits and harms less pain at 3 months and 1 to 2 years and better mobility score at
12 months with the older designs of cemented hemiarthroplasties.
There was no significant difference for length of stay, failure to
return to the same place of residence and failure to regain
mobility. None of the reported outcomes showed any advantage of
uncemented arthroplasty over cemented.
More evidence is available for older designs than newer designs of
arthroplasty. Only one study was identified in newer designs. This
showed no statistical difference for any reported outcomes. The
direction of effect varies depending on the outcome: cemented
implants are favoured for mortality, number of reoperations,
length of stay, ability to walk unaided at 12 months; uncemented
for need for pain medication at 12 months and Barthel index. The
Eq-5d visual analogue score also favours uncemented. However,
the Eq-5d index score shows no difference with tight confidence
intervals. In light of this uncertainty in newer designs, the
increased costs and lack of evidence or clinical reason to suggest a
difference between the use of cement in newer and older stem
120 HIP FRACTURE
designs the GDG considered that cemented implants should be

recommended for all arthroplasties.
There is no direct evidence comparing the use of cemented and
uncemented stems in total hip replacement for displaced
intracapsular fractures. However, the GDG did not consider there
would be a difference in the performance of cemented stems
between outcomes for total hip replacement and
hemiarthroplasty. Also, all the studies which looked at total hip
replacements in other comparisons (section 10.3.2) used cemented
femoral stems for total hip replacement.
No RCT evidence was found to raise concerns about the safety of
the use of cement.
Economic considerations One study with potentially serious limitations and partial
applicability found that the older cemented hemiarthroplasty are
cost saving compared to uncemented hemiarthroplasty.
The NCGC cost analysis on cemented stems versus uncemented
stems for newer designs of arthroplasty has considered several
cost components, such as the cost of the implants, length of stay in
hospital, rate of re-operations, accessories costs for the cemented
implants.
As the clinical evidence did not show any advantage of
uncemented over cemented arthroplasty in the newer design, and
as the cost of new designs of cemented implants was shown to be
lower than that of uncemented implants, the GDG consider
cemented implants cost-effective based on the outcomes reported
though these are not statistically significant.
One outcome reported in Figved showed a higher level of blood
loss with cemented hemiratrhoplasty. However, the GDG did not
consider the higher level of blood loss reported in Figved et al
(2009)94 for patients receiving cemented implants (89mL) to be
significant in terms of both patients’ outcomes and costs.
Quality of evidence The evidence was of low or moderate quality. All but one of the
studies comparing older arthroplasty designs used a Thompson or
Austin Moore hemiarthroplasty (these are the first generation of
implants to be used). The other study used an unspecified bipolar
hemiarthroplasty. The evidence for modern stem designs is low
quality mainly due to the lack of certainty around the effect size
and only evidence being identified in bipolar hemiarthroplasty.
Overall, the GDG felt there was sufficient evidence to recommend
the use of cemented arthroplasties over uncemented.
Other considerations All studies comparing the effectiveness of internal fixation with
THR and hemiarthroplasty with THR used cemented THR (see
section 10.3.2)
10.5 Surgical approach to hemiarthroplasty

Hemiarthroplasties are usually inserted using one of two approaches, either an
anterolateral or a posterior approach. The choice of surgical approach for a surgeon is
often dictated by local custom and practice and personal experience. This review looks at
the evidence to see if one is better than the other. RCTs and cohorts adjusted for
confounders were included.
In patients having surgical treatment for intracapsular hip fracture with hemiarthroplasty
what is the clinical and cost effectiveness of anterolateral compared to posterior surgical
approach on mortality, number of reoperations, dislocation, functional status, length of
hospital stay, quality of life and pain.
One systematic review269 including 1 RCT with 114 participants and one cohort study
involving 720 participants were identified. See Evidence Table 9, Appendix E.
Table 10-52: Posterior vs. anterolateral approach to hemiarthroplasty – Clinical study

characteristics
Numbe Other
Mortality309 1 RCT serious Unable to serious serious
limitations (a, b) assess this (f) indirectness (c) imprecision (d)
Number of 1 RCT serious no serious serious serious
patients with limitations (a, b) inconsistency indirectness (c) imprecision (d)
impairment of
mobility at 6
months
compared to
prefracture309
Dislocation at 0 1 RCT Very serious no serious serious serious
to 2 years309 limitations (a, b) inconsistency indirectness (c) imprecision (d)
Dislocation at 0 1 Cohor serious no serious no serious no serious
to 10 years85 t limitations (e) inconsistency indirectness imprecision
Pain at 1 1 RCT serious no serious serious serious
month309 limitations (a) inconsistency indirectness (c) imprecision (d)
(a) Unclear allocation concealment and randomisation method
(b) Patients allocated to the posterior approach were nursed flat in bed for two weeks after surgery as
a precaution against dislocation.
(c) Most operations performed by surgical trainees
(d) The wide confidence intervals make the estimate of effect imprecise.
(e) Only a limited number of confounders were included in the analysis. No adjustment or mention of
the anaesthetists experience or grade.
(f) Actual event rates were not provided for this, mortality was given as percentages in a graph. The
percentages were estimated usingthis. Mortality was significantly higher at three months, six
122 HIP FRACTURE
months, 12 months and two years in the posterior group _p<0.05. The rate was around double for
all these time points.
Table 10-53: Posterior vs. anterolateral approach to hemiarthroplasty - Clinical summary of

findings
Mortality at 6 Significantly higher in
Not Not
months, 12 months & posterior group Not estimable Very low
reported reported
2 years (p<0.05)
Number of patients
220 fewer per
with impairment of
5/34 15/41 1000 (from 4
mobility at 6 months RR 0.40 (0.16 to 0.99) Very Low
(14.7%) (36.6%) fewer to 307
compared to
fewer)
prefracture
Dislocation at 0 to 2 0 fewer per 1000
RR 1.00 (0.06 to
years 1/57 (1.8%) 1/57 (1.8%) (from 16 fewer Very Low
15.60)
to 256 more)
Dislocation at 0 to 10 87 more per
years (posterior 15/176 multivariate odds 1000 (from 18
13/431 (3%) Very Low
approach with (8.5%) ratio 3.9 (1.6 to 9.8) more to 265
posterior repair) more)
Dislocation at 0 to 10 178 more per
years (posterior 17/129 multivariate odds 1000 (from 48
13/431 (3%) Very Low
approach without (13.2%) ratio 6.9 (2.6 to 19) more to 543
posterior repair) more)
Pain at 1 month 218 more per
6/55 1000 (from 40
2/55 (3.6%) RR 3.0 (0.63, 14.22) Very low
(10.9%) fewer to 1442
more)
No cost-effectiveness evidence was identified.
Clinical Two studies of different designs showed different effects for dislocation rates.
One old RCT showed no statistically significant difference in dislocation rate
between approaches. (VERY LOW QUALITY). One recent cohort which
adjusted for confounders showed a statistically and clinically significant
higher dislocation rate with the posterior approach compared to the
anterolateral approach. (VERY LOW QUALITY)
Significantly fewer patients had impaired mobility at 6 months with a
posterior approach to hemiarthroplasty compared to an anterior approach
when the procedure was performed by surgical trainees. (VERY LOW
QUALITY)
One study reported a significantly higher mortality with a posterior approach
at 6 months, 12 months and two years but did not provide the event rates.
(VERY LOW QUALITY]
Economic No evidence was identified regarding the cost-effectiveness of posterior vs.

anterolateral approach to hemiarthroplasty.
Recommendation Consider an anterolateral approach in favour of a posterior

approach when inserting a hemiarthroplasty.
Relative values of different Functional status, reoperation rate, and quality of life were
outcomes considered the main outcomes. Pain, wound infection,
dislocations, length of stay in secondary care and mortality were
also considered.
Trade off between clinical The cohort study showed a significantly higher dislocation rate
benefits and harms with a large effect size with the posterior approach compared to
the anterolateral approach. This reduces the potential
complications of re-operation or revision surgery. An old RCT data
showed a significantly lower impaired mobility at 6 months with a
posterior approach, a doubling of mortality and no difference in
dislocations compared to an anterolateral approach. However, the
operations had been carried out by trainees with varying degrees
of experience. Also, the group operated on with an antrolateral
approach were allowed to mobilise straight away and the group
operated on with a posterior approach had two weeks
postoperatively bed rest.
None of the other outcomes were reported.
Economic considerations An anterolateral approach is likely to result in cost savings
because of their lower dislocation rates, and hence less revision
surgery.
Quality of evidence Both the studies available are of very low quality. The RCT is an old
study where the operations were mostly carried out by surgical
trainees. This RCT also treated patients differently, with those
receiving a posterior approach being nursed flat in bed for two
weeks after surgery as a precaution against dislocation and had a
much higher mortality in the posterior group. The cohort study,
which adjusted for important factors in their results, is a recent
study and shows a large effect size in favour of an anterolateral
approach.
Other considerations The GDG considered this evidence along with the GDG opinions
and decided the recent evidence is more relevant. They therefore
recommend the anterolateral approach over the posterior. It is
also recognized that the posterior approach may well be as safe in
preventing dislocation in those surgeons with a large experience of
using it. However, the GDG believe the majority of surgeons who
perform the surgery do not regularly perform posterior
approaches. It is also noted that all the RCTs comparing
hemiarthroplasty and total hip replacement utilized the
124 HIP FRACTURE
anterolateral approach in all of the studies.
10.6 Extracapsular fracture fixation

In the extracapsular fractures the femoral head blood supply is unaffected and the proximal
fragment large enough to allow secure fixation, therefore internal fixation is the norm. The
surgical decision in this group is which of the various available methods of fracture fixation
is most effective for each pattern. When treating the extracapsular fractures around the
trochanter it is necessary to stabilise the intact femoral head and neck onto the shaft of the
femur. The head portion is stabilised by one or more screws up the neck and into the head.
This screw is attached to either a plate on the outside of the bone (called extramedullary
fixation) or a metal rod which is inserted down the middle of the femoral shaft
(intramedullary fixation). The rod can either be short, spanning approximately a third of the
length of the femur, or long spanning the whole length of the femur. The generic term for
the plate and screw used for the extramedullary fixation is a sliding hip screw and the term
for the intramedullary fixation is the intramedullary nail.
Extracapsular fractures are split into pertochanteric (also called intertrochanteric), reverse
oblique and subtrochanteric (see Introduction, Figure 1).
10.6.1 Intramedullary versus extramedullary implants for fixation of

trochanteric extracapsular fractures
There are numerous studies comparing intramedullary and extramedullary results. The
intramedullary nails can vary in size and shape, with most evolving from the initial nail
design which was changed due to an increase in per-and postoperative fractures of the
femur. When reviewing the evidence, the trochanteric fractures were divided into stable
fractures, (those with an intact lesser trochanter – AO/ OTA A1), unstable fractures (those
with a fracture between the trochanters with displacement of the lesser trochanter –
AO/OTA A2 fractures) and reverse oblique fractures (AO/OTA A3). Historically and presently
there have been numerous implants used to treat these and we have divided them into
intramedullary (those which have a rod down the shaft of the bone) and extramedullary
where the device sits on the outside of the bone. Commonly these are called
intramedullary nails and sliding hip screws respectively. The intramedullary nails can come
in various designs from different manufacturers. Their size and shape have evolved over the
last twenty years. The design of the sliding hip screw has not changed over the last thirty
years and sliding hip screws are generally very similar between the different manufacturers.
In patients undergoing repair for trochanteric extracapsular hip fractures what is the clinical
and cost effectiveness of extramedullary sliding hip screws compared to intramedullary
nails on mortality, surgical revision, functional status, length of stay, quality of life, pain and
place of residence after hip fracture?
21 studies met the inclusion criteria for this review question with a total of 4,336 patients.
See evidence table 5.8, Appendix E and forest plots G96 to G106 in Appendix G.
Table 10-54: Intramedullary vs. extramedullary implants for trochanteric extracapsular fracture
– Clinical study characteristics
Numbe Other
Mortality – 30 9 RCT no serious no serious no serious no serious
days14,37,128,137,191,1 limitations inconsistency indirectness imprecision
95,244,279,337 (a,b)
Mortality – 3 3 RCT serious no serious no serious serious

months128,134,251 limitations inconsistency indirectness imprecision (c)
(d,e)
Mortality – 1 11 RCT no serious no serious no serious no serious

year3,14,37,77,134,191, limitations (f) inconsistency indirectness imprecision
195,251,294,300,337
Reoperation – 16 RCT no serious no serious no serious no serious

within follow up limitations inconsistency indirectness(h) imprecision
period of
study9,14,77,128,134,14
7,191,195,214,244,251,254,
279,294,300,337
Operative or 17 RCT no serious no serious serious no serious

postoperative limitations inconsistency indirectness (k) imprecision
(h, j)
fracture - within
follow up period
of study
3,9,37,77,128,134,137,147,
191,214,244,251,258,279,3
00,337,364
Cut-out (at latest 20 RCT no serious no serious no serious serious

follow up) limitations(l) inconsistency indirectness imprecision(c)
9,14,37,77,128,134,137,147
,191,195,214,244,251,254,2
58,279,294,300,337,364
Infection (deep 14 RCT no serious no serious no serious serious

infection or limitations(m inconsistency indirectness (n) imprecision (c)
)
requires
reoperation)128,134
,147,191,195,214,244,251,2
54,258,279,294,300,337
Non-union (at 9 RCT no serious no serious no serious serious

latest follow limitations inconsistency indirectness (p) imprecision (c)
up)77,137,191,251,258,2 (o)
79,294,300,364
Pain (at latest 4 RCT no serious no serious serious no serious

follow limitations inconsistency indirectness(q) imprecision
up)134,147,191,337
Length of stay in 8 RCT no serious serious (g) no serious serious
hospital137,147,191,2 limitations indirectness imprecision (c)
44,251,254,294,300
Mean mobility 4 RCT no serious no serious no serious no serious

(Parker – Palmer limitations inconsistency indirectness imprecision
score. At 1
year)134,294,300,337
126 HIP FRACTURE
(a) Unclear allocation concealment in 4 out of 9 studies.

(b) Loss to follow up not reported or more than 5% in 4 out of 9 studies
(c) The relatively few events and few patients give wide confidence intervals around the estimate of
(d) Unclear allocation concealment in 2 out of 3 studies.
(e) Loss to follow of not reported or more than 5%, in 2 out os 3 studies.
(f) Unclear allocation concealment in 3 out of 11 studies.
(g) There is significant statistical heterogeneity in the results
(h) The definition of reoperation varies between studies to include minor or major revisions.
(i) Unclear allocation concealment in 7 out of 15 studies.
(j) Loss to follow up not reported more than 5% in 8 out of 16 studies.
(k) All fractures of the femur that were reported have been combined.
(l) Loss to follow up not reported or more than 5% in 8 out 19 studies.
(m) Loss to follow up not reported or more than 5% in 5 out of 15 studies
(n) Inclusion of reported infection varied between studies and included deep infection and infection
that required reoperation.
(o) Loss to follow up not reported or more than 5% in 4 out of 10 cases.
(p) All cases of non-union were combined using data at latest follow up.
(q) Different definitions of patient reported pain combined.
Table 10-55: Intramedullary vs. extramedullary implants for trochanteric extracapsular fracture
- Clinical summary of findings
Intramedull Extramedull
Outcome ary ary Relative risk Absolute effect Quality
Mortality – 30 days 34 more per
78/712 56/729 RR 1.44 (1.04 to 1000 (from 3
High
(11%) (7.7%) 1.99) more to 76
more)
Mortality – 3 months 12 fewer per
19/173 21/173 RR 0.9 (0.52 to 1000 (from 58
Low
(11%) (10%) 1.59) fewer to 72
more)
Mortality – 1 year 15 more per
186/1005 175/1021 RR 1.09 (0.91 to 1000 (from 15
High
(18.5%) (17.1%) 1.31) fewer to 53
more)
Reoperation – within 15 more per
follow up period of 69/1261 50/1312 RR 1.39 (0.87 to 1000 (from 5
High
study (5.5%) (3.8%) 2.23) fewer to 47
more)
Operative or 16 more per
postoperative fracture 54/1334 RR 5.61 (2.98 to 1000 (from 7
5/1380 (0%) Low
- within follow up (4%) 10.59) more to 33
period of study more)
Cut-out (at latest 1 fewer per 1000
39/1446 42/1508 RR 0.95 (0.63 to
follow up) (from 10 fewer Moderate
(2.7%) (2.8%) 1.45)
to 13 more)
Infection (deep 1 fewer per 1000
8/922 RR 0.86 (0.38 to
infection or requires 10/943 (1%) (from 7 fewer to Moderate
(0.9%) 1.93)
reoperation) 10 more)
Non-union (at latest 0 more per 1000
3/610 3/621 RR 1.01 (0.3 to
follow up) (from 3 fewer to Moderate
(0.5%) (0.5%) 3.46)
12 more)
Pain (at latest follow 9 more per 1000
90/278 90/285 RR 1.03 (0.81 to
up) (from 60 fewer Low
(32.4%) (25.9%) 1.30)
to 95 more)
Length of stay in MD 0.54 lower
hospital 474 482 N/A (1.93 lower to Moderate
0.84 higher)
Mean mobility (Parker MD 0.17 higher
– Palmer score. At 1 274 281 N/A (0.17 lower to High
year) 0.51 higher)
Three economic studies were indentified 110,114,179. All these studies have been excluded. 114
is a cost-consequence analysis based on a retrospective cohort study set in the US
comparing trochanteric fixation nail with sliding hip screw. This study was excluded due to
poor methodological design and to the limited applicability to the UK NHS. 179 compared
proximal femoral nail with long-stem cementless calcar-replacement prosthesis which was
not an included intervention. Another study 110 was excluded as no cost figures were
reported.
128 HIP FRACTURE
The GDG was informed of the prices of implants produced by all major orthopaedic
suppliers in the UK. At 2010 prices, the average cost for a sliding hip screw was estimated at
£252.51, of a short intramedullary nail at £760.08, and of a long intramedullary nail at
£1,175.40. Please see section 20.3 in Appendix H for further details.
Clinical There is a statistically significant and clinically significant increase in operative

or postoperative fracture of the femur with intramedullary implants
compared to extramedullary implants for fixation of trochanteric
extracapsular fractures. (LOW QUALITY)
There is no statistically significant difference in mortality, reoperation, and
mean mobility score with intramedullary implants compared to
extramedullary implants for fixation of trochanteric extracapsular fractures.
(HIGH QUALITY)
There is no statistically significant difference in cut-out, infection, non-union
and length of hospital stay with intramedullary implants compared to
extramedullary implants for fixation of trochanteric extracapsular fractures.
(MODERATE QUALITY)
There is no statistically significant difference in pain, with intramedullary
implants compared to extramedullary implants for fixation of trochanteric
extracapsular fractures. (LOW QUALITY)
No studies were identified investigating reverse oblique trochanteric
extracapsular fractures.
Economic No applicable evidence was identified regarding the cost-effectiveness of
Intramedullary vs. extramedullary implants.

evidence
Recommendation Use extramedullary implants such as a sliding hip screw in

preference to an intramedullary nail in patients with trochanteric
fractures above and including the lesser trochanter (AO
classification types A1 and A2).
Relative values of different The most important outcomes considered by the GDG include early
outcomes and late mortality, re-operation, postoperative fracture, length of
hospital stay and post fracture mobility.
Trade off between clinical None of the studies reported have shown any advantage of
benefits and harms intramedullary devices over extramedullary devices.
Intramedullary devices had been shown to have a higher re-
operation rate due to an increased incidence of periprosthetic
fracture both in the perioperative period and the postoperative
period (risk ratio 5.61). This may be due to the inclusion of studies
with original nail designs no longer implanted. All other outcomes
have been reported as similar. An additional meta-analysis is

included in Appendix G, page 503. By grouping studies using a cut
off of publication after 2000, no changes to the existing evidence
statement are made.
Economic considerations In patients with trochanteric fractures above and including the
lesser trochanter (AO classification types A1 and A2) the price of
intramedullary fixation devices varies but on average is three times
the price of sliding hip screws for short nails and five times the
price for long nails. As pointed out in the clinical evidence
statement, no significant benefit has been proven of the
advantages of intramedullary devices over extramedullary devices,
so that the GDG agreed to consider extramedullary implants cost-
effective for hip fracture patients.
Quality of evidence The level of evidence is high with numerous studies producing very
similar findings.
Other considerations All patients should be allowed to be mobilised full weight bearing
are designed to be load sharing devices to facilitate this. Full
weight bearing allows early mobilisation and rehabilitation.
The GDG highlighted this recommendation as a key priority for
implementation.
10.6.2 Intramedullary versus extramedullary implants for fixation of reverse

oblique trochanteric extracapsular fractures
In the reverse oblique fractures, which lie anatomically between the trochanteric and the
subtrochanteric fractures there is loss of this lateral stabilizing cortical buttress. Such
fractures are difficult to adequately reduce and fix at the time of the surgery. It is then the
more unpredictable as to whether that adequate reduction will be retained during the
healing process whilst allowing early mobilisation of the patient
In patients undergoing repair for reverse oblique trochanteric extracapsular hip fractures
what is the clinical and cost effectiveness of extramedullary sliding hip screws compared to
intramedullary nails on mortality, surgical revision, functional status, length of stay, quality
of life, pain and place of residence after hip fracture?
No RCT evidence was identified.
No cost-effectiveness evidence was identified.
10.6.2.4 Research recommendations

130 HIP FRACTURE
Intramedullary versus extramedullary fixation
➢ What is the clinical and cost effectiveness of intramedullary versus extramedullary

fixation on mortality, functional status and quality of life in patients with reverse
oblique trochanteric hip fracture?
Reverse oblique trochanteric fractures account for approximately 5 % of all trochanteric hip
fractures. This means it affects approximately over 1000 patients per year in the UK.
Presently there is little evidence as to which is the preferable implant (which can be either
extramedullary – outside the bone, or intramedullary - inside the bone). The potential
biomechanical advantage of intramedullary advantage may be offset by increased cost
(which can be over £1000 more expensive). A randomised trial comparing the two implants
using patient mobility, function and re-operation would allow a more informed choice of
treatment for this injury.
10.6.3 Intramedullary versus extramedullary implants for fixation of

subtrochanteric extracapsular fractures
Subtrochanteric fractures involve the shaft of the femur somewhere between the base of
the lesser trochanter and a point 5 cm distal to this. They may extend proximally or distally.
They have been considered as a separate group for practical purposes. Many of the
implants available for treating a standard trochanteric fracture are not long enough to
reach the intact bone distal to a subtrochanteric fracture. Thus whilst the general principles
of extra and intramedullary fixation described earlier still apply a different inventory of
implants to deal with these fractures is required.
It is noted that subtrochanteric fractures can often occur as a result of a metastatic

pathological deposit affecting the strength of the bone. The presence of pathological
deposits may not be obvious on the initial radiographs.
In patients undergoing repair for subtrochanteric extracapsular hip fractures, what is the
clinical and cost effectiveness of extramedullary sliding hip screws compared to
intramedullary nails on mortality, surgical revision, functional status, length of stay, quality
Four studies met the inclusion criteria for this review question with a total of 149 patients.
See evidence table 5.8, Appendix E and forest plots G107 to G111 in Appendix G.

Table 10-56: Intramedullary vs. extramedullary implants for subtrochanteric extracapsular

fracture – Clinical study characteristics
Numbe Other
Mortality – 1 2 RCT no serious serious (c) serious (b)serious (a)
year77,281 limitations
Reoperation – 4 RCT no serious serious (c) no serious serious (a)
within follow up limitations indirectness
period of
study9,77,214,281
Cut-out (at latest 1 RCT serious (d) no serious no serious serious (a)
77
follow up) inconsistency indirectness
Infection (deep 2 RCT no serious no serious serious (b) serious (a)
infection or limitations inconsistency
requires
reoperation)214,281
Non-union (at 2 RCT no serious no serious serious (b) no serious
latest follow limitations inconsistency imprecision
up)77,281
(a) The relatively few events and few patients give wide confidence intervals around the estimate of
(b) These studies are comparing intramedullary nailing to a Medoff sliding plate or fixed angle blade
plate.
(c) There is significant statistical heterogeneity in the results.
(d) Only one study with a small sample size.
Table 10-57: Intramedullary vs. extramedullary implants for subtrochanteric extracapsular

fracture - Clinical summary of findings
Intramedull Extramedull
Outcome ary ary Relative risk Absolute effect Quality
Mortality – 1 year 10 fewer per
7/48 RR 0.93 (0.08 to 1,000 (from 138
5/42 (15%) Very low
(14.6%) 11.52) fewer to 1578
more)
Reoperation – within 55 fewer per
follow up period of 11/71 RR 0.56 (0.06 to 1,000 (from 117
4/78 (5.1%) Low
study (12.5%) 5.47) fewer to 559
more)
Cut-out (at latest 25 fewer per
follow up) RR 0.68 (0.05 to 1,000 (from 73
1/19 (5.3%) 1/13 (7.7%) Low
9.98) fewer to 691
more)
Infection (deep 16 more per
infection or requires RR 1.27 (0.28 to 1,000 (from 42
3/45 (6.7%) 2/41 (5.9%) Low
reoperation) 5.88) fewer to 288
more)
Non-union (at latest 150 fewer per
follow up) 9/42 RR 0.15 (0.03 to 1,000 (from 32
1/48 (2.1%) Moderate
(17.6%) 0.82) fewer to 171
fewer)
132 HIP FRACTURE
No economic evidence was identified.
Clinical There is a statistically significant and clinically significant decrease in non-

union with intramedullary implants compared to extramedullary implants for
fixation of subtrochanteric extracapsular fractures. (MODERATE QUALITY)
There is no statistically significant difference in reoperation, cut-out and
infection with intramedullary implants compared to extramedullary implants
for fixation of subtrochanteric extracapsular fractures. (LOW QUALITY)
There is no statistically significant difference in mortality, with intramedullary
implants compared to extramedullary implants for fixation of subtrochanteric
extracapsular fractures. (VERY LOW QUALITY)
Economic No economic evidence was identified.

evidence
Recommendation Use an intramedullary nail to treat patients with a

subtrochanteric fracture.
Relative values of different The GDG considered the most important outcomes to be
outcomes functional status, pain, requirement for reoperations and wound
healing complications.
Trade off between clinical There was no evidence of a difference except for non-union of
benefits and harms fracture. It is accepted by expert opinion that the treatment of
choice is intramedullary fixation which allows splinting of the
whole of the femoral shaft.
Economic considerations Although intramedullary nails are more expensive than
extramedullary implants, the latter lead to more patients with non-
union of fracture, which would require more re-operation.
Quality of evidence There were few studies investigating this type of fracture. Several
studies were excluded as the population was from road traffic
accidents, therefore high energy trauma fractures, which were
excluded from the scope. The reported outcomes were
predominantly of low quality.
Other considerations Surgeons should use a technique where they are happy for the
patient to mobilise fully weight bearing (see section 10.2). When
patients suffer from subtrochanteric fractures it is advised to
consider whether there is a pathological process which would
increase the fracture risk (suck as a metastatic deposit).
As noted in the introduction subtrochanteric fractures may occur
as a result of a pathological process in the bone such as metastatic
disease. This pre-existing pathology may not always be recognised

on the initial radiographs. It is considered to be an additional
advantage of using a long intramedullary device that it provides
mechanical protection to a potentially diseased bone.
11 Mobilisation strategies
11.1 Introduction
Mobilisation is the process of re-establishing the ability to move between postures (for
example sit to stand), maintain an upright posture, and to ambulate with increasing levels
of complexity (speed, changes of direction, dual and multi-tasking).
Early restoration of mobility after surgery for hip fracture has been suggested as an
essential part of high quality care since the early 1980s309,310. The suggested benefits are
minimisation of hospital stay, avoiding complications of prolonged bed confinement, and
re-establishing people into their normal environments168,168.
Early restoration of mobility is an aspiration of many clinical services, although guidance on

the optimal time to re-mobilise patients and strategies that can be used to accelerate and
optimise recovery of mobility are less clear. Good quality clinical care, in particular effective
pain management should be considered essential components of early mobilisation and a
rehabilitation programme, as discussed in Chapter 7.
Specific therapeutic procedures, such as those implemented by physiotherapists and

occupational therapists have the potential to accelerate the recovery of mobility. Timing of
the intervention examined evidence about early (within 48 hours of surgery) mobilisation
and physiotherapy assessment, as opposed to later mobilisation (> 48 hours). Within the
type of intervention the GDG considered regimes that tested protocols delivering more
than one short session of physiotherapy per day (the benchmark for usual care), or more
intensive protocols than would comprise usual care. These protocols included intensive
strength training regimes (characterised by prescription and progression using recognised
American College of Sports Medicine criteria), intensive weight bearing exercise regimes
(supplemented by treadmill training), and increased numbers of physiotherapy usual care
sessions. Usual care was taken to be prescription of walking aids, gait re-education, and bed
exercises247,247.
Mobility can be measured in a range of different ways. The most simple and basic mobility
indicators, are the ability to transfer independently. This is usually taken to be between a
134 HIP FRACTURE
bed and a chair, but not all investigators report the exact definition they have used. Chair
rise ability and time to complete chair rises, along with timed tests of walking and balance
have a long established history for measuring mobility. In addition, the GDG considered
muscle strength, length of stay, discharge destination, independence in activity of daily
living (such as washing, bathing) and more complex tasks (for example, meal preparation),
and mortality as outcomes. Measurement of falls, and time to first fall are considered good
safety indicators for interventions like early mobilisation, but no studies reported these
outcomes.
11.2 Early vs. delayed mobilisation
In patients who have undergone surgery for hip fracture, what is the clinical and cost
effectiveness of early mobilisation (<48 hours after surgery) compared to late mobilisation
on functional status, mortality, place of residence/discharge, pain and quality of life?
See Evidence Table 10, Appendix E and forests G123 to 126).
Only one, small randomised controlled trial was identified with 60 patients.
Table 11-58: Early vs. delayed mobilisation – Clinical study characteristics

Numbe Other
Independent to 1 RCT Serious no serious no serious no serious
transfer at day 7 limitations inconsistency indirectness imprecision
247 (b)
Independent to 1 RCT Serious no serious no serious no serious

step at day limitations(b) inconsistency indirectness imprecision
7 247
Discharged to 1 RCT Serious no serious no serious Serious
home 247 limitations(b) inconsistency indirectness imprecision(a)
fast stream rehab limitations(b) inconsistency indirectness imprecision(a)
247

slow stream limitations(b) inconsistency indirectness imprecision(a)
rehab 247
nursing home 247 limitations (b)
inconsistency indirectness imprecision (a)
Mortality 247 1 RCT Serious no serious no serious Serious
limitations(b) inconsistency indirectness imprecision (a)
Mean walking 1 RCT Serious no serious no serious -(c)
247 (b)
distance limitations inconsistency indirectness
(a) The relatively few events and few patients give wide confidence intervals around the estimate of
(b) Unclear blinding and allocation concealment, also the small sample size makes it difficult to know
the true effect size for this outcome.
(c) The data is a mean with a range and therefore no relative risk was calculated. The wide range
around the mean indicates that the result may be imprecise.
MOBILISATION STRATEGIES 135
Table 11-59: Early vs. delayed mobilisation - Clinical summary of findings

Independent to 423 more per
transfer at day 7 16/29 4/31 1000 (from 80
RR 4.28 (1.62 to 11.3) Moderate
(55.2%) (12.9%) more to 1329
more)
Independent to step 401 fewer per
at day 7 10/29 23/31 1000 (from 148
(34.5%) (74.2%) fewer to 542
fewer)
Discharged to home 140 more per
5/29 RR 5.34 (0.66 to 1000 (from 11
1/31 (3.2%) Low
(17.2%) 43.06) fewer to 1357
more)
Discharged to fast 176 fewer per
stream rehab 8/29 14/31 1000 (from 316
RR 0.61 (0.3 to 1.24) Low
(27.6%) (45.2%) fewer to 108
more)
Discharged to slow 31 fewer per
stream rehab 14/29 16/31 1000 (from 227
RR 0.94 (0.56 to 1.55) Low
(48.3%) (51.6%) fewer to 284
more)
Discharged to nursing 0 more per 1000
home 1/29 (3.4%) 0/31 (0%) RR 3.2 (0.14 to 75.55) (from 0 fewer to Low
0 more)
Mortality 0 more per 1000
1/29 (3.4%) 0/31 (0%) RR 3.2 (0.14 to 75.55) (from 0 fewer to Low
0 more)
Mean walking
distance, metres 82.55 (0.5- 34.7 (5-103) N/A -(a) Moderate
400)
(a) An absolute effect could not be calculated as the study did not provide a mean, only a range.

The GDG was informed of the hourly cost of physiotherapy in a hospital setting for England
and Wales, which corresponds to £23 61. Physiotherapist sessions delivered during the
weekends and during public holidays would be paid at an enhanced rate of pay of time and
a third (BMA contract, 2008).
Clinical There is a statistically significant and clinically significant increase in

independence to transfer at day 7 for patients who had early mobilisation
compared to delayed mobilisation. (MODERATE QUALITY)
There is a doubling in the distance walked at day 7 for patients who had early
mobilisation compared to delayed mobilisation. (MODERATE QUALITY)
There is no statistically significant difference between early versus delayed
mobilisation for discharge destination or mortality. (LOW QUALITY)
136 HIP FRACTURE
There is a statistically significant and clinically significant decrease in

independence to step at day 7 for patients who had early mobilisation
compared to delayed mobilisation. (MODERATE QUALITY)
Economic No studies were identified on the cost-effectiveness of early vs. delayed
mobilisation.
Recommendation Offer patients a physiotherapy assessment and, unless medically

or surgically contraindicated, mobilisation on the day after
surgery.
Early mobilisation with a physiotherapist appears safe and is

Relative values of different
effective in promoting early recovery of ability to transfer without
outcomes
help of a person or walking aid. These outcomes are important
markers of early recovery of mobility. See also, chapter 10 section
10.2 where the recommendation is made that surgeons should
operate on patients with the aim to allow them to fully weight bear
(without restriction) in the immediate postoperative period.
Trade off between clinical The only outcome relating to harm or safety was mortality, which
benefits and harms showed no statistically significant difference. If safety issues were a
concern it is likely that they would be reflected in the overall
functional outcomes, all of which improved or had no significant
effect, therefore the GDG do not believe that harm is caused in
relation to this evidence. If any attempt at mobilisation is
supervised by a physiotherapist it should in any case be sensitive to
limitations imposed by individuals' pre-fracture abilities and
postoperative pain and fatigue. Thus a policy of early mobilisation
with a physiotherapist should be seen as beneficial, and delayed
only when individuals' clinical circumstances indicate this as
appropriate.
Economic considerations Evidence on the cost effectiveness of early mobilisation treatments
is lacking. The GDG acknowledged that early mobilisation
strategies will generally involve higher personnel costs (linked to
the provision of physiotherapy sessions over the entire week, thus
also during weekends and public holidays). However, the GDG
considered the cost-savings associated with an earlier recovery of
ability to transfer and step without help of a person or walking aid,
and agreed that early mobilisation strategy represent a cost-
effective intervention for our population.
Quality of evidence There is only one RCT of low to moderate quality with a relatively
small sample size (n = 60) and therefore the findings were
interpreted with caution by the GDG.
Other considerations Early mobilisation protocols may require new service delivery
models for weekend or 7 day physiotherapy services.
The GDG also noted that albeit the intervention should be

overseen by physiotherapists it is also important for nurses to re-
enforce and encourage patients’ mobility at all other times, under
the guidance of the physiotherapist.

implementation
11.3 Intensity of physiotherapy
In patients who have undergone surgery for hip fracture, what is the clinical and cost
effectiveness of intensive physiotherapy compared to non intensive physiotherapy on
functional status, mortality, place of residence/discharge, pain and quality of life?
See evidence table 5.10, Appendix E and forest plots G127 to G139.
Three randomised studies were found with a total of 288 patients, comparing three
different types of intensive physiotherapy/physical medicine programme. Hauer et al
(2002)139,140 investigated intensive, progressive strength training. Moseley et al (2009)216,216
tested an intensive weight bearing exercise programme supplemented by treadmill gait re-
training programme, and Karumo (1977)171,171 investigated twice daily physiotherapy (of
one hours duration) in comparison to usual care (<=30 mins, once daily).
Table 11-60: Intensive exercise or physiotherapy vs. usual care – Clinical study characteristics
Numbe Other
Intensive physiotherapy (strength training)
Leg-press 1 RCT no serious no serious no serious serious(b)
strength limitations inconsistency indirectness
fractured side
(kg) 140
Leg extensor 1 RCT no serious no serious no serious no serious
strength limitations inconsistency indirectness imprecision
fractured side
(Newtons) 140
Ankle plantar 1 RCT no serious no serious no serious serious(b)
flexion strength limitations inconsistency indirectness
fractured side
(Newtons) 140
Walking speed – 1 RCT no serious no serious no serious no serious
3 months 140 limitations inconsistency indirectness imprecision
Tinetti’s POMA(d) 1 RCT no serious no serious no serious no serious
– overall 140 limitations inconsistency indirectness imprecision
138 HIP FRACTURE
Numbe Other
Tinetti’s POMA – 1 RCT no serious no serious no serious no serious
part 1 (balance) limitations inconsistency indirectness imprecision
140
Tinetti’s POMA – 1 RCT no serious no serious no serious no serious

part 2 (gait) 140 limitations inconsistency indirectness imprecision
Timed up-and-go 1 RCT no serious no serious no serious serious(b)
(seconds) 140 limitations inconsistency indirectness
Chair rise 1 RCT no serious no serious no serious serious(b)
(seconds) 140 limitations inconsistency indirectness
Barthel’s ADL 140 1 RCT no serious no serious no serious serious(b)
limitations inconsistency indirectness
Lawton’s IADL 140 1 RCT no serious no serious no serious no serious
limitations inconsistency indirectness imprecision
Intensive physiotherapy (weight bearing exercise and treadmill training)
Knee extensor 1 RCT serious(a) no serious no serious serious (b)
strength – 4 inconsistency indirectness
weeks 216
Knee extensor 1 RCT serious(a) no serious no serious serious(b)
strength – 16 inconsistency indirectness
weeks 216
Walking speed – 1 RCT serious(a) no serious no serious no serious
216
4 weeks inconsistency indirectness imprecision
Walking speed – 1 RCT serious(a) no serious no serious no serious
8 weeks 216 inconsistency indirectness imprecision
(a)
Sit-to-stand test 1 RCT serious no serious no serious no serious
at 4 weeks 216 inconsistency indirectness imprecision
Sit-to-stand test 1 RCT serious(a) no serious no serious no serious
at 16 weeks 216 inconsistency indirectness imprecision
Quality of life – 4 1 RCT serious(a) no serious no serious no serious
weeks 216 inconsistency indirectness imprecision
(a)
Quality of life – 1 RCT serious no serious no serious no serious
16 weeks 216 inconsistency indirectness imprecision
Pain – 4 weeks 216 1 RCT serious(a) no serious no serious no serious
inconsistency indirectness imprecision
Pain – 16 weeks 1 RCT serious(a) no serious no serious no serious
216
inconsistency indirectness imprecision
(a)
Length of 1 RCT serious no serious no serious serious(b)
216
hospital stay inconsistency indirectness
Intensive (more frequent) physiotherapy
Adductor muscle 1 RCT serious(c) no serious no serious serious(b)
strength (kp) at 9 inconsistency indirectness
weeks 171
Length of 1 RCT serious(c) no serious no serious serious(b)
171
hospital stay inconsistency indirectness
(a) Low number of subjects in each arm (N = 24) therefore the study may be underpowered.
(c) Method of randomisation, blinding and allocation concealment is unclear.
(d) POMA: Tinetti’s performance oriented mobility assessment
Table 11-61: Intensive exercise or physiotherapy vs. usual care - Clinical summary of findings
Intensive physiotherapy (strength training)
Leg-press strength 12 12 N/A MD 21 higher Low
fractured side (kg) (2.09 lower to
44.09 higher)
Leg extensor strength 12 12 N/A MD 17 higher Moderate
fractured side (2.54 to 31.46
(Newtons) higher)
Ankle plantar flexion 12 12 N/A MD 23 higher Low
strength fractured (2.23 lower to
side (Newtons) 48.23 higher)
Walking speed – 3 12 12 N/A MD 0.23 higher Moderate
months (0.05 to 0.41
higher)
Tinetti’s POMA - 12 12 N/A MD 3 higher Moderate
overall (0.41 lower to
6.41 higher)
Tinetti’s POMA – part 12 12 N/A MD 1.3 higher Moderate
1 (balance) (0.54 lower to
3.14 higher)
Tinetti’s POMA – part 12 12 N/A MD 1.7 higher Moderate
2 (gait) (0.15 lower to
3.55 higher)
Timed up-and-go 12 12 N/A MD 0.8 lower Low
(seconds) (12.3 lower to
10.7 higher)
Chair rise (seconds) 12 12 N/A MD 1.8 lower Low
(6.61 lower to
3.01 higher)
Barthel’s ADL 12 12 N/A MD 3.1 lower Low
(9.66 lower to
3.46 higher)
Lawton’s IADL 12 12 N/A MD 0.4 higher Moderate
(0.68 lower to
1.48 higher)
Intensive physiotherapy (weight bearing exercise and treadmill training)
Knee extensor 80 80 N/A MD 0.1 higher Moderate
strength – 4 weeks (1.12 lower to
1.32 higher)
Knee extensor 80 80 N/A MD 1 higher Moderate
strength – 16 weeks (0.46 lower to
2.46 higher)
Walking speed – 4 80 80 N/A MD 0.05 higher High
weeks (0.02 lower to
0.12 higher)
Walking speed – 8 80 80 N/A MD 0.03 higher High
0.13 higher)
Sit-to-stand test at 4 80 80 N/A MD 0.05 higher High
weeks (0.01 to 0.09
higher)
Sit-to-stand test at 16 80 80 N/A MD 0.04 higher High
weeks (0 to 0.08
higher)
140 HIP FRACTURE

Quality of life – 4 80 80 N/A MD 0 higher High
0.08 higher)
Quality of life – 16 80 80 N/A MD 0 higher High
0.09 higher)
Pain – 4 weeks 36 more per High
41/80 1000 (from 102
44/80 (55%) RR 1.07 (0.8 to 1.44)
(51.3%) fewer to 226
more)
Pain – 16 weeks 11 more per High
30/80 29/80 1000 (from 112
RR 1.03 (0.69 to 1.55)
(37.5%) (36.3%) fewer to 199
more)
Length of hospital 80 80 N/A MD 3 higher (1.5 Moderate
stay (Moseley) lower to 7.5
higher)
Adductor muscle 38 49 N/A MD 0.76 lower Low
strength (kp) at 9 (2.42 lower to
weeks 0.9 higher)
Length of hospital 39 39 N/A MD 2.8 lower Low
stay (12.09 lower to
6.49 higher)
No studies were identified. A cost analysis was conducted based on the resources used in
the studies included in the clinical review, which is reported in section 20.4of Appendix Hof
this guideline.
Table 11-62: Intensive exercise or physiotherapy vs. usual care – Economic study characteristics
NCGC cost analysis Minor limitations (a) Partially applicable (b) Cost analysis based on
resources used in the
studies included in the
clinical review 140,171,216
(c) No sensitivity analysis.

(d) UK study but does not estimate QALYs. One study171 quite outdated. All studies not UK based and
therefore may not reflect current NHS practice.
Table 11-63: Intensive exercise or physiotherapy vs. usual care - Economic summary of findings
Incremental
Study Incremental cost (£) effects ICER Uncertainty
NCGC cost - £12 (strength training programme N/A N/A N/R
analysis vs. usual care 140) (a)
- £180.18 (more intensive
physiotherapy vs usual care 171) (b)
- £827.62 (inpatient-based part of
the weight bearing and treadmill
exercise programme 216) (c)
(b) Intervention group slightly more costly than the control group because of the use of ad-hoc
exercise equipment.
(c) Intervention group more costly because of longer physiotherapy sessions
(d) It was not possible to estimate the outpatient costs of the rehabilitation programme as
insufficient information was given in the study.
Evidence statement (s)
Clinical Strength training

Additional, progressive strength training produces a statistically significant
and clinically significant increase in leg extensor power, hip flexor strength
and walking speed compared to placebo motor training (control) at 3 months
after surgery. (HIGH QUALITY)
There is no statistically significant difference in basic or extended activities of
daily living or gait and balance as measure by the Performance Orientated
Mobility Assessment with strength training compared to placebo motor
training (control) at 3 months after surgery. (HIGH QUALITY)
There is no statistically significant difference in timed up and go test and chair
rises with strength training compared to placebo motor training (control) at 3
months after surgery. (MODERATE QUALITY)
Weight bearing exercise and treadmill training
There is no statistically significant difference in functional performance tests,
quality of life, walking speed or pain with weight bearing exercise and
treadmill gait training compared to the control. (HIGH QUALITY)
There is no statistically significant difference in length of hospital stay with
weight bearing exercise and treadmill gait training compared to the control.
(MODERATE QUALITY)
There is no statistically significant difference in knee extensor strength
adductor muscle strength or length of stay in hospital with an increased
number of physiotherapy sessions per day compared to the control. (LOW
QUALITY)
Economic All intensive exercise and physiotherapy programmes are more expensive
than usual care, albeit the strength programme is only slightly more costly
compared to usual care.
142 HIP FRACTURE
This evidence has minor limitations and partial applicability.
Recommendation Offer patients mobilisation at least once a day and ensure regular
physiotherapy review.
The outcomes considered most important were mobility,

Relative values of different
functional status, pain, quality of life and length of hospital stay.
outcomes
There is evidence of training effects in muscle strength and other
variables which are known to be important determinants of ability
to walk, and hence live independently. Further research is needed
to confirm effects on outcomes including return to independent
living, quality of life, health service resource, and time to discharge.
The evidence shows that there was no difference in once a day or

twice a day physiotherapy for length of hospital stay and adductor
muscle strength171, and thus the GDG are recommending at least
once a day mobilisation.
Trade off between clinical GDG consensus was that mobilisation at least once a day has
benefits and harms potential benefits of improved mobility and balance, increased
independence, and reduced need for institutional and social care.
The included studies failed to show improvements for these
outcomes, but are all small low quality studies. There is no
evidence of harm from mobilisation once a day. There is potential
to exacerbate pain and induce excessive fatigue, and training
should be prescribed and overseen by a physiotherapist.
There is insufficient evidence to suggest what the exact dosing of
physiotherapy should be, and this will vary according to the
physical capabilities of each patient. Those who are very ill will not
tolerate as much physical activity as those who are progressing
well. The dosing should be based on a physiotherapist assessment.
Hence the issue is one of professional judgement as we have no
evidence to guide us any further. However, an additional
observation is that the principles of management should not be
any different for people with dementia, than those without.
Economic considerations The GDG acknowledged the lack of cost-effective evidence on this
question, and agreed that intensive rehabilitation sessions are
likely to be more expensive than usual care. The GDG also noted
that intensive rehabilitation can bring some benefits in terms of
strength and on other factors affecting the ability to walk and live
independently.
The GDG agreed that daily mobilisation sessions and regular
physiotherapy review represent a cost-effective intervention for
our patients.
Quality of evidence Although 3 RCTs were included, the interventions were not
comparable and could not be combined in a meta-analysis. The

studies were all considered individually and the evidence base is
limited. The quality of the evidence ranged from low to high, but
due to few studies being identified the GDG considered the overall
quality to be poor.
The economic evidence is based on the resources described in the
programmes in the three RCTs included in the clinical review. Only
the costs of the interventions and of the usual care programme
were considered. The analysis is also only partially applicable in
that, even current NHS unit costs were used, the actual level of
resources reported in the trials may not reflect the current practice
in the UK NHS.
Other considerations GDG expert opinion indicates that patients may benefit from more
intensive protocols of rehabilitation therapy (including
occupational and physiotherapy), but that more evidence is
needed.
implementation
11.3.3 Research recommendations on mobilisation
11.3.3.1 Frequency of physiotherapy
➢ What is the clinical and cost effectiveness of additional intensive physiotherapy

and/or occupational therapy (for example progressive resistance training) after hip
fracture?
The rapid restoration of physical and self care functions is critical to recovery from hip
fracture, particularly where the goal is to return the patient to preoperative levels of
around, 400 to 500 patients, and the primary outcome should be physical function and
144 HIP FRACTURE
12 Multidisciplinary management
12.1 Introduction
Multidisciplinary care is central to the management of frail older people with multiple
medical, psychological and social problems. Since these are the people who typically suffer
hip fracture every Trauma Unit will provide some form of multidisciplinary care. Although
the prevalence of comorbidity is generally lower in younger patients, the key principles of
multidisciplinary intervention are applicable across the adult age spectrum and the same
skills and organisational approaches derived within the development of a focus on the older
population should be provided irrespective of chronological age.
In this chapter the evidence for the different models of enhanced inpatient and community
management were considered that have evolved to meet the specific needs of patients
with hip fracture.
Secondary prevention of fracture by means of the assessment and management of both

osteoporosis234-236 and risk of falling 227 are covered in separate NICE guidance. It is,
however, important in practice that the elements of multidisciplinary management covered
in this guidance relate in an organized manner closely and reliably with these secondary
prevention programmes to deliver all the elements of comprehensive care required by each
patient. The precise organizational approach to this differs amongst centres. In some there
is considerable overlap and/or cross-representation between the secondary prevention
programmes and the service models covered in this guideline.
Units across the UK have adopted a variety of multidisciplinary service models, but most
have at least some form of access to geriatrician input into the care of these patients. Local
circumstances and expertise have determined the precise model developed in different
centres, but in general these are variations on the following four approaches.
The traditional model of orthopaedic care - 'usual care'.
• The patient with hip fracture is admitted to a trauma ward where the orthopaedic
surgical team lead both their surgical care and subsequent rehabilitation.
Geriatrician input to such wards may be limited, with referrals and medical queries
being dealt with on a consultative basis by the on-call medical registrar or on
occasional geriatrician visits, but without a proactive geriatrician lead to the
multidisciplinary team.
A more collaborative model of trauma ward working is formal 'orthogeriatric' care - with
trauma patients admitted to a specialised ward under the joint care of both geriatricians
MULTIDISCIPLINARY MANAGEMENT 145
and orthopaedic surgeons. Surgical and geriatrician ward rounds may happen
independently, or be combined in multidisciplinary ward rounds.
• This collaborative model is particularly relevant to hip fracture patients. Such joint
working can thus lead to the development of a formal 'Hip Fracture Programme'
(HFP), with the geriatric medical team contributing to joint preoperative patient
assessment, and increasingly taking the lead for postoperative medical care,
multidisciplinary rehabilitation (MDR) and discharge planning.
Both 'traditional' and 'orthogeriatric' models of the acute trauma ward may continue to
care for patients throughout their recovery and rehabilitation following hip fracture, or
each may be followed by a transfer of some patients to other models of rehabilitation.
• In some centres, surgical care and initial mobilisation is followed by early

postoperative transfer to a 'Geriatric Orthopaedic Rehabilitation Unit' (GORU) - a
separate geriatrician-led rehabilitation ward. The extent of surgical input to the
GORU varies, depending on how early patients are moved from the acute trauma
wards.
• In other centres, similar patients would be transferred to a generic 'Mixed

Assessment and Rehabilitation Unit' (MARU), able to accept patients with a variety
of medical, surgical and orthopaedic conditions.
A further service model is some form of community rehabilitation.
• One approach is 'Early Supported Discharge' (ESD) or ‘Intermediate Care' at home.

Patients are discharged home from the acute trauma ward, or in some cases a
rehabilitation ward within the hospital, with a supported 4-6 week rehabilitation
package. This may include patients living in care homes but in many parts of the
country is limited to patients returning to live independently in their own homes.
• Alternatively, patients with more complex needs may be moved for rehabilitation
to an Intermediate Care facility outside the hospital setting, such as a care home, or
a community hospital. Again this will vary depending on the provision of services
available locally.
146 HIP FRACTURE
12.2 Hospital-based multidisciplinary rehabilitation versus usual care

Multidisciplinary rehabilitation (MDR) after hip fracture has been taken by the GDG to
incorporate medicine, nursing, physiotherapy, occupational therapy and social care as core
components of assessment and management. Additional components may include
dietetics, pharmacy and clinical psychology.
The GDG also assumes:
• The required degree of relevant specialist expertise in each case.
• Formal arrangements for co-ordination/teamwork, and
• Regular on-going multidisciplinary assessment.
'Usual care' will be taken to imply the traditional model, with ad hoc or selective referral to
some or all of the separate MDR components listed above, but without formal
arrangements for co-ordinated multidisciplinary teamwork.
In contrast, the different models of 'orthogeriatric care' all assume the involvement of a
geriatrician, in addition to the orthopaedic surgical team, in the development and
supervision of a formal process of coordinated multidisciplinary care.
Such orthogeriatric models have been sub-divided into:
• Those focused predominantly or exclusively on the acute trauma ward; typified by

the HFP model43.
• Those provided in a subsequent inpatient rehabilitation setting (with GORU and

MARU having been combined because no evidence has addressed a comparison of
these models).
• Those with a community focus (the focus of Section 12.4).
12.2.1 Review questions
In this section two review questions were combined as the evidence overlapped and could
not be separated in a useful way. The questions were:
In patients with hip fracture what is the clinical and cost effectiveness of hospital-based
multidisciplinary rehabilitation on functional status, length of stay in secondary care,
mortality, place of residence/discharge, hospital readmission and quality of life?
All the published studies included in the analysis of hospital-based MDR are of models that
include geriatrician input. The results of a collective analysis of all such studies therefore
reflect both the effectiveness of hospital-based MDR, and the overall value of
orthogeriatrician involvement in hip fracture care.
In addition, the benefits of different models of hospital-based MDR can be considered by

comparing 'usual care' with the two general sub-types of orthogeriatric care:
• Hip Fracture Programme (HFP)
• Geriatric Orthopaedic Rehabilitation Unit (GORU), or near equivalents such as a

Mixed Assessment and Rehabilitation Unit (MARU).
In patients with hip fracture what is the clinical and cost effectiveness of orthogeriatrician
involvement in the whole pathway of assessment, peri-operative care and rehabilitation on
functional status, length of stay in secondary care, mortality, place of residence/discharge,
hospital readmission and quality of life?
The geriatrician is increasingly seen as having a key role in the integration of initial
assessment and peri-operative care with the coordinated MDR (in whatever setting) which
follows it.
The usefulness of this early element of orthogeriatric input has been assessed; an element
that it is central to the first of the two models (HFP), but lacking from the second
(GORU/MARU). In the absence of trials directly comparing the two models the impact of
early geriatrician involvement can only be inferred from any differences that might be
apparent when each is compared to ‘usual care’.
11 studies met the inclusion criteria for this question, with a total of 2214 patients. See
Evidence Table 11, Appendix E and forest plots G129 to 138 in Appendix G.
148 HIP FRACTURE
Table 12-64: Hospital based multidisciplinary rehabilitation vs. usual care – Clinical
study characteristics
Numb
er of Other
studie considerations/
Outcome s Design Limitations Inconsistency Indirectness imprecision
Mortality at 6 2 RCT no serious no serious no serious no serious
months – limitations inconsistency indirectness imprecision
GORU/MARU113,2
22
Mortality at 12 4 RCT serious(a, b, c) no serious no serious no serious

months – inconsistency indirectness(d) imprecision
GORU/MARU107,1
58,176,319
Mortality at 12 4 RCT serious(e, f) no serious no serious no serious

months – inconsistency indirectness imprecision
HFP44,305,325,344
Mortality (at 6 RCT serious(a, b, c, no serious no serious no serious
g)
discharge) – inconsistency indirectness(d) imprecision
GORU/MARU107,1
13,158,176,222,319
Mortality (at 2 RCT no serious serious(h) no serious serious(h)

discharge) – limitations(f) indirectness
HFP325,344
Non- 1 RCT no serious no serious no serious no serious
recovery/decline limitations inconsistency indirectness imprecision
in walking at 6
months –
GORU/MARU222
Decline in 1 RCT no serious no serious no serious no serious
transfers (bed to limitations inconsistency indirectness imprecision
chair etc) at –
GORU/MARU222
More dependent 2 RCT serious(b, g) no serious no serious serious(k)
(based on Katz inconsistency indirectness(d)
index) at 1 year –
GORU/MARU176,3
19
Non-recovery in 1 RCT no serious no serious no serious serious(k)

activities of daily limitations(g) inconsistency indirectness
living (ADL) at 1
year –
GORU/MARU319
Non-recovery of 2 RCT no serious no serious no serious serious(k)
ADL/decline in limitations(e, inconsistency indirectness
f)
walking at 1 year
– HFP305,344
Chinese Barthel 1 RCT no serious no serious no serious serious(k)
Index at 6 limitations inconsistency indirectness
months - HFP305
Modified Barthel 1 RCT no serious no serious no serious serious(k)
Index at 6 limitations inconsistency indirectness
months – HFP325
Numb
er of Other
studie considerations/
Outcome s Design Limitations Inconsistency Indirectness imprecision
Length of 5 RCT no serious serious(I, j) no serious serious(k)
hospital stay - limitations indirectness(d)
GORU/MARU107,1
13,176,222,319
Length of 3 RCT no serious serious(l) no serious serious(h)

hospital stay - limitations indirectness
HFP44,305,325
Pressure sores344 1 RCT no serious no serious no serious no serious
limitations(f) inconsistency indirectness imprecision
Heart failure344 1 RCT no serious no serious no serious serious(k)
limitations(f) inconsistency indirectness
Pneumonia344 1 RCT no serious no serious no serious serious(h)
limitations(f) inconsistency indirectness
Confusion344 1 RCT no serious no serious no serious no serious
limitations(f) inconsistency indirectness imprecision
Chest infection, 1 RCT no serious no serious no serious serious(k)
cardiac problem, limitations inconsistency indirectness
bedsore325
Stroke, emboli325 1 RCT no serious no serious no serious serious(h)
limitations inconsistency indirectness
Delirium203 1 RCT no serious no serious serious(m) serious(k)
limitations inconsistency
Severe 1 RCT no serious no serious serious(m) serious(k)
delirium203 limitations inconsistency
Readmitted to 2 RCT serious(c, g) serious(n) no serious no serious
hospital during indirectness imprecision
follow-up –
GORU/MARU107,3
19
Readmitted to 4 RCT serious(f, g) no serious no serious no serious

hospital during inconsistency indirectness imprecision
follow-up –
HFP44,305,325,344
(a) Intervention group in Huusko 2002157,158 had greater number of patients with dementia (32/120 vs.
20/123); fewer were functionally independent in ADL before hip fracture (41 vs. 66).
(b) Kennie 1988176,176: difference in age mental state. Control group average age higher and with more
moderate and severe impairment.
(c) In Galvard 1995107,107, the intervention group were older than usual care (79.1 vs. 73.6), and there
were a higher proportion of patients with subtrochanteric fractures, which often require longer
rehab (12% vs. 4%).
(d) Kennie 1988176,176 is an all female population.
(e) In Shyu 2008305 the patient's insurance policy determined the number of physiotherapy sessions in
the control group.
(f) In Vidan 2005344,344 there is potential for contamination bias given both groups were on the same
ward and had the same staff.
(g) In Stenvall 2007a319,320, outpatient rehabilitation was not standardised.
(h) The relatively few events and few patients give wide confidence intervals around the estimate of
(i) Galvard 2002107,107, author's note that geriatric department had less experience with hip fracture
patients than the orthopaedic ward, which may have contributed to increased length of stay in
intervention group.
(j) The intervention in Naglie 2002222,222 was expected to increase the length of stay in hospital.
150 HIP FRACTURE
(k) The wide confidence intervals around the estimate make it difficult to determine and effect size for
this outcome.
(l) There is significant statistical heterogeneity between the studies. This could be due to the variation
in intervention and country of study.
(m) The intervention in Marcantonio 2001203,203 does not examine multidisciplinary rehabilitation in the
form of an HFP, but focuses on the value of early comprehensive geriatric assessment and targeted
intervention.
(n) There is significant statistical heterogeneity between the studies. However, this could be due to
differences in access to hospital services and follow up procedures.
Table 12-65: Hospital based multidisciplinary rehabilitation vs. Usual care - Clinical summary of findings
Mortality at 6 months 35 fewer per
– GORU/MARU 31/238 44/263 1,000 (from 80
RR 0.79 (0.52 to 1.21) High
(13%) (16.8%) fewer to 35
more)
Mortality at 12 10 fewer per
months – 89/455 96/466 1000 (from 54
RR 0.95 (0.74 to 1.23) Moderate
GORU/MARU (19.6%) (19.7%) fewer to 47
more)
months – HFP 72/400 90/404 1000 (from 87
RR 0.81 (0.61 to 1.06) Moderate
(18%) (21%) fewer to 13
more)
Mortality (at 19 fewer per
discharge) – 46/693 62/729 1000 (from 39
RR 0.78 (0.54 to 1.13) Moderate
more)
Mortality (at 41 fewer per
discharge) – HFP 3/193 11/197 1000 (from 2
RR 0.27 (0.07 to 0.96) Low
(1.6%) (5.8%) fewer to 52
fewer)
Non-recovery/decline
5 fewer per 1000
in walking at 6 59/124 56/117
RR 0.99 (0.76 to 1.29) (from 115 fewer Moderate
months – (47.6%) (47.9%)
to 139 more)
GORU/MARU
Decline in transfers 15 fewer per
(bed to chair etc) at – 45/124 44/117 1000 (from 117
RR 0.96 (0.69 to 1.34) Moderate
more)
More dependent 250 fewer per
(based on Katz index) 57/127 77/111 1000 (from 132
RR 0.64 (0.51 to 0.81) Low
at 1 year – (44.9%) (72.2%) fewer to 340
GORU/MARU fewer)
Non-recovery in 171 fewer per
activities of daily 51/84 59/76 1000 (from 31
RR 0.78 (0.63 to 0.96) Moderate
living (ADL) at 1 year - (60.7%) (77.6%) fewer to 287
GORU/MARU fewer)
Non-recovery in 171 fewer per
ADL/decline in 86/207 108/207 1000 (from 31
RR 0.79 (0.65 to 0.97) Moderate
walking at 1 year – (41.5%) (52.2%) fewer to 287
HFP fewer)
Chinese Barthel Index MD 6.17 (0.86
73 75 N/A Moderate
at 6 months - HFP to 13.2)
Modified Barthel
MD 6.3 (0.53 to
Index at 6 months – 33 27 N/A Moderate
13.13)
HFP
Length of hospital MD 1.32 (-12.83
572 606 N/A Low
stay - GORU/MARU to 15.47)
Length of hospital MD -6.06 (-14.5
245 240 N/A Low
stay - HFP to 2.38)
Pressure sores 114 fewer per
8/155 27/164 1000 (from 54
RR 0.31 (0.15 to 0.67) High
(5.2%) (16.5%) fewer to 140
fewer)
Heart failure 47 more per
12/155 5/164 1000 (from 2
RR 2.54 (0.92 to 7.04) Moderate
(7.7%) (3.1%) fewer to 184
more)
Pneumonia 2 more per 1000
6/155 6/164
RR 1.06 (0.35 to 3.21) (from 24 fewer Moderate
(3.9%) (3.7%)
to 81 more)
Confusion 65 fewer per
53/155 67/164 1000 (from 151
RR 0.84 (0.63 to 1.11) High
(34.2%) (40.9%) fewer to 45
more)
Chest infection, 236 fewer per
cardiac problem, 6/38 13/33 1000 (from 24
bedsore (15.8%) (39.4%) fewer to 327
fewer)
stroke, emboli 75 more per
4/38 RR 3.47 (0.41 to 1000 (from 18
1/33 (3%) Moderate
(10.5%) 29.56) fewer to 865
more)
Delirium 175 fewer per
20/62
32/64 (50%) RR 0.65 (0.42 to 1) 1000 (from 290 Low
(32.3%)
fewer to 0 more)
Severe delirium 169 fewer per
7/62 18/64 1000 (from 31
RR 0.4 (0.18 to 0.89) Low
(11.3%) (28.1%) fewer to 231
fewer)
Readmitted to 46 fewer per
hospital during 74/256 87/262 1000 (from 110
RR 0.86 (0.67 to 1.12) Low
follow-up - (28.9%) (33.2%) fewer to 40
GORU/MARU more)
Readmitted to 29 more per
hospital during 86/373 78/378 1000 (from 27
RR 1.14 (0.87 to 1.48) Moderate
follow-up – HFP (23.1%) (17%) fewer to 99
more)
The included studies for hospital-based MDR consisted of Cameron (1994)42,45, Galvard
(1995)107,107, Farnworth (1994)91,91 and Huusko (2002)157,158. Further details on the studies
are available in Evidence Table 16 of Appendix F. An HTA by Cameron (2000)41 was
excluded because the studies were grouped in a different way to that considered for our
clinical review, and therefore its cost analysis was not applicable for our review question.
152 HIP FRACTURE
An original decision analysis has been conducted comparing the cost-effectiveness of the
HFP vs. GORU/MARU vs. usual care. A Markov model was developed, adopting a life-time
horizon.
An indirect comparison between the HFP and GORU/MARU models of care was made as no
evidence was available which compares directly the two rehabilitation programmes. The
usual care arms in the trials of HFP vs. usual care and of GORU/MARU vs. usual care were
combined for this purpose.
Treatment effects were based on the findings of the clinical review and applied only up to 1
year from follow-up. Resource use was determined from the NHS and PSS perspective.
Effectiveness was measured in QALYs. Costs and QALYs were discounted at a rate of 3.5%.
Please see section 8.6 of Appendix H for further details.
Table 12-66: Hospital based multidisciplinary rehabilitation vs. usual care - Economic study
characteristics
Cameron 1994 45 – Potentially serious Partial applicability (b) Accelerated rehab was
HFP limitations (a) compared to usual care.
The follow up time was 4
months.
Farnworth 199491 – Potentially serious limitations Partial applicability (b) Fractured Hip Management
(c)
HFP Program (FHMP) was
The follow up time was 6
months.
107 (e)
Galvard 1995 Potentially serious Partial applicability Rehabilitation in a geriatric
- GORU limitations (d) department was compared
to usual care. The follow up
time was 1 year.
Huusko (2002)158 Potentially serious Partial applicability (g) Intensive multidisciplinary
- MARU limitations (f) geriatric team rehabilitation
versus usual care. Follow up
was 1 year.
NCGC economic Minor limitations (h) Direct applicability Cost-effectiveness analysis
model of HFP vs. GORU/MARU vs.
usual care based on the
meta-analysis of the trails
included in the clinical
review of this guideline
(a) Patients in the intervention and control group treated in the same ward, so that results could be biased
due to an underestimation of the cost effectiveness of accelerated rehab.
(b) Study conducted in Australia. Not a CUA.
(c) The year in which cost date were collected is not clear. The duration of follow up is not clear. HRQoL not
calculated. The statistical significance of the outcome and cost measures between the two groups was
not reported. Outcome at 1 year was not known for 12% of the intervention and 14% of the control
group.
(d) No sensitivity analysis was performed to test robustness of findings. HRQoL not calculated. The source
used to estimate the unit cost of resources was unclear.
(e) Study conducted in Sweden. Not a CUA.
(f) Not a cost-effectiveness analysis. No sensitivity analysis was performed. 38 patients were lost during
follow up. The year(s) at which cost data refer to is not clear. Imbalance of baseline characteristics.
Intervention group had a more patients with dementia (32/120 vs. 20/123, and fewer who were
functionally independent in ADL before hip fracture (41 vs. 66).
(g) Study conducted in Finland. Not a CUA.
(h) Treatment effects from meta-analysis of clinical trials available up to 1 year from follow-up.
154 HIP FRACTURE
Table 12-67: Hospital based multidisciplinary rehabilitation vs. usual care - Economic
summary of findings

Cameron 1994 – -£956(a) Several outcomes Accelerated Threshold sensitivity analysis:
HFP were reported (b) rehabilitation is results not sensitive to
the dominant changes in % of patients
strategy (less recovering nor to the
costly and definition of recovery.
more effective) Accelerated rehab becomes
more costly than usual care
when difference in LOS less
than 1.5-2 days and when
cost of treatment more than
40% per bed day.
Farnworth 1994 £784 (c) Several outcomes N/A Deterministic sensitivity
– HFP were reported (d) analysis showed that results
were robust to changes in the
time spent to get patients to
surgery more quickly; to the
proportion of nursing home
patients and to the average
cost of the final days of a
patient’s stay
Galvard 1995 -£665 (e) Several outcomes N/A N/R
- GORU were reported (f)
Huusko 2002 £1310(g) Several outcomes N/A N/R
- MARU were reported (h)
NCGC economic -£ 2,000 (HFP vs. -0.13 QALYs (HFP HFP is the Deterministic sensitivity
model – HFP vs. GORU/MARU) vs. GORU/MARU) dominant analysis showed that results
GORU/MARU -£25,000 (HFP vs. -1.01 QALYs (HFP strategy were sensitive to changes in
vs. usual care usual care) (i) vs. usual care) (j) compared to the proportion of patients
(Appendix H) both discharged to their own
GORU/MARU home following
and usual care rehabilitation.
A probabilistic sensitivity
analysis showed that there is
no uncertainty that hospital
MDR is better than usual
care. However, there is some
uncertainty over the cost-
effectiveness of HFP vs.
GORU/MARU.(k)
95% CI (HFP vs usual care and

GORU/MARU vs usual care):
usual care dominated.
95% CI (HFP vs.

GORU/MARU): HFP dominant
– GORU dominant.
(a) Accelerated rehab is cost saving. A$ converted using the PPP of 1990. p=0.186. The cost
components estimated were: inpatients hospital costs, readmissions, community support services,
institutional care.
156 HIP FRACTURE
(b) No. of patients recovered at 4 months from surgery (mean Barthel index score): 63 (49.6%) vs. 52
(41.6%); 95% CI (-3% to 21%). Median length of stay (days, interquartile range): 13 (7-25) vs. 15 (8-
44).
(c) Fractured Hip Management Program (FHMP) is cost saving.
(d) FHMP entails lower mortality and readmission at 1 year, and lower length of stay.
(e) Swedish Krona (SEK) converted using the PPP of 1989; Rehabilitation in geriatric department more
expensive than usual care (£665 per patient)
(f) The intervention had a lower level of readmissions to hospital than usual care (36 vs. 57; p value
NR) but it had a higher mortality at 1 year (45 vs. 40, p value NR) and a higher mean length of stay
in hospital (53.3 vs. 28 days, p value NR).
(g) The study expressed costs in Euros (values of 1999). T he intervention is more costly than usual care
(p value NR).
(h) Intervention did not statistically differ from usual care in terms of mortality at 12 months (15% vs.
16%); mortality at discharge (5 vs. 5) and length of stay in hospital during 1 year (80 vs. 80 days),
and number of patients reporting complications (51% vs. 46%, p=0.4). Patients in the intervention
group regained their independency in the IADL functions faster (p=0.005) than usual care at 3
months (but after 1 year there was no significant difference between the two groups).
(i) The mean costs associated with HFP were estimated to be £34,000, for GORU/MARU £36,000 and
for usual care £59,000.
(j) The mean effectiveness corresponded to 3.74 QALYs for HFP, 3.61 QALYs for GORU/MARU and 2.73
QALYs for usual care.
(k) Usual care was never the most cost-effective strategy. At a willingness to pay of £20k per
incremental QALY, HFP was found to be the most cost-effective option in 70% of the 10,000
simulations run in the PSA, while GORU/MARU was the most cost-effective option in 30% of the
simulations. At a willingness to pay of £30K per incremental QALY, HFP was found to be the most
cost-effective option in 80% of the 10,000 simulations run in the PSA, while GORU/MARU was the
most cost-effective option in 20% of simulations.
12.2.2 Evidence statement (s)
Clinical Hospital-based MDR (GORU/MARU)

There is a statistically significant and clinically significant reduction in pressure
sores with hospital-based MDR (GORU/MARU) compared to usual care. (HIGH
QUALITY)
There is a statistically significant, but not clinically significant improvement in
recovery of activities of daily living at 1 year with hospital-based MDR
(GORU/MARU) compared to usual care. (MODERATE QUALITY)
There is a statistically significant, but not clinically significant improvement in
transfers (bed to chair) and being more dependent (Katz index) at 1 year with
hospital-based MDR (GORU/MARU) compared to usual care. (LOW QUALITY)
There is a statistically significant, but not and clinically significant reduction in
severe delirium with hospital-based MDR (GORU/MARU) compared to usual
care. (LOW QUALITY)
There is no statistically significant difference in mortality at 6 months and
functional outcomes at 6 months between hospital-based MDR
(GORU/MARU) and usual care. (MODERATE QUALITY)
mortality at discharge between hospital-based MDR (GORU/MARU) and usual
care. (MODERATE QUALITY)

There is no statistically significant difference in length of hospital stay and
readmission to hospital between hospital-based MDR (GORU/MARU) and
usual care. (LOW QUALITY)
Hip fracture programme (HFP)
There is a statistically significant and clinically significant improvement in
functional outcomes at 1 year with HFP compared to usual care. (MODERATE
QUALITY)
There is a statistically significant and clinically significant reduction in
mortality at discharge between HFP and usual care. (LOW QUALITY)
readmission to hospital, between HFP and usual care. (MODERATE QUALITY)
There is no statistically significant difference in length of hospital stay,
between HFP and usual care. (LOW QUALITY)
Economic HFP is the dominant strategy (less costly and more effective) than both
GORU/MARU and usual care as a hospital based multidisciplinary
rehabilitation of hip fracture patients. This evidence has minor limitations and
direct applicability.
Recommendation From admission, offer patients a formal, acute orthogeriatric or

orthopaedic ward-based Hip Fracture Programme that includes
all of the following:
• early identification of individual goals for multidisciplinary
rehabilitation to recover mobility and independence, and to
facilitate return to prefracture residence and long-term
wellbeing.
• continued, coordinated, orthogeriatric and multidisciplinary
review
• liaison or integration with related services, particularly
mental health, falls prevention, bone health, primary care
and social services.
• clinical and service governance responsibility for all stages of

the pathway of care and rehabilitation, including those
delivered in the community.
Relative values of different Patients, clinical staff and health services share the objective of
outcomes safely returning patients to their original functional state and
residence as quickly as possible. However, these objectives are
often in conflict – for instance earlier discharge may be at the
158 HIP FRACTURE
expense of functional improvement, while length of stay may

increases if mortality is prevented among frailer individuals.
Therefore the most important outcomes considered by the GDG
were functional status, length of stay, discharge destination and
mortality. All these outcomes were incorporated into an original
economic decision analysis.
Trade off between clinical Studies of MDR show no significant evidence of harm and a trend
benefits and harms towards improved outcomes across all outcomes. There is no
suggestion of harm resulting from orthogeriatric collaboration in
the HFP literature.
Evidence to support the effectiveness of coordinated hospital-
based orthogeriatric MDR is derived from studies of both HFP and
GORU/MARU models.
Taken together these studies suggest:
• improvement in functional outcome at 1 year, though this has
not been shown to lead to greater success in achieving
patients' objective of returning to their original residence.
• trend toward reduced mortality at discharge, 1, 6 and 12
months, which must reflect an effect in reducing medical
and/or surgical complications (problems with diagnosis,
definition and ascertainment leave this issue unclear).
• reduced hospital length of stay, though some studies only
examined orthopaedic ward length of stay, so the preferred
measure of 'super-spell' (the total time until return home) was
inconsistently characterised.
Additional evidence supporting the effectiveness of a hospital-
based model incorporating continuous orthogeriatrician
supervision is derived from studies of Hip Fracture Programmes
which suggest:
• reduced patient mortality at discharge and follow-up
• improved functional outcomes
• reduced hospital LOS
• reduced risk of delirium203.
Both HFP and GORU/MARU proved markedly more cost-effective
than usual care, although HFP emerged as the dominant strategy.
The GDG took the view that HFP approach is also preferable
because of its provision of a more extensive programme of
multidisciplinary care that:
• supports admission assessment and peri-operative care, in
addition to rehabilitation, discharge planning and follow-up
• addresses the needs of all patients, including those who might
be viewed as inappropriate for a GORU/MARU (because of
ongoing orthopaedic, medical or psychiatric problems)
• provides a coordinated multidisciplinary structure that will
support other recommendations in this guideline (eg. early

operation).
Economic considerations There were no published economic studies on hospital-based MDR

for hip fracture patients, so an original decision analysis was
developed to determine the cost-effectiveness of HFP vs.
GORU/MARU vs. usual inpatient rehabilitation (usual care).
The cost-effectiveness model was based on an indirect comparison
of randomised trials, but clearly showed that usual care was not
the optimal approach.
The increased costs of hospital MDR were more than offset by:
• reduction in the acute hospital stay costs, including those
associated with complications such as delirium and pressure
sores.
• a reduction in the level of domiciliary social care costs as a
result of increased probability of regaining pre-fracture
independence in activities of daily living.
• reduction in costs for patients who avoid the need for long-
term care in a residential or a nursing home.
HFP was the strategy with the highest incremental net benefit
averaged across all the probabilistic simulations, and appeared to
be the optimal strategy in the cost-effectiveness analysis both in
comparison to usual care, and in comparison to GORU/MARU.
However, there remains some uncertainty about the relative cost-
effectiveness of HFP and GORU/MARU. In particular, the results
were sensitive to the proportion of patients returning home after
completing the rehabilitation programme. Sensitivity analysis
suggested that if the probability of returning home in the
GORU/MARU programme was increased to 83% (instead of 79% as
in the base case) then GORU/MARU would become the optimal
strategy.
Quality of evidence The GDG noted that the precision of the cost-effectiveness analysis
was partially limited by the lack of clinical trials directly comparing
HFP vs. GORU/MARU, and by the heterogeneous patient
population in the meta-analysis of clinical trials on which the cost-
effectiveness analysis is based.
However, the GDG agreed that the outcomes used in the economic
analysis were overall of moderate quality and that the decision
model is likely to provide a relatively unbiased estimate of cost
effectiveness.
There are consistent trends towards benefit across all outcomes,
but the small size of individual trials with a highly heterogeneous
patient population means that statistical significance is difficult to
achieve.
Inconsistency in definition of outcome (variable length of follow-
up, differing functional outcome measures, and poor definition of
160 HIP FRACTURE
'super-spell’) result in several similar outcomes reported separately

which could not be combined in a meta-analysis.
There are no studies in which orthogeriatrician input is confined to
initial assessment and peri-operative medical care, (without then
leading into orthogeriatric MDR). Therefore, the value of such early
orthogeriatrician involvement can only be inferred from the
outcome of HFP studies.
The quality of the studies ranges from low to high, with the
majority of outcomes obtaining a moderate score.
Other considerations The orthogeriatric assessment that would be provided to individual
patients by a multidisciplinary HFP team will vary according to
individual circumstances, and it was not felt appropriate to specify
these in detail in this Guideline.
Assumptions – all papers included an orthogeriatrician, but the
outcomes are most plausibly those of coordinated hospital-based
multidisciplinary team working, with orthogeriatricians playing a
medical and supervisory role within the team.
An important function of the HFP is to ensure the required liaison
with, or cross-coverage of, the programmes in place for the
secondary prevention of fracture by means of the assessment and
treatment of osteoporosis and risk of falling (see NICE Clinical
Guideline 21 & Technology Appraisal 161 and 204 227,234-236). In
some centres HFP staff (including the orthogeriatrician) have
common or parallel commitments within these programmes, with
the resulting potential to achieve additional economies over and
above those identified in the model.
implementation.
Recommendation If a hip fracture complicates or precipitates a terminal illness, the

multidisciplinary team should still consider the role of surgery as
part of a palliative care approach that:
• minimises pain and other symptoms and
• establishes patients' own priorities for rehabilitation and
• considers patients' wishes about their end-of-life care.
Relative values of different Patients with advanced, life-threatening cardiorespiratory,

outcomes neurological, and malignant disease make up a substantial
proportion of those presenting with hip fracture.
In addition the trauma of suffering a hip fracture, and orthopaedic
and medical complications of the injury, immobility and surgery
can themselves precipitate a deterioration in the health of
individuals.
In these circumstances such individuals and their families may view
relief of pain, restoration of function and return home as a higher
priority than survival. Taking this into consideration the GDG

prioritised pain, functional status and discharge destination as the
most important outcomes.
Sometimes this may make it necessary to move from an active
surgical and rehabilitative approach to a palliative focus that
ensures that the patient can die with dignity, with appropriate
attention pain and other symptoms, and all the support necessary
to minimise their and their family's distress.
Trade off between clinical Pain, immobility, continence, pressure ulcer risk and dignity are all
benefits and harms improved if the hip fracture can be addressed surgically, and
perioperative risk should not preclude consideration of surgical
management as an integral component of palliative care.
The prognosis for an individual patient's recovery, mobility and
return home can change markedly and multidisciplinary
assessment is necessary if patients, their families and carers are
given information with which to make informed decisions about
their priorities for care (see chapter 13 Patient and carer views and
information).
High quality palliative and terminal care requires a multi-
disciplinary approach, which should be provided as a key part of
the support that the Hip Fracture Programme offers. Early
orthogeriatric assessment and ongoing multidisciplinary working
will help in:
• avoidance of complications such as pressure sores 344 and
delirium 203
• expediting discharge.
Economic considerations No cost-effectiveness evidence was identified on this sub-group of
patients. Additional time spent in counseling and supporting
patients and their families will clearly carry a cost. While
improvements in a patient's symptoms and quality of life may be of
only short duration, sensitively handled palliative care can
substantially improve their relatives' distress both before and for
many years after bereavement.
Quality of evidence There is no evidence directly relating to this very frail sub-group.
Terminally ill patients were often excluded from these papers and
if included were not reported in specific sub group analysis. This
recommendation was based on GDG consensus opinion.
Other considerations For patients whose hip fracture occurs in the context of advanced
or terminal cancer-related illness, please see NICE Clinical
Guideline “Improving supportive and palliative care for adults with
cancer”227.
162 HIP FRACTURE
Recommendation Healthcare professionals should deliver care that minimises the

patient’s risk of delirium and maximises their independence, by:
• actively looking for cognitive impairment when patients
first present with hip fracture
• reassessing patients to identify delirium that may arise

during their admission
• offering individualised care in line with ‘Delirium’ (NICE

clinical guideline 103)
Relative values of different Patients with memory problems make up a substantial proportion
outcomes of admissions, and face increased risk of delirium, medical
complications, mortality, prolonged length of stay, and failure to
return to pre-fracture independence.
The GDG considered medical complications, mortality, length of
stay and discharge destination as the most important outcomes.
Trade off between clinical Patients with memory problems are known to benefit from acute
benefits and harms comprehensive geriatric assessment and targeted intervention as a
means of reducing their risk of delirium and severe delirium, which
are significant contributors to increased length of stay and
increased risk of morality at 6 months 150,150, as well as being a
source of profound distress for patients, their families and carers
203,203
.
In addition, intensive rehabilitation has been shown to be effective
in improving outcome in terms of independent living among
patients with mild to moderate cognitive impairment 157,157.
No evidence of harm was found and the GDG would not expect
harm. Although no evidence met our inclusion criteria for this area,
GDG consensus is that the potential benefits include avoidance of
the distress that delirium causes to patients, their family, carers,
and other inpatients, along with avoidance of the persistent
reduction in cognitive function that can follow an episode of
delirium, and of the increased length of stay and mortality
associated with delirium.
The avoidance and management of delirium in patients with hip
fracture is specifically addressed in the NICE Guideline on
Delirium224.
Economic considerations The decision model from the NICE guideline on Delirium (CG103)
found that the tailored multi-component intervention package was
cost-effective for hip fracture patients (£8,000 per QALY gained), as
this care would lead to a reduced risk of long-term institutional
care placement, lower incidence of other medical complications
and lower length of hospital stay for these patients.
Quality of evidence Patients with cognitive impairment are usually a group excluded
from studies. Over 60% of the papers reviewed either excluded
patients with cognitive impairment and/ or dementia, or made no
specific comments relating to this subgroup. The studies that
specifically analysed this subgroup157,203 are of moderate quality.
Other considerations For patients whose hip fracture occurs in the context of dementia,
please see the NICE guidance on dementia224.
Identification of cognitive impairment is a key part of assessment,
and a number of tools have been used in patients with hip fracture.
The Abbreviated Mental Test (AMT) score is often used, and forms
part of the National Hip Fracture Database's dataset, but the GDG
did not examine the choice of tool or approach to assessment.
Assessment of mental state can be complex in patients who are in
pain, or who have received strong analgesia at the time of
presentation. Approaches to the prevention and management of
delirium require much more than screening for cognitive
impairment at admission, and must include a sensitivity to changes
in mental state and an awareness that delirium may arise at any
stage of a patient's stay.
Delirium is not confined to patients with pre-existing cognitive
problems, and its incidence will be reduced most effectively by the
provision of continuous orthogeriatric support to all patients203.
Evidence on the effectiveness of models to prevent and manage
delirium following hip fracture were key to the recommendations
made in the NICE Guideline on Delirium224, and that Guideline
should be read alongside our own when developing services for
patients with hip fracture.
12.3 Research recommendations on hospital multidisciplinary

rehabilitation
12.3.1 Hip fracture unit
➢ What is the clinical and cost effectiveness of a designated hip fracture unit within
the trauma ward compared to units integrated into acute trusts on mortality,
quality of life and functional status in patients with hip fracture?
The increasingly structured approach to hip fracture care has led to a number of UK units
considering or establishing a specific ‘hip fracture ward’ as a specialist part of their acute
orthopaedic service.
Designated hip fracture wards may prove an effective means of delivering the whole programme
of coordinated perioperative care and multidisciplinary rehabilitation which this NICE Guidance
164 HIP FRACTURE
has proposed, but at present there is no high quality evidence of their clinical effectiveness when
compared to such care within general orthopaedic or trauma beds.
It may not be practical to run an RCT within a trauma unit, but there is certainly potential for
cohort studies to explore the effect of such units on individual patients' mobility, discharge
residence, mortality and length of stay. Units considering the establishment of hip fracture wards
should be encouraged to consider performing such trials.
12.4 Community-based multidisciplinary rehabilitation versus usual care

In addition or as an alternative to hospital based multidisciplinary rehabilitation (MDR), a
number of studies have evaluated the role of community based MDR.
Community-based MDR includes approaches that are:
➢ based in the patient’s own home - Early Supported Discharge (ESD)
➢ based within a residential care unit or community hospital
➢ based within a Social Care Unit (SC) - or their near equivalents.
The many versions of these services across the country are named differently (for example
‘intermediate care at home’, ‘intermediate care residential rehabilitation’), but each
consists of a rehabilitation component delivered in one of the above settings.
In patients with hip fracture what is the clinical and cost effectiveness of community-based
multidisciplinary rehabilitation on functional status, length of stay in secondary care,
mortality, place of residence/discharge, hospital readmission and quality of life?
Two studies met the inclusion criteria for this review question, with a total of 168 patients.
See evidence table 11, Appendix E and forest plots G140 to G149 Appendix G.
Table 12-68: Home-based multidisciplinary early supported discharge vs. usual care – Clinical
study characteristics
Numbe Other
Mortality at 12 1 RCT serious(a) no serious no serious serious(b)
months 59 inconsistency indirectness
Moved to a 1 RCT serious(a) no serious no serious serious(b)
higher level of inconsistency indirectness
care 59
Unable to walk 1 RCT serious(a) no serious no serious serious(b)
59
inconsistency indirectness
SF-36 scores at 12 1 RCT serious(a) no serious no serious no serious
months (0: worst inconsistency indirectness imprecision
to 100: best) -
Physical
component
summary scores
59
166 HIP FRACTURE
Numbe Other
SF-36 scores at 12 1 RCT serious(a) no serious no serious serious(c)
months (0: worst inconsistency indirectness
to 100: best) -
Mental
component
summary scores
59
Length of 1 RCT serious(a) no serious no serious no serious

hospital stay 59,360 inconsistency indirectness imprecision
Lengths of 1 RCT serious(a) no serious no serious no serious
hospital or inconsistency indirectness imprecision
rehabilitation
stays (days) -
Length of
rehabilitation
(hospital + home)
59
Readmission to 1 RCT serious(a) no serious no serious serious(b)

hospital during 4 inconsistency indirectness
months follow-up
59
Degree of 1 RCT no serious no serious no serious no serious

independence limitations inconsistency indirectness imprecision
(Functional
Independent
Measure) - FIM
Self-care – 1
month
360

(Functional
Independent
Measure) - FIM
Mobility
360

(Functional
Independent
Measure) - FIM
Locomotion
360
Mobility and 1 RCT no serious no serious no serious serious(b)

strength tests - limitations inconsistency indirectness
Up and go test
360
Mobility and 1 RCT no serious no serious no serious no serious

strength tests - limitations inconsistency indirectness imprecision
Sit-to-stand test
360
(a) Baseline data for Crotty et al., 200359 each study arm not given.
(b) The relatively few events and few patients give wide confidence intervals around the
estimate of effect. This makes it difficult to know the true effect size for this outcome
(c) The wide confidence intervals around the measurement make the result imprecise. This
makes it difficult to know the true effect size for this outcome.
Table 12-69: Home-based multidisciplinary early supported discharge vs. usual care - Clinical
summary of findings
months 4/32 1000 (from 104
3/34 (8.8%) RR 0.71 (0.17, 2.91) Low
(12.5%) fewer to 239
more)
Moved to a higher 33 fewer per
level of care 1000 (from 60
1/34 (2.9%) 2/32 (6.3%) RR 0.47 (0.04 to 4.94) Low
fewer to 246
more)
Unable to walk 51 fewer per
1000 (from 62
0/34 (0%) 2/32 (6.3%) RR 0.19 (0.01 to 3.78) Low
fewer to 174
more)
SF-36 scores at 12
months (0: worst to
MD 4.7 (0.04 to
100: best) - Physical 34 32 N/A Moderate
9.44)
component summary
scores
SF-36 scores at 12
months (0: worst to
MD 1.5 (2.54 to
100: best) - Mental 34 32 N/A Low
5.54)
component summary
scores
Length of hospital MD -2.96 (-5.50
82 86 N/A Moderate
stay (days) to -0.42)
Lengths of hospital or
rehabilitation stays
MD 2.96 (5.5 to
(days) - Length of 34 32 N/A Moderate
0.42)
rehabilitation
(hospital + home)
Readmission to 18 more per
hospital during 4 7/32 1000 (from 123
8/34 (23.5%) RR 1.08 (0.44, 2.62) Low
months follow-up (21.9%) fewer to 354
more)
Degree of
independence
(Functional MD 4.90 (2.81,
48 54 N/A High
Independent 6.99)
Measure) - FIM Self-
care
Degree of
independence
48 54 N/A High
Independent 2.98)
Measure) - FIM
Mobility – 1 month
168 HIP FRACTURE
Degree of
independence
48 54 N/A High
Independent 3.99)
Measure) - FIM
Locomotion
Mobility and strength MD 5.9 lower
tests - Up and go test 48 54 N/A (12 lower to 0.2 Moderate
higher)
Mobility and strength MD 1.5 lower
tests - Sit-to-stand 48 54 N/A (2.49 to 0.51 High
test lower)
Our search identified five studies on community MDR versus usual care. Of these, one 55,55
was excluded as it included a mixed population with only 31% hip fracture patients. Van
Balen et al., 2002340,340 was excluded as patients in the early supported discharge scheme
were only discharged to a nursing home with rehabilitation facilities and not to their own
home.
The following studies were included as economic evidence on the cost-effectiveness of

home-based multidisciplinary early supported discharge vs. usual care: Hollingworth
(1993)148,148 O’Cathain (1994)245 and Parker (1991)270,270. Hollingworth (1993)148,148 is a cost
analysis based on a case series. O’Cathain (1994)245 is a cost-consequences analysis based
on a non-randomised trail with concurrent controls. Parker (1991)270,270 is a cost-
consequences analysis based on a prospective observational study. For further details on
these studies please refer to the Evidence Table 16 in Appendix F.
An original decision analysis has been conducted comparing the cost-effectiveness of the
community MDR vs. usual care. A decision tree model with Markov states was developed,
adopting a life-time horizon.
Treatment effects and EQ-5Ds scores were based on the findings of Crotty (2002) 60 and
applied only up to 4 months from follow-up. Resource use was determined from the NHS
and PSS perspective. Effectiveness was measured in QALYs. Costs and QALYs were
discounted at a rate of 3.5%. Please see section 20.7 in Appendix H for further detail.
Table 12-70: Home-based multidisciplinary early supported discharge vs. usual care -
Economic study characteristics
Hollingworth 1993148 Potentially serious limitations Partial applicability A community-based MDR
(a)
at home scheme was
The MDR at home
programme consisted of:
care from trained nurses,
nursing auxiliaries,
physiotherapists, and
occupational therapists in
the patient’s home for up
to 24 hrs a day under the
medical supervision of
the general practitioner
O’Cathain 1994245 Potentially serious limitations Partial applicability MDR at home compared
(b)
to usual care. MDR team
consisted of district
nurses, physiotherapists,
occupational therapists
and generic workers, all
working under the clinical
responsibility of a GP for
a maximum of 12 days.
Parker 1991270 Potentially serious limitations Partial applicability MDR at home scheme
(c)
MDR team consisted of
trained nurses, nursing
auxiliaries,
physiotherapists, and
occupational therapists.
NCGC economic Minor limitations (d) Direct applicability Cost-effectiveness
model analysis of community
MDR – ESD versus usual
care based on the RCT by
Crotty et al (2002) 60
included in the clinical
review.
(a) Unclear follow up time. HRQoL not calculated. Information on costs obtained from hospital records,
not national statistics. Not an RCT.
(b) The length of time during which costs are calculated is unclear. No sensitivity analysis was
conducted. Not based on a RCT. Not a CUA.
(c) Not based on a RCT. No sensitivity analysis. Cost data from hospital source, not national statistics.
Only patients admitted from their own home were then discharged under the HAH scheme.
(d) The analysis consists of a decision tree with Markov states which spans a life-time horizon.
Treatment effects based on the findings of the paper by Crotty in the clinical review and applied
only up to 4 months from follow-up. Resource use determined from the NHS and PSS perspective,
Effectiveness measured in QALYs. QALYs discounted at a rate of 3.5%.
170 HIP FRACTURE
Table 12-71: Home-based multidisciplinary early supported discharge vs. usual care - Economic
summary of findings
Study (£) effects ICER Uncertainty
Hollingworth -£722 LOS; readmissions N/A One way sensitivity
(l)
1993 analysis: costs of MDR
scheme at home would still
be lower than usual care if
inpatients costs 50% lower
and MDR at home costs
50% higher than predicted.
O’Cathain 1994 -£370 Several outcomes N/A N/R
reported (m)
Parker 1991 -£799.80(n) Several outcomes N/A N/R
reported (o)
NCGC economic £434.6(p) 0.0456 QALYs(q) £9533/QALYs 95% CI: Community MDR
model dominant –usual care
dominant (r)
(l) LOS for MDR at home vs. usual care: 32.5 vs. 41.7 days (p<0.001); readmission rates at 1 year: 6.8%
(53 patients) vs. 2.7% (8 patients), p=0.008
(m) Several outcomes were reported: HRQoL measured with the Nottingham Health Profile
questionnaire (14 vs. 24, p<0.05); Mortality (5.3% vs. 5.9%; p = NR); readmission rates at 3 months:
(15.8% vs. 8.8%, p=0.187); LOS (median no of days): 10 vs. 17, p<0.001
(n) Costs based on the following resource use: hospital length of stay; sessions with hospital
occupational therapist; readmission days; MDR ESD staff time; other NHS or social services (GP
visits, day care, meals on wheels, community services)
(o) LOS (mean, days): 29 vs. 38 (p value: 0.035). Mortality (at 90 days): 40 (14%) vs. 14 (11%)
(p) The mean costs associated with community MDR were estimated to be £6901.20 and for usual care
£6466.60
(q) The mean effectiveness corresponded to 3.1283 QALYs and 3.0827 QALYs for usual care.
(r) Deterministic sensitivity analysis showed that findings were sensitive to the length of stay spent in
hospital and during rehabilitation at home. Community MDR was found to be the most cost-
effective option in 50% of the 10,000 simulations run in the PSA at a willingness to pay of £20k, and
in 60% of the simulations at a willingness to pay of 30k per QALY.
12.4.3 Evidence statement (s)
Clinical There is a statistically significant and clinically significant reduction in hospital

length of stay, but an increase in total length of rehabilitation (hospital +
home) with home-based multidisciplinary early supported discharge (ESD)
compared with usual care. (MODERATE QUALITY)
There is a statistically significant and clinically significant increase in
functional independence measures with home-based multidisciplinary ESD
compared with usual care. (HIGH QUALITY)
readmission to hospital at 4 months with home-based multidisciplinary ESD
compared with usual care. (LOW QUALITY)
Economic Home-based MDR – ESD is cost-effective in the rehabilitation of patients with
hip fracture. This evidence has minor limitations and direct applicability.
Recommendation Consider early supported discharge as part of the Hip Fracture

Programme, provided the Hip Fracture Programme
multidisciplinary team remains involved, and the patient:
• has the mental ability to participate in continued
rehabilitation and
• has not yet achieved their full rehabilitation potential, as
discussed with the patient, carer and family.
Relative values of different Length of hospital stay, functional outcomes and re-admission
outcomes rates were considered the primary outcomes of interest. All these
outcomes were used in the decision analytical model.
Trade off between clinical Multidisciplinary ESD at home in selected patients reduces hospital
benefits and harms length of stay but may result in overall prolonged rehabilitation
(hospital + home) compared to hospital MDR. Selected patients
were defined from the studies as medically stable, cognitively
intact, able to transfer independently, and mobilise short
distances.
Despite only a few low quality studies being identified the GDG
consensus was that multidisciplinary ESD at home is beneficial to a
specific patient group, as defined above. The evidence reviewed
showed an increase in functional independence measures with ESD
Our decision analysis found QALYs were 0.0456 higher in the
community MDR arm of the study compared to usual care.
Economic considerations No cost-effectiveness studies were identified for this clinical
question. An original decision analytical model was developed,
which was based on the findings of an RCT included in our clinical
review 58,60. The analysis showed that there is uncertainty as to
whether MDR ESD at home is cost-effective compared to usual
care. In particular, findings were sensitive to the length of hospital
stay and length of the home-based rehabilitation programme.
However, the GDG noted that the ICER of £9533/QALYs is well
below the £20,000 threshold.
It is also important to note that our model did not find community
MDR to be cost saving compared to usual care. This was because
patients in the community MDR branch of the model underwent
rehabilitation in their own home for a relatively longer period of
time than those of the other studies included in the economic
evidence profile in section 8.7 in Appendix H.
Quality of evidence There were few studies identified, which ranged from low to high
quality with often only one study per outcome. Therefore our
172 HIP FRACTURE
confidence in the results is low.

Studies were undertaken in medically stable and cognitively intact
patients and there were no studies that evaluated multidisciplinary
ESD at home in cognitively impaired patients or patients living in
care/nursing homes. This recommendation was therefore partly
based on evidence and partly GDG consensus opinion.
Other considerations Patient selection, as defined above is very important for
multidisciplinary ESD at home and may represent a very small
number of eligible patients.
The benefits of MDR ESD in patient with mild to moderate
cognitive impairment living at home alone or with a relative /carer
are unknown. MDR ESD in this context may be beneficial and
should be considered.
The benefits of MDR ESD in patients living in care /nursing homes
are unknown. MDR ESD in these patients, undertaken alongside
the care/nursing homes may be beneficial.
Interaction with any key carer and evaluation of his/her ability and
willingness to provide support and care is in all cases an essential
and normative element of the decision making process in
considering the appropriateness or otherwise of early supported
discharge
implementation.
Recommendation Only consider intermediate care (continued rehabilitation in a

community hospital or residential care unit) if all of the following
criteria are met:
• intermediate care is included in the Hip Fracture

Programme and
• the Hip Fracture Programme team retains the clinical

lead, including patient selection, agreement of length of
stay and ongoing objectives for intermediate care and
• the Hip Fracture Programme team retains the managerial

lead, ensuring that intermediate care is not resourced as
a substitute for an effective acute hospital Programme.
Relative values of different The GDG considered the most important outcomes to be length of
outcomes stay in hospital (in particular superspell) and return to pre fracture
residence.
Trade off between clinical There are risks that transfer to intermediate care may prematurely
benefits and harms move a co-morbid patient group from a diagnostically supported
environment, impair continuity, and prolong the superspell.
In certain settings and specific circumstances, proximity to home

with access for relatives/carers visiting and a more relaxed and
“homely” atmosphere for continued rehabilitation than the acute
hospital might be considered advantageous.
Economic considerations The average weekly cost of the social care received in an
intermediate care setting based in residential homes varies from a
minimum of £412 to a maximum of £840 for schemes run by local
authorities. The average weekly cost of social and health care
services in the same setting but for schemes run by the local
authority in conjunction with primary care trusts amounts to £574
(source: PSSRU 200961). Subject to the criteria in the
recommendation above, intermediate care may be feasible for our
population, but there is currently no evidence on its cost-
effectiveness.
Quality of evidence There is no evidence on the effectiveness or cost-effectiveness of
rehabilitation within a community hospital or residential care unit
in hip fracture rehabilitation. This recommendation was based on
GDG consensus opinion.
Other considerations Intermediate care rehabilitation for hip fracture remains ill-defined
and highly variable in the UK in terms of its admission criteria,
multidisciplinary composition, intervention components and
mechanisms for shared outcome and resource accountability
within a comprehensive hip fracture programme.
Recommendation Patients admitted from care or nursing homes should not be

excluded from rehabilitation programmes in the community or
hospital, or as part of an early supported discharge programme.
Relative values of different The GDG considered the most important outcomes to be
outcomes functional status, readmission to hospital and return to pre-
fracture residence.
Early assessment and MDR offered as part of a hip fracture
programme with continued rehabilitation for patients admitted
from care/nursing homes is likely to improve/maintain the
patient’s functional ability with regard to mobility, transfers from
bed to chair and activities of daily living. This is in the interests of
both patients and care/nursing home staff. In addition patient
status as a care home resident as opposed to a nursing home
resident may be maintained and equality for patients in
care/nursing homes is maintained with regard to access to
rehabilitation.
Trade off between clinical There is no evidence of harm accruing to care/nursing home
benefits and harms residents from the provision of appropriately individualised
rehabilitation programmes.
For some patients admitted from care/nursing homes there may
be advantages (and no particular risks) in completing their
rehabilitation after hospital MDR within that home (subject to the
174 HIP FRACTURE
recommended criteria above), recognising that their rehabilitation

goals may be more complex and must be shared by the HFP team
on a continuing basis with the care/nursing home staff.
The potential benefits of ESD for patients admitted from
care/nursing homes include the possibility of functional recovery
within the patient’s familiar environment, shared communication,
goal setting and collaboration between care/nursing home staff
and HFP team resulting in improved functional outcome, and the
possibility of reduced hospital stay and inappropriate hospital
readmission.
This subgroup is considered at particular risk of premature
discharge because of ease of access to the care/nursing home
environment and the corresponding perception that functional
recovery matters less. Failure to undertake adequate
rehabilitation carries the subsequent risk of inappropriate
functional decline and/or levels of dependency, reduced quality of
life, unnecessary hospital readmission, and premature mortality.
Provision of part of a patient's continuing rehabilitation
programme in the care or nursing home of origin is correctly
categorised as either early supported discharge or intermediate
care, and the continued involvement of the Hip Fracture
Programme team in liaison with the community-based component
is therefore correspondingly a requirement.
Economic considerations There was no cost-effectiveness evidence. The GDG believe that
any increase in the cost of hospital bed days from the avoidance of
premature discharge should be at least partially offset by the
avoidance of inappropriate readmissions and reduction in
subsequent care costs resulting from optimised functional status.
Quality of evidence No RCTs were identified regarding patients admitted from care or
nursing homes undergoing community ESD, as this patient
subgroup has typically been excluded from clinical trials. The
recommendation is based on GDG opinion and consensus that this
group of patients would benefit from ESD.
Other considerations There is a high prevalence of cognitive impairment in this
population, therefore realistic rehabilitation goals need to be
defined, but not at the expense of excluding rehabilitation.
12.5 Research recommendations on community multidisciplinary

rehabilitation
12.5.1 Early supported discharge
➢ What is the clinical and cost effectiveness of early supported discharge on

mortality, quality of life and functional status in patients with hip fracture who are
admitted from a care home?
Given the patient frailty and comorbidities, rehabilitation may have a limited effect on
clinical outcomes for this group. However, the fact that they already live in a home where
they are supported by trained care staff, clearly provides an opportunity for a systematic
approach to rehabilitation. Early care/nursing home based multidisciplinary rehabilitation
would take advantage of the day-to-day care arrangements already in place in homes and
provide additional NHS support to deliver naturalistic rehabilitation, where problems are
tackled in the setting in which the patient lives.
care required.
The research would follow a two-stage design: (1) An initial feasibility study to refine the
collaboration between the HFP team, care-home staff and other community-based
professionals, and (2) A cluster randomized controlled comparison (with two or more
intervention units and matched control units) set against agreed outcome criteria. The
latter should include those specified above, together with measures of the impact on care-
home staff activity and cost, as well as qualitative data from patients on relevant quality-of-
life variables.
12.5.2 Care/nursing home residents

176 HIP FRACTURE
➢ Do patients admitted to hospital with a fractured hip who live permanently in a

care/nursing home have equal access to multidisciplinary rehabilitation as patients
admitted from their own homes?
The existing literature on the effectiveness of multidisciplinary rehabilitation typically

excludes patients who live in care/nursing homes. From an equality perspective it
hypothesised that this group of people do not have access to the same multidisciplinary
rehabilitation as patients who are returning home as it is assumed patients returning to
care/nursing homes will have their care needs met by the home. The research design would
be a prospective observational cohort study to determine the extent and quality of
rehabilitation services available to this group in comparison to patients returning to their
own homes.
13 Patient and carer views and information
13.1 Introduction
Patient views about their hip fracture and its management, and the way patients are
provided with information are important elements of the natural recovery and treatment of
hip fracture. Care givers also have need for information, and can influence the recovery
process. Timely and clear information could reduce stress and uncertainty for patients and
potentially improve their outcome. This section examines the literature on patient views
and the provision of information to patients.
13.2 Patient and carer views

A systematic literature review was conducted into the views of patients and carers about
their experience of hip fracture management from hospital admission until discharge from
rehabilitation. Studies examining areas not covered by the guideline scope were not
included. For example, hip protectors for falls management, nutrition support or patient
views relating to the time after discharge from rehabilitation programmes.
The aim of this review was to provide:
• Supplementary evidence to clinical questions addressed in the guideline

PATIENT AND CARER VIEWS AND INFORMATION 177
• A general overview of patients views’ on hip fracture and hip fracture

management
• Evidence relating to the provision of information to patients and carers
Eleven qualitative studies are included here, only two of which are UK based studies. More
details about the studies are presented in the evidence table (Evidence table 12 in
Appendix E). Studies were assessed using the NICE methodology checklist for qualitative
studies233.
13.2.1 Summary of studies
Table 13-72 Patient views study quality

Study Population Methods Analysis Relevance to guideline population
Archibald Adequately Adequately Adequately Community hospital in UK
20038 reported reported reported, 4 patients interviewed during
credible rehabilitation
Borkan 1991 Adequately Well Well 4 hospitals in USA
& 199228,29 reported reported reported, 80 patients interviewed during hospital
credible stay
Bowman Adequately Poorly Poorly Teaching hospital in Canada
199733 reported reported reported, 17 patients interviewed on day of
credible admission
Furstenberg Adequately Poorly Poorly Urban hospital in USA
1986105 reported reported reported, 11 patients interviewed at one or more
credible points during hospital stay
Olsson Well Well Well Geriatric/ orthopaedic ward in Sweden
2007249 reported reported reported, 13 patients interviewed soon after the
credible operation
Pownall Well Poorly Adequately Trauma/ orthopaedic ward in UK
2004274 reported reported reported, 1 patient interviewed prior to discharge
credible from acute trauma and orthopaedic
ward
Slauenwhite Poorly Adequately Poorly Hospital in Canada
1998314 reported reported reported, 23 ‘caregivers’ for 23 patients
credible interviewed 4 to 6 weeks after discharge
William Poorly Poorly Poorly Hospital in USA
1994354 reported reported reported, 120 patients interviewed before hospital
credible discharge and followed up at 2, 8 & 14
weeks
Wykes Well Well Well Rehabilitat-ion hospital in Australia
2009355 reported reported reported, 5 patients interviewed during
credible rehabilitation
Young Adequately Well Well Rehabiliitat-ion centre in USA
2009358 reported reported reported, 62 patients interviewed after 12 month
credible follow up meeting
178 HIP FRACTURE
Study Population Methods Analysis Relevance to guideline population

Ziden Well Well Well Hospital in Sweden
2010362 reported reported reported, 18 patients interviewed at 1 month
credible follow up meeting and 15 at 1 year
follow up
Archibald et al (2003)8 conducted a qualitative study of 5 hip fracture patients in a

community hospital in the UK. Their aim was to explore experiences of individuals who had
suffered a hip fracture. Interviews with open ended questions were conducted during their
stay in hospital.
Four main themes were identified: injury experience, pain experience, recovery experience,
disability experience. Only the pain and recovery experience relate to their time in hospital
and rehabilitation. Most patients described the pain they experienced, one mentioned
being in a lot of pain in the orthopaedic unit despite pain killers. Another mentioned they
thought the pain went with rest after a while, but not completely. Only 1 person was still
having pain at time of interview. The recovery experience was split into 3 sequential
categories: the operation, beginning the struggle and regaining independence. Only 1
person described the operation, they had a “horrendous” recollection of a noisy operating
theatre, like being in an engineering shop or something”. Three patients remembered
‘beginning the struggle’: they reported not being able to do anything; struggling to get to
the toilet and into a chair; and hating using a bed pan. The comments relating to regaining
independence were all positive. Motivation, be it getting to the toilet, the dining room or
smoke room was found to be a key factor in the recovery of the patients.
Borkan et al (1991 & 1992)28,29 conducted a qualitative study of 80 hip fracture patients in 4
hospitals in the USA. Their aim was to investigate the meanings of hip fracture to older
patients, and to identify potentially important prognostic indicators or risk factors for
rehabilitation outcomes. Patients were interviewed during the first week after hip fracture
with a combination of open-ended and multiple choice questions.
The study reports how patients perceive their fracture, their perception of their disability
and whether they were hopeful for the future (see evidence table). Also reported were
patient expectations of recovery (43 expected full recovery, 14 partial recovery and the rest
did not know or did not give an answer) and patient expectations about their living
situation (61% predicted going home, 15% into a nursing home though none came from
one, 9% predicted being discharged to their children’s houses and 15% did not know or did
not respond). The actual figures showed that 43% were discharged to long term care
institutions, of these 38% remained in the institution at 1 year, 53% returned home and 9%
died.
Bowman (1997)33 conducted a quantitative study of 43 patients undergoing surgery on the
hip in a hospital in Canada, 17 of these had a hip fracture. The main aim was to describe
sleep satisfaction, pain perceptions and psychological concerns of patients undergoing hip
operations. Also two open ended questions were asked at the time of admission to
elucidate the patient’s biggest concerns about their injury and forthcoming surgery, and
whether they had concerns about their ability to recover fully and quickly. The mean age of
hip fracture patients was 80 years old and, unlike most the other studies, it also included
patients with delirium (8 out of 17). Six out of 17 patients feared being unable to walk
again, an additional 3 out of 17 were concerned about their recovery and managing on
their own.
Fustenberg (1986)105 conducted a qualitative study of 11 patients of hospitalised patients

with hip fracture in a hospital in the USA. The aim of the study was to “construct a natural
history of the hip fracture”, from the events surrounding the hip fracture through the
hospitalisation period. Ethnographic interviews were carried out at one or more points
during their hospital stay.
The findings were split into two main sets: immediate patient expectations about their
recovery and “contextual factors” to the evolving expectations about their recovery. The
immediate expectations mostly included expressions of despair and discouragement: hip
fracture was going to result in extended period of slow recovery of function, with attendant
dependency, postponement or relinquishment of plans and changed living situation with
the threat of permanent loss of independent living. Participants also suffered uncertainty
about timing and completeness of return to full recovery
As time progressed participants commented that although progress was slow they could
see improvements. They also took encouragement from other people’s recovery. The study
notes that while patients could focus on positive and negative points, the participants only
focused on encouraging examples.
The study also reports that healthcare professionals’ cues, encouragement and feedback
guided the participants’ perceptions about their own progress. However, some participants
“referred to the elusiveness of the doctors and their own unanswered questions.”
Olsson et al (2007)249 conducted a qualitative study of 13 hospitalised patients in Sweden.
The aim of the study was to describe patient’s own perceptions of their situation and views
of their responsibility in the rehabilitation process. Interviews were conducted with semi-
structured questions as soon after the operation as the patients felt strong enough.
The study categorised the findings into different conceptions: ‘autonomous’ – responses
from people who appeared confident and accustomed to managing on their own; ‘modest’
– responses from people who gave the impression of being vulnerable and dependent on
others, this group worried about their future more than the others; ‘heedless’ – responses
from people who appeared to have a sense of detachment. The heedless did not doubt
they would recover and that people around them would care for them. This group was
characterised predominantly by a reluctance to reflect on their own situation, by a refusal
to accept responsibility and by their need for information.
The study also identified some common traits: a lack of awareness - most patients lacked
awareness about their condition, what to do and how to act, and needed more
information; a shocking event - although several suspected they had a fracture all were
distressed by the diagnosis. The period before surgery was mostly blurred and filled with
fear and pain. The participants worried about how they would function postoperatively;
zest for life - all expressed a strong desire to recuperate although, while confined to bed
they worried about the pain, their inability to move their leg, their forthcoming operation
and the fear of being unable to walk again.
Pownall 2004274 conducted a critical appraisal of a 60 year old women’s experience with hip
fracture in a UK hospital. The study was undertaken in an effort to understand further the
nature of personal experience. Narrative was acquired as part of a routine nursing
evaluation and helped to illuminate nursing care issues through the eyes of the patient. The
participant was interviewed prior to discharge with four open-ended questions.
The study identified three areas for improvement within the hospital: better
communication skills; time management for staff so time spent with the patient is used
effectively; and better pain management. The participant’s comments included not
understanding why they had to wait so long in the Emergency department after the x-ray as
180 HIP FRACTURE
they had already been told their hip was fractured; staff were so busy, no one had time to
sit and explain things to her; concern that the operation was explained to her son but not
her; shock at being mobilised the day after surgery.
Slauenwhite and Simpson (1998)314 conducted a qualitative study of 23 “caregivers” for 23
patients who had experienced hip fracture in Canada. The purpose of the study was to
investigate the impact of enhanced early discharge on families experiencing a repaired hip
fracture in an older adult. “Caregivers” were interviewed 4 to 6 weeks after discharge.
The length of stay was considered too long by the patient with the fracture and too short by
the carer for families. 15 out of the 23 families found length of stay not an issue. 20 of the
families stated pain management was not a problem in hospital or at home. Several
families thought the transition from hospital to home was a problem as it took several
hours to days for all the information to be relayed to home care system. This went hand in
hand for those with comorbidities. Many caregivers had stories of dissatisfaction which was
suggested to be related to health care system and mismatched care. Mismatched care was
not well defined.
Williams et al (1994)354 conducted a study into patient recovery and views for 120 patients
after hospital discharge in the USA. Participants were asked what advice they would offer
to other patients who had just fractured their hip. Patients were interviewed at 14 weeks
after discharge.
The advice offered was grouped into categories: 94 patients emphasised the importance of
mental attitude with comments such as patients should “maintain hope” and “look to the
future”; 76 patients suggested that following experts’ advice; 34 advised mobility was key
with comments such as keep mobile, rest before getting up to walk, use walker to help get
up; 15 advised maintain healthy lifestyle; 7 said use caution and be careful not to fall; 3
suggested limiting stay in institution and get help to be at home if possible; and 6 gave no
specific advice as they commented that everyone is different.
Wykes et al (2009)355 conducted a qualitative pilot study to explore the impact of hip
fracture on the lives of previously independent women and to identify their concerns when
participating in inpatient rehabilitation. Five patients were interviewed during their stay in a
rehabilitation hospital in Australia.
The impact of the fracture was an issue for all five women as others had to assume
responsibility for things they had done previously. The study categorised the women’s
concerns into four categories: the behaviour of others; what was happening to them; the
impact of their injury on others; and other health issues. A few comments were raised
about the behaviour of others including things others said and did, friends and family doing
things without asking first, the family not being told when one woman had moved hospital,
concern that staff expect one woman’s daughter to look after her until rehabilitation
started. Concerns about what was happening to them included a possible loss of
independence, possible accommodation changes after discharge and money issues. The
women were also concerned about inconveniencing or upsetting others by telling them
what they were feeling or asking too many questions. Two women had pre-existing health
issues which, combined with their hip fracture, had adverse effects on their outcome. These
overshadowed specific concerns about their hip fracture.
Young and Resnick (2009)358 conducted a qualitative study to explore the perceptions of 62
older adults regarding their functional recovery 1 year after hip fracture and after
participating in rehabilitation programme in the USA. Participants were asked whether they
were satisfied with their functional recovery, what helped or hindered recovery, what
would improve recovery and what one piece of advice they would offer other hip fracture
patients. The themes identified are listed below.
53 participants were satisfied with their functional recovery. The main factors they listed as
facilitators of recovery were seeing health care professionals and their positive attitude (40
respondents); social support, particularly from family and friends (13 respondents); and
their own determination (12 respondents). Other factors mentioned included lifestyle
factors or an environment that encourage healthy living, individualised care & verbal
encouragement; spirituality and identifying goals. The nine people who were dissatisfied
with their recovery listed medical complications or comorbidities, unpleasant sensations
and age as factors that hindered their recovery.
The respondents also identified areas that would facilitate recovery: more direct physical &
occupational therapy and more education about the recovery process and ways to optimise
physical function (26 respondents); better follow up and care in the home setting after
discharge from rehabilitation (9 respondents); spirituality (3 respondents), social support (2
respondents); additional information (8 respondents); elimination of unpleasant sensations
(4 respondents) and policy (1 respondent).
The patients also offered the following advice on how to facilitate recovery to anyone with
a hip fracture: listen to healthcare instructions (19 respondents) and participate as much as
possible in rehabilitation activities (48 respondents); participants strongly recommended
that older adults who sustain hip fractures maintain a positive attitude (20 respondents)
and remain determined throughout the recovery experience (13 respondents); be careful
to avoid subsequent trauma and prevent anything that would impede recovery (8
respondents); push through the pain and use all medication offered (6 respondents); and
don’t worry (4 respondents).
Ziden et al (2008 & 2010)362,363 conducted a qualitative study to explore and describe the
consequences of an acute hip fracture among home dwelling elderly people shortly after
discharge from hospital in Sweden. Patients, who had participated in a randomised
controlled trial investigating rehabilitation360 included in the rehabilitation chapter (Section
12.2), attended semi-structured interviews at 1 month and 1 year after hip fracture.
The study identified different responses or perceptions over time. At 1 month patients:
found they were limited in movement and have lost confidence in their body (18 people);
had become humble and grateful (7 people); respected themselves and their own needs (2
people); had become more dependent on others (12 people); gain more human contact
and are treated in a friendly way by others (2 people); were secluded and trapped at home
(4 people); were old, closer to death and have lost your zest for life (4 people); were taking
one day at a time and were uncertain about the future (7 people). At 1 year after discharge
patients felt: more insecure and afraid (11 patients); they had more limited ability to move
(12 patients); disappointed and sad that identity and life have changed (8 patients);
satisfied with the situation or felt even better than before their fracture (5 patients).
The study also identified some patient views about determinants of hip fracture recovery:
10 patients stated their own mind and actions influenced recovery; 4 patients stated that
treatment and the actions from others influenced recovery; whereas 6 patients stated you
cannot influence recovery.
13.2.2 Common themes
The following themes have been identified from the studies:

182 HIP FRACTURE
Initial outlook in hospital

Five studies with 126 participants reported views from this period 8,28,29,33,105,249. One of the
studies reported the responses varied "from stubborn optimism to despair“28,29. Another
study also reported all 13 participants expressed a strong desire to recuperate 249. However,
most of the expressions were negative with no positive comments reported in the papers.
The concerns covered:
• pain and the inability to move their leg while confined to bed
• the fear of being unable to walk
• not being able to do anything
• hating using a bed pan
• struggling to get out of the chair or bed
• concern about recovery and managing on their own
• return to independent living
• limitations on their functioning and consequent implications
• being burden on their “caretakers” [families and carers]
• further falls
• uncertainty about timing and completeness of return to full recovery
Attitude as patients began to regain independence
Two studies reported comments relating to this period.
• Archibald (2003)8 with 5 participants reported motivation to be key factor in recovery.
All comments in the study were positive about regaining independence during their
rehabilitation.
• In Furstenberg (1986)105 (11 participants) participants commented that although
progress was slow they could see improvement. Participants also took encouragement
from others’ progress.
Management by health care professionals
Positive and negative comments were reported about healtcare professionals:
• Encouragement and positive attitude - Furstenberg (1986)105 (11 participants) reported
that healthcare professionals’ cues, encouragement and feedback guided the
informants’ perceptions about their own progress. 40 out of the 62 participants in
Young (2009)358 identified that communication and a positive attitude by professionals
were seen as a facilitator of recovery.
• Provision of information to patients - Two studies also noted some negative points,
some patients “referred to the elusiveness of the doctors and their own unanswered
questions.” in Furstenberg (1986)105. The woman with a hip fracture in the individual
patient narrative 274 was unhappy that things were not explained to her. One of her
comments highlighted this where she reported that the “staff were so busy no one has
time to sit and explain things to you”.
• Explaining directly to patients - The patient from the individual narrative 274 was also
unhappy that she could hear the nurse explaining the operation to her son, but
nothing was explained to her.
Overall, little evidence was identified that provided direct comments relating to our review
questions. Where applicable data were identified, reference to the evidence has been
made in the link to evidence of the relevant recommendations. These related to:
• Several comments were identified that fed into our recommendation relating to the
provision of information to patients (see next section 13.3).
• Some supplementary evidence was identified relating to pain that fed into our
analgesia recommendations (see section 7.2.2).
13.3 Information for patients
This section covers structured health education approaches, advice, information and
reassurance. In addition to qualitative literature the search conducted for patient views
included terms relating to patient education interventions. This also aimed to identify
randomised controlled trials investigating the effectiveness of different ways of providing
information to patients with hip fracture in improving outcomes.
13.3.1 Evidence
No randomised evidence was identified. However, good quality advice, reassurance,

information and education were highlighted by patients as important to the recovery
process in the qualitative review presented above.
The evidence above suggests that
• The positive attitude of and encouragement by health professionals is important
• Patients value time spent with them, and the advice and explanation given. This seems
important in the recovery process
• Patients should be treated with dignity, and provided with an explanation about their
condition and information about recovery.
Two studies asked participants to suggest what advice they would offer other hip fracture
patients based on their experiences. The main advice by participants in the studies to other
patients with hip fracture was:
• Maintain a positive attittude
• Follow experts advice and participate as much as possible in the rehabilitation process
• Keep mobile

184 HIP FRACTURE
Recommendation Offer patients (or, as appropriate, their carer and/or family)

verbal and printed information about treatment and care
including:
• diagnosis
• choice of anaesthesia
• choice of analgesia and other medications
• surgical procedures
• possible complications
• postoperative care
• rehabilitation programme
• long-term outcomes
• healthcare professionals involved.
Relative values of different Patient views on their satisfaction with the management of their
outcomes condition were the main outcomes.
Trade off between clinical The data highlighted examples where information was not
benefits and harms provided to individual patients. Patients were unhappy when
things were not explained to them. Patients were also unhappy
when issues about their fracture were discussed with their family
members instead of directly to them.
The themes that came out of the evidence suggest that: a positive
attitude of healthcare professionals is important; patients value
time spent with them, and the advice and explanation given; and
patients should be treated with dignity, and provided with an
explanation about their condition and information about recovery.
The GDG were unanimous in their view that discussion with
patients (and where necessary their carers) about all aspects of the
management of their hip fracture in is an important contributory
factor in the recovery process.
Economic considerations Although staff time is a scarce resource, information can be passed
on to patients in the course of usual care and therefore needn’t
increased costs. Furthermore there may be benefits from greater
adherence to treatment plans.
Quality of evidence The qualitative evidence identified was of mixed quality. Data were
not identified covering all the points mentioned above.
Other considerations No comments were identified in the studies mentioning that
adequate or good information was provided. However, the studies
did not specifically ask about the quality of the information
provided.
13.4 Carer involvment

In patients who have been discharged after hip fracture repair, what is the effectiveness of
having a non paid carer (e.g. spouse, relative or friends) on mortality, length of stay, place
of residence/discharge, functional status, hospital readmission and quality of life?
No published evidence was identified. The GDG recognised the often crucial and sometimes
major contribution made by involved relatives and other non-professional carers to
successful rehabilitation. Early discussion with carers of prognosis and discharge planning
avoids misunderstanding of rehabilitation objectives, enables those involved to prepare in
an informed and timely manner for a patient’s return home, consequently averts
inappropriate delay in discharge, and may reduce both length of stay and the likelihood of
inappropriate readmission to hospital.
There is the potential for the delay of some decisions with this approach and it remains
incumbent on clinicians with the agreement of patients (and/or any nominated proxy) to
ensure that their best interests are correctly identified and not compromised, particularly
(but not exclusively) in any urgent decision-making situation.
13.4.1 Clinical evidence
No relevant studies were identified.
No relevant studies were identified.
13.5 Research recommendations
13.5.1 Quality of life

➢ What quality of life value do individual patients and their carers place on different
mobility, independence and residence states following rehabilitation?
It is important in evaluating future priorities for intervention to determine whether the

perceived clinical and health economic benefits of rehabilitation outcomes in the research
literature are matched over the same time-frame by the quality of life judgements,
aspirations and expectations of patients themselves and their carers. There is currently no
evidence.
13.5.2 Patient experience
➢ What is the patient’s experience of being admitted to hospital with a hip fracture in
relation to surgery, pain management, timeliness of information given, and
rehabilitation?
186 HIP FRACTURE
No studies from NHS populations were identified where patients commented specifically on
their surgery, their pain management and rehabilitation programme. There were comments
in the patient views studies about not being kept informed about the management of their
condition, however, there was no information identified about the appropriate time to be
told. It may be that different patients want the information at different times. The studies
suggest that patients suffer from fear, pain and delirium until after surgery and it is
important to learn what (if anything) can be done to alleviate this which for many will be
considered the worst stage in their treatment.
Glossary
Abstract Summary of a study, which may be published alone or as an

introduction to a full scientific paper.
Algorithm (in guidelines) A flow chart of the clinical decision pathway described in the guideline,
where decision points are represented with boxes, linked with arrows.
Allocation concealment The process used to prevent advance knowledge of group assignment
in a RCT. The allocation process should be impervious to any influence
by the individual making the allocation, by being administered by
someone who is not responsible for recruiting participants.
AO classification Classification system used to describe stable trochanteric fractures

(type A1), unstable trochanteric (type A2), and transtrochanteric which
includes those fracture lines at the level of the lesser trochanter and
reversed fracture lines (type A3) 219.
Applicability The degree to which the results of an observation, study or review are
likely to hold true in a particular clinical practice setting.
Arm (of a clinical study) Sub-section of individuals within a study who receive one particular
intervention, for example placebo arm.
Association Statistical relationship between two or more events, characteristics or

other variables. The relationship may or may not be causal.
Baseline The initial set of measurements at the beginning of a study (after run-in
period where applicable), with which subsequent results are compared.
Before-and-after study A study that investigates the effects of an intervention by measuring

particular characteristics of a population both before and after taking
the intervention, and assessing any change that occurs.
Bias Systematic (as opposed to random) deviation of the results of a study

from the ‘true’ results that is caused by the way the study is designed
or conducted.
Blinding Keeping the study participants, caregivers, researchers and outcome

assessors unaware about the interventions to which the participants
have been allocated in a study.
Carer (caregiver) Someone other than a health professional who is involved in caring for
a person with a medical condition.
Case-control study Comparative observational study in which the investigator selects

individuals who have experienced an event (For example, developed a
disease) and others who have not (controls), and then collects data to
determine previous exposure to a possible cause.
Case-series Report of a number of cases of a given disease, usually covering the

course of the disease and the response to treatment. There is no
comparison (control) group of patients.
Clinical efficacy The extent to which an intervention is active when studied under
controlled research conditions.
Clinical effectiveness The extent to which an intervention produces an overall health benefit
in routine clinical practice.
Clinician A healthcare professional providing direct patient care, for example

doctor, nurse or physiotherapist.
Cochrane Review The Cochrane Library consists of a regularly updated collection of

evidence-based medicine databases including the Cochrane Database
of Systematic Reviews (reviews of randomised controlled trials
prepared by the Cochrane Collaboration).
Cohort study A retrospective or prospective follow-up study. Groups of individuals to

be followed up are defined on the basis of presence or absence of
exposure to a suspected risk factor or intervention. A cohort study can
be comparative, in which case two or more groups are selected on the
basis of differences in their exposure to the agent of interest.
Comorbidity Co-existence of more than one disease or an additional disease (other

than that being studied or treated) in an individual.
A local hospital, unit or centre providing an appropriate range and
Community hospital
format of accessible health care facilities and resources. These are
typically small, and provide non-emergency services.
188 HIP FRACTURE
Comparability Similarity of the groups in characteristics likely to affect the study

results (such as health status or age).
Concordance This is a recent term whose meaning has changed. It was initially
applied to the consultation process in which doctor and patient agree
therapeutic decisions that incorporate their respective views, but now
includes patient support in medicine taking as well as prescribing
communication. Concordance reflects social values but does not
address medicine-taking and may not lead to improved adherence.
Confidence interval (CI) A range of values for an unknown population parameter with a stated
‘confidence’ (conventionally 95%) that it contains the true value. The
interval is calculated from sample data, and generally straddles the
sample estimate. The ‘confidence’ value means that if the method used
to calculate the interval is repeated many times, then that proportion
of intervals will actually contain the true value.
Confounding In a study, confounding occurs when the effect of an intervention on an

outcome is distorted as a result of an association between the
population or intervention or outcome and another factor (the
‘confounding variable’) that can influence the outcome independently
of the intervention under study.
Consensus methods Techniques that aim to reach an agreement on a particular issue.

Consensus methods may used when there is a lack of strong evidence
on a particular topic.
Control group A group of patients recruited into a study that receives no treatment, a
treatment of known effect, or a placebo (dummy treatment) - in order
to provide a comparison for a group receiving an experimental
treatment, such as a new drug.
Cost benefit analysis A type of economic evaluation where both costs and benefits of
healthcare treatment are measured in the same monetary units. If
benefits exceed costs, the evaluation would recommend providing the
treatment.
Cost-consequences analysis A type of economic evaluation where various health outcomes are
(CCA) reported in addition to cost for each intervention, but there is no
overall measure of health gain.
Cost-effectiveness analysis An economic study design in which consequences of different

(CEA) interventions are measured using a single outcome, usually in ‘natural’
units (For example, life-years gained, deaths avoided, heart attacks
avoided, cases detected). Alternative interventions are then compared
in terms of cost per unit of effectiveness.
Cost-effectiveness model An explicit mathematical framework, which is used to represent clinical

decision problems and incorporate evidence from a variety of sources
in order to estimate the costs and health outcomes.
Cost-utility analysis (CUA) A form of cost-effectiveness analysis in which the units of effectiveness
are quality-adjusted life-years (QALYs).
Credible Interval The Bayesian equivalent of a confidence interval.
Lag screw cut-out A complication in which the implant may protrude into the surrounding
tissue or penetrate into the acetabulum. Symptoms include increasing
pain and impaired mobility; and treatment depends on the severity of
the symptoms as well as the fitness of the patient to undergo what may
be major revision surgery. It may take the form of re-fixation of the
fracture, replacement arthroplasty, or simple removal of the implant.
Decision analysis An explicit quantitative approach to decision making under uncertainty,

based on evidence from research. This evidence is translated into
probabilities, and then into diagrams or decision trees which direct the
clinician through a succession of possible scenarios, actions and
outcomes.
Discounting Costs and perhaps benefits incurred today have a higher value than
costs and benefits occurring in the future. Discounting health benefits
reflects individual preference for benefits to be experienced in the
present rather than the future. Discounting costs reflects individual
preference for costs to be experienced in the future rather than the
present.
Dominance An intervention is said to be dominated if there is an alternative

intervention that is both less costly and more effective.
Drop-out A participant who withdraws from a trial before the end.
Early Supported Discharge Patients are discharged home from the acute trauma ward, or in some
(ESD) cases a subsequent rehabilitation ward within the hospital, with a
supported 4-6 week rehabilitation package.
Economic evaluation Comparative analysis of alternative health strategies (interventions or

programmes) in terms of both their costs and consequences.
Effect (as in effect measure, The observed association between interventions and outcomes or a
treatment effect, estimate statistic to summarise the strength of the observed association.
of effect, effect size)
Effectiveness See ‘Clinical effectiveness’.
Efficacy See ‘Clinical efficacy’.
Epidemiological study The study of a disease within a population, defining its incidence and
prevalence and examining the roles of external influences (For
example, infection, diet) and interventions.
190 HIP FRACTURE
EQ-5D (EuroQol-5D) A standardise instrument used to measure a health outcome. It

provides a single index value for health status.
Evidence Information on which a decision or guidance is based. Evidence is

obtained from a range of sources including randomised controlled
trials, observational studies, expert opinion (of clinical professionals
and/or patients).
Exclusion criteria (literature Explicit standards used to decide which studies should be excluded
review) from consideration as potential sources of evidence.
Exclusion criteria (clinical Criteria that define who is not eligible to participate in a clinical study.
study)
Extended dominance If Option A is both more clinically effective than Option B and has a
lower cost per unit of effect, when both are compared with a do-
nothing alternative then Option A is said to have extended dominance
over Option B. Option A is therefore more efficient and should be
preferred, other things remaining equal.
Extramedullary implant Implants used to fix extracapsular fractures. Examples of

extramedullary implants include the sliding hip screw and the Medoff
plate.
Extrapolation In data analysis, predicting the value of a parameter outside the range
of observed values.
Follow-up Observation over a period of time of an individual, group or initially

defined population whose appropriate characteristics have been
assessed in order to observe changes in health status or health-related
variables.
Generalisability The extent to which the results of a study based on measurement in a

particular patient population and/or a specific context hold true for
another population and/or in a different context. In this instance, this is
the degree to which the guideline recommendation is applicable across
both geographical and contextual settings. For instance, guidelines that
suggest substituting one form of labour for another should
acknowledge that these costs might vary across the country.
Gold standard See ‘Reference standard’.
Geriatric Orthopaedic A separate geriatrician-led trauma ward. The extent of surgical input to
Rehabilitation Unit (GORU) the GORU varies, depending on how early patients are moved from the
acute trauma wards.
GRADE / GRADE profile A system developed by the GRADE Working Group to address the
shortcomings of present grading systems in healthcare. The GRADE
system uses a common, sensible and transparent approach to grading
the quality of evidence. The results of applying the GRADE system to
clinical trial data are displayed in a table known as a GRADE profile.
Harms Adverse effects of an intervention.
Health economics The study of the allocation of scarce resources among alternative
healthcare treatments. Health economists are concerned with both
increasing the average level of health in the population and improving
the distribution of health.
Health-related quality of A combination of an individual’s physical, mental and social well-being;

life (HRQoL) not merely the absence of disease.
Heterogeneity Or lack of homogeneity. The term is used in meta-analyses and

systematic reviews when the results or estimates of effects of
treatment from separate studies seem to be very different – in terms of
the size of treatment effects or even to the extent that some indicate
beneficial and others suggest adverse treatment effects. Such results
may occur as a result of differences between studies in terms of the
patient populations, outcome measures, definition of variables or
duration of follow-up.
Hip fracture programme Formal 'orthogeriatric' care - with the geriatric medical team
(HFP) contributing to joint preoperative patient assessment, and increasingly
taking the lead in postoperative medical care, MDR and discharge
planning.
Imprecision Results are imprecise when studies include relatively few patients and
few events and thus have wide confidence intervals around the
estimate of effect.
Inclusion criteria (literature Explicit criteria used to decide which studies should be considered as
review) potential sources of evidence.
Incremental analysis The analysis of additional costs and additional clinical outcomes with
different interventions.
Incremental cost The mean cost per patient associated with an intervention minus the
mean cost per patient associated with a comparator intervention.
Incremental cost The difference in the mean costs in the population of interest divided
effectiveness ratio (ICER) by the differences in the mean outcomes in the population of interest
for one treatment compared with another.
(Cost A − Cost B )
ICER =
(Effectiveness A − EffectivenessB )
Incremental net benefit The value (usually in monetary terms) of an intervention net of its cost
(INB) compared with a comparator intervention. The INB can be calculated
for a given cost-effectiveness (willingness to pay) threshold. If the
threshold is £20,000 per QALY gained then the INB is calculated as:
(£20,000 x QALYs gained) – Incremental cost.
192 HIP FRACTURE
Indirectness The available evidence is different to the review question being

addressed, in terms of PICO (population, intervention, comparison and
outcome).
A strategy for analysing data from a randomised controlled trial. All
Intention to treat analysis
participants are included in the arm to which they were allocated,
(ITT)
whether or not they received (or completed) the intervention given to
that arm. Intention-to-treat analysis prevents bias caused by the loss of
participants, which may disrupt the baseline equivalence established by
randomisation and which may reflect non-adherence to the protocol.
Intermediate care Care provided in community hospitals or residential care units as an

intermediate step between hospital care and care in a person’s own
home
Intervention Healthcare action intended to benefit the patient, for example, drug
treatment, surgical procedure, psychological therapy.
Intraoperative The period of time during a surgical procedure.
Intramedullary implant Implants used to fix extracapsular fractures. Examples of intramedullary

implants are the Gamma nail, the intramedullary hip screw and the
proximal femoral nail.
Kappa statistic A statistical measure of inter-rater agreement that takes into account
the agreement occurring by chance.
Length of stay The total number of days a participant stays in hospital.
Licence See ‘Product licence’.
Life-years gained Mean average years of life gained per person as a result of the
intervention compared with an alternative intervention.
Likelihood ratio The likelihood ratio combines information about the sensitivity and
specificity. It tells you how much a positive or negative result changes
the likelihood that a patient would have the disease. The likelihood
ratio of a positive test result (LR+) is sensitivity divided by 1- specificity.
Long-term care Care in a home that may include skilled nursing care and help with
everyday activities. This includes nursing homes and care homes.
Loss to follow-up Also known as attrition. The loss of participants during the course of a
study. Participants that are lost during the study are often call
dropouts.
Markov model A method for estimating long-term costs and effects for recurrent or
chronic conditions, based on health states and the probability of
transition between them within a given time period (cycle).
Meta-analysis A statistical technique for combining (pooling) the results of a number

of studies that address the same question and report on the same
outcomes to produce a summary result. The aim is to derive more
precise and clear information from a large data pool. It is generally
more reliably likely to confirm or refute a hypothesis than the
individual trials.
Mixed Assessment and A rehabilitation unit able to accept patients with a variety of medical,
Rehabilitation Unit (MARU) surgical and orthopaedic conditions.
Mobilisation Mobilisation is the process of re-establishing the ability to move

between postures (for example sit to stand), maintain an upright
posture, and to ambulate with increasing levels of complexity (speed,
changes of direction, dual and multi-tasking).
Multidisciplinary Rehabilitation after hip fracture incorporating the following core

rehabilitation (MDR) components of assessment and management: medicine; nursing;
physiotherapy; occupational therapy; social care. Additional
components may include: dietetics, pharmacy, clinical psychology.
Multivariate model A statistical model for analysis of the relationship between two or more
predictor (independent) variables and the outcome (dependent)
variable.
Negative predictive value [In screening/diagnostic tests:] A measure of the usefulness of a

(NPV) screening/diagnostic test. It is the proportion of those with a negative
test result who do not have the disease, and can be interpreted as the
probability that a negative test result is correct. It is calculated as
follows:
Non-union The terms non-union, pseudarthrosis or delayed union are used for
those fractures that fail to heal after a few months.
Number needed to treat The number of patients that who on average must be treated to
(NNT) prevent a single occurrence of the outcome of interest.
Observational study Retrospective or prospective study in which the investigator observes

the natural course of events with or without control groups; for
example, cohort studies and case–control studies.
Odds ratio A measure of treatment effectiveness. The odds of an event happening

in the treatment group, expressed as a proportion of the odds of it
happening in the control group. The 'odds' is the ratio of events to non-
events.
194 HIP FRACTURE
Opportunity cost The loss of other health care programmes displaced by investment in or
introduction of another intervention. This may be best measured by
the health benefits that could have been achieved had the money been
spent on the next best alternative healthcare intervention.
Orthogeriatrician A care of the elderly physician with an interest in fracture care.
Outcome Measure of the possible results that may stem from exposure to a
preventive or therapeutic intervention. Outcome measures may be
intermediate endpoints or they can be final endpoints. See
‘Intermediate outcome’.
P-value The probability that an observed difference could have occurred by

chance, assuming that there is in fact no underlying difference between
the means of the observations. If the probability is less than 1 in 20, the
P value is less than 0.05; a result with a P value of less than 0.05 is
conventionally considered to be ‘statistically significant’.
Perioperative The period from admission through surgery until discharge,

encompassing the preoperative and postoperative periods.
Placebo An inactive and physically identical medication or procedure used as a

comparator in controlled clinical trials.
Polypharmacy The use or prescription of multiple medications.
Positive predictive value In screening/diagnostic tests: A measure of the usefulness of a

(PPV) screening/diagnostic test. It is the proportion of those with a positive
test result who have the disease, and can be interpreted as the
probability that a positive test result is correct. It is calculated as
follows:
Postoperative Pertaining to the period after patients leave the operating theatre,
following surgery.
Post-test probability For diagnostic tests. The proportion of patients with that particular test
result who have the target disorder (post test odds/[1 + post-test
odds]).
Power (statistical) The ability to demonstrate an association when one exists. Power is
related to sample size; the larger the sample size, the greater the
power and the lower the risk that a possible association could be
missed.
Preoperative The period before surgery commences.

Pre-test probability For diagnostic tests. The proportion of people with the target disorder
in the population at risk at a specific time point or time interval.
Prevalence may depend on how a disorder is diagnosed.
Primary care Healthcare delivered to patients outside hospitals. Primary care covers
a range of services provided by general practitioners, nurses, dentists,
pharmacists, opticians and other healthcare professionals.
Primary outcome The outcome of greatest importance, usually the one in a study that
the power calculation is based on.
Product licence An authorisation from the MHRA to market a medicinal product.
Prognosis A probable course or outcome of a disease. Prognostic factors are

patient or disease characteristics that influence the course. Good
prognosis is associated with low rate of undesirable outcomes; poor
prognosis is associated with a high rate of undesirable outcomes.
Prospective study A study in which people are entered into the research and then
followed up over a period of time with future events recorded as they
happen. This contrasts with studies that are retrospective.
Publication bias Also known as reporting bias. A bias caused by only a subset of all the
relevant data being available. The publication of research can depend
on the nature and direction of the study results. Studies in which an
intervention is not found to be effective are sometimes not published.
Because of this, systematic reviews that fail to include unpublished
studies may overestimate the true effect of an intervention. In
addition, a published report might present a biased set of results (e.g.
only outcomes or sub-groups where a statistically significant difference
was found.
Quality of life See ‘Health-related quality of life’.
Quality-adjusted life year An index of survival that is adjusted to account for the patient’s quality
(QALY) of life during this time. QALYs have the advantage of incorporating
changes in both quantity (longevity/mortality) and quality (morbidity,
psychological, functional, social and other factors) of life. Used to
measure benefits in cost-utility analysis. The QALYs gained are the
mean QALYs associated with one treatment minus the mean QALYs
associated with an alternative treatment.
Quick Reference Guide An abridged version of NICE guidance, which presents the key priorities
for implementation and summarises the recommendations for the core
clinical audience.
Randomisation Allocation of participants in a research study to two or more alternative

groups using a chance procedure, such as computer-generated random
numbers. This approach is used in an attempt to ensure there is an
even distribution of participants with different characteristics between
groups and thus reduce sources of bias.
196 HIP FRACTURE
Randomised controlled trial A comparative study in which participants are randomly allocated to
(RCT) intervention and control groups and followed up to examine
differences in outcomes between the groups.
Residential care unit A unit or centre where care is given outside of the patient's home. Care
can be 24 hour care or partial care depending on the person's needs.
RCT See ‘Randomised controlled trial’.
Receiver operated A graphical method of assessing the accuracy of a diagnostic test.

characteristic (ROC) curve Sensitivity Is plotted against 1-specificity. A perfect test will have a
positive, vertical linear slope starting at the origin. A good test will be
somewhere close to this ideal.
Reference standard The test that is considered to be the best available method to establish
the presence or absence of the outcome – this may not be the one that
is routinely used in practice.
Relative risk (RR) The number of times more likely or less likely an event is to happen in
one group compared with another (calculated as the risk of the event in
group A/the risk of the event in group B).
Reporting bias See publication bias.
Resource implication The likely impact in terms of finance, workforce or other NHS
resources.
Retrospective study A retrospective study deals with the present/ past and does not involve
studying future events. This contrasts with studies that are prospective.
Review question In guideline development, this term refers to the questions about
treatment and care that are formulated to guide the development of
evidence-based recommendations.
Secondary outcome An outcome used to evaluate additional effects of the intervention

deemed a priori as being less important than the primary outcomes.
Selection bias A systematic bias in selecting participants for study groups, so that the
groups have differences in prognosis and/or therapeutic sensitivities at
baseline. Randomisation (with concealed allocation) of patients
protects against this bias.
Sensitivity Sensitivity or recall rate is the proportion of true positives which are
correctly identified as such. For example in diagnostic testing it is the
proportion of true cases that the test detects.
See the related term ‘Specificity’
Sensitivity analysis A means of representing uncertainty in the results of economic

evaluations. Uncertainty may arise from missing data, imprecise
estimates or methodological controversy. Sensitivity analysis also
allows for exploring the generalisability of results to other settings. The
analysis is repeated using different assumptions to examine the effect
on the results.
One-way simple sensitivity analysis (univariate analysis): each
parameter is varied individually in order to isolate the consequences of
each parameter on the results of the study.
Multi-way simple sensitivity analysis (scenario analysis): two or more
parameters are varied at the same time and the overall effect on the
results is evaluated.
Threshold sensitivity analysis: the critical value of parameters above or
below which the conclusions of the study will change are identified.
Probabilistic sensitivity analysis: probability distributions are assigned
to the uncertain parameters and are incorporated into evaluation
models based on decision analytical techniques (For example, Monte
Carlo simulation).
Significance (statistical) A result is deemed statistically significant if the probability of the result
occurring by chance is less than 1 in 20 (p <0.05).
Specificity The proportion of true negatives that a correctly identified as such. For
example in diagnostic testing the specificity is the proportion of non-
cases incorrectly diagnosed as cases.
See related term ‘Sensitivity’.
In terms of literature searching a highly specific search is generally
narrow and aimed at picking up the key papers in a field and avoiding a
wide range of papers.
Stakeholder Those with an interest in the use of the guideline. Stakeholders include
manufacturers, sponsors, healthcare professionals, and patient and
carer groups.
Subtrochanteric Subtrochanteric fractures are those in which the fracture is

extracapsular fracture predominantly in the 5cms of bone immediately distal to the lesser
trochanter.
Superspell Total time in NHS care.

198 HIP FRACTURE
Systematic review Research that summarises the evidence on a clearly formulated

question according to a pre-defined protocol using systematic and
explicit methods to identify, select and appraise relevant studies, and
to extract, collate and report their findings. It may or may not use
statistical meta-analysis.
Time horizon The time span over which costs and health outcomes are considered in
a decision analysis or economic evaluation.
Treatment allocation Assigning a participant to a particular arm of the trial.
Trochanteric extracapsular Extracapsular fractures occur outside or distal to the hip joint capsule
fracture and include basal, trochanteric and subtrochanteric fractures.
Trochanteric fractures may be further subdivided into two part
fractures, which are also termed stable fractures, and those that are
comminuted or multi-fragmentary, which may be termed unstable
fractures.
Univariate Analysis which separately explores each variable in a data set.
The management of hip fracture in

adults
Appendices A – J
APPENDICES
APPENDICES
Produced by the National Clinical Guideline Centre
Appendices
CONTENTS ..................................................................................................................................................... 3
GUIDELINE DEVELOPMENT GROUP MEMBERS ......................................................................................................... 1
NCGC STAFF MEMBERS OF THE GUIDELINE DEVELOPMENT GROUP .............................................................................. 2
EXPERT ADVISORS ........................................................................................................................................... 2
ACKNOWLEDGEMENTS ..................................................................................................................................... 3
ACRONYMS AND ABBREVIATIONS ........................................................................................................................ 4
1 INTRODUCTION ...................................................................................................................................... 6
2 DEVELOPMENT OF THE GUIDELINE ............................................................................................................ 10
2.1 WHAT IS A NICE CLINICAL GUIDELINE? ....................................................................................................... 10
2.2 REMIT .................................................................................................................................................. 11
2.3 WHO DEVELOPED THIS GUIDELINE? ............................................................................................................ 11
2.4 WHAT THIS GUIDELINE COVERS.................................................................................................................. 12
2.5 WHAT THIS GUIDELINE DOES NOT COVER ..................................................................................................... 13
2.6 RELATIONSHIPS BETWEEN THE GUIDELINE AND OTHER NICE GUIDANCE ............................................................. 13
3 METHODS .......................................................................................................................................... 15
3.1 DEVELOPING THE REVIEW QUESTIONS AND OUTCOMES .................................................................................. 15
3.2 SEARCHING FOR EVIDENCE ........................................................................................................................ 20
3.2.1 Clinical literature search .......................................................................................................... 20
3.2.2 Health economic literature search ........................................................................................... 21
3.3 EVIDENCE OF EFFECTIVENESS..................................................................................................................... 21
3.3.1 Inclusion/exclusion ................................................................................................................... 22
3.3.2 Methods of combining clinical studies ..................................................................................... 22
3.3.3 Appraising the quality of evidence by outcomes...................................................................... 23
3.3.4 Grading the quality of clinical evidence ................................................................................... 24
3.3.5 Study limitations ...................................................................................................................... 25
3.3.6 Inconsistency ............................................................................................................................ 25
3.3.7 Indirectness .............................................................................................................................. 26
3.3.8 Imprecision ............................................................................................................................... 26
3.4 EVIDENCE OF COST-EFFECTIVENESS............................................................................................................. 27
200 APPENDICES - CONTENTS
3.4.1 Literature review ...................................................................................................................... 27

3.4.2 NICE economic evidence profiles .............................................................................................. 28
3.4.3 Undertaking new health economic analysis ............................................................................ 29
3.4.4 Cost-effectiveness criteria ........................................................................................................ 30
3.5 DEVELOPING RECOMMENDATIONS ............................................................................................................. 30
3.5.1 Research recommendations ..................................................................................................... 31
3.6 VALIDATION PROCESS .............................................................................................................................. 31
3.7 UPDATING THE GUIDELINE ........................................................................................................................ 31
3.8 DISCLAIMER .......................................................................................................................................... 31
3.9 FUNDING .............................................................................................................................................. 31
4 GUIDELINE SUMMARY ........................................................................................................................... 32
4.1 MAP OF RECOMMENDATIONS ................................................................................................................... 32
AN ALGORITHM WILL BE ADDED BEFORE PUBLICATION.KEY PRIORITIES FOR IMPLEMENTATION ......................................... 32
KEY PRIORITIES FOR IMPLEMENTATION ................................................................................................................. 33
4.2 FULL LIST OF RECOMMENDATIONS.............................................................................................................. 35
4.2.1 Imaging options in occult hip fracture ..................................................................................... 35
4.2.2 Timing of surgery ..................................................................................................................... 35
4.2.3 Analgesia.................................................................................................................................. 35
4.2.4 Anaesthesia .............................................................................................................................. 36
4.2.5 Planning the theatre team ....................................................................................................... 36
4.2.6 Surgical procedures .................................................................................................................. 36
4.2.7 Mobilisation strategies ............................................................................................................ 37
4.2.8 Multidisciplinary management ................................................................................................ 37
4.2.9 Patient and carer information ................................................................................................. 39
4.3 RESEARCH RECOMMENDATIONS ................................................................................................................ 39
4.3.1 Research recommendation on imaging options in occult hip fracture .................................... 39
4.3.2 Research recommendation on anaesthesia ............................................................................. 40
4.3.3 Research recommendation on displaced intracapsular hip fracture ....................................... 40
4.3.4 Research recommendation on early supported discharge ....................................................... 41
4.3.5 Research recommendation on physiotherapy.......................................................................... 41
4.3.6 Additional research recommendations .................................................................................... 42
5 IMAGING OPTIONS IN OCCULT HIP FRACTURE............................................................................................... 45
5.1 INTRODUCTION ...................................................................................................................................... 45
5.2 REVIEW QUESTION .................................................................................................................................. 46
5.3 RADIOGRAPHS ....................................................................................................................................... 46
5.3.1 What is the diagnostic accuracy of additional radiographs (X-Rays) after 48 hours compared
to MRI in the diagnosis of occult hip fractures ....................................................................................... 46
5.4 RADIONUCLIDE BONE SCAN (RNS) ............................................................................................................. 47
5.4.1 What is the diagnostic accuracy of RNS compared to MRI in the diagnosis of occult hip
fractures ................................................................................................................................................. 47
5.4.2 Economic evidence ................................................................................................................... 48
5.4.3 Recommendations and link to evidence................................................................................... 48
5.5 ULTRASOUND (US) ................................................................................................................................. 48
5.5.1 Diagnostic accuracy of ultrasound (US) compared to MRI in the diagnosis of occult hip
fractures ................................................................................................................................................. 49
5.6 COMPUTED TOMOGRAPHY (CT) ................................................................................................................ 50
5.7 RESEARCH RECOMMENDATION ON IMAGING OPTIONS IN OCCULT HIP FRACTURE ................................................. 53
6 TIMING OF SURGERY ............................................................................................................................. 53
6.1 INTRODUCTION ...................................................................................................................................... 53
6.1.1 Review question ....................................................................................................................... 55
6.2 RESEARCH RECOMMENDATIONS ON TIMING OF SURGERY ................................................................................ 68
APPENDICIES - CONTENTS 201
6.2.1 Surgery within 36 hours ........................................................................................................... 68

7 ANALGESIA ......................................................................................................................................... 68
7.1 INTRODUCTION ...................................................................................................................................... 68
7.2 SYSTEMIC ANALGESIA .............................................................................................................................. 69
7.2.1 Review question ....................................................................................................................... 69
7.2.2 Recommendations and link to evidence ................................................................................... 70
7.3 NERVE BLOCKS COMPARED TO SYSTEMIC ANALGESIA ...................................................................................... 70
7.3.1 Review question ....................................................................................................................... 70
7.4 RESEARCH RECOMMENDATIONS ON ANALGESIA ............................................................................................ 81
8 REGIONAL (SPINAL OR EPIDURAL) VERSUS GENERAL ANAESTHESIA ................................................................... 82
8.1 INTRODUCTION ...................................................................................................................................... 82
8.2 REGIONAL VERSUS GENERAL ANAESTHESIA ................................................................................................... 82
8.2.1 Review question ....................................................................................................................... 82
8.3 RESEARCH RECOMMENDATION ON ANAESTHESIA .......................................................................................... 87
9 SURGEON SENIORITY ............................................................................................................................. 88
9.1 INTRODUCTION ...................................................................................................................................... 88
9.2 SURGEON SENIORITY ............................................................................................................................... 89
9.2.1 Review question ....................................................................................................................... 89
9.2.2 Clinical evidence ....................................................................................................................... 89
9.3 RECOMMENDATIONS AND LINK TO EVIDENCE ............................................................................................... 91
10 SURGICAL PROCEDURES ......................................................................................................................... 94
10.1 INTRODUCTION ................................................................................................................................. 94
10.2 SURGERY WITH REGARD TO EARLY MOBILISATION ..................................................................................... 94
10.2.1 Recommendations and link to evidence .............................................................................. 95
10.3 DISPLACED INTRACAPSULAR FRACTURES ................................................................................................. 95
10.3.1 Internal fixation versus hemiarthroplasty ........................................................................... 96
10.3.2 Internal fixation versus total hip replacement .................................................................. 100
10.3.3 Hemiarthroplasty versus total hip replacement ................................................................ 103
10.3.4 Recommendations and link to evidence ............................................................................ 107
10.3.5 Research recommendations on displaced intracapsular fractures ................................... 111
10.4 USE OF CEMENT IN ARTHROPLASTY ...................................................................................................... 112
10.4.1 Use of cement in original Thompson and Austin Moore designs of arthroplasty ............. 112
10.4.2 Use of cement in newer designs of arthroplasty ............................................................... 116
10.5 SURGICAL APPROACH TO HEMIARTHROPLASTY ....................................................................................... 121
10.6 EXTRACAPSULAR FRACTURE FIXATION................................................................................................... 124
10.6.1 Intramedullary versus extramedullary implants for fixation of trochanteric extracapsular
fractures 124
10.6.2 Intramedullary versus extramedullary implants for fixation of reverse oblique trochanteric
extracapsular fractures ........................................................................................................................ 129
10.6.3 Intramedullary versus extramedullary implants for fixation of subtrochanteric
extracapsular fractures ........................................................................................................................ 130
11 MOBILISATION STRATEGIES .................................................................................................................. 133
11.1 INTRODUCTION ............................................................................................................................... 133
11.2 EARLY VS. DELAYED MOBILISATION ...................................................................................................... 134
11.2.1 Review question ................................................................................................................ 134
11.3 INTENSITY OF PHYSIOTHERAPY ............................................................................................................ 137
11.3.1 Review question ................................................................................................................ 137

11.3.3 Research recommendations on mobilisation .................................................................... 143
12 MULTIDISCIPLINARY MANAGEMENT ........................................................................................................ 144
12.1 INTRODUCTION ............................................................................................................................... 144
12.2 HOSPITAL-BASED MULTIDISCIPLINARY REHABILITATION VERSUS USUAL CARE ................................................ 146
12.2.1 Review questions ............................................................................................................... 146
12.2.2 Evidence statement (s) ...................................................................................................... 156
12.3 RESEARCH RECOMMENDATIONS ON HOSPITAL MULTIDISCIPLINARY REHABILITATION ...................................... 163
12.3.1 Hip fracture unit ................................................................................................................ 163
12.4 COMMUNITY-BASED MULTIDISCIPLINARY REHABILITATION VERSUS USUAL CARE ............................................ 165
12.4.1 Review question ................................................................................................................ 165
12.4.2 Economic evidence ............................................................................................................ 168
12.4.3 Evidence statement (s) ...................................................................................................... 170
12.5 RESEARCH RECOMMENDATIONS ON COMMUNITY MULTIDISCIPLINARY REHABILITATION.................................. 175
12.5.1 Early supported discharge ................................................................................................. 175
12.5.2 Care/nursing home residents ............................................................................................ 175
13 PATIENT AND CARER VIEWS AND INFORMATION......................................................................................... 176
13.1 INTRODUCTION ............................................................................................................................... 176
13.2 PATIENT AND CARER VIEWS ................................................................................................................ 176
13.2.1 Summary of studies ........................................................................................................... 177
13.2.2 Common themes ............................................................................................................... 181
13.3 INFORMATION FOR PATIENTS ............................................................................................................. 183
13.3.1 Evidence ............................................................................................................................ 183
13.4 CARER INVOLVMENT......................................................................................................................... 185
13.4.1 Clinical evidence ................................................................................................................ 185
13.4.2 Economic evidence ............................................................................................................ 185
13.5 RESEARCH RECOMMENDATIONS ......................................................................................................... 185
13.5.1 Quality of life ..................................................................................................................... 185
13.5.2 Patient experience ............................................................................................................. 185
GLOSSARY ................................................................................................................................................. 186
APPENDICES ............................................................................................................................................... 199
APPENDIX A: SCOPE .................................................................................................................................... 206
13.6 GUIDELINE TITLE .............................................................................................................................. 206
13.6.1 Short title ........................................................................................................................... 206
13.7 THE REMIT ..................................................................................................................................... 206
13.8 CLINICAL NEED FOR THE GUIDELINE...................................................................................................... 206
13.8.1 Epidemiology ..................................................................................................................... 206
13.8.2 Current practice................................................................................................................. 207
13.9 THE GUIDELINE ................................................................................................................................ 209
13.9.1 Population ......................................................................................................................... 209
13.9.2 Healthcare setting ............................................................................................................. 210
13.9.3 Clinical management ........................................................................................................ 210
13.9.4 Main outcomes.................................................................................................................. 211
13.9.5 Economic aspects .............................................................................................................. 212
13.9.6 Status ................................................................................................................................ 212
13.10 RELATED NICE GUIDANCE ................................................................................................................. 212
13.10.1 Published ........................................................................................................................... 212
13.10.2 Guidance under development ........................................................................................... 213
13.11 FURTHER INFORMATION .................................................................................................................... 213
14 APPENDIX B: DECLARATIONS OF INTEREST ............................................................................................... 214

14.1 INTRODUCTION ............................................................................................................................... 214
14.2 DECLARATIONS OF INTERESTS OF THE GDG MEMBERS ............................................................................. 215
14.2.1 Professor Cameron Swift ................................................................................................... 215
14.2.2 Professor Opinder Sahota ................................................................................................. 216
14.2.3 Dr Antony Johansen........................................................................................................... 217
14.2.4 Mr Tim Chesser .................................................................................................................. 218
14.2.5 Mr Bob Handley ................................................................................................................. 219
14.2.6 Ms Karen Hertz .................................................................................................................. 220
14.2.7 Dr Richard Griffiths ............................................................................................................ 221
14.2.8 Professor Sallie Lamb ........................................................................................................ 222
14.2.9 Mrs Heather Towndrow .................................................................................................... 223
14.2.10 Dr Sally Hope ..................................................................................................................... 224
14.2.11 Ms Tessa Somerville .......................................................................................................... 225
14.2.12 Mr Anthony Field ............................................................................................................... 226
14.2.13 Mr Martin Wise ................................................................................................................. 227
14.3 DECLARATIONS OF INTERESTS OF THE NCGC MEMBERS ........................................................................... 228
14.4 DECLARATIONS OF INTERESTS OF THE EXPERT ADVISORS .......................................................................... 229
14.4.1 Mr Martin Parker .............................................................................................................. 229
14.4.2 Mrs Pamela Holmes .......................................................................................................... 229
14.4.3 Professor Judith Adams ..................................................................................................... 229
15 APPENDIX C: REVIEW PROTOCOLS.......................................................................................................... 229
15.1 REVIEW PROTOCOL – IMAGING IN OCCULT HIP FRACTURE......................................................................... 229
15.2 REVIEW PROTOCOL – TIMING OF SURGERY ............................................................................................ 231
15.3 REVIEW PROTOCOL – ANALGESIA- SYSTEMIC MEDICATIONS ...................................................................... 233
15.4 REVIEW PROTOCOL – ANALGESIA- NERVE BLOCKS COMAPARED TO SYSTEMIC ANALGESICS .............................. 235
15.5 REVIEW PROTOCOL - ANAESTHESIA ..................................................................................................... 237
15.6 REVIEW PROTOCOL – SURGEON SENIORITY ............................................................................................ 239
15.7 REVIEW PROTOCOL – CEMENT............................................................................................................ 240
15.8 REVIEW PROTOCOL – INTRACAPSULAR FRACTURES.................................................................................. 242
15.9 REVIEW PROTOCOL – SURGICAL APPROACH ........................................................................................... 244
15.10 REVIEW PROTOCOL – HEMIARTHROPLASTY STEM DESIGN ......................................................................... 246
15.11 REVIEW PROTOCOL – EXTRACAPSULAR FRACTURES ................................................................................. 248
15.12 REVIEW PROTOCOL – MOBILISATION STRATEGIES ................................................................................... 250
15.13 REVIEW PROTOCOL – MULTIDISCIPLINARY REHABILITATION ...................................................................... 252
15.14 REVIEW PROTOCOL – CARER INVOLVEMENT .......................................................................................... 258
15.15 REVIEW PROTOCOL – HEALTH ECONOMICS ........................................................................................... 260
16 APPENDIX D: LITERATURE SEARCH STRATEGIES.......................................................................................... 262
16.1 SEARCH STRATEGIES ......................................................................................................................... 262
16.2 SEARCH TERMS ................................................................................................................................ 264
Anaesthesia .......................................................................................................................................... 264
Analgesia .............................................................................................................................................. 265
Animal/publication filter ...................................................................................................................... 266
Carer involvement ................................................................................................................................ 266
Diagnostic filter .................................................................................................................................... 267
Early Surgery ........................................................................................................................................ 268
Economic .............................................................................................................................................. 268
Hip Fracture Terms ............................................................................................................................... 270
Orthogeriatrician .................................................................................................................................. 271
Patient education ................................................................................................................................. 272
Patient views ........................................................................................................................................ 273
Radiological Imaging ............................................................................................................................ 274
RCT filter ............................................................................................................................................... 275
Rehabilitation ....................................................................................................................................... 276
Surgeon seniority .................................................................................................................................. 282

Surgical Interventions ........................................................................................................................... 282
Systematic review filter ........................................................................................................................ 283
17 APPENDIX E: EVIDENCE TABLES - CLINICAL STUDIES .................................................................................... 284
17.1 EVIDENCE TABLE 1: IMAGING OPTIONS IN OCCULT HIP FRACTURE ............................................................. 286
17.2 EVIDENCE TABLE 2: TIMING OF SURGERY ............................................................................................. 289
17.3 EVIDENCE TABLE 3: OPTIMAL ANALGESIA ............................................................................................ 307
17.4 EVIDENCE TABLE 4: ANAESTHESIA ...................................................................................................... 310
17.5 EVIDENCE TABLE 5: SURGEON SENIORITY ............................................................................................. 313
17.6 EVIDENCE TABLE 6: DISPLACED INTRACAPSULAR FRACTURES .................................................................... 317
17.7 EVIDENCE TABLE 7: SURGERY – CEMENT VERSUS NO CEMENT.................................................................. 326
17.8 EVIDENCE TABLE 8: EXTRACAPSULAR FIXATION ..................................................................................... 328
17.9 EVIDENCE TABLE 9: SURGICAL APPROACH TO HEMIARTHROPLASTY............................................................ 376
17.10 EVIDENCE TABLE 10: MOBILISATION STRATEGIES .................................................................................. 378
17.12 EVIDENCE TABLE 12: PATIENT VIEWS .................................................................................................. 411
18 APPENDIX F: EVIDENCE TABLES - ECONOMIC STUDIES ................................................................................. 427
18.1 EVIDENCE TABLE 13: GENERAL VERSUS REGIONAL ANAESTHESIA .............................................................. 428
18.2 EVIDENCE TABLE 14: DISPLACED INTRACAPSULAR FRACTURES.................................................................. 430
18.3 EVIDENCE TABLE 15: CEMENTED ARTHROPLASTIES ................................................................................ 434
19 APPENDIX G: FOREST PLOTS ................................................................................................................. 449
19.1 RADIOLOGY .................................................................................................................................... 452
19.2 TIMING OF SURGERY......................................................................................................................... 453
19.3 ANALGESIA ..................................................................................................................................... 459
19.4 ANAESTHESIA.................................................................................................................................. 465
19.5 SURGICAL INTERVENTIONS ................................................................................................................. 471
19.5.1 Surgeon seniority............................................................................................................... 471
19.5.2 Cement in older designs of arthroplasty ........................................................................... 472
19.5.3 Cement in newer designs of arthroplasty.......................................................................... 480
19.5.4 Internal fixation versus hemiarthroplasty ......................................................................... 482
19.5.5 Internal fixation versus total hip replacement .................................................................. 487
19.5.6 Hemiarthroplasty versus total hip replacement ................................................................ 490
19.5.7 Trochanteric extracapsular fracture – all studies .............................................................. 495
19.5.8 Trochanteric extracapsular fracture – studies from 2000 ................................................. 503
19.5.9 Subtrochanteric extracapsular fracture. ........................................................................... 508
19.6 MOBILISATION STRATEGIES ................................................................................................................ 510
19.6.1 Timing of mobilisation ....................................................................................................... 510
19.7 INTENSIVE EXERCISE OR PHYSIOTHERAPY VS. USUAL CARE ......................................................................... 512
19.7.1 Intensive physiotherapy (Strength training) ...................................................................... 512
19.7.2 Intensive physiotherapy (treadmill training) ..................................................................... 514
19.7.3 Intensive (more frequent) physiotherapy .......................................................................... 516
19.8 MULTIDISCIPLINARY REHABILITATION ................................................................................................... 517
19.8.1 Hospital-based MDR.......................................................................................................... 517
19.9 HOME-BASED MDR VERSUS USUAL INPATIENT REHABILITATION ................................................................ 522
20 APPENDIX H: HEALTH ECONOMIC ANALYSIS.............................................................................................. 524
20.1 COST ANALYSIS OF NERVE BLOCKS, NON-OPIOID ANALGESICS AND OPIOID ANALGESICS ................................... 524
20.1.1 Nerve block cost analysis .................................................................................................. 524
20.1.2 Non-opioid analgesics ....................................................................................................... 526
20.1.3 Opioid analgesics .............................................................................................................. 526
20.2 HOURLY WAGE COSTS FOR A PLANNED TRAUMA LIST ............................................................................... 527
20.3 PRICES FOR SLIDING HIP SCREWS AND SHORT AND LONG INTRAMEDULLARY NAILS ......................................... 530
20.4 COST ANALYSIS OF THE INTERVENTIONS FOR INTENSIVE MOBILISATION STRATEGIES. ...................................... 532
20.5 COST-EFFECTIVENESS ANALYSIS OF HOSPITAL INVESTMENT VERSUS NO HOSPITAL INVESTMENT FOR EARLY SURGERY
535
20.5.1 Introduction ....................................................................................................................... 535
20.5.2 Background ....................................................................................................................... 535
20.5.3 Population and time horizon ............................................................................................. 536
20.5.4 Software ............................................................................................................................ 536
20.5.5 Methods ............................................................................................................................ 536
20.5.6 Treatment effects .............................................................................................................. 537
20.5.7 Quality of life ..................................................................................................................... 539
20.5.8 Cost analysis ...................................................................................................................... 540
20.5.9 Cost-effectiveness analysis ................................................................................................ 544
20.5.10 Discussion .......................................................................................................................... 550
20.6 COST-EFFECTIVENESS ANALYSIS OF HOSPITAL MDR VS USUAL CARE .......................................................... 552
20.6.1 Introduction ....................................................................................................................... 552
20.6.3 Software ............................................................................................................................ 553
20.6.4 Structure of the model....................................................................................................... 553
20.6.5 Evidence and treatment effects on mortality .................................................................... 559
20.6.6 Utilities data ...................................................................................................................... 560
20.6.7 Calculating QALYs gained .................................................................................................. 563
20.6.8 Cost data ........................................................................................................................... 564
20.6.9 Cost-effectiveness findings for base-case analysis ............................................................ 574
20.6.10 Discussion .......................................................................................................................... 585
20.7 COST-EFFECTIVENESS ANALYSIS OF COMMUNITY MDR VS USUAL CARE ...................................................... 586
20.7.1 Introduction ....................................................................................................................... 586
20.7.3 Software ............................................................................................................................ 586
20.7.4 Economic evaluation type ................................................................................................. 586
20.7.5 Time horizon, Perspective, Discount rates used ................................................................ 587
20.7.6 Structure of the model....................................................................................................... 587
20.7.7 Utility data......................................................................................................................... 587
20.7.8 Mortality ........................................................................................................................... 588
20.7.9 Calculating QALYs gained .................................................................................................. 588
20.7.10 Cost analysis ...................................................................................................................... 589
20.7.11 Cost effectiveness findings ................................................................................................ 591
20.7.12 Discussion .......................................................................................................................... 598
20.8 COST ANALYSIS OF CEMENTED VS. UNCEMENTED IMPLANTS (NEWER DESIGNS OF ARTHROPLASTY) ......................... 598
a) Cost of implants .......................................................................................................................... 598
c) Cost of length of stay in hospital ................................................................................................ 600
d) Re-operation costs ...................................................................................................................... 601
e) Theatre time costs ...................................................................................................................... 601
Summary of costs components............................................................................................................. 602
21 APPENDIX I: HIGH PRIORITY RESEARCH RECOMMENDATIONS....................................................................... 603
21.1 IMAGING OPTIONS IN OCCULT HIP FRACTURE ......................................................................................... 603
21.2 ANAESTHESIA .................................................................................................................................. 606
21.3 DISPLACED INTRACAPSULAR HIP FRACTURES .......................................................................................... 609
21.4 INTENSIVE REHABILITATION THERAPIES AFTER HIP FRACTURE ..................................................................... 613
21.5 EARLY SUPPORTED DISCHARGE IN CARE HOME PATIENTS......................................................................... 616
22 APPENDIX J: EXCLUDED STUDIES ............................................................................................................ 619
BIBLIOGRAPHY............................................................................................................................................ 628
206 APPENDIX A
Appendix A: Scope
13.6 Guideline title

Hip fracture: the management of hip fracture in adults
13.6.1 Short title
Hip fracture
13.7 The remit

The Department of Health has asked NICE: ”to prepare a clinical guideline on the
management of fractured neck of femur”.
13.8 Clinical need for the guideline
13.8.1 Epidemiology
a) About 70–75,000 hip fractures (proximal femoral fractures) occur annually in

the UK. Hip fracture is the commonest reason for admission to an orthopaedic
ward, and is usually a ‘fragility’ fracture 2 caused by a fall affecting an older
person with osteoporosis or osteopaenia (a lesser degree of bone reduction
and weakness due to the same process as in osteoporosis). The average age of
a person with hip fracture is 77 years. The annual cost of medical and social
2 The strict definition of a fragility fracture is one caused by a fall from standing height or less. For
the purposes of this guidance, the definition will be slightly more flexible to encompass all hip
fractures judged to have an osteoporotic or osteopaenic basis
APPENDIX A 207
care for all the hip fracture cases in the UK amounts to about £2 billion.
Demographic projections indicate that the UK annual incidence will rise to
91,500 by 2015 and 101,000 in 2020, with an associated increase in annual
expenditure that could reach £2.2 billion by 2020. The majority of this
expenditure will be accounted for by hospital bed days and a further
substantial contribution will come from health and social aftercare. About a
quarter of patients with hip fracture are admitted from institutional care.
About 10–20% of those admitted from home ultimately move to institutional
care.
b) Mortality is high – about 10% of people with a hip fracture die within
1 month, and about one third within 12 months. However, fewer than half of
deaths are attributable to the fracture. This reflects the high prevalence of
comorbidity in people with hip fractures; often the combination of fall and
fracture brings to light underlying ill health. This presents major challenges for
anaesthetic, surgical, postoperative and rehabilitative care.
13.8.2 Current practice
a) The primary and secondary prevention of fragility fractures by treating

osteoporosis and reducing the risk of falls are of key importance to the
current and future epidemiology of hip fracture. These are, or will be, covered
by related NICE guidance (see section 5).
b) The diagnosis and management of hip fracture itself and of any comorbidity
before, during and after surgery, have a profound effect on outcome, both for
individuals and for services.
c) Patients with hip fracture need immediate referral to hospital (other than in
exceptional circumstances). Their assessment and management on admission
commonly involve a range of specialties and disciplines, but it is not always
clear how and when this involvement should take place. Prompt surgery is
important but is sometimes delayed for administrative or clinical reasons. It is
208 APPENDIX A
essential that mobilisation and rehabilitation after surgery are undertaken

according to individual need, but this does not always happen.
d) In spite of a significant body of evidence, hip fracture management and the

resulting length of hospital stay vary markedly among centres across England
and Wales.
e) Existing UK guidance from other sources includes:
• Scottish Intercollegiate Guidelines Network (2002) Prevention and

management of hip fracture in older people. Available from
www.sign.ac.uk/guidelines/fulltext/56/index.html
• British Orthopaedic Association (2007) The care of patients with fragility
fracture. Available from www.nhfd.co.uk
• Department of Health (2001) National service framework for older people3.
Available from www.dh.gov.uk
f) This clinical guideline will provide guidance on the emergency, preoperative,

operative and postoperative management of hip fracture, including
rehabilitation, in adults. It will not cover those aspects of hip fracture
addressed by related NICE guidance, but will refer to them as appropriate.
g) At all stages of hip fracture management, and especially during rehabilitation,

the importance of optimal communication with, and support for, patients
themselves and those who provide or will provide care – including unpaid
care family members or others – will be a fundamental tenet of guidance
development.
3Elaborates on relevant (but not specific) standards of contextual importance (intermediate care,
general hospital care and falls).
APPENDIX A 209
13.9 The guideline

The guideline development process is described in detail on the NICE website (see section
6, ‘Further information’).
This scope defines what the guideline will (and will not) examine, and what the guideline
developers will consider. The scope is based on the referral from the Department of
Health.
The areas that will be addressed by the guideline are described in the following sections.
13.9.1 Population
13.9.1.1 Groups that will be covered
a) Adults aged 18 years and older presenting to the health service with a clinical
diagnosis (firm or provisional) of fragility fracture of the hip.
b) People with the following types of hip fracture:4
• intracapsular (undisplaced and displaced)

• extracapsular (trochanteric and subtrochanteric).
c) Those with comorbidity strongly predictive of outcome, and those without

such comorbidity. The influence (if any) of advanced age or gender on clinical
decision-making, management and outcome will be specifically evaluated.
13.9.1.2 Groups that will not be

covered
People younger than 18 years.
People with fractures caused by specific pathologies other than osteoporosis or

osteopaenia (because these would require more condition-specific guidance).
4 These terms explain where the bone has fractured, which can be either near or within the hip
joint.
210 APPENDIX A
13.9.2 Healthcare setting
a) Secondary care settings where preoperative, operative, and postoperative

acute and subacute care are undertaken.
b) Primary, secondary and social care settings, as well as an individual’s own

home, where rehabilitation is undertaken.
13.9.3 Clinical management
13.9.3.1 Key clinical issues that will be

covered
a) Using alternative radiological imaging to confirm or exclude a suspected hip

fracture in patients with a normal X-ray.
b) Involving a physician or orthogeriatrician in the care of patients presenting

with hip fracture.
c) Early surgery (within 48 hours).
d) Optimal preoperative and postoperative analgesia (pain relief), including the

use of nerve blockade.
e) Regional (spinal – also known as ‘epidural’) versus general anaesthesia in

patients undergoing surgery for hip fracture.
f) Does surgeon experience reduce the incidence of mortality, the need for
repeat surgery, and poor outcome in terms of mobility?
g) For displaced intracapsular fracture:
• internal fixation versus arthroplasty (hip replacement surgery)

• total hip replacement versus hemiarthroplasty (replacing the head of the
femur only) .
h) Choice of surgical implants - Sliding hip screw versus intramedullary nail for
trochanteric extracapsular fracture.
APPENDIX A 211
i) Choice of surgical implants - Sliding hip screw versus intramedullary nail for
subtrochanteric extracapsular fracture.
j) Cemented versus non-cemented arthroplasty implants.
k) Hospital-based multidisciplinary rehabilitation for patients who have

undergone hip fracture surgery.
l) Early transfer to community-based multidisciplinary rehabilitation for patients

who have undergone hip fracture surgery.
13.9.3.2 Clinical issues that will not be

covered
The following will not be directly covered in this guideline, but related NICE guidance will
be referred to if appropriate:
a) Primary and secondary prevention of fragility fracture.
b) Prevention and management of pressure sores.
c) Prophylaxis for venous thromboembolism.
d) Prevention and management of infection at the surgical site.
e) Nutritional support.
f) Selection of prostheses for hip replacement.
g) Complementary and alternative therapies.
13.9.4 Main outcomes
a) Requirement for surgical revision.
b) Short-term and long-term mortality.
c) Length of stay in secondary care.
d) Length of time before community resettlement/discharge.

212 APPENDIX A
e) Place of residence (compared with baseline) 12 months after fracture.
f) Short-, medium- and long-term functional status.
g) Short-, medium- and long-term quality of life.
13.9.5 Economic aspects
Developers will take into account both clinical and cost effectiveness when making
recommendations involving a choice between alternative interventions. A review of the
economic evidence will be conducted and analyses will be carried out as appropriate. The
preferred unit of effectiveness is the quality-adjusted life year (QALY), and the costs
considered will usually be only from an NHS and personal social services (PSS)
perspective. Further detail on the methods can be found in 'The guidelines manual' (see
‘Further information’).
13.9.6 Status
13.9.6.1 Scope
This is the final scope.
13.9.6.2 Timing
The development of the guideline recommendations will begin in June 2010.
13.10 Related NICE guidance
13.10.1 Published
• Surgical site infection. NICE clinical guideline 74 (2008). Available from

• Alendronate, etidronate, risedronate, raloxifene, strontium ranelate and teriparatide
for the secondary prevention of osteoporotic fragility fractures in postmenopausal
women. NICE technology appraisal guidance 161 (2008). Available from
www.nice.org.uk/TA161
APPENDIX A 213
• Alendronate, etidronate, risedronate, raloxifene and strontium ranelate for the

primary prevention of osteoporotic fragility fractures in postmenopausal women. NICE
technology appraisal guidance 160 (2008). Available from www.nice.org.uk/TA160
• Venous thromboembolism. NICE clinical guideline 46 (2007). Available from
• Delirium: diagnosis, prevention and management of delirium. NICE clinical guideline
103 (2010). Available from www.nice.org.uk/guidance/CG103
• Venous thromboembolism –prevention. NICE clinical guideline 92 (2010). Available
from www.nice.org.uk/guidance/CG92Minimally invasive hip replacement. NICE
interventional procedure guidance (2010). Available from
www.nice.org.uk/guidance/IPG363
• Nutrition support in adults. NICE clinical guideline 32 (2006). www.nice.org.uk/CG32
• The management of pressure ulcers in primary and secondary care. NICE clinical
guideline 29 (2005). Available from www.nice.org.uk/CG29
• Falls. NICE clinical guideline 21 (2004). Available from www.nice.org.uk/CG21
• Preoperative tests. NICE clinical guideline 3 (2003). Available from
www.nice.org.uk/CG3
• The selection of prostheses for primary total hip replacement. NICE technology
appraisal guidance 2 (2000). Available from www.nice.org.uk/TA2
13.10.2 Guidance under development
NICE is currently developing the following related guidance (details available from the
NICE website).
• Osteoporosis. NICE clinical guideline. Publication date to be confirmed.
13.11 Further information

Information on the guideline development process is provided in:
• ‘How NICE clinical guidelines are developed: an overview for stakeholders, the public
and the NHS’
• ‘The guidelines manual’.
214 APPENDIX A
These are available from the NICE website (www.nice.org.uk/guidelinesmanual).

Information on the progress of the guideline will also be available from the NICE website
(www.nice.org.uk).
14 Appendix B: Declarations of Interest
14.1 Introduction
All members of the GDG and all members of the NCGC staff were required to make
formal declarations of interest at the outset of each meeting, and these were
updated at every subsequent meeting throughout the development process. No
interests were declared that required actions.
APPENDIX B 215
14.2 Declarations of interests of the GDG members
14.2.1 Professor Cameron Swift
GDG meeting Declaration of Interests

First GDG meeting No interests to declare
(1st July 2009)
Second GDG Meeting No interests to declare
(17th July 2009)
Third GDG Meeting No interests to declare
(15th September 2009)
Fourth GDG Meeting No interests to declare
(8th December 2009)
Fifth GDG Meeting No interests to declare
(14th December 2009)
Sixth GDG Meeting No interests to declare
(Subgroup meeting)
(18th January 2010)
Seventh GDG Meeting No interests to declare
(9th March 2010)
Eighth GDG Meeting No interests to declare
(26th April 2010)
Tenth GDG Meeting No interests to declare
(11th June 2010)
Eleventh GDG Meeting Declared a non personal non pecuniary interest: has been invited to join
(30th June 2010) the Department of Health Board on Fragility Fractures Programme
Twelfth GDG Meeting No change
(29th July 2010)
Thirteenth GDG Meeting No change
Fourteenth GDG Meeting No change
(18th January 2011)
Fifteenth GDC Meeting No change
(24th March 2011)
Actions None required
216 APPENDIX B
14.2.2 Professor Opinder Sahota

(1st July 2009)
Second GDG Meeting Did not attend
(17th July 2009)
(8th December 2009)
(Subgroup workshop)
(18th January 2010)
Seventh GDG Meeting Declared a non personal, non pecuniary interest regarding involvement in
(9th March 2010) the Map of Medicine project with the Department of Health.
Eighth GDG Meeting No change
(26th April 2010)
Tenth GDG Meeting Did not attend
11th June 2010)
Eleventh GDG Meeting Declared a non personal non pecuniary interest: has been invited to join
(30th June 2010) the Department of Health Board on Fragility Fractures Programme
(29th July 2010)
(18th January 2011)
Fifteenth GDC Meeting Did not attend
(24th March 2011)
APPENDIX B 217
14.2.3 Dr Antony Johansen

(1st July 2009)
(17th July 2009)
(8th December 2009)
(Subgroup Workshop)
(18th January 2010)
(9th March 2010)
(26th April 2010)
(11th June 2010)
Eleventh GDG Meeting No interests to declare
(30th June 2010)
Twelfth GDG Meeting No interests to declare
(29th July 2010)
Thirteenth GDG Meeting No interests to declare
Fourteenth GDG Meeting No interests to declare
(18th January 2011)
Fifteenth GDC Meeting No interests to declare
(24th March 2011)
218 APPENDIX B
14.2.4 Mr Tim Chesser

First GDG meeting Performed consultancy work with orthopaedic manufacturer for unrelated
(1st July 2009) orthopaedic implants (locking plates for particular fractures)- compliance
and worded guidelines. His Department receives research support from
orthopaedic manufacturers including DePuy, Smith and Nephew, Biomet
and Stryker. Department have research fellows funded by orthopaedic
manufacturer. Publishing RCT on surgical treatment for peri-articular
fractures which was not funded by industry.
(17th July 2009)
Third GDG Meeting Performed consultancy work with orthopaedic manufacturer for unrelated
(15th September 2009) orthopaedic implants (locking plates for particular fractures)- compliance
and worded guidelines. His Department receives research support from
orthopaedic manufacturers including DePuy, Smith and Nephew, Biomet
and Stryker. Department have research fellows funded by orthopaedic
manufacturer. Publishing RCT on surgical treatment for peri-articular
fractures which was not funded by industry.
Fourth GDG Meeting Travel and accommodation funded by the Orthopaedic Trauma
(8th December 2009) Association in the US to present a poster on outcomes in Pelvic Fractures
at an Experts in Pelvic Trauma meeting (sponsored by Stryker Trauma).
Fifth GDG Meeting No change

Sixth GDG Meeting Did not attend
(Subgroup workshop)
(18th January 2010)
Seventh GDG Meeting Did not attend
(9th March 2010)
(26th April 2010)
Tenth GDG Meeting No change
(11th June 2010)
Eleventh GDG Meeting Did not attend meeting
(30th June 2010)
Twelfth GDG Meeting Declared a personal non pecuniary interest- invited to teach on a hip
(29th July 2010) fracture surgical techniques course organised by Stryker who paid his
travel expenses. No other payment was received.
Fourteenth GDG Meeting Declared that he has a contract with an orthopaedic company (Stryker) to
(18th January 2011) design reduction clamps, instrumentation and update for pelvic ring and
acetabular fractures. He was also invited to be on the NHS Map of
Medicine Commissioners’ toolkit.
(24th March 2011)
APPENDIX B 219
14.2.5 Mr Bob Handley

First GDG meeting Non personal pecuniary interest: responsibility for – Synthes Fellows in the
(1st July 2009) Trauma Department at the John Radcliffe hospital- 2 week fellowships
usually 3-4 per year.
Second GDG Meeting No change
(17th July 2009)
Third GDG Meeting No change
Fourth GDG Meeting No change
(8th December 2009)
(Subgroup workshop)
(18th January 2010)
Seventh GDG Meeting No change
(9th March 2010)
(26th April 2010)
(11th June 2010)
(30th June 2010)
(29th July 2010)
(18th January 2011)
(24th March 2011)
220 APPENDIX B
14.2.6 Ms Karen Hertz

(1st July 2009)
(17th July 2009)
Fourth GDG Meeting Miss Karen Hertz- funding for flights and accommodation by a Chinese
(8th December 2009) university to attend a conference in Hong Kong.

Sixth GDG Meeting No change
(Subgroup workshop)
(18th January 2010)
Seventh GDG Meeting KH declared a non personal, non pecuniary interest regarding involvement
(9th March 2010) in the Map of Medicine project with the Department of Health.
(26th April 2010)
(11th June 2010)
Eleventh GDG Meeting No change
(30th June 2010)
Twelfth GDG Meeting Declared a non personal non pecuniary interest: has been invited to join
(29th July 2010) the Department of Health Board on Fragility Fractures Programme
(18th January 2011)
(24th March 2011)
APPENDIX B 221
14.2.7 Dr Richard Griffiths

(1st July 2009)
(17th July 2009)
Third GDG Meeting Did not attend
Fourth GDG Meeting Did not attend
(8th December 2009)
(Subgroup workshop)
(18th January 2010)
(9th March 2010)
(26th April 2010)
(11th June 2010)
(30th June 2010)
Twelfth GDG Meeting Did not attend
(29th July 2010)
Thirteenth GDG Meeting Did not attend
Fourteenth GDG Meeting Did not attend
(18th January 2011)
(24th March 2011)
222 APPENDIX B
14.2.8 Professor Sallie Lamb

First GDG meeting Did not attend
(1st July 2009)
Second GDG Meeting Declared a non personal pecuniary interest: NIHR funded research grant.
(17th July 2009) One trial is in the final stages of finding approval in primary care- using
peripheral fracture (including hip fracture). The second- potential trial-
ideas unclear as to whether they will be submitted. Vitamin D in Hip
fracture; anaemia in hip fracture.
Third GDG Meeting No change

(8th December 2009)
(Subgroup workshop)
(28th January 2010)
(9th March 2010)
(26th April 2010)
(11th June 2010)
(30th June 2010)
Twelfth GDG Meeting Did not attend
(29th July 2010)
(18th January 2011)
(24th March 2011)
APPENDIX B 223
14.2.9 Mrs Heather Towndrow

(1st July 2009)
(17th July 2009)
(8th December 2009)
(Subgroup workshop)
(18th January 2010)
(9th March 2010)
(26th April 2010)
(11th June 2010)
(30th June 2010)
(29th July 2010)
(18th January 2011)
(24th March 2011)
224 APPENDIX B
14.2.10 Dr Sally Hope

First GDG meeting Did not attend
(1st July 2009)
Second GDG Meeting Declared a personal pecuniary interest- MSD paid for hotel in Manchester
(17th July 2009) for NOS Conference (approx £200) in July 2009: in accordance with NOS
policy to reduce costs for speakers.
Third GDG Meeting Did not attend

(8th December 2009)
Sixth GDG Meeting Did not attend meeting
(Subgroup workshop)
(28th January 2010)
Seventh GDG Meeting Declared a non personal, non pecuniary interest regarding involvement in
(9th March 2010) the Map of Medicine project with the Department of Health.
Eighth GDG Meeting Did not attend
(26th April 2010)
(11th June 2010)
(30th June 2010)
(29th July 2010)
(18th January 2011)
(24th March 2011)
APPENDIX B 225
14.2.11 Ms Tessa Somerville

(1st July 2009)
(17th July 2009)
(8th December 2009)
Sixth GDG Meeting Did not attend meeting
(Subgroup workshop)
(18th January 2010)
(9th March 2010)
(26th April 2010)
(11th June 2010)
(30th June 2010)
(29th July 2010)
(18th January 2011)
(24th March 2011)
226 APPENDIX B
14.2.12 Mr Anthony Field

(1st July 2009)
(17th July 2009)
Fourth GDG Meeting Did not attend
(8th December 2009)
Fifth GDG Meeting Did not attend
(Subgroup workshop)
(18th January 2010)
(9th March 2010)
(26th April 2010)
(11th June 2010)
(30th June 2010)
(29th July 2010)
(18th January 2011)
(24th March 2011)
APPENDIX B 227
14.2.13 Mr Martin Wise

(1st July 2009)
(17th July 2009)
(8th December 2009)
(Subgroup workshop)
(18th January 2010)
(9th March 2010)
Eighth GDG Meeting Did not attend
(26th April 2010)
(11th June 2010)
(30th June 2010)
(29th July 2010)
(18th January 2011)
(24th March 2011)
228 APPENDIX B
14.3 Declarations of interests of the NCGC members

GDG meeting Declaration of Interests of the NCGC members
(1st July 2009)
(17th July 2009)
(8th December 2009)
Fifth GDG Meeting Antonia Morga declared her husband works for Novartis
Sixth GDG Meeting No change
(Subgroup workshop)
(18th January 2010)
Seventh GDG Meeting No change
(9th March 2010)
(26th April 2010)
(11th June 2010)
(30th June 2010)
(29th July 2010)
(18th January 2011)
(24th March 2011)
APPENDIX B 229
14.4 Declarations of interests of the Expert Advisors
14.4.1 Mr Martin Parker
Mr Martin Parker only attended the first and second GDG meetings. He declared
that he had received and may in the future receive money for advising implant
manufacturing companies about their products and advising on implant design. He
has produced research papers with different conclusions and publically presented
the results. No actions were required as the first two meetings were introductory
and did not involve any discussions about the evidence or formulating
recommendations.
14.4.2 Mrs Pamela Holmes
Mrs Pamela Holmes had no interests to declare and did not attend any GDG
meetings
14.4.3 Professor Judith Adams
Professor Judith Adams only attended the twelfth GDG meeting on July 29th 2010
and did not have any interests to declare.
15 Appendix C: Review protocols
15.1 Review protocol – Imaging in occult hip fracture

Component Description
Review question In patients with a continuing clinical suspicion of hip fracture, despite
negative radiographic findings, what is the clinical and cost-
effectiveness of additional imaging (radiography after at least 48
hours), Radionuclide scanning (RNS), ultrasound (US) and cpmputed
tomography (CT), compared to magnetic resonance imaging (MRI), in
confirming, or excluding, a hip fracture?
230 APPENDIX C
Objectives To identify an alternative method of diagnosis of occult hip fractures

when MRI is not available.
Population Patients >18 years old with a hip fracture undergoing different types of
surgery for hip fracture repair
People with fractures caused by specific pathologies other than

osteoporosis or osteopaenia, and patients under 18 years old are
excluded from the scope.
Intervention ▪ Computed tomography

▪ Radionuclide scanning (also known as isotope scanning or
scintigraphy).
Comparison ▪ Magnetic resonance imaging
Outcomes ▪ Sensitivity
▪ Specificity
▪ Positive and negative predictive values
▪ Positive and negative likelihood ratios
Search strategy The databases to be searched are Medline, Embase, The Cochrane
Library, CINAHL and AMED.
Randomised controlled trials (RCTs) will be considered. If no RCTs are

found for certain outcomes such as adverse events, well conducted
cohort studies and observational studies may also be considered.
Studies will be restricted to English language only
No date restriction will be applied. Databases will be searched from

their date of origin
The review strategy Meta-analysis will not be conducted for diagnostic studies. Ranges of
results will be reported.
If there is heterogeneity the following subgroups will be analysed

separately:
▪ Comorbidities strongly predictive of outcome (as mentioned in
the scope but will need the GDG to list them)
▪ Concurrent medication
▪ Age
▪ Gender
▪ Cognitive impairment
▪ Palliative care patients
APPENDIX C 231
15.2 Review protocol – Timing of surgery

Review question In patients with hip fractures what is the clinical and cost effectiveness
of early surgery (within 24, 36 or 48 hours) on the incidence of
complications such as mortality, pneumonia, pressure sores, cognitive
dysfunction and increased length of hospital stay?
Objectives To investigate whether early surgery improves patient outcomes.

Intervention Early surgery (within the cut off of 24, 36 and 48 hours of admission to
hospital)
Comparison Late surgery (after the cut off of 24, 36 and 48 hours of admission)
Outcomes Mortality (30 days, 3 months, 1 year)

Length of stay in secondary care
Length of time before community resettlement/discharge.
Place of residence (compared with baseline) 12 months after fracture.
Functional status (30 days, 3 months, 1 year)
Quality of life (30 days, 3 months, 1 year)
Complications (including pressure ulcers)

found well conducted cohort studies and observational studies may
also be considered. In particular, cohort studies using logistic regression
to adjust for confounders such as comorbidity and age, which is a
particular bias in this area.

The review strategy
Meta-analyses will be conducted where possible.
separately:
▪ Reason for delay to surgery (administrative or medical reasons)
▪ Age
232 APPENDIX C
▪ Gender
APPENDIX C 233
15.3 Review protocol – Analgesia- systemic medications

Review question In patients who have or are suspected of having a hip fracture, what is
the comparative effectiveness and cost effectiveness of systemic
analgesics in providing adequate pain relief and reducing side effects
and mortality?
Objectives To identify the most effective systemic analgesia medication for pain
relief in hip fracture patients

Intervention Systemic:
▪ Opioids e.g
o Buprenorphine
o Codeine
o Dihydrocodeine
o Hydromorphone
o Morphine
o Oxycodone
o Papaveretum (no, has been withdrawn)
o Pentazocine
o Pethidine (?) causes delirium in elderly
o Tramadol (potent cause of delirium in elderly)
▪ Non Opioid e.g.
o Paracetamol, iv, PR, oral
o Non steroidal anti inflammatory (NSAIDs)
Comparison Systemic:
▪ Opioids e.g
o Buprenorphine
o Codeine
o Dihydrocodeine
o Hydromorphone
o Morphine
o Oxycodone
o Pentazocine
▪ Non Opioid e.g.
234 APPENDIX C

o Non steroidal anti inflammatory (NSAIDs)
Outcomes ▪ Pain (generally measured by visual analogue scale or verbal

rating)
▪ Need for ‘breakthrough’ analgesia
▪ Mortality
▪ Adverse effects
o Paracetamol
▪ Virtually none but may decrease blood
pressure with iv
o Opioids
▪ Itching/histamine release,
▪ PONV,
▪ respiratory depression,
▪ decrease in blood pressure,
▪ delerium


The review strategy


separately:
▪ Comorbidities strongly predictive of outcome
▪ Age
▪ Gender
▪ Type of fracture
▪ Type of surgery
o THR vs. hemiarthroplasty
o THR vs. internal fixation
APPENDIX C 235
15.4 Review protocol – Analgesia- Nerve blocks comapared to systemic

analgesics
Review question In patients who have or are suspected of having a hip fracture, what is
the clinical and cost effectiveness of nerve blocks compared to systemic
analgesia in providing adequate pain relief and reducing side effects
and mortality?
Objectives To identify an optimal analgesia protocol including the use of nerve

blocks which may help reduce usage of systemic analgesics with strong
side effects in this patient group.
Population Patients over 18 years old with a hip fracture undergoing different
types of surgery for hip fracture repair

Intervention Nerve blocks (any type: lateral cutaneous, femoral, triple, psoas, 3-in-1
[includes femoral, obturator, lateral femoral cutaneous nerves], fascia
iliaca, with ultrasound guidance for localisation)
Comparison Pharmacological (systemic):

▪ Opioids e.g
o Buprenorphine
o Codeine
o Dihydrocodeine
o Hydromorphone
o Morphine
o Oxycodone
o Pentazocine
▪ Non Opioid e.g.
▪ NSAIDs
o upper gastrointestinal bleeding
o renal, hepatic and cardiovascular side effects
Outcomes ▪ Pain (generally measured by visual analogue scale or verbal

rating)
▪ Need for ‘breakthrough’ analgesia
▪ Mortality
236 APPENDIX C
▪ Adverse effects
o Nerve Block:
▪ Nerve damage
▪ Pressure necrosis following motor block
▪ Postoperative nausea and vomiting (PONV)
o Paracetamol
▪ Virtually none but may decrease blood
pressure with iv
o Opioids
▪ Itching/histamine release,
▪ PONV,
▪ respiratory depression,
▪ decrease in blood pressure,
▪ delirium
o NSAIDs
▪ upper gastrointestinal bleeding
▪ renal, hepatic and cardiovascular side
effects


The review strategy


separately:
▪ Age
▪ Gender
▪ Type of fracture
▪ Type of surgery
APPENDIX C 237
15.5 Review protocol - Anaesthesia

Review question In patients undergoing surgical repair for hip fractures, what is the
clinical and cost effectiveness of regional (spinal/epidural) anaesthesia
compared to general anaesthesia in reducing complications such as
mortality, cognitive dysfunction thromboembolic events, postoperative
respiratory morbidity, renal failure and length of stay in hospital?
Objectives To identify whether regional anaesthesia confers any benefit compared

to general anaesthesia with regards to reducing complications and
improving patient outcomes after surgery.
Population Patients over 18 years old with a hip fracture undergoing different
types of surgery for hip fracture repair

Intervention General anaesthesia for different types of surgery
Comparison • Regional anaesthesia for the same type of surgery

▪ Spinal/epidural without nerve block
▪ Spinal/epidural with nerve block
Outcomes ▪ Patient preference

▪ Mortality at 30 days
▪ Functional status up to 1 year
▪ Pain (generally measured by visual analogue scale or verbal
rating)
▪ Adverse effects
o General:
▪ postoperative lung complications
▪ Pulmonary emboli
▪ Pneumonia
▪ Myocardial infarction
▪ Renal failure
▪ Postoperative nausea and vomiting (PONV)
o Regional
▪ Neural damage
▪ Spinal haematoma
238 APPENDIX C


The review strategy


separately (where possible):
▪ Age
▪ Gender
▪ Type of surgery
▪ Duration of anaesthesia
APPENDIX C 239
15.6 Review protocol – surgeon seniority

Review question Does surgeon seniority (consultant or equivalent) reduce the incidence
of mortality, operative revision and poor functional outcome?
Objectives To investigate whether senior surgeons lead to better outcomes for hip
fracture patients

Intervention ▪ Consultant grade or equivalent
Comparison ▪ Below consultant grade or equivalent

▪ Trainee
Outcomes ▪ Mortality (30 days, 3 months, 1 year)
▪ Length of stay in secondary care
▪ Dislocations
▪ Wound infection
Library and CINAHL.

found well conducted cohort studies and observational studies may
also be considered.

The review strategy


separately:
▪ Age
240 APPENDIX C
15.7 Review protocol – Cement

Review question In hip fracture patients undergoing total hip replacement what is the
clinical and cost effectiveness of cemented total hip replacement versus
uncemented total hip replacement on mortality, surgical revision,
functional status, length of stay, quality of life, pain and place of
residence after hip fracture?
Objectives To examine the effectiveness of cement when inserting arthroplasty for

surgical repair
Population Patients >18 years old with a hip fracture undergoing surgical repair

Intervention Cemented arthroplasty
Comparison Uncemented arthroplasty
Outcomes ▪ Perioperative mortality

▪ Mortality at 30 days, 3 months & 1 year or longer
rating)
▪ Quality of life
▪ Length of stay in hospital/acute care
▪ Length of stay in to community or resettlement (i.e.
superspell)
▪ Place of residence 12 months after fracture
Randomised controlled trials (RCTs) will be considered.

All questions relating to surgical repair for hip fractures will be searched
together.
The review strategy

Studies will be restricted to English language articles

APPENDIX C 241
separately:
▪ Comorbidities
▪ Age
▪ Ideally “younger and fitter” patients compared to the “older
and frailer” patients. Could be a combination of age and
comorbidities
▪ Type of arthroplasty
242 APPENDIX C
15.8 Review protocol – Intracapsular fractures

Review question In patients undergoing repair for intracapsular hip fractures what is the
clinical and cost effectiveness of internal fixation compared to
hemiarthroplasty compared to total hip replacement on mortality,
surgical revision, functional status, length of stay, quality of life, pain
and place of residence after hip fracture?
Objectives To examine the effectiveness of the 3 different techniques for fixing

displaced intracapsular fractures
Population Patients >18 years old with a hip fracture undergoing surgical repair

Intervention • Internal fixation

• Hemiarthroplasty
• Total hip replacement
Comparison All of the above are compared to each other.
Outcomes ▪ Mortality at 30 days, 3 months & 1 year or longer

rating)
▪ Quality of life
superspell)

together.
APPENDIX C 243
The review strategy


separately:
comorbidities
▪ Type of internal fixation or arthroplasty
▪ Use of cement in arthroplasty
244 APPENDIX C
15.9 Review protocol – surgical approach

Review question In patients having surgical treatment for intracapsular hip fracture with
hemiarthroplasty what is the clinical and cost effectiveness of
anterolateral compared to posterior surgical approach on mortality,
number of reoperations, dislocation, functional status, length of
hospital stay, quality of life and pain?
Objectives To investigate whether one surgical approach is better than the other
when inserting a hemiarthroplasty.
Population Patients >18 years old with a hip fracture undergoing replacement
arthroplasty with a hemiarthroplasty

Intervention ▪ Anterolateral approach
Comparison ▪ Posterior approach

▪ Length of hospital stay
▪ Dislocations
▪ Functional status
▪ Quality of life
▪ Pain
Library and CINAHL.
Randomised controlled trials (RCTs) and well conducted cohort studies

and observational studies that adjust for confounders will be
considered.

APPENDIX C 245
The review strategy


separately:
▪ Type of procedure
246 APPENDIX C
15.10 Review protocol – Hemiarthroplasty stem design

Review question In patients undergoing surgery for hip fracture what is the clinical and
cost effectiveness of ‘OEDP 10A rating’ designs of stems in preference
to Austin Moore or Thompson stems when inserting a hemiarthroplasty
on mortality, surgical revision, functional status, length of stay, quality
Objectives To examine the effectiveness of modern design stems (‘OEDP 10A

rating’) compared to Austin Moore or Thompson stems.
Population Patients >18 years old with a hip fracture undergoing hemiarthroplasty

Intervention Hemiarthroplasty with a modern design stem (‘OEDP 10A rating’)
Comparison Hemiarthroplasty with an Austin Moore or Thompson

rating)
▪ Quality of life
superspell)

together.
APPENDIX C 247
The review strategy


separately:
comorbidities
248 APPENDIX C
15.11 Review protocol – extracapsular fractures
Review question In patients undergoing repair for trochanteric extracapsular hip
fractures what is the clinical and cost effectiveness of extramedullary
sliding hip screws compared to intramedullary nails on mortality,
surgical revision, functional status, length of stay, quality of life, pain
and place of residence after hip fracture?
In patients undergoing repair for subtrochanteric extracapsular hip

fractures, what is the effectiveness of extramedullary sliding hip screws
compared to intramedullary nails on mortality, surgical revision,
functional status, length of stay, quality of life, pain and place of
residence after hip fracture?
Objectives To examine the effectiveness of extramedullary implants, including

sliding hip screws, compared to intramedullary implants, including nails,
in fixing trochanteric and subtrochanteric fractures.
Population Patients >18 years old with a extracapsular hip fracture undergoing
surgical repair

Intervention Extramedullary sliding hip screws
Comparison Intramedullary nails

rating)
▪ Quality of life
▪ Requirement for reoperation (operative or postoperative
fracture of the femur, cut-out and non-union)
superspell)

APPENDIX C 249
together.
The review strategy


separately:
▪ Stability of fracture
▪ Comorbidities
▪ Age
▪ Previous fracture or surgery to femur
250 APPENDIX C
15.12 Review protocol – Mobilisation strategies

Review question In patients who have undergone surgery for hip fracture, what is the
clinical and cost effectiveness of early mobilisation (<48 hours after
surgery) compared to late mobilisation on functional status, mortality,
place of residence/discharge, pain and quality of life?
Objectives To examine the effectiveness of early mobilisation on functional
outcomes compared to delayed mobilisation
Population Patients >18 years old that have had surgery for a hip fracture.

Intervention Mobilisation (physiotherapy) within 48 hours of surgery.
Comparison Mobilisation (physiotherapy) after 48 hours of surgery.

rating)
▪ Quality of life

together.
The review strategy


separately:
▪ Comorbidities
▪ Age
APPENDIX C 251
Component
Description
Review question
In patients who have undergone surgery for hip fracture, what is the
clinical and cost effectiveness of intensive physiotherapy compared to
non intensive physiotherapy on functional status, mortality, place of
residence/discharge, pain and quality of life?
Objectives
To examine the effectiveness of intensity of mobilisation on functional
outcomes.
Patients >18 years old that have had surgery for a hip fracture.
Population
Intervention
Intensive physiotherapy, defined by an increased number of sessions or
an increase in intensity (strength) of exercise.
Comparison
Fewer sessions of physiotherapy or usual care ad defined by the paper.

rating)
▪ Quality of life
▪ Mobility
Search strategy
The databases to be searched are Medline, Embase, The Cochrane
together.
The review strategy
separately:
▪ Type or component of exercise programme
▪ Comorbidities
▪ Age
252 APPENDIX C
15.13 Review protocol – Multidisciplinary rehabilitation

Review question In patients with hip fracture what is the clinical and cost effectiveness
of 'orthogeriatrician' involvement in the whole pathway of assessment,
peri-operative care and rehabilitation on functional status, length of
stay in secondary care, mortality, place of residence/discharge, hospital
readmission and quality of life?
Objectives To identify the benefit of an orthogeriatrician involved early in the care

pathway to patient outcomes.

Intervention Involvement of an orthogeriatrician/physician throughout patient care,

starting from admission
Comparison No involvement of an orthogeriatrician/physician throughout the care

pathway (e.g. only present in rehabilitation).
▪ Length of time before community resettlement/discharge.
▪ Place of residence (compared with baseline) 12 months after
fracture.
▪ Functional status (30 days, 3 months, 1 year)
▪ Quality of life (30 days, 3 months, 1 year)


APPENDIX C 253
The review strategy

separately:
▪ Age
▪ Gender
▪ Patients from nursing homes
254 APPENDIX C
of hospital-based multidisciplinary rehabilitation on functional status,
length of stay in secondary care, mortality, place of
residence/discharge, hospital readmission and quality of life?
Objectives To identify the effectiveness of hospital-based multidisciplinary

rehabilitation compared to usual care.

Intervention Multidisciplinary hospital-based rehabilitation. Multidisciplinary

rehabilitation after hip fracture will be assumed if the following core
components are present: medicine; nursing; physiotherapy;
occupational therapy; social care. Additional components may include:
nutrition, pharmacy, clinical psychology. Additional criteria include
formal arrangements for co-ordination/teamwork and regular on-going
multidisciplinary assessment.
Types of multidisciplinary hospital-based rehabilitation include Geriatric

orthopaedic rehabilitation unit (GORU); mixed assessment and
rehabilitation unit (MARU); geriatric hip fracture programme (GHFP).
Comparison Usual hospital-based care (not multidisciplinary)

fracture.


APPENDIX C 255
The review strategy


separately:
▪ Type of hospital-based MDR
▪ Age
▪ Gender
256 APPENDIX C
of community-based multidisciplinary rehabilitation on functional
status, length of stay in secondary care, mortality, place of
residence/discharge, hospital readmission and quality of life?
Objectives To compare community-based programmes with each other and usual

care.

Intervention Community-based multidisciplinary rehabilitation, including

intermediate care unit-based, home-based (early supported discharge)
and social care unit-based. Any programme starting more than 1 week
postoperatively will be excluded.
Comparison Usual hospital-based care (not multidisciplinary)

fracture.


APPENDIX C 257
The review strategy


separately:
▪ Type of community rehabilitation programme
▪ Age
▪ Gender
258 APPENDIX C
15.14 Review protocol – Carer involvement

Review question In patients who have been discharged after hip fracture repair, what is
the clinical and cost effectiveness of having a non paid carer (e.g.
spouse, relative, friends) on mortality, length of stay, place of
residence/discharge, functional status, hospital readmission and quality
of life?
Objectives To compare the effectiveness of hospital-based multidisciplinary

rehabilitation with involvement of a carer versus without a carer.

Intervention Hospital-based multidisciplinary rehabilitation with involvement of a

non paid carer (e.g. spouse, relative, friends).
Comparison Hospital-based multidisciplinary rehabilitation without involvement of a
non paid carer (e.g. spouse, relative, friends).
fracture.


The review strategy
separately:
APPENDIX C 259
▪ Age
▪ Gender
260 APPENDIX C
15.15 Review protocol – Health Economics

Objectives The aim is to identify economic studies relevant to the review questions
set out above.
Criteria Populations, interventions and comparators as specified in the review

protocols above. Must be a relevant economic study design (cost-utility
analysis, cost-benefit analysis, cost-effectiveness analysis, cost-
consequence analysis, comparative cost analysis).
Search strategy See Appendix D, section 4.2
The review strategy

Each study is assessed using the NICE economic evaluation checklist –
NICE (2009) Guidelines Manual, Appendix H.
Inclusion/exclusion criteria
• If a study is rated as both ‘Directly applicable’ and ‘Minor
limitations’ (using the NICE economic evaluation checklist) then it
should be included in the guideline. An evidence table should be
completed and it should be included in the economic profile.
• If a study is rated as either ‘Not applicable’ or ‘Very serious
limitations’ then it should be excluded from the guideline. It should
not be included in the economic profile and there is no need to
include an evidence table.
• If a study is rated as ‘Partially applicable’ and/or ‘Potentially serious
limitations’ then there is discretion over whether it should be
included. The health economist should make a decision based on
the relative applicability and quality of the available evidence for
that question, in discussion with the GDG if required. The ultimate
aim being to include studies that are helpful for decision making in
the context of the guideline. Where exclusions occur on this basis,
this should be noted in the relevant section of the guideline with
references.
Also exclude:
• unpublished reports unless submitted as part of the call for
evidence
• abstract-only studies
• letters
APPENDIX C 261
• editorials
• reviews of economic evaluations5
• foreign language articles
Where there is discretion

The health economist should be guided by the following hierarchies.
Setting:
1. UK NHS
2. OECD countries with predominantly public health insurance
systems (e.g. France, Germany, Sweden)
3. OECD countries with predominantly private health insurance
systems (e.g. USA, Switzerland)
4. Non-OECD settings (always ‘Not applicable’)
Economic study type:

1. Cost-utility analysis
2. Other type of full economic evaluation (cost-benefit analysis,
cost-effectiveness analysis, Cost-consequence analysis)
3. Comparative cost analysis
4. Non-comparative cost analyses including cost of illness studies
(always ‘Not applicable’)
Year of analysis:
• The more recent the study, the more applicable it is
Quality of effectiveness data used in the economic analysis:

• The more closely the effectiveness data used in the economic
analysis matches with the studies included for the clinical review
the more useful the analysis will be to decision making for the
guideline.
5Recent reviews will be ordered although not reviewed. The bibliographies will be checked for relevant studies, which
will then be ordered.
262 APPENDIX D
16 Appendix D: Literature search strategies
16.1 Search Strategies

Searches were constructed by using the following groups of terms. These groups are
expanded in full in Section 1.2 below.
All searches were run in Medline, Embase and the Cochrane Library. Additionally CINAHL and
PsychINFO were searched where this was deemed appropriate. Economic searches were
conducted in Medline, Embase, NHS EED and the HTA (Health Technology Reports) database
from the Cochrane Library.
Anaesthesia search
Hip fracture terms

AND
Anaesthesia terms
AND
RCT filter or systematic review filter
NOT
Animal/publication filter
Analgesia search
Hip fracture terms

AND
Analgesia terms
AND
NOT
Carer involvement search
Hip fracture terms

AND
Carer involvement terms
NOT
APPENDIX D 263
Early surgery search
Hip fracture terms

AND
Early surgery terms
NOT
Economic searches (Medline and Embase)
Hip fracture terms

AND
Economic filter
NOT
Economic searches (NHS EED and HTA)
Hip fracture terms
Orthogeriatrician search
Hip fracture terms

AND
Orthogeriatrician terms
NOT
Patient education search
Hip fracture terms

AND
Patient education terms
NOT
Patient views search
Hip fracture terms

AND
Patient view terms
NOT
Radiological imaging search

264 APPENDIX D
Hip fracture terms

AND
Radiological imaging terms
AND
RCT filter or systematic review filter or diagnostic filter
NOT
Rehabilitation search
Hip fracture terms

AND
Rehabilitation terms
NOT
Surgeon seniority search
Hip fracture terms

AND
Surgeon seniority terms
NOT
Surgical interventions search
Hip fracture terms

AND
Surgical intervention terms
AND
NOT
16.2 Search terms
Anaesthesia
Anaesthesia terms – Cochrane Library
1 MeSH descriptor Anesthesia explode all trees
2 ((an?esthet* or an?esthesia) NEAR/4 (regional* or local* or general or spinal or
epidural)):ti,ab,kw
3 #1 OR #2
Anaesthesia terms - OVID Embase

APPENDIX D 265
1 exp Anesthesia/
2 ((an?esthet$ or an?esthesia) adj4 (regional$ or local$ or general or spinal or
epidural)).ti,ab.
3 1 or 2
Anaesthesia terms - OVID Medline

1 exp Anesthesia/
2 ((an?esthet$ or an?esthesia) adj4 (regional$ or local$ or general or spinal or
epidural)).ti,ab.
3 1 or 2
Analgesia
Analgesia terms – Cochrane Library
1 MeSH descriptor Analgesia explode all trees
2 MeSH descriptor Analgesics explode all trees
3 MeSH descriptor Nerve Block explode all trees
4 (analg$ or (pain* NEAR/3 relie*) or ((nerve* or neural*) NEAR/3 block*)):ti,ab,kw
5 (opioid* or opiate*):ti,ab,kw
6 (paracetamol or propacetamol or acetaminophen or co-codamol):ti,ab,kw
7 (morphine or buprenorphine or codeine or diphenoxylate or dipipanone or
diamorphine or dihydrocodeine or alfentanil or fentanyl or remifentanil or
meptazinol or methadone or oxycodone or papaveretum or pentazocine or
pethidine or tramadol):ti,ab,kw
8 MeSH descriptor Opiate Alkaloids explode all trees
9 MeSH descriptor Acetaminophen explode all trees
10 #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9
Analgesia terms - OVID Embase

1 exp analgesia/
2 exp Nerve Block/
3 (analg$ or (pain$ adj3 relie$) or ((nerve$ or neural$) adj3 block$)).ti,ab.
4 exp analgesic agent/
pethidine or tramadol).ti,ab.
6 (paracetamol or propacetamol or acetaminophen or co-codamol).ti,ab.
7 (opioid$ or opiate$).ti,ab.
8 or/1-7
Analgesia terms - OVID Medline

1 exp Analgesia/
2 exp Nerve Block/
3 exp Analgesics/
4 (analg$ or (pain$ adj3 relie$) or ((nerve$ or neural$) adj3 block$)).ti,ab.
5 (opioid$ or opiate$).ti,ab.
266 APPENDIX D
6 (paracetamol or propacetamol or acetaminophen or co-codamol).ti,ab.

pethidine or tramadol).ti,ab.
8 exp Opiate Alkaloids/
9 acetaminophen/
10 or/1-9
Animal/publication filter - OVID Embase
1 Case-Study/ or Abstract-Report/ or Letter/ or (case adj report).tw.
2 (exp Animal/ or Nonhuman/ or exp Animal-Experiment/) not exp Human/
3 or/1-2
Animal/publication filter - OVID Medline

1 ((Case-Reports not Randomized-Controlled-Trial) or Letter or Historical-Article or
Review-Of-Reported-Cases).pt.
2 exp Animal/ not Human/
3 or/1-2
Carer involvement
Carer involvement terms – Cochrane Library
1 MeSH descriptor Family explode all trees
2 MeSH descriptor Caregivers, this term only
3 MeSH descriptor Friends, this term only
4 MeSH descriptor Voluntary Workers, this term only
5 (carer* or caregiver* or care giver* or ((care* or caring) NEAR/5 (child* or parent*
or husband* or wife* or wives or relative* or relation* or spous* or partner* or
offspring or son* or daughter* or famil* or brother* or sister* or sib* or friend* or
volunteer*))):ti,ab,kw
6 #1 or #2 or #3 or #4 or #5
Carer involvement terms – EBSCO CINAHL

1 mh Family+ or mh caregivers or mh friends or mh voluntary workers
2 carer* or caregiver* or care giver* or care* n5 child* or care* n5 parent* or care*
n5 husband* or care* n5 wife* or care* n5 wives or care* n5 relative* or care* n5
relation* or care* n5 spous* or care* n5 partner*
3 care* n5 offspring or care* n5 son* or care* n5 daughter* or caring n5 child* or
caring n5 parent* or caring n5 husband* or caring n5 wife* or caring n5 wives or
caring n5 relative* or caring n5 relation* or caring n5 spous* or caring n5
partner*
4 care* n5 famil* or care* n5 brother* or care* n5 sister* or caring n5 offspring or
caring n5 son* or caring n5 daughter* or caring n5 famil* or caring n5 brother* or
caring n5 sister* or caring n5 sib* or caring n5 friend* or caring n5 volunteer*
5 care* n5 sib* or care* n5 friend* or care* n5 volunteer*
6 S1 or S2 or S3 or S4 or S5
APPENDIX D 267
1 Case-Study/ or Abstract-Report/ or Letter/ or (case adj report).tw. or ((exp Animal/

or Nonhuman/ or exp Animal-Experiment/) not exp Human/)
Carer involvement terms – Ovid Embase

1 (carer$ or caregiver$ or care giver$ or ((care$ or caring) adj5 (child$ or parent$ or
husband$ or wife$ or wives or relative$ or relation$ or spous$ or partner$ or
offspring or son$ or daughter$ or famil$ or brother$ or sister$ or sib$ or friend$ or
volunteer$ or voluntary))).ti,ab.
2 exp family/ or friend/ or caregiver/ or volunteer/
3 or/1-2
Carer involvement terms – Ovid Medline

1 exp Family/ or caregivers/ or friends/ or voluntary workers/
2 (carer$ or caregiver$ or care giver$ or ((care$ or caring) adj5 (child$ or parent$ or
husband$ or wife$ or wives or relative$ or relation$ or spous$ or partner$ or
offspring or son$ or daughter$ or famil$ or brother$ or sister$ or sib$ or friend$ or
volunteer$ or voluntary))).ti,ab.
3 or/1-2
Diagnostic filter
Diagnostic filter - OVID Embase
1 exp "SENSITIVITY AND SPECIFICITY"/
2 (sensitivity or specificity).tw.
3 (predictive adj3 value$).tw.
4 ((false adj positiv$) or (false adj negativ$)).tw.
5 (observer adj variation$).tw.
6 (roc adj curve$).tw.
7 (likelihood adj3 ratio$).tw.
8 *Diagnostic Accuracy/
9 exp *hip fracture/di
10 or/1-9
Diagnostic filter - OVID Medline

1 exp "Sensitivity and Specificity"/
2 (sensitivity or specificity).tw.
3 (predictive adj3 value$).tw.
4 exp diagnostic errors/
5 ((false adj positiv$) or (false adj negativ$)).tw.
6 (observer adj variation$).tw.
7 (roc adj curve$).tw.
8 (likelihood adj3 ratio$).tw.
9 likelihood functions/
10 exp *hip fractures/di, ra, ri, us
11 or/1-10
268 APPENDIX D
Early Surgery
Early surgery terms – Cochrane Library
1 MeSH descriptor Time Factors explode all trees
2 (((early or time* or delay*) NEAR/3 (surger* or operat*)) or (fast NEAR/2 track*) or
(rapid NEAR/2 transit*) or (time* NEAR/2 factor*)):ti,ab,kw
3 #1 OR #2
Early surgery terms – EBSCO CINAHL

1 early n3 surger* or early n3 operat* or time* n3 surger* or time* n3 operat* or
delay* n3 surger* or delay* n3 operat* or fast n2 track* or rapid n2 transit* or
time* n2 factor*
2 mh time factors+ or mh treatment delay+
3 S1 or S2
Early surgery terms - OVID Embase

1 (((early or time$ or delay$) adj3 (surger$ or operat$)) or (fast adj2 track$) or (rapid
adj2 transit$) or (time$ adj2 factor$)).ti,ab.
2 Therapy Delay/
3 1 or 2
Early surgery terms - OVID Medline

1 time factors/
2 (((early or time$ or delay$) adj3 (surger$ or operat$)) or (fast adj2 track$) or (rapid
adj2 transit$) or (time$ adj2 factor$)).ti,ab.
3 1 or 2
Economic
Economic filter - OVID Embase
1 exp economic aspect/
2 cost$.tw.
3 (price$ or pricing$).tw.
4 (fee or fees).tw.
5 (financial or finance or finances or financed).tw.
6 (value adj2 (money or monetary)).tw.
7 resourc$ allocat$.tw.
8 expenditure$.tw.
9 (fund or funds or funding or fundings or funded).tw.
10 (ration or rations or rationing or rationings or rationed).tw.
11 (saving or savings).tw.
12 or/1-11
13 Quality of Life/
14 quality of life.tw.
15 life quality.tw.
16 quality adjusted life.tw.
17 (qaly$ or qald$ or qale$ or qtime$).tw.
18 disability adjusted life.tw.
19 daly$.tw.
APPENDIX D 269
20 (sf36 or sf 36 or short form 36 or shortform 36 or sf thirtysix or sf thirty six or

shortform thirtysix or shortform thirty six or short form thirtysix or short form
thirty six).tw.
21 (sf6 or sf 6 or short form 6 or shortform 6 or sf six or sfsix or shortform six or short
form six).tw.
22 (sf12 or sf 12 or short form 12 or shortform 12 or sf twelve or sftwelve or
shortform twelve or short form twelve).tw.
23 (sf16 or sf 16 or short form 16 or shortform 16 or sf sixteen or sfsixteen or
shortform sixteen or short form sixteen).tw.
24 (sf20 or sf 20 or short form 20 or shortform 20 or sf twenty or sftwenty or
shortform twenty or short form twenty).tw.
25 (euroqol or euro qol or eq5d or eq 5d).tw.
26 (hql or hqol or h qol or hrqol or hr qol).tw.
27 (hye or hyes).tw.
28 health$ equivalent$ year$.tw.
29 (hui or hui1 or hui2 or hui3).tw.
30 health utilit$.tw.
31 disutilit$.tw.
32 rosser.tw.
33 (quality of wellbeing or quality of well being).tw.
34 qwb.tw.
35 willingness to pay.tw.
36 standard gamble$.tw.
37 time trade off.tw.
38 time tradeoff.tw.
39 tto.tw.
40 factor analy$.tw.
41 preference based.tw.
42 (state adj2 valu$).tw.
43 Life Expectancy/
44 life expectancy$.tw.
45 ((duration or length or period of time or lasting or last or lasted) adj4
symptom$).tw.
46 or/13-46
47 exp model/
48 exp Mathematical Model/
49 markov$.tw.
50 Monte Carlo Method/
51 monte carlo.tw.
52 exp Decision Theory/
53 (decision$ adj2 (tree$ or anlay$ or model$)).tw.
54 model$.tw.
55 or/47-55
56 12 or 46 or 55
Economic filter - OVID Medline

1 exp "Costs and Cost Analysis"/
2 Economics/
3 Economics, Nursing/ or Economics, Medical/ or Economics, Hospital/ or Economics,
Pharmaceutical/
4 exp "Fees and Charges"/
5 exp Budgets/
6 budget$.tw.
270 APPENDIX D
7 cost$.ti.
8 (cost$ adj2 (effective$ or utilit$ or benefit$ or minimi$)).ab.
9 (economic$ or pharmacoeconomic$ or pharmaco-economic$).ti.
10 (price$ or pricing$).tw.
11 (financial or finance or finances or financed).tw.
12 (fee or fees).tw.
13 (value adj2 (money or monetary)).tw.
14 Value of Life/
15 quality adjusted life.tw.
16 (qaly$ or qald$ or qale$ or qtime$).tw.
17 disability adjusted life.tw.
18 daly$.tw.
19 Health Status Indicators/
20 (sf36 or sf 36 or short form 36 or shortform 36 or sf thirtysix or sf thirty six or
shortform thirtysix or shortform thirty six or short form thirtysix or short form
thirty six).tw.
21 (sf6 or sf 6 or short form 6 or shortform 6 or sf six or sfsix or shortform six or short
form six).tw.
22 (sf12 or sf 12 or short form 12 or shortform 12 or sf twelve or sftwelve or
shortform twelve or short form twelve).tw.
23 (sf16 or sf 16 or short form 16 or shortform 16 or sf sixteen or sfsixteen or
shortform sixteen or short form sixteen).tw.
24 (sf20 or sf 20 or short form 20 or shortform 20 or sf twenty or sftwenty or
shortform twenty or short form twenty).tw.
25 (euroqol or euro qol or eq5d or eq 5d).tw.
26 (hql or hqol or h qol or hrqol or hr qol).tw.
27 (hye or hyes).tw.
28 (hui or hui1 or hui2 or hui3).tw.
29 utilit$.tw.
30 disutilit$.tw.
31 rosser.tw.
32 quality of wellbeing.tw.
33 qwb.tw.
34 willingness to pay.tw.
35 standard gamble$.tw.
36 time trade off.tw.
37 time tradeoff.tw.
38 tto.tw.
39 exp models, economic/
40 models, theoretical/ or models, organizational/
41 economic model$.tw.
42 markov chains/
43 markov$.tw.
44 Monte Carlo Method/
45 monte carlo.tw.
46 exp Decision Theory/
47 (decision$ adj2 (tree$ or anlay$ or model$)).tw.
48 or/1-47
Hip Fracture Terms

Hip fracture terms – Cochrane Library
1 MeSH descriptor Hip Fractures explode all trees
APPENDIX D 271
2 ((hip* or pertrochant* or intertrochant* or trochant* or subtrochant* or

intracapsular* or extracapsular* or ((femur* or femoral*) NEAR/3 (neck or
proximal))) NEAR/4 fracture*):ti,ab,kw
3 #1 OR #2
Hip fracture terms – EBSCO CINAHL

1 mh hip fractures+
2 femur* n3 proximal n4 fracture* or femur* n3 neck n4 fracture* or femoral* n3
proximal n4 fracture* or femoral* n3 neck n4 fracture* or pertrochant* n4
fracture* or intertrochant* n4 fracture* or trochanteric n4 fracture* or
subtrochanteric n4 fracture* or extracapsular* n4 fracture* or hip* n4 fracture*
3 intracapsular* n4 fracture* or femur* n4 fracture* or femoral* n4 fracture*
4 S1 or S2 or S3
Hip fracture terms - OVID Embase

1 exp Hip Fracture/
2 ((femur$ or femoral$) adj3 (head or neck or proximal) adj4 fracture$).ti,ab.
3 ((hip$ or femur$ or femoral$ or trochant$ or pertrochant$ or intertrochant$ or
subtrochant$ or intracapsular$ or extracapsular$) adj4 fracture$).ti,ab.
4 1 or 2 or 3
Hip fracture terms - OVID Medline

1 exp Hip Fractures/
4 1 or 2 or 3
Hip fracture terms - OVID PsychInfo

1 hips/
4 1 or 2 or 3
Orthogeriatrician
Orthogeriatrician terms – Cochrane Library
1 (geriatr*-orthop* or orthop?edic-geriatr* or ortho*-geriatr* or
orthogeriatr*):ti,ab,kw
2 (orthop* NEAR/2 geriatr*):ti,ab,kw
3 MeSH descriptor Physicians, this term only
4 MeSH descriptor Geriatrics explode all trees
5 #1 or #2 or #3 or #4
Orthogeriatrician terms – EBSCO CINAHL

272 APPENDIX D
1 orthop* n2 geriatr*
2 geriatr*-orthop* or orthogeriatr* or ortho*-geriatr* or orthop?edic-geriatr*
3 (MH "Physicians")
4 (MH "Geriatrics")
5 (MH "Multidisciplinary Care Team")
6 S1 or S2 or S3 or S4 or S5
Orthogeriatrician terms - OVID Embase

1 (geriatr$-orthop$ or orthop?edic-geriatr$ or ortho$-geriatr$ or
orthogeriatr$).ti,ab.
2 (orthop$ adj2 geriatr$).ti,ab.
3 geriatric care/
4 geriatrics/
5 physician/
6 or/1-5
Orthogeriatrician terms - OVID Medline

1 (geriatr$-orthop$ or orthop?edic-geriatr$ or ortho$-geriatr$ or
orthogeriatr$).ti,ab.
3 Physicians/
4 Geriatrics/
5 or/1-4
Patient education
Patient education – EBSCO CINAHL
1 mh Patients or mh Inpatients or mh Outpatients
2 mh Caregivers or mh Family+ or mh Parents+ or mh Guardianship, Legal
3 patients or carer* or famil*
4 S1 or S2 or S3
5 mh Information Services+ or mh Books+ or mh Pamphlets or mh Counseling
6 S4 and S5
7 patient n3 education or patient n3 educate or patient n3 educating or patient n3
information or patient n3 literature or patient n3 leaflet* or patient n3 booklet* or
patient n3 pamphlet*
8 patients n3 education or patients n3 educate or patients n3 educating or patients
n3 information or patients n3 literature or patients n3 leaflet* or patients n3
booklet* or patients n3 pamphlet*
9 mh Patient Education+
10 S6 or S7 or S8 or S9
Patient education - OVID Embase

1 Patient/ or Hospital patient/ or Outpatient/
2 Caregiver/ or exp Family/ or exp Parent/
3 (patients or carer$ or famil$).tw.
4 or/1-3
APPENDIX D 273
5 Information Service/ or Information center/ or Publication/ or Book/ or

Counseling/ or Directive counseling/
6 4 and 5
7 ((patient or patients) adj3 (education or educate or educating or information or
literature or leaflet$ or booklet$ or pamphlet$)).ti,ab.
8 Patient information/ or Patient education/
9 or/6-8
Patient education – OVID Medline

1 Patients/ or Inpatients/ or Outpatients/
2 Caregivers/ or exp Family/ or exp Parents/ or exp Legal-Guardians/
4 or/1-3
5 Popular-Works-Publication-Type/ or exp Information-Services/ or Publications/ or
Books/ or Pamphlets/ or Counseling/ or Directive-Counseling/
6 4 and 5
8 Patient-Education/ or Patient-Education-Handout-Publication-Type/
9 or/6-8
Patient education – Ovid PsychInfo

1 exp patients/
2 caregivers/ or exp family/ or exp parents/ or exp guardianship/
4 or/1-3
5 exp information services/ or exp printed communications media/ or reading
materials/ or exp counseling/
6 4 and 5
8 client education/
10 or/6-9
Patient views
Patient views – EBSCO CINAHL
1 mh Consumer Satisfaction+ or mh Consumer Attitudes or mh Personal Satisfaction
or mh Consumer Participation or mh Patient Rights+ or mh Questionnaires+ or mh
Interviews+ or mh Focus groups or mh surveys
2 patient* n3 view* or patient* n3 opinion* or patient* n3 awareness or patient* n3
tolerance or patient* n3 perception or patient* n3 persistenc* or patient* n3
attitude* or patient* n3 compliance or patient* n3 satisfaction or patient* n3
concern* or patient* n3 belief* or patient* n3 feeling*
3 patient* n3 position or patient* n3 idea* or patient* n3 preference* or patient* n3
choice*
4 discomfort or comfort or inconvenience or bother* or trouble or fear* or anxiety
or anxious or embarrass*
274 APPENDIX D
Patient views - OVID Embase

1 Consumer attitude/ or patient satisfaction/ or patient compliance/ or patient right/
or health survey/ or questionnaire/ or interview/
2 (patient$ adj3 (view$ or opinion$ or awareness or tolerance or perception or
persistenc$ or attitude$ or compliance or satisfaction or concern$ or belief$ or
feeling$ or position or idea$ or preference$ or choice$)).tw.
3 (Discomfort or comfort or inconvenience or bother$4 or trouble or fear$ or anxiety
or anxious or embarrass$4).tw.
4 or/1-3
Patient views - OVID Medline

1 exp Consumer-Satisfaction/ or Personal-Satisfaction/ or exp Patient-Acceptance-
Of-Health-Care/ or exp Consumer-Participation/ or exp Patient-Rights/ or Health
Care Surveys/ or Questionnaires/ or Interview/ or Focus groups/
feeling$ or position or idea$ or preference$ or choice$)).tw.
3 (Discomfort or comfort or inconvenience or bother$4 or trouble or fear$ or anxiety
or anxious or embarrass$4).tw.
4 or/1-3
Patient views - OVID PsychInfo

1 exp consumer satisfaction/ or exp client attitudes/ or client participation/ or exp
client rights/ or treatment compliance/ or consumer surveys/ or exp
questionnaires/ or interviews/ or expectations/
feeling$ or position or idea$ or preference$ or choice$ or expect$)).tw.
3 ((Discomfort or comfort or inconvenience or bother$4 or trouble or fear$ or
anxiety or anxious or embarrass$4).tw..
4 or/1-3
Radiological Imaging
Radiological imaging terms – Cochrane Library
1 MeSH descriptor Magnetic Resonance Imaging, this term only
2 ((MR or NMR) NEAR/2 tomograph*):ti,ab,kw
3 (MRI):ti,ab,kw
4 ((magnetic resonance or MR or NMR) NEAR/2 imag*):ti,ab,kw
5 MeSH descriptor Tomography, X-Ray Computed, this term only
6 MeSH descriptor Tomography, Spiral Computed, this term only
7 mdct:ti,ab,kw
8 (ct or compute* tomograph* or compute*-tomograph* or cat):ti,ab,kw
9 MeSH descriptor Radionuclide Imaging, this term only
10 (((radionuclide or radioisotope or isotope) NEAR (imag* or scan*)) or rns or
scintigraph* or scintiphotograph*):ti,ab,kw
11 MeSH descriptor Ultrasonography, this term only
12 (ultrason* or ultrasound* or sonograph* or echograph*):ti,ab,kw
13 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12
APPENDIX D 275
Radiological imaging terms – EBSCO CINAHL

1 mh Magnetic Resonance Imaging or magnetic resonance n2 imag* or MR n2 imag*
or NMR n2 imag* or MR n2 tomograph$ or NMR n2 tomograph$ or MRI
2 mdct or compute* tomograph* or cat or MH "Tomography, X-Ray Computed" or
mh Tomography, Spiral Computed or compute*-tomograph* or "ct"
3 mh Radionuclide Imaging or radionuclide n1 imag* or radioisotope n1 imag* or
isotope n1 imag* or radionuclide n1 scan* or radioisotope n1 scan* or isotope n1
scan* or rns or scintigraph* or scintiphotograph*
4 mh Ultrasonography or ultrason* or sonograph* or echograph* or ultrasound*
Radiological imaging terms – OVID Embase

1 nuclear magnetic resonance imaging/
2 ((magnetic resonance or MR or NMR) adj2 imag$).ti,ab.
3 (((MR or NMR) adj2 tomograph$) or MRI).ti,ab.
4 computer assisted tomography/
5 spiral computer assisted tomography/
6 mdct.ti,ab.
7 (ct or compute$ tomograph$ or compute$-tomograph$ or cat).ti,ab.
8 scintiscanning/ or scintigraphy/
9 (((radionuclide or radioisotope or isotope) adj1 (imag$ or scan$)) or rns or
scintigraph$ or scintiphotograph$).ti,ab.
10 (ultrason$ or ultrasound$ or sonograph$ or echograph$).ti,ab.
11 echography/
12 or/1-11
Radiological imaging terms – OVID Medline

1 Magnetic Resonance Imaging/
2 ((magnetic resonance or MR or NMR) adj2 imag$).ti,ab.
3 (((MR or NMR) adj2 tomograph$) or MRI).ti,ab.
4 Tomography, X-Ray Computed/
5 Tomography, Spiral Computed/
6 mdct.ti,ab.
7 (ct or compute$ tomograph$ or compute$-tomograph$ or cat).ti,ab.
8 Radionuclide Imaging/
9 (((radionuclide or radioisotope or isotope) adj1 (imag$ or scan$)) or rns or
scintigraph$ or scintiphotograph$).ti,ab.
10 Ultrasonography/
11 (ultrason$ or ultrasound$ or sonograph$ or echograph$).ti,ab.
12 or/1-11
RCT filter
RCT filter Embase
1 Clinical-Trial/ or Randomized-Controlled-Trial/ or Randomization/ or Single-Blind-
Procedure/ or Double-Blind-Procedure/ or Crossover-Procedure/ or Prospective-
Study/ or Placebo/
276 APPENDIX D
2 (((clinical or control or controlled) adj (study or trial)) or ((single or double or triple)

adj (blind$3 or mask$3)) or (random$ adj (assign$ or allocat$ or group or grouped
or patients or study or trial or distribut$)) or (crossover adj (design or study or
trial)) or placebo or placebos).ti,ab.
3 1 or 2
RCT filter Medline

1 Randomized-Controlled-Trials/ or Random-Allocation/ or Double-Blind-Method/ or
Single-Blind-Method/ or exp Clinical-Trials as topic/ or Cross-Over-Studies/ or
Prospective-Studies/ or Placebos/
2 (Randomized-Controlled-Trial or Clinical-Trial or Controlled-Clinical-Trial).pt.
3 (((clinical or control or controlled) adj (study or trial)) or ((single or double or triple)
adj (blind$3 or mask$3)) or (random$ adj (assign$ or allocat$ or group or grouped
or patients or study or trial or distribut$)) or (crossover adj (design or study or
trial)) or placebo or placebos).ti,ab.
4 or/1-3
Rehabilitation
Rehabilitation terms - Cochrane Library
1 MeSH descriptor Rehabilitation explode all trees
2 MeSH descriptor Rehabilitation Centers explode all trees
3 MeSH descriptor Rehabilitation Nursing explode all trees
4 MeSH descriptor Patient Care Team explode all trees
5 MeSH descriptor Patient Care Management explode all trees
6 MeSH descriptor Occupational Therapy explode all trees
7 MeSH descriptor Physical Therapy Modalities explode all trees
8 MeSH descriptor Physical Therapy Department, Hospital explode all trees
9 MeSH descriptor Physical Therapy (Specialty) explode all trees
10 MeSH descriptor Critical Pathways explode all trees
11 MeSH descriptor Therapy, Computer-Assisted explode all trees
12 MeSH descriptor Exercise Therapy explode all trees
13 MeSH descriptor Social Work explode all trees
14 MeSH descriptor Social Support explode all trees
15 MeSH descriptor Pain Clinics explode all trees
16 MeSH descriptor Patient Education as Topic explode all trees
17 MeSH descriptor Health Education explode all trees
18 MeSH descriptor Recovery of Function, this term only
19 MeSH descriptor Subacute Care, this term only
20 MeSH descriptor Residential Facilities explode all trees
21 MeSH descriptor Day Care, this term only
22 MeSH descriptor Home Care Services, this term only
23 MeSH descriptor Home Care Services, Hospital-Based, this term only
24 MeSH descriptor Home Nursing, this term only
25 MeSH descriptor Hospital Units, this term only
26 MeSH descriptor Nursing Homes explode all trees
27 MeSH descriptor Walking explode all trees
28 MeSH descriptor Caregivers, this term only
29 (rehab* or habilitat* or recover*):ti,ab,kw
30 (multidisciplinar* or interdisciplinar* or multiprofessional* or multimodal* or mdt
or mdr):ti,ab,kw
31 (social NEAR (work* or support or care)):ti,ab,kw
APPENDIX D 277
32 (pain clinic* or pain service* or pain relief unit* or (pain center* or pain
centre*)):ti,ab,kw
33 ((treatment* or therap* or training or education* or healthcare) NEAR/10
(program* or intervention* or approach*)):ti,ab,kw
34 (early NEAR (mobil* or discharg* or ambulat*)):ti,ab,kw
35 (occupational therap* or physical therap* or physiotherap* or physio):ti,ab,kw
36 (exercis* NEAR/3 therap*):ti,ab,kw
37 ((early or earli* or immediat* or initial* or begin* or first* or first-line or first line
or first choice or primar* or preceed* or original*) NEAR/3 (interven* or treat* or
therap* or care or medicine* or technique* or strateg* or activit* or
mobili*)):ti,ab,kw
38 (walk or walks or walking):ti,ab,kw
39 mobili?ation strateg*:ti,ab,kw
40 (ambulate* or ambulation* or ambulating*):ti,ab,kw
41 (exerci* NEAR/3 (rehab* or habilitat* or recover* or therap* or treat* or
medicine* or intervention* or technique* or strateg*)):ti,ab,kw
42 ((walk* or mobil* or mov* or motor* or physi*) NEAR/3 (rehab* or habilitat* or
recover* or therap* or treat* or medicine* or intervention* or technique* or
strateg*)):ti,ab,kw
43 (extend* NEAR/2 care* NEAR/3 (facilit* or service* or unit* or center* or clinic* or
program* or residen* or home* or hous*)):ti,ab,kw
44 ((residen* or intermediate* or assist* liv*) NEAR/3 (facilit* or care* or service* or
unit* or center* or clinic* or program* or residen* or home* or hous*)):ti,ab,kw
45 ((halfway or transition*) NEAR/3 (home* or hous* or facilit* or care* or residen*
or service* or unit* or center* or clinic* or program*)):ti,ab,kw
46 (nurs* NEAR/2 home*):ti,ab,kw
47 (geriatr*-orthop* or orthop?edic-geriatr* or ortho*-geriatr* or orthogeriatr* or
goru):ti,ab,kw
48 (orthop* NEAR/2 geriatr*):ti,ab,kw
49 rehabilitation unit*:ti,ab,kw
50 (mixed assessment or maru):ti,ab,kw
51 (geriatric hip fracture program* or ghfp):ti,ab,kw
52 (day NEAR (hospital* or care or unit*)):ti,ab,kw
53 ((home-based or home based) NEAR care):ti,ab,kw
54 carer* involve*:ti,ab,kw
55 (esd or early supported discharge):ti,ab,kw
56 sequential care:ti,ab,kw
57 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14
or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26
or #51 or #52 or #53 or #54 or #55 or #56
Rehabilitation terms – EBSCO CINAHL

1 (MH "Rehabilitation+")
2 (MH "Rehabilitation Nursing")
3 (MH "Recovery")
4 (MH "Subacute Care")
5 (MH "Rehabilitation Centers+")
6 mh residential facilities or mh Assisted Living Facilities or mh Halfway Houses
7 mh Day Care or mh home care services or mh home care services, hospital-based
or mh home nursing or mh Hospital Units
8 mh Nursing Homes+ or mh Patient Care Team+ or mh Patient Care Management+
or mh Physical Therapy Techniques+ or mh Physical Therapy Department,
278 APPENDIX D
Hospital+
9 mh Critical Pathways+ or mh Therapy, Computer-Assisted+ or mh Exercise
Therapy+ or mh Walking+
10 mh Social Work+ or mh Social Support+ or mh Pain Clinics+ or mh Patient
Education+ or mh Health Education+ or mh Caregivers
11 (MH "Multidisciplinary Care Team+")
12 rehab* or habilitat* or recover*
13 multidisciplinar* or mdr or mdt or multimodal* or multiprofessional* or
interdisciplinar*
14 social n1 work* or social n1 support or social n1 care
15 pain clinic* or pain service* or pain relief unit* or pain center* or pain centre*
16 treatment* n10 program* or treatment* n10 intervention* or treatment* n10
approach* or therap* n10 program* or therap* n10 intervention* or therap* n10
approach* or training n10 program* or training n10 intervention* or training n10
approach* or education* n10 program* or education* n10 intervention* or
education* n10 approach*
17 healthcare n10 program* or healthcare n10 intervention* or healthcare n10
approach*
18 early n1 mobil* or early n1 discharg* or early n1 ambulat*
19 occupational therap* or physical therap* or physiotherap* or physio
20 exercis* n3 therap*
21 early n3 interven* or early n3 treat* or early n3 therap* or early n3 care or early n3
medicine* or early n3 technique* or early n3 strateg* or early n3 activit* or early
n3 mobili*
22 earli* n3 interven* or earli* n3 treat* or earli* n3 therap* or earli* n3 care or
earli* n3 medicine* or earli* n3 technique* or earli* n3 strateg* or earli* n3
activit* or earli* n3 mobili*
23 immediat* n3 interven* or immediat* n3 treat* or immediat* n3 therap* or
immediat* n3 care or immediat* n3 medicine* or immediat* n3 technique* or
immediat* n3 strateg* or immediat* n3 activit* or immediat* n3 mobili*
24 initial* n3 interven* or initial* n3 treat* or initial* n3 therap* or initial* n3 care or
initial* n3 medicine* or initial* n3 activit* or initial* n3 technique* or initial* n3
strateg* or initial* n3 mobili*
25 begin* n3 interven* or begin* n3 treat* or begin* n3 therap* or begin* n3 care or
begin* n3 medicine* or begin* n3 technique* or begin* n3 strateg* or begin* n3
activit* or begin* n3 mobili*
26 first* n3 interven* or first* n3 treat* or first* n3 therap* or first* n3 care or first*
n3 medicine* or first* n3 technique* or first* n3 strateg* or first* n3 activit* or
first* n3 mobili*
27 first-line n3 interven* or first-line n3 treat* or first-line n3 therap* or first-line n3
care or first-line n3 medicine* or first-line n3 technique* or first-line n3 strateg* or
first-line n3 activit* or first-line n3 mobili*
28 primar* n3 interven* or primar* n3 treat* or primar* n3 therap* or primar* n3
care or primar* n3 medicine* or primar* n3 technique* or primar* n3 strateg* or
primar* n3 activit* or primar* n3 mobili*
29 original* n3 interven* or original* n3 treat* or original* n3 therap* or original* n3
care or original* n3 medicine* or original* n3 technique* or original* n3 strateg*
or original* n3 activit* or original* n3 mobili*
30 preceed* n3 interven* or preceed* n3 treat* or preceed* n3 therap* or preceed*
n3 care or preceed* n3 medicine* or preceed* n3 technique* or preceed* n3
strateg* or preceed* n3 activit* or preceed* n3 mobili*
31 walk or walks or walking
323 mobili?ation strateg*
33 ambulate* or ambulation* or ambulating*
34 exerci* n3 rehab* or exerci* n3 habilitat* or exerci* n3 recover* or exerci* n3
APPENDIX D 279
therap* or exerci* n3 treat* or exerci* n3 medicine* or exerci* n3 intervention* or

exerci* n3 technique* or exerci* n3 strateg*
35 walk* n3 rehab* or walk* n3 habilitat* or walk* n3 recover* or walk* n3 therap*
or walk* n3 treat* or walk* n3 medicine* or walk* n3 intervention* or walk* n3
technique* or walk* n3 strateg*
36 mov* n3 rehab* or mov* n3 habilitat* or mov* n3 recover* or mov* n3 therap* or
mov* n3 treat* or mov* n3 medicine* or mov* n3 intervention* or mov* n3
technique* or mov* n3 strateg*
37 motor* n3 rehab* or motor* n3 habilitat* or motor* n3 recover* or motor* n3
therap* or motor* n3 treat* or motor* n3 medicine* or motor* n3 intervention*
or motor* n3 technique* or motor* n3 strateg*
38 physi* n3 rehab* or physi* n3 habilitat* or physi* n3 recover* or physi* n3
therap* or physi* n3 treat* or physi* n3 medicine* or physi* n3 intervention* or
physi* n3 technique* or physi* n3 strateg*
39 extend* n2 care* n3 facilit* or extend* n2 care* n3 service* or extend* n2 care*
n3 unit* or extend* n2 care* n3 center* or extend* n2 care* n3 clinic* or extend*
n2 care* n3 program* or extend* n2 care* n3 residen* or extend* n2 care* n3
home* or extend* n2 care* n3 hous*
40 residen* n3 facilit* or residen* n3 care* or residen* n3 service* or residen* n3
unit* or residen* n3 center* or residen* n3 clinic* or residen* n3 program* or
residen* n3 residen* or residen* n3 home* or residen* n3 hous*
41 intermediate* n3 facilit* or intermediate* n3 care* or intermediate* n3 service* or
intermediate* n3 unit* or intermediate* n3 center* or intermediate* n3 clinic* or
intermediate* n3 program* or intermediate* n3 residen* or intermediate* n3
home* or intermediate* n3 hous*
42 assist* liv* n3 facilit* or assist* liv* n3 care* or assist* liv* n3 service* or assist*
liv* n3 unit* or assist* liv* n3 center* or assist* liv* n3 clinic* or assist* liv* n3
program* or assist* liv* n3 residen* or assist* liv* n3 home* or assist* liv* n3
hous*
43 halfway n3 home* or halfway n3 hous* or halfway n3 facilit* or halfway n3 care*
or halfway n3 residen* or halfway n3 service* or halfway n3 unit* or halfway n3
center* or halfway n3 clinic* or halfway n3 program*
44 transition* n3 home* or transition* n3 hous* or transition* n3 facilit* or
transition* n3 care* or transition* n3 residen* or transition* n3 service* or
transition* n3 unit* or transition* n3 center* or transition* n3 clinic* or
transition* n3 program*
45 nurs* n2 home* or geriatr*-orthop* or orthop?edic-geriatr* or ortho*-geriatr* or
orthogeriatr* or goru or orthop* n2 geriatr* or rehabilitation unit* or mixed
assessment or maru
46 geriatric hip fracture program* or ghfp or day n1 hospital* or day n1 care or day n1
unit* or home-based n1 care or home based n1 care or carer* involve* or esd or
early supported discharge or sequential care
47 S1 or S2 or S3 or S4 or S5 or S6 or S7 or S8 or S9 or S10 or S11 or S12 or S13 or S14
or S15 or S16 or S17 or S18 or S19 or S20 or S21 or S22 or S23 or S24 or S25 or S26
or S27 or S28 or S29 or S30 or S31 or S32 or S33 or S34 or S35 or S36 or S37 or S38
or S39 or S40 or S41 or S42 or S43 or S44 or S45 or S46
Rehabilitation terms - OVID Embase

1 exp Rehabilitation/ or exp Rehabilitation Nursing/ or exp daily life activity/
2 assisted living facility/ or nursing home/ or pain clinic/ or rehabilitation center/ or
residential home/ or halfway house/
3 day hospital/ or home care/ or home health agency/ or home physiotherapy/ or
home rehabilitation/ or patient care/ or patient care planning/ or rehabilitation
care/
4 exp mobilization/ or exp Occupational Therapy/ or exp Physiotherapy/ or exp
280 APPENDIX D
kinesiotherapy/ or walking/
5 exp clinical pathway/ or social care/ or caregiver support/ or social support/ or
caregiver/
6 (rehab$ or habilitat$ or recover$).ti,ab.
7 (multidisciplinar$ or interdisciplinar$ or multiprofessional$ or multimodal$ or mdt
or mdr).ti,ab.
8 (social adj1 (work$ or support or care)).ti,ab.
9 (pain clinic$ or pain service$ or pain relief unit$ or (pain center$ or pain
centre$)).ti,ab.
10 ((treatment$ or therap$ or training or education$ or healthcare) adj10 (program$
or intervention$ or approach$)).ti,ab.
11 (early adj1 (mobil$ or discharg$ or ambulat$)).ti,ab.
12 (occupational therap$ or physical therap$ or physiotherap$ or physio).ti,ab.
13 (exercis$ adj3 therap$).ti,ab.
14 ((early or earli$ or immediat$ or initial$ or begin$ or first$ or first-line or first line
or first choice or primar$ or preceed$ or original$) adj3 (interven$ or treat$ or
therap$ or care or medicine$ or technique$ or strateg$ or activit$ or
mobili$)).ti,ab.
15 (walk or walks or walking).ti,ab.
16 mobili?ation strateg$.ti,ab.
17 (ambulate$ or ambulation$ or ambulating$).ti,ab.
18 (exerci$ adj3 (rehab$ or habilitat$ or recover$ or therap$ or treat$ or medicine$ or
intervention$ or technique$ or strateg$)).ti,ab.
19 ((walk$ or mobil$ or mov$ or motor$ or physi$) adj3 (rehab$ or habilitat$ or
recover$ or therap$ or treat$ or medicine$ or intervention$ or technique$ or
strateg$)).ti,ab.
20 (extend$ adj2 care$ adj3 (facilit$ or service$ or unit$ or center$ or clinic$ or
program$ or residen$ or home$ or hous$)).ti,ab.
21 ((residen$ or intermediate$ or assist$ liv$) adj3 (facilit$ or care$ or service$ or
unit$ or center$ or clinic$ or program$ or residen$ or home$ or hous$)).ti,ab.
22 ((halfway or transition$) adj3 (home$ or hous$ or facilit$ or care$ or residen$ or
service$ or unit$ or center$ or clinic$ or program$)).ti,ab.
23 (nurs$ adj2 home$).ti,ab.
24 (geriatr$-orthop$ or orthop?edic-geriatr$ or ortho$-geriatr$ or orthogeriatr$ or
goru).ti,ab.
26 rehabilitation unit$.ti,ab.
27 (mixed assessment or maru).ti,ab.
28 (geriatric hip fracture program$ or ghfp).ti,ab.
29 (day adj (hospital$ or care or unit$)).ti,ab.
30 ((home-based or home based) adj care).ti,ab.
31 carer$ involve$.ti,ab.
32 (esd or early supported discharge).ti,ab.
33 sequential care.ti,ab.
34 or/1-33
Rehabilitation terms - OVID Medline

1 exp rehabilitation/ or exp rehabilitation nursing/ or "Recovery of Function"/ or
Subacute Care/
2 exp rehabilitation centers/ or Residential Facilities/ or Assisted Living Facilities/ or
Halfway Houses/
3 Day Care/ or home care services/ or home care services, hospital-based/ or home
APPENDIX D 281
nursing/ or Hospital Units/

4 exp Nursing Homes/ or exp Patient Care Team/ or exp Patient Care Management/
or exp Occupational Therapy/ or exp Physical Therapy Techniques/ or exp Physical
Therapy Department, Hospital/
5 exp "Physical Therapy (Specialty)"/ or exp Critical Pathways/ or exp Therapy,
Computer-Assisted/ or exp Exercise Therapy/ or exp Walking/
6 exp Social Work/ or exp Social Support/ or exp Pain Clinics/ or exp Patient
Education/ or exp Health Education/ or Caregivers/
7 (rehab$ or habilitat$ or recover$).ti,ab.
8 (multidisciplinar$ or interdisciplinar$ or multiprofessional$ or multimodal$ or mdt
or mdr).ti,ab.
9 (social adj1 (work$ or support or care)).ti,ab.
10 (pain clinic$ or pain service$ or pain relief unit$ or (pain center$ or pain
centre$)).ti,ab.
11 ((treatment$ or therap$ or training or education$ or healthcare) adj10 (program$
or intervention$ or approach$)).ti,ab.
12 (early adj1 (mobil$ or discharg$ or ambulat$)).ti,ab.
13 (occupational therap$ or physical therap$ or physiotherap$ or physio).ti,ab.
14 (exercis$ adj3 therap$).ti,ab.
15 ((early or earli$ or immediat$ or initial$ or begin$ or first$ or first-line or first line
or first choice or primar$ or preceed$ or original$) adj3 (interven$ or treat$ or
therap$ or care or medicine$ or technique$ or strateg$ or activit$ or
mobili$)).ti,ab.
16 (walk or walks or walking).ti,ab.
17 mobili?ation strateg$.ti,ab.
18 (ambulate$ or ambulation$ or ambulating$).ti,ab.
19 (exerci$ adj3 (rehab$ or habilitat$ or recover$ or therap$ or treat$ or medicine$ or
intervention$ or technique$ or strateg$)).ti,ab.
20 ((walk$ or mobil$ or mov$ or motor$ or physi$) adj3 (rehab$ or habilitat$ or
recover$ or therap$ or treat$ or medicine$ or intervention$ or technique$ or
strateg$)).ti,ab.
21 (extend$ adj2 care$ adj3 (facilit$ or service$ or unit$ or center$ or clinic$ or
program$ or residen$ or home$ or hous$)).ti,ab.
22 ((residen$ or intermediate$ or assist$ liv$) adj3 (facilit$ or care$ or service$ or
unit$ or center$ or clinic$ or program$ or residen$ or home$ or hous$)).ti,ab.
23 ((halfway or transition$) adj3 (home$ or hous$ or facilit$ or care$ or residen$ or
service$ or unit$ or center$ or clinic$ or program$)).ti,ab.
24 (nurs$ adj2 home$).ti,ab.
25 (geriatr$-orthop$ or orthop?edic-geriatr$ or ortho$-geriatr$ or orthogeriatr$ or
goru).ti,ab.
27 rehabilitation unit$.ti,ab.
28 (mixed assessment or maru).ti,ab.
29 (geriatric hip fracture program$ or ghfp).ti,ab.
30 (day adj (hospital$ or care or unit$)).ti,ab.
31 ((home-based or home based) adj care).ti,ab.
32 carer$ involve$.ti,ab.
33 (esd or early supported discharge).ti,ab.
34 sequential care.ti,ab.
35 or/1-34
282 APPENDIX D
Surgeon seniority
Surgeon seniority terms – Cochrane Library
1 MeSH descriptor Clinical Competence explode all trees
2 (surgeon* NEAR/3 (senior* or experience* or supervision* or volume* or
grade*)):ti,ab,kw
3 (consultant* or registrar* or spr or staff grade or trust grade or associate
specialist*):ti,ab,kw
4 (surg* NEAR (team* or list*)):ti,ab,kw
5 (list* NEAR (organise* or organize* or consultant-led or consultant led)):ti,ab,kw
6 #1 or #2 or #3 or #4 or #5
Surgeon seniority terms – EBSCO CINAHL

1 surgeon* n3 senior* or surgeon* n3 volume* or surgeon* n3 supervision* or
surgeon* n3 experience* or surgeon* n3 grade* or surg* n1 team* or surg* n1
list* or list* n1 organise* or list* n1 organize* or list* n1 consultant-led or list* n1
consultant led
2 consultant* or spr or registrar* or staff grade or trust grade or associate specialist*
or mh clinical competence+
3 S1 or S2
Surgeon seniority terms - OVID Embase

1 exp clinical competence/
2 (surgeon$ adj3 (senior$ or experience$ or supervision$ or volume$ or
grade$)).ti,ab.
3 (consultant$ or registrar$ or spr or staff grade or trust grade or associate
specialist$).ti,ab.
4 (surg$ adj1 (team$ or list$)).ti,ab.
5 (list$ adj1 (organise$ or organize$ or consultant-led or consultant led)).ti,ab.
6 or/1-5
Surgeon seniority terms - OVID Medline

1 Clinical Competence/
2 (surgeon$ adj3 (senior$ or experience$ or supervision$ or volume$ or
grade$)).ti,ab.
3 (consultant$ or registrar$ or spr or staff grade or trust grade or associate
specialist$).ti,ab.
4 (surg$ adj1 (team$ or list$)).ti,ab.
5 (list$ adj1 (organise$ or organize$ or consultant-led or consultant led)).ti,ab.
6 or/1-5
Surgical Interventions
Surgical Interventions terms – Cochrane Library
1 MeSH descriptor Fracture Fixation, Internal explode all trees
2 MeSH descriptor Internal Fixators explode all trees
3 MeSH descriptor Bone Nails explode all trees
4 MeSH descriptor Bone Screws explode all trees
APPENDIX D 283
5 MeSH descriptor Bone Plates explode all trees

6 MeSH descriptor Bone Cements explode all trees
7 MeSH descriptor Arthroplasty explode all trees
8 (pin* or nail* or screw* or plate* or arthroplast* or fix* or prosthes* or ((cement*
or glue* or paste*) NEAR/3 bone*)):ti,ab,kw
9 #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8
Surgical interventions terms - OVID Embase

1 (pin$ or nail$ or screw$ or plate$ or arthroplast$ or hemiarthroplast$ or fix$ or
prosthes$).ti,ab.
2 arthroplasty/ or hip arthroplasty/
3 ((cement$ or glue$ or paste$) adj3 bone$).ti,ab.
4 Fracture Treatment/ or Hip Surgery/ or Femur Intertrochanteric Osteotomy/ or Hip
Osteotomy/ or exp Fracture Fixation/ or Bone Screw/ or Bone Plate/ or Bone Nail/
or ender Nail/ or Interlocking Nail/ or Osteosynthesis Material/ or external fixator/
or exp bone cement/
5 or/1-4
Surgical interventions terms - OVID Medline

1 (pin$1 or nail$ or screw$1 or plate$1 or arthroplast$ or fix$ or prosthes$).ti,ab.
2 Internal Fixators/ or Bone Screws/ or Fracture Fixation, Internal/ or Bone Plates/ or
Bone Nails/ or Bone Cements/
3 ((cement$ or glue$ or paste$) adj3 bone$).ti,ab.
4 Arthroplasty/ or Arthroplasty, Replacement, Hip/
5 or/1-4
Systematic review filter

Systematic review filter - OVID Medline
1 meta-analysis/
2 (metaanalys$ or meta-analys$ or meta analys$).tw.
3 exp "review literature"/
4 (systematic$ adj3 (review$ or overview$)).tw.
5 (selection criteria or data extraction).ab. and review.pt.
6 (cochrane or embase or psychlit or psyclit or psychinfo or psycinfo or cinahl or
cinhal or science citation index or bids or cancerlit).ab.
7 (reference list$ or bibliograph$ or hand search$ or hand-search$ or manual
search$ or relevant journals).ab.
8 or/1-7
Systematic review filter - OVID Embase

1 meta analysis/
2 (metaanalys$ or meta-analys$ or meta analys$).tw.
3 systematic review/
4 (systematic$ adj3 (review$ or overview$)).tw.
5 (selection criteria or data extraction).ab. and Review.pt.
6 (cochrane or embase or psychlit or psyclit or psychinfo or psycinfo or cinahl or
cinhal or science citation index or bids or cancerlit).ab.
7 (reference list$ or bibliograph$ or hand search$ or manual search$ or relevant
journals).ab.
284 APPENDIX D
8 or/1-7
17 Appendix E: Evidence tables - Clinical studies
17.1 Evidence Table 1: Imaging options in occult hip fracture............................................................................................. 286

17.2 Evidence Table 2: Timing of surgery .................................................................................................................................. 289
17.3 Evidence Table 3: Optimal analgesia ................................................................................................................................. 307
17.4 Evidence Table 4: Anaesthesia ............................................................................................................................................ 310
17.5 Evidence Table 5: Surgeon seniority .................................................................................................................................. 313
17.6 Evidence Table 6: Displaced intracapsular fractures ..................................................................................................... 317
17.7 Evidence Table 7: Surgery – Cement versus no cement ............................................................................................... 326
17.8 Evidence Table 8: Extracapsular fixation .......................................................................................................................... 328
17.9 Evidence Table 9: Surgical approach to hemiarthroplasty ........................................................................................... 376
17.10 Evidence Table 10: Mobilisation strategies ..................................................................................................................... 378
17.11 Evidence Table 11: Multidisciplinary rehabilitation ....................................................................................................... 387
17.12 Evidence Table 12: Patient views ....................................................................................................................................... 411
18.1 Evidence Table 13: General versus regional anaesthesia ............................................................................................. 428
18.2 Evidence Table 14: Displaced intracapsular fractures ................................................................................................... 430
18.3 Evidence Table 15: Cemented arthroplasties .................................................................................................................. 434
18.4 Evidence Table 16: Multidisciplinary rehabilitation ....................................................................................................... 436
APPENDIX E 285
Abbreviations
CI Confidence interval
ITT Intention to treat analysis
LOS Length Of Stay
LR+ Positive likelihood ratio
M/F Male/female
N Total number of patients randomised
NA Not Applicable
NR Not reported
QALY Quality-Adjusted Life Years
QoL Quality of life
RR Relative risk
SD Standard Deviation
SE Standard Error
Sig Statistically significant at 5%
286 APPENDIX E
17.1 Evidence Table 1: Imaging options in occult hip fracture

Study Patients Diagnostic tools Measure of Disorders Results Comments
details
Study name: Patient group: Assessment tool under investigation: Funding:
Patients with painful hips after Sonography (HDI 5000 ultrasound device) Sensitivity 100% Not reported
Safran et al., low energy trauma (e.g. fall Bilateral hips were examined and saggital, axial
2009296,297 from a sitting or standing and coronal planes and particular attention Specificity 65% Limitations:
position) was paid to the hip joint and greater Sonographic examinations
Study design: trochanteric regions searching for fracture performed by 2
PPV 59%
Prospective cross- Inclusion criteria: lines, joint and bursal effusions and musculoskeletal
sectional study • Difficulty or inability to peritrochaneric oedema radiologists who may not
bear weight after a fall The findings were recorded before the MRI NPV 100% always be available at
Duration of • Tenderness around the examination community hospitals
follow up: hip with painful hip
LR+ 2.85
motion Reference standard: 72 hours delay before MRI
Not reported • Negative pelvic and hip MRI within 72 hours of admission on a 1.5-T was given
radiographic finding Sigma scanner or a 1.5-T Avanto scanner. Scans
LR- 0
were performed in the axial and coronal planes The time from injury to
Exclusion criteria: with a T1 weighted fast spin echo sequence admission ranged from 0
• Prior ipsilateral hip and with Short Tau inversion recovery with Prevalence 33% to 14 days (average 1.7
fractures or surgery magnitude display sequence. The scans were days)
• Contraindications to MRI performed in the axial plane from the level of
the anterior superior iliac spine to 5 cm below Notes:
All patients the level of the lesser trochanter. In the
N: 30 coronal plane, the scans were performed from An overall well conducted
Mean age (range): 73 (26-94) the symphysis pubis to the sacrum. and well reported study
M/F: 6/24 with low risk of bias
The MRI scans were read by a radiologist with
Drop outs: 0 15 years experience in musculoskeletal MRI,
who was blinded to the sonographic findings
APPENDIX E 287
Evidence tables – imaging

details
Rizzo et al., Patients whose history Sensitivity 97.3% None
1993286,286 and clinical examination bone scanning 72 hours after
suggestive of a hip admission using a technetium-99m Limitations:
Study design: fracture but whose bone scan Patients had MRI within 24
Prospective Cross radiographs were Specificity 100% hours of admission
sectional negative Reference standard: whereas bone scanning
MRI within 24 hours after admission. PPV 100 was carried out 72 hours
Inclusion/exclusion Only T1-weighted coronal spin-echo after admission
Duration of criteria: pulse sequences were obtained
NPV 95.8
follow-up: Not reported
6 months
All patients LR+ 0 Notes:
N: 62
1 patient had an initial
Mean age (range): 73 LR- 0.02 negative CT scan bit a
(26-93) positive MRI scan. CT
scanning after 6 days
M/F: 23/39 Prevalence 60 showed a positive result.
This patient has been
Drop outs: 0 considered as a false
negative in this analysis
288 APPENDIX E
Evidence tables – imaging

details
Evans et al., Elderly patients Isotope scanning Sensitivity 75% None
199487,88 admitted to hospital Tecnitium 99m, 48 hours after MRI
with hip pain after a fall scan Limitations:
Study design: and whose radiographs Relatively small patient
Prospective cross were normal or showed Reference standard: Specificity 100% numbers
sectional study a fracture of the greater MRI, 5 minute sequence of T1-
trochanter weighted coronal images. Where PPV 100 Isotope scans given 48
necessary Short tau inversion hours after the fall to avoid
Duration of Inclusion/exclusion recovery and/or T2 weighted images false positives
NPV 93
follow-up: criteria: were also obtained
Not reported Not clear who interpreted
3 months LR+ 0 the results and whether
All patients they were blind to the
N: 37 results of the reference
Mean age (range): not LR- 0.25 standard test
reported
Authors did not report any
Drop outs: 0 Prevalence 22 information on patient
demographics
Notes:
APPENDIX E 289
17.2 Evidence Table 2: Timing of surgery
Study details Patients Interventions Outcome measures Effect size Comments

Alani et al., Patient group: Return to independent Unadjusted (patients without Funding:
20084 Patients with hip fracture Group 1 living dementia): One or more authors
Early surgery. ≤48 Adjusted odds ratio Group 1: 320/375 received, in any one
Country of hours adjusted for age, sex, Group 2: 43/59 year, outside funding
study: Setting: Danderd and Huddinge prefracture walking Missing data: 22 (5%)
hospitals, Stockholm, Sweden. or grants in excess of
Sweden Group 2 ability, whether patient $10,000 from the
Late surgery. >48 was living with someone, <24 hours: 178/209 Stockholm County
Inclusion criteria:
Study design: hours ASA score, treatment ≥24 hours: 185/225 Council Research
• Patients with acute hip fracture modality, reoperation, Missing data: 22 (5%) Fund for clinical
Prospective aged 50 years or older and reason for delay of studies. No benefits
cohort surgery. <36 hours: 282/329 received from
Exclusion criteria: ≥36 hours: 81/105 commercial entities.
• Patients with a pathological Missing data: 22 (5%)
fracture and patients who arrived Limitations:
Duration of
at the hospital one calendar day Impact of
follow-up: Adjusted odds ratio:
after the time of injury. Delay >24h: 0.86 (0.45 to 1.65) NS comorbidity on
mortality (unadjusted
Hospital stay Delay >36h: 0.44 (0.21 to 0.90) P<0.05
All patients data).
Delay >48h: 0.33 (0.14 to 0.78) P<0.01
N: 744
Lost to follow up: 22 patients (missing Pressure ulcers Unadjusted:
data for return to independent living) Adjusted odds ratio Group 1: 41/646
adjusted for age, Group 2: 20/98 Additional outcomes
Age (mean +SD): 81
prefracture walking reported:
M/F: 200/544
Diagnosis of dementia: 209 (28%) ability, dementia, ASA <24 hours: 53/354 None
score, and duration of ≥24 hours: 60/345
N for time to surgery: surgery. p<0.05
≤24h = 359 Notes:
>24 = 385 <36 hours: 31/550 None
≤36 = 550 ≥36 hours: 30/194
>36 = 194 p<0.0001
290 APPENDIX E

≤48 = 646 Adjusted odds ratio:
>48 = 98 Delay >24h: 2.19 (1.21 to 3.96)
P<0.01
Group 1 Early Delay >36 hours: 3.42 (1.94 to 6.04)
No.: 646 P<0.001
No. of dropouts: not stated Delay >48 hours: 4.34 (2.34 to 8.04)
Age (mean): 81 P<0.001
M/F: 166/480 Length of hospital stay – Unadjusted:
Other factors: median (including rehab) Group 1: 15
Diagnosis of dementia: 181 (28%) Group 2: 21
Group 2 Late <24 hours: 14

No. : 98 ≥24 hours: 18
No. of dropouts: not stated p <0.001
Age (mean): 81
M/F: 34/64 <36 hours: 15
Other factors: ≥36 hours: 19
Diagnosis of dementia: 28 (29%) p <0.001
Delay due to: Length of hospital stay – Unadjusted:
Patient related (e.g. medical): 57 (58%) median (including Group 1: 13
System related (e.g. no available rehab), excluding days Group 2: 16
operating room): 41 (42%) prior to surgery p <0.01
<24 hours: 14
≥24 hours: 17
p <0.05
<36 hours: 15
≥36 hours: 18
p <0.05
APPENDIX E 291

Mortality rate – 4 Adjusted odds ratio:
months Delay >24h: 1.07 (0.67 to 1.70) NS
Adjusted odds ratio Delay >36h: 1.05 (0.63 to 1.74) NS
adjusted for age, sex, Delay >48h: 0.86 (0.44 to 1.69) NS
prefracture walking
ability, dementia and ASA
score.
292 APPENDIX E
Evidence tables – timing of surgery

Bergeron et al., Patient group: In hospital mortality All Funding:
200619,19 Patients with hip fracture Group 1 Group 1: 99/848 Not stated
Setting: Analysis of hospital Early surgery. ≤48 Group 2: 20/129 Limitations:
Country of administrative database. hours Comparison is >48h
study: With comorbidity
vs. 0-24 h time to
Canada Inclusion criteria: Group 2 Group 1: 93/600
Group 2: 20/99 surgery
• Consecutive patients aged 15 Late surgery. > 48
Study design: hours
years and older admitted with a
<24 hours: 53/354
diagnosis of fracture of the
Retrospective ≥24 hours: 60/345
proximal femur from April 1,
cohort
1993 to March 31, 2003.
Without comorbidity
• Patients with a low velocity fall
Group 1: 6/248
from a maximum of standing
Duration of Group 2: 0/30
height.
follow-up:
<24 hours: 6/169
Exclusion criteria:
Hospital stay ≥24 hours: 0/109
• A preadmission delay >24 hours,
no surgery, other associated
Adjusted Odds ratio:
injuries with Abbreviated Injury
24-48hs (vs.24h): 0.88 (0.55-1.41)
Scale of 2 or more, and inter
>48 hours (vs. 24h): 1.16 (0.64-2.13)
hospital transfers.
Postoperative length of All
All patients stay in days (median) Group 1: <24 hrs: 18
N: 977 24-48 hrs: 19
Age (mean +SD): 81.4 (32 – 104) Group 2: 28
M/F: 332/645
Comorbidity: With comorbidity
Cardiac disease: 40.1% Group 1: <24 hrs: 20
Neurologic disease and dementia: 24-48 hrs: 22
36.5% Group 2: 30
Pulmonary disease: 20.6%
Without comorbidity
APPENDIX E 293

Diabetes: 16.4% Group 1: <24 hrs: 16
Anticoagulation:6.6% 24-48 hrs: 15
Chronic renal dialysis: 2.1% Group 2: 20
Active cancer:2.1% Severe complications All
Cirrhosis: 0.3% (Cerebrovascular Group 1: 147/848
Fall occurred at: accident, cardiovascular Group 2: 40/129
Home: 58.2% complication, digestive
Nursing home: 21.5% complication – except <24 hours: 88/523
Outdoor: 19% unspecified paralytic ≥24 hours: 90/454
In-hospital: 1.2% ileus- dialysis)
Time of surgery: Adjusted Odds ratio:
<24h: 523 24-48hs (vs. 24h): 0.87 (0.58-1.29)
24-48h: 325 >48 hours (vs. 24h): 1.32 (0.79-2.20)
>48h: 129
Group 1 Early
No.: 848
No. of dropouts: not stated
Age (mean): <24 hrs: 79
24-48 hrs: 80
M/F: <24 hrs: 25%/75%
24-48 hrs: 21.5%/78.5%
Sever complications: 17.2%
Dementia: 308/848
Group 2 Late
No. : 129
Age (mean): 80
M/F: 24%/76%
Sever complications: 24.8%
Dementia: 49/129
294 APPENDIX E

Bottle et al., Patient group: Patients underwent 30 day mortality Group 1: 6366/90551 Funding:
200630,31 Patients with hip fracture one of 4 types of Group 2: 2625/24391 The unit is funded by
surgery: fixation, a grant from Dr Foster
Country of
Setting: NHS hospital trusts in England prosthetic 30 day mortality >1 day vs. ≤1 day: 1.25 (1.19 to 1.31) Ltd (an independent
study: replacement of health service
with at least 100 admissions for Adjusted Odds ratios >2 day vs. ≤2 day: 1.36 (1.29 to 1.43)
England head of femur, research
fractured neck of femur (adjusted for age, sex,
other procedure deprivation fifth and organisation).
Study design: (including non-
Inclusion criteria: comorbidity)
orthopaedic) and no
Retrospective • Patients aged ≥65 admitted with a Emergency readmission >1 day vs. ≤1 day: 1.04 (0.99 to 1.08) Limitations:
procedure recorded
cohort primary diagnosis of fractured within 28 days (adjusted >2 day vs. ≤2 day: 1.04 (0.99 to 1.10) Baseline
(medical
neck of femur admitted from their for age, sex, deprivation characteristics given
management).
own home. fifth and comorbidity) for entire cohort,
• Patients with a first hip fracture Group 1 which includes
Duration of Early surgery. <
follow-up: only were included. patients who did not
2days receive surgery.
1 year Exclusion criteria:
• Patients admitted from nursing Group 2
and residential homes Late surgery. > 2
days Additional outcomes
All patients reported:
N: 114,942 Adjusted effect of
operative delay on
Group 1 Early mortality, excess risk
No.: 90551 of death
Age (mean +SD): not stated
Group 2 Late
No. : 24391
APPENDIX E 295

Grimes et Patient group: Time from 30 day mortality Group 1: 175 Funding:
al.,2002125 Patients with hip fracture admission to Group 2: Not stated
surgery. Active medical problems: 56 Limitations:
Country of No medical problems:166
Setting: 20 hospitals in New Brunswick, No baseline data
study:
New Jersey; San Antonio, Texas; Group 1 provided
USA 30 day mortality >48-72h: 0.71 (0.45-1.10)
Philadelphia, Pennsylvania; and Early surgery
Richmond, Virginia – and represented (adjusted odds ratio) n = 3805
Study design:
university, community, and Veterans Group 2 Decubitus Ulcer >48-72h: 1.2 (0.9-1.6)
Affairs medical centers. Late surgery (adjusted odds ratio) n = 3579
Retrospective
cohort
Inclusion criteria:
• Consecutive patients with hip
Duration of fracture who were aged 60 years
follow-up: or older and who underwent
surgical repair between 1983 and
5 – 10 years 1993.
Exclusion criteria:
• Patients were excluded if they
had metastatic cancer, trauma
resulting in multiple injuries
requiring surgery, or declined
blood transfusion for religious
reasons.
• Patients with a fracture occurring
>48 hours before admission to
the hospital.
All patients
N: 8383
Lost to follow up: Not stated
296 APPENDIX E

Age (mean +SD): 80.4 ±8.6
M/F: 1751/6632
Group 1 Early (≤ 24 hours)

No.: 4578
Age (mean +SD):
60-69: 590
70-79: 1356
80-89: 1972
≥90: 3683
M/F: 895/3683
Other factors:
ASA class:
1 or 2: 1341
3: 2852
4 or 5: 385
Group 2 Late (≥ 24 hours)

No. : 3805
Age (mean +SD):
60-69: 485
70-79: 1089
80-89: 1683
≥90: 549
M/F: 858/2949
Other factors:
ASA class:
1 or 2: 974
3: 2279
4 or 5: 552
APPENDIX E 297

Lefaivre et al., Patient group: Pre-existing medical Logistic regression model Death Funding:
2009189,189 Patients with hip fracture comorbidity was (adjusted for medical 0.82 (0.42 to 1.62) None
quantified by listing comorbidity age, gender p = 0.5713
Country of the pre-injury and fracture type)
Setting: Limitations:
study: medical diagnoses Major medical complication
Canada Vancouver General Hospital 24 to 48h 0.96 (0.52 to 1.75)
by a body system
such as cardiac, p = 0.8868 Notes:
Inclusion criteria:
Study design: Odds ratio (95% CI)
All patients over the age of 65 who had pulmonary,
Minor medical complication
Retrospective been admitted with an isolated fracture autoimmune, 1.53 (1.05 to 2.22)
690 patients added to
of the proximal femur between 1998 substance the database, of
cohort dependence etc. p = 0.0257 these they were only
and 2001.
Patients were able to review the
catagorised into no Pressure sores complete medical
All patients
major comorbidity, 1.23 (0.71 to 2.12) records of 607
N: 607
those with one to p = 0.4700 patients.
M/F: 125/482
Duration of two body systems Logistic regression model Death
follow-up: with major
Delay to surgery (adjusted for medical 0.93 (0.38 to 2.33)
<24h: 245 comorbidity and comorbidity age, gender p = 0.8840
In hospital those with ≥3 body
24 to 48: 264 and fracture type)
>48: 98 systems with major Major medical complication
comorbidities. > 48h 2.21 (1.01 to 4.34)
Age: p = 0.0260
<75: 102, 76 – 85: 262
86 – 95: 212, 96 – 105: 30 Minor medical complication
106 – 115: 1 2.27 (1.38 to 3.72)
p = 0.0012
Medical comorbidities:
0: 141 Pressure sores
1 to 2: 405 2.29 (1.19 to 4.40)
≥3: 61 p = 0.0128
298 APPENDIX E

Majumdar et Patient group: Timing of surgery In hospital mortality Group 1: 160/3200 Funding:
al., Patients with hip fracture was based on the Group 2: 66/664 None
2006200,200 calendar date of
hospital admission <24 hours: 5/1046
Country of and calendar date ≥24 hours: 36/2933
Tertiary care hospitals in Edmonton, Adjusted odds ratios
study: of surgical repair.
Canada Alberta, Canada Adjusted odds ratio: compare <24h to
24 -48hr vs. <24: 0.90 (0.85-1.99) >48h time to surgery.
Inclusion criteria: Group 1
Study design: P = 0.59
Consecutive patients with hip fracture Early surgery.
during March 1994 to February 2000 Within 48 hours of
Retrospective >48hr vs. <24h: 1.30 (0.86-2) Additional outcomes
Patients aged 60 years or older admission
cohort p = 0.21 reported:
Hip fracture patients included femoral
neck, intertrochanteric, Group 2 1 year mortality Group 1: 970/3200 Type of fracture, %
subtrochanteric or subcapital Late surgery. Group 2: 219/664 with dementia,
fractures. After 48 hours of prefracture
admission <24 hours: 5/1046 comorbidities
Duration of ≥24 hours: 35/497
follow-up: Exclusion criteria: Notes:
Patients with multiple traumatic Length of stay (after Group 1: <24h: 7 (1-13)
fractures, pathologic hip fractures, or surgery) (in days, median, 24-48h: 8 (2-14)
30 days
bilateral hip fractures. with interquartile range) Group 2: 11 (0-24)
All patients
N: 3981 (3846 – had surgery)
Age (mean +SD): 82 (±8.52) Complications Group 1: 614/3200
M/F: 1154/2827 (Myocardial infarction, Group 2: 130/664
Time of surgery: heart failure, cardiac
<24h: 1048 arrhythmia, electrolytes <24 hours: 235/1046
24 – 48h: 2152 abnormal, anaemia, ≥24 hours: 509/497
>48h: 664 pneumonia, urinary tract
infection).
Group 1 Early
No.: 3200
APPENDIX E 299

Age (mean +SD): 82
M/F: 892/2308
Group 2 Late
No. : 664
Age (mean +SD): 81
M/F: 214/450
300 APPENDIX E

Moran et al., Patient group: Group 1 30 day mortality of No delay: 85/982 Funding:
2005215,215 Patients with hip fracture Early surgery. patients fit for surgery: Delay 1 day: 85/1166 Not stated
No delay, surgery p = 0.51
Country of performed in less
study: than one day of No delay: 134/1651
University hospital Nottingham No protocol for
UK admission Delay 2 day: 36/497
determining which
Inclusion criteria: patients were unfit for
Study design: Group 2 No delay: 158/1978
• All adult patients with a fracture Delay 3 day: 12/170 surgery and anaesthesia,
Late surgery.
Prospective of the femoral neck. therefore variation
Surgery after 1 day
cohort No delay: 166/2092 between clinicians.
or more from
Exclusion criteria: Delay 4 day: 4/56
admission
• Isolated femoral head fractures
and acetabular fractures
• 140 patients who did not have Notes:
Duration of
surgery were excluded Delay to surgery was most
follow-up:
frequently due to acute
All patients medical comorbidity (206
30 days
N: 2148 patients). The subgroup of
Lost to follow up: patients who were fit for
Age (mean +SD): 80 surgery is given; any delay
M/F: 684/2219 here is due to logistical
reasons.
Group 1 Early
No.: 982
Group 2 Late
No. : 1166
APPENDIX E 301

Orosz et al., Patient group: Patients enrolled as Major postoperative Adjusted OR = 0.26 ( 0.07to 0.95) Funding:
2004250,250 Patients with hip fracture early in the complications (those that p = 0.04 Grants were received
admission as pose a threat to life or from the Agency for
Country of possible (69% on or bodily functions and that Healthcare Research
Setting:
study: before the day of typically are treated with and Quality
4 hospitals in the New York City
USA surgery). parenteral medications,
Study design: metropolitan area (an academic procedures, or intensive
medical centre, an urban teaching Limitations:
Prospective monitoring e.g.
hospital, and a suburban hospital) Group 1 Baseline data given
cohort pneumonia or
Surgery within 24 arrhythmias. Data for for study arms, but
Inclusion criteria: hours not for reported
List who was patients enrolled in 1st 12
Patients with hip fracture aged 50 and separately for the
masked to months only.
over. Group 2 restricted cohort.
interventions: Mean pain scores over Group 1: 2.52
Nurses Surgery after 24
Exclusion criteria: hours the first 5 hospital days. Group 2: 2.90
identifying Data for patients enrolled
Patients aged younger than 50 years,
complications in 1st 12 months only. Additional outcomes
fractures that occurred as an inpatient,
were not aware Score from 1 (none) - 5 Difference (95% CI) = -0.38 (-0.61 to - reported:
of the study transfers from another hospital, Adjustments to odd
multiple trauma, pathological fractures, ratios were based (very severe pain). 0.16) Notes:
hypothesis, but p = 0.001
distal and femoral shaft fractures, on age, sex, nursing Restricted cohort
physicians
bilateral hip fractures, or previous home residence, Number of days of severe Group 1: 0.50 excluded patients
categorising
complications
fracture or surgery on the currently needing a proxy for pain over hospital days 1- Group 2: 0.80 who might not be
fractured site. consent, delirium 5 (assessed by asking if candidates for early
were not
on admission, they were experiencing Difference (95% CI) = -0.30 (-0.50 to - surgery because of
blinded.
All patients prefracture FIM no pain, or mild, 0.08) markedly abnormal
N: 1203 locomotion score, moderate or severe pain). clinical findings or the
Age (mean +SD): fracture type, Data for patients enrolled p = 0.007 need for additional
Duration of st
follow-up:
M/F: history of diabetes, in 1 12 months only. time for preoperative
COPD, stroke Length of stay, mean stay Group 1: 6.94 evaluation. This, the
Group 1 Early syndrome, in days and adjusted Group 2: 7.85 restricted cohort
6 months
No.: 398 dementia, cardiac odds ratio excludes patients
No. of dropouts: not stated disease, Difference (95% CI) = -0.91 (-1.81 to - admitted with
302 APPENDIX E

Age (mean +SD): 82 (9.2) hypertension, 0.01) abnormal clinical
M/F: 82/316 hospitalisation p = 0.05 findings, aortic
Delirium at admission: 10 within 6 months, FIM locomotion score at Group 1: 9.94 stenosis, dementia,
Admitted from nursing home : 63 hospital site, day 6 months (2-item Group 2: 9.97 and endstage renal
and time of subscale focusing on disease on dialysis.
Group 2 Late admission and walking and climbing Difference (95% CI) = -0.03 (-0.60 to
No. : 780 abnormal clinical stairs) 0.54)
No. of dropouts: findings. p = 0.91
Age (mean +SD): 82 (8.6)
FIM self care (6 item scale Group 1: 34.8
M/F: 147/633
of self-care activities Group 2: 35.4
Delirium at admission: 20
including bathing and
Admitted from nursing home : 90
dressing) Difference (95% CI) = -0.60 (-1.98 to
The restricted cohort is a subset of the
0.65)
groups shown above, which is
p = 0.32
described in the notes section.
FIM transferring (3 item Group 1: 15.7
scale focusing on Group 2: 15.7
transfers from the bed,
toilet and bath tub) Difference (95% CI) = 0 (-0.64 to 0.77)
p = 0.85
Dead or needing total Adjusted OR = 0.62 (0.35 to 1.08)

assistance in locomotion p = 0.09
at 6 months
APPENDIX E 303

Siegmeth et al., Patient group: Surgical treatment Mean hospital stay in Group 1: 21.6 Funding:
2005A308,308 Patients with hip fracture involved either days (95% CI) Group 2: 36.5 No benefits in any form
Setting: internal fixation (includes time spent on (5.7-16) were/will be received
Country of Peterborough District Hospital with cannulated orthopaedic ward and P value(s): <0.0001 from a commercial party
study: screws or any other hospital wards related directly or
England Inclusion criteria: hemiarthroplasty or convalescent units indirectly to the subject of
Patients with hip fracture admitted to for intracapsular until eventual discharge the article.
Study design: the Peterborough Hip fracture service fixation. Those with to a permanent place of
extracapsular residence)
Prospective Limitations:
Exclusion criteria: fractures were Return to original Group 1: 2974 (86.1%)
cohort Baseline data reported for
• Patients aged younger than 60 operated on with a residence (%) Group 2: 128 (73.6%)
6 individual groups, but
years, those treated dynamic hip screw P value(s): <0.0001
not split according to <48
conservatively and those with a or an Change in residence Group 1: 240 (6.9%)
Duration of or >48 hours delay.
pathological fracture or a fracture intramedullary nail (admitted to a more Group 2: 22 (12.6%)
follow-up: device. Outcomes not reported:
of the shaft or distal femur. dependent P value(s): <0.0007
• Patients who were delayed for accommodation) List the outcomes in which
1 year we are interested that are
any medical reason when Group 1 Mortality at 1 year Group 1: 238 (6.9%)
orthopaedic or anaesthetic staff Early surgery not reported here
Group 2 24 (13.8%)
felt that operation should be P value(s): <0.001
delayed in order to improve the Group 2 Additional outcomes
patient’s fitness for surgery Late surgery reported: N/A
All patients
Notes:
N: 3628
Lost to follow up: 2 Delay for non-medical
Age (mean +SD): 81 (8.06) reasons was because of
lack of operating theatre
Group 1 Early (≤ 48 hours) space, equipment or
No.: 3454 available staff.
Age (mean +SD):
M/F: 656/2798
304 APPENDIX E

Group 2 Late (> 48 hours)
No. : 174
Age (mean +SD):
M/F: 39/135
APPENDIX E 305

Weller et al., Patient group: Group 1 In-hospital mortality Group 1: 3509 (6.6%) Funding:
2005351,351 Patients with hip fracture Early surgery < 2 Group 2: 433 (10%) N/R
days Limitations:
Country of <24hr: 1177/20303
Setting: One aim of the study
study: Group 2 ≥24hr: 2765/37012
Inclusion criteria: was to determine
Canada Late surgery
• Patients aged over 50 years who Adjusted Odds Ratio: whether mortality
>2 days after hip fracture is
Study design: were admitted to hospital in 1 day: 1.17 (1.08-1.26)
Ontario, Canada between 1993 related to type of
2 days: 1.36 (1.23 – 1.52)
Retrospective and 1999 for surgical treatment >2 days: 1.60 (1.42 to 1.80) hospital (teaching or
of a hip fracture from the non teaching and
cohort 3 -month mortality Group 1: 7277 (13.7%)
Canadian Institute for Health urban or rural) in
Information Discharge Abstracts Group 2: 790 (18%) which the patient is
Duration of Database treated.
<24hr: 2552/20303
follow-up: ≥24hr: 5515/37012
Exclusion criteria:
1 year • Delay to surgery ≥ 7 days.
Adjusted Odds Ratio: Notes:
All patients 1 day: 1.11 (1.05 – 1.17) A modified Charlson-
2 days: 1.27 (1.17 – 1.37) Deyo index was used
N: 57,315
>2 days: 1.40 (1.27 to 1.53) to adjust for
Lost to follow up: Not stated
Age (mean +SD): Men: 77.7 ±10.2 6-month mortality Group 1: 9441 (17.8%) comorbidity. An
Woman: : 81.4 ±8.8 Group 2: 1038 (24%) algorithm was used in
M/F: 14,329/42,986 order to identify any
<24hr: 3361/20303 major complications
Group 1 Early (≤ 2 days) ≥24hr: 7118/37012 after hip fracture
No.: 52,937 surgery, including
No. of dropouts: not stated Adjusted Odds Ratio: infection deep vein
Age (mean +SD): not stated 1 day: 1.09 (1.04 – 1.15) thrombosis, intra-
M/F: not stated 2 days: 1.20 (1.12 – 1.29) operative surgical
>2 days: 1.42 (1.31 to 1.55) complications and
306 APPENDIX E

Other factors: 1-Year mortality Group 1: 12233 (23.1%) significant medical
Group 2: 1313 (30%) complications.
Group 2 Late (> 2 days)
No. : 4378 <24hr: 4366/20303
No. of dropouts: not stated ≥24hr: 9180/37012
Age (mean +SD): not stated Adjusted Odds Ratio:
M/F: not stated 1 day: 1.13 (1.05 – 1.22)
2 days: 1.26 (1.11 – 1.44)
Data given by type of hospital, not by >2 days: 1.58 (1.26 to 1.99)
delay to surgery.
APPENDIX E 307
17.3 Evidence Table 3: Optimal analgesia

Study Patients Interventions Outcome measures Effect size Comments
details
Parker et Patient group: Group 1 Pain Group 1: 106 Funding:
al.,2002262,270 Hip fracture Nerve blocks (any type, Group 2: 104 Supported internally by
subcostal, lateral SMD -0.52 (-0.8 to -0.25) Peterborough and
Study design: Inclusion criteria: cutaneous, femoral, p value: p = 0.0002 Stamford NHS
Cochrane Skeletally mature patients with a triple, psoas) Unsatisfactory pain Group 1: 18/150 (12%) Foundation Trust, UK
systematic proximal femoral fracture control preoperatively Group 2: 47/148 (31.8%) and externally by
review. The undergoing nerve blocks (including Group 2 or need for Relative risk: 0.37 Scottish Home and
review epidurals) versus no nerve blocks. no block (either systemic ‘breakthrough’ 95% CI: (0.23-0.61) Health Department, UK.
includes 17 analgesics or placebo) analgesia p value: p<0.0001
randomised
Nausea and/or Group 1: 18/141 (12.8%)
and quasi Exclusion criteria:
vomiting Group 2: 25/159 (15.7%)
randomised Not stated
Relative risk: 1.05
studies Additional outcomes:
95% CI: (0.63-1.75)
All patients Length of operation,
p value: 0.84
Setting: N (range): 888 (19-100) operative hypotension,
Hospitals in Age range: 59-86 Need for anti-emetics Group 1: 0/20 (0%) intra-operative blood
Europe, M/F: 70-95% Group 2: 5/20 (25%) gases, complications
Turkey, South Relative risk: 0.09 specific to methods of
Africa and Drop outs: 95% CI: (0.01-1.54) treatment, allergic
Israel. Most trials report 0%. 1 trial p value: not reported reactions,
reported 2% and 3 did not state the Wound infection Group 1: 0/28 (0%) cerebrovascular
number lost to follow up. Group 2: 2/27(7.4%) accident, congestive
Duration of Relative risk: 0.019 cardiac failure, renal
follow-up: 95% CI: (0.01-3.85) failure
Range: 24 p value: p= 0.14
hours-6 Pneumonia Group 1: 12/129 (9.3%) Notes:
months. Also Group 2: 25/130 (19.2%)
includes: Relative risk: 0.49
length of 95% CI: (0.26-0.94)
308 APPENDIX E

details
hospital stay p value: 0.03
and duration
of time in
Any cardiac Group 1: 3/62 (4.8%)
emergency
complication Group 2: 12/62 (19.4%)
department
Relative risk: 0.25
95% CI: (0.07-0.84)
p value: 0.02
Myocardial infarction Group 1: 1/34
Group 2: 4/34
Relative risk: 0.25
95% CI: (0.03-2.12)
p value: Not significant
Puritis Group 1: 0/20
Group 2: 5/20
Relative risk: 0.09
95% CI: (0.01-1.54)
p value:
Pulmonary embolism Group 1: 1/53 (1.9%)
Group 2: 2/52 (3.8%)
Relative risk: 0.66
95% CI: (0.11-3.86)
p value: 0.64
Deep vein thrombosis Group 1: 7/116 (6%)
Group 2: 7/137 (5.1%)
Relative risk: 1.12
95% CI: (0.43-2.93)
p value: 0.82
Mortality Group 1: 9/189 (4.8%)
Group 2: 19/205 (9.3%)
Relative risk: 0.59 d
95% CI: (0.29-1.21)
APPENDIX E 309

details
p value: 0.15
Pressure sores Group 1: 3/86 (3.5%)

Group 2: 9/106 (8.5%)
Relative risk: 0.51
95% CI: (0.11-2.39)
p value: 0.39
Confusional state Group 1: 15/77 (19.5%)
Group 2: 34/101 (33.7%)
Relative risk: 0.63
95% CI: (0.37-1.06)
p value: 0.08
310 APPENDIX E
17.4 Evidence Table 4: Anaesthesia

details
Parker et al., Patient group: Group 1 Mortality (early up to 1 Group 1: 64/912 (7%) Funding:
2004266,270 Hip fracture patients Regional (spinal or month) Group 2: 93/966 (9.6%) Supported internally
epidural) anaesthesia Relative risk: RR 0.73 University of Teesside,
Study design: 95% CI: (0.54-0.99) Middlesbrough, UK and
Cochrane Inclusion criteria Group 2 p value: 0.04 Peterborough and
systematic Skeletally mature patients General anaesthesia Mortality at 1 month Group 1: 56/811 (6.9%) Stamford Hospitals NHS
review. undergoing hip fracture surgery Group 2: 86/857 (10%) Foundation Trust,
Includes 22 Relative risk: 0.69 Peterborough, UK.
randomised Exclusion criteria 95% CI: (0.50-0.95)
and quasi Not stated p value: 0.02 Limitations:
randomised
Mortality at 3 months Group 1: 86/726 (12%)
controlled All patients
Group 2: 98/765 (13%)
trials N (range): 2567 Additional outcomes:
Relative risk: 0.92
Age range: 60-91 Length of operation,
95% CI: (0.92-1.21)
operative hypotension,
p value: 0.55
Duration of Drop outs: operative blood loss,
follow-up: 0-7%. Not stated Mortality at 6 months Group 1: 103/613 (17%) patients receiving blood
Range: 2 days Group 2: 105/651 (16%) transfusion, transfusion
to 30 months Setting: Relative risk: 1.04 requirements,
Hospitals in Europe, Hong Kong, 95% CI: (0.81-1.33) postoperative hypoxia,
New Zealand, Japan p value: 0.76 cerebrovascular
Mortality at 12 months Group 1: 80/354 accident, congestive
Group 2: 78/372 cardiac failure, renal
Relative risk: 1.07 failure, urine retention.
95% CI: (0.82-1.33)
p value: 0.61 Notes:
Length of stay in Group 1: n=108 All results reported in
hospital Group 2: n=110 this table have been
Mean Difference: -0.21 obtained using a fixed
APPENDIX E 311

details
95% CI: -5.21-4.78 effect model. Where
p value: (If no p-value: Sig/Not sig/NR) there was
Vomiting Group 1: 2/46 (4.3%) heterogeneity a
Group 2: 3/49 (6.1%) random effects model
Relative risk: 0.7 was used the results of
95% CI: (0.12-3.94) which have not been
p value: 0.68 reported here (please
refer to forest plots).
Acute confusional state Group 1: 11/117 (9.4%)
Group 2: 23/120 (19.2%)
Relative risk: 0.5
95% CI: (0.26-0.95)
p value: 0.03
Pneumonia Group 1: 21/574 (3.7%)
Group 2: 29/612 (4.7%)
Relative risk: 0.76
95% CI: (0.44-1.3)
p value:0.32
Myocardial infarction Group 1: 5/502 (1%)
Group 2: 11/531 (2.1%)
Relative risk: 0.55
95% CI: (0.22-1.37)
p value: 0.2
Pulmonary embolism Group 1: 9/605 (1.5%)
Group 2: 13/640 (2%)
Relative risk: 0.88
95% CI: (0.32-2.39)
p value: 0.8
Deep vein thrombosis Group 1: 39/129 (30.2%)
Group 2: 61/130 (36.9%)
Relative risk: 0.64
95% CI: (0.48-0.86)
312 APPENDIX E

details
p value: 0.003
APPENDIX E 313
17.5 Evidence Table 5: Surgeon seniority
Study details Patients Exposure Outcome measures Effect size Comments

Enocson et al., Patient group: Surgeon experience Number of dislocations Group 1: 37/404 (9.2%) Funding:
200885,85 Consecutive patients who had a Group 2: 8/135 (5.9%) None reported
hemiarthroplasty for non-pathological Group 1
Country of displaced femoral neck fracture Post registrar: 604
Dislocation by ‘post Odds ratio: 1.0 (0.4, 2.2) Limitations:
study: operations
registrars’ compared to P=0.9 No details about
Sweden
Setting: ‘Registrars’. Logistic surgeons and the
Group 2
Orthopaedics department regression univariate number in each
Study design: Registrar: 135
analysis group.
Historical operations
Inclusion criteria: Dislocation by ‘post Odds ratio: 1.3 (0.6, 3.0) Not reported how
cohort
• Not reported registrars’ compared to P=0.5 patients were
List who was 59 surgeons in total ‘Registrars’. Logistic allocated to
masked to Exclusion criteria: - number of regression multivariate surgeons, no mention
interventions: • None reported surgeons by grade analysis adjusted for age, of anaesthetists
Not applicable not reported sex, indication for grade/experience
All patients surgery, surgical involved in
N: 739 hips in 720 patients approach and type of operations.
Duration of No. of dropouts: not reported hemiarthroplasty
follow-up: Age (mean +SD): women: 84 (54-103) ,
men 82 (55-97) years Outcomes not
Median 2.3 (0- reported:
10) years M/F: 147/592
Mortality, length of
stay in secondary
care, reoperations,
quality of life,
functional status,
wound infection.
314 APPENDIX E
Evidence tables – surgeon seniority

Enocson et al., Patient group: Surgeon experience Number of dislocations Group 1: 38*/636 (6%) Funding:
200983,85 Consecutive patients who had a Group 2: 3*/77 (3.9%) None reported
primary total hip replacement for non- Group 1
Country of pathological displaced femoral neck Post registrar: 636
Dislocation by ‘post Hazard ratio: 1.4 (0.4, 4.5) Limitations:
study: fracture (Garden III or IV) or secondary operations
registrars’ compared to P=0.6 No details about
Sweden total hip replacement due to a fracture ‘Registrars’. Cox surgeons and the
healing complication (non-union or Group 2
regression univariate number in each
Study design: avascular necrosis) after internal Registrar: 77
analysis group.
Historical fixation. operations
cohort Dislocation by ‘post Hazard ratio: 0.9 (0.3, 2.8) Not reported how
54 surgeons in total registrars’ compared to P=0.8 patients were
Setting: ‘Registrars’. Cox allocated to
List who was - number of
Orthopaedics department
masked to surgeons by grade regression multivariate surgeons, no mention
interventions: not reported analysis adjusted for age, of anaesthetists
Inclusion criteria: sex, indication for grade/experience
Not applicable
• Not reported surgery, surgical involved in
approach and femoral operations.
Duration of Exclusion criteria: head size
follow-up: • None reported * number calculated by NCGC Outcomes not
Median 4.3 (0- reported:
11) years All patients
N: 713 hips in 698 patients Mortality, length of
No. of dropouts: not reported stay in secondary
Age (mean +SD): women: 78 +8.6 (46- care, reoperations,
96) , men 74 +9.8 (45-90) years quality of life,
M/F: 140/573 functional status,
wound infection.
APPENDIX E 315
Evidence tables – surgeon seniority

Palm et al., Patient group: Surgeon Reoperation at 6 months Group 1: 16/56 (29%) Funding:
2007256,257 Consecutive patients with proximal experience. for technical demanding Group 2: 47/309 (15%) Supported by grant
fracture of the femur. Various Number of fractures (unadjusted for P=0.015 from IMK Fonden
Country of classifications of fracture. surgeons not other factors)
study: reported. Reoperation at 6 months for Odds ratio 2.01 (1.01, 4.02)
technical demanding P=0.048 Limitations:
Denmark
All patients fractures (multivariate Not stated how
N: 600 Group 1 analysis combining age >85, patients were
Study design: female gender, ASA score III-
No. of dropouts: none Unsupervised allocated to
Prospective IV, Pre fracture New Mobility
cohort orthopaedic junior score 0-5 (poor score), time surgeons, no mention
Group 1 surgeon (<3 years to surgery >1 day from of anaesthetists
No.: 137 orthopaedic admission & type of implant grade/experience
List who was (arthroplasty or
No. of dropouts: 0 surgical experience) osteosynthesis)). involved in
masked to
interventions: Age (mean +SD): 81 (72-87) 137 operations (56 Prefracture New Mobility operations.
M/F: 12/44 Group 1: 173/309 (56%)
None classified as Score of 0-5 (scale 0f 0-9, Group 2: 21/56 (38%) Senior surgeons
Types of fracture:: technically score of 0 means patient is P=0.011 operated on
Technically demanding fractures demanding). unable do any of the significantly more
following: to get around the
Duration of o Posterior angulated Garden I-II house, get out of the house or patients with a poor
follow-up: (n=8) Group 2 go shopping. Score of 9 prefracture mobility
6 months o Garden III-IV (n=23) means the patient can do all score
Experienced 3 with no difficulty)
o Petrotrochanteric (Evans type 5) surgeon (> 3 years Number of patients Group 1: 166/309 (54%)
(n=23) orthopaedic receiving arthroplasty Group 2: 12/56 (21%) Outcomes not
o Per-/subtrochanteric (n=2) surgical experience) P<0.0001 reported:
o Subtrochanteric (n=0) 463 operations (309 Mortality, length of
o Pathological (n=0) classified as stay in secondary
technically care, requirement for
demanding. surgical revision,
Technically undemanding fractures wound infection.
o Garden I-II (n=13)
o Basocervical (n=4) Additional outcomes
o Petrotrochanteric (Evans type 1-4) reported:
316 APPENDIX E

(n=64) multivariate analysis
for age >85, female
Group 2 gender, ASA score III-
No.: 463 IV, Pre fracture New
No. of dropouts: 0 Mobility score 0-5
Age (mean +SD): 83 (77-88) (poor score), time to
M/F: 63/246 surgery >1 day from
Types of fracture: admission & type of
Technically demanding fractures implant.
o Posterior angulated Garden I-II
(n=18) Notes:
o Garden III-IV (n=176) Only technically
o Petrotrochanteric (Evans type 5) demanding fractures
(n=73) were analysed by
logistic regression.
o Per-/subtrochanteric (n=18)
o Subtrochanteric (n=20)
o Pathological (n=4)
Technically undemanding fractures

o Garden I-II (n=43)
o Basocervical (n=11)
o Petrotrochanteric (Evans type 1-4)
(n=100)
APPENDIX E 317
17.6 Evidence Table 6: Displaced intracapsular fractures

Parker et al., Patient group: Group 1 Outcomes extracted Results reported in forest plots for: Funding: supported
2010263,265,270 Skeletally mature patients with a Hemiarthroplasty - Mortality at 3 to 4 months, 1 year internally at
proximal femoral fracture. (cemented or & 2 to 4 years Peterborough and
Country of uncemented) - Number of reoperations Stamford Hospitals
study: - Pain – residual pain and Harris Hip NHS Trust, UK. No
Setting: Hospital
Group 2 Score for pain at 1 year external source of
Total hip - Failure to regain mobility at final funding.
Study design: replacement follow up
Systematic - Functional scores: Oxford Hip
Limitations:
review Score, Harris Hip Score, Barthel
including 6 out Additional non- Score, Hip Rating Questionnaire,
of the 19 RCTs comparative Short Form 36 physical function Outcomes not
from the review score reported:
prophylaxis:
with 734 - Self reported walking distance at
Not applicable
participants. end of study. Additional outcomes
The remaining - Quality of Life – Eq-5d index score reported: length of
RCTs were not - All medical complications surgery, hypotension
relevant to this - Length of hospital stay during surgery,
comparison. operative blood loss,
postoperative blood
transfusion, cost of
Duration of treatment, leg
follow-up: shortening, external
Average ranged rotation deformity
from 6 months
to 4 years
Notes:
318 APPENDIX E
Evidence tables – displaced intracapsular fractures

Parker et al., Patient group: Group 1 Outcomes extracted Results in forest plots for: Funding: supported
2006264,270 Skeletally mature patients with a Internal fixation - Mortality at 1 month, 3 months, 1 internally at
intracapsular proximal femoral year & 2 to 4 years Peterborough and
Country of fracture. Group 2 - Number of reoperations split into Stamford Hospitals
study: a. Hemi- major, moderate, minor and total NHS Trust, UK. No
arthroplasty number of reoperations external source of
Setting: Hospital - Pain at 1 year and 2 to 3 years
b. total hip funding.
Study design: replacement - Failure to return to same place of
12 trials involving 1973 participants
Systematic residence by final follow up
review compared internal fixation to - Failure to regain mobility at final Limitations:
hemiarthroplasty.
including 17 Additional non- follow up
RCTs with 2694 comparative - All medical complications Outcomes not
6 trials involving 881 participants
participants. prophylaxis: - Length of hospital stay reported:
compared internal fixation to total hip
replacement. Not applicable
Additional outcomes
The numbers do not add up to 17 trials reported: length of
Duration of surgery, hypotension
follow-up: and 2694 participants as: 1 trial of 409
patients was not included in our during surgery,
Average ranged operative blood loss,
analysis as it did not distinguish
from 1 to 13 postoperative blood
years between hemiarthroplasty and total hip
replacement; and two trials transfusion, cost of
investigated a three way comparison of treatment, leg
internal fixation, hemiarthroplasty and shortening, external
total hip replacement . rotation deformity
Notes:
APPENDIX E 319

Frihagen et al., Patient group: Group 1 Mortality at 30 days Group 1: 7/112 Funding: Norwegian
2007102,103 Patients with a intracapsular femoral Closed reduction Group 2: 10/110 Foundation for
neck fracture with angular and internal P value(s): 0.42 Health and
Country of displacement in either radiographic fixation with two Mortality at 90 days Group 1: 16/112 Rehabilitation
study: plane. parallel cannulated Group 2: 20/110 through the
Norway screws (Olmed, P value(s): 0.43 Norwegian
DePuy/Johnson and Mortality at 12 months Group 1: 24/112 Osteoporosis Society
Study design: Inclusion criteria
Johnson, Sweden) Group 2: 29/110 and the Norwegian
RCT - age >60 Research Council,
P value(s): 0.39
- ability for independent ambulation Nycomed, Smith and
List who was before fracture Group 2 Mortality at two years Group 1: 39/112 Nephew, and
masked to Charnley-Hastings Group 2: 39/110 OrtoMedic
- displaced femoral neck fracture
interventions: bipolar cemented P value(s): 0.92
Investigators of hemiarthroplasty Any medical Group 1: 28/111
Exclusion criteria Limitations:
functional (DePuy/Johnson complication Group 2: 30/109
outcomes were - unfit for arthroplasty according to and Johnson, P value(s): 0.70 Functional outcome
blinded to anaesthesiologist ata not available for
Sweden). Total number of Group 1: 70/111
interventions. - previous symptomatic hip all patients.
reoperations at 24 Group 2: 13/108
Unclear if pathology such as arthritis months P value(s): <0.001
anyone else - pathological fracture Outcomes not
was masked to Total number of hips Group 1: 47/111
- delay of more than 96 hours from reported: length of
the intervention with any reoperation at Group 2: 11/108
injury to treatment superspell, place of
after 24 months P value(s): <0.001
residence 12 months
randomisation. - living outside hospital’s designated Total number of hips Group 1: 44/111 after fracture, pain
area with major reoperation Group 2: 11/108
at 24 months P value(s): <0.001
Setting: Hospital Additional outcomes
Duration of Length of hospital stay Group 1: 8.2 +7.35 (n= 111)
reported: time from
follow-up: (mean +SD) Group 2: 10.2 +11.95 (n= 109)
All patients admission to surgery,
24 months P value(s): 0.14
N: 222 time in operation
Harris hip score (mean Group 1: 59.6 +19.5 (n= 89) theatre, time of
No. of dropouts: 0
+SD) at 4 months Group 2: 67.7 +15.8 (n= 84) surgery,
P value(s): 0.003
320 APPENDIX E

Group 1: internal fixation Harris hip score (mean Group 1: 65.8 +15.9 (n= 87) intraoperative blood
No. randomised: 112 +SD) at 12 months Group 2: 72.6 +17.5 (n= 74) loss, main surgeons
No. of dropouts: 0 P value(s): 0.01 with >3 years
Mean age (SD): 83.2 (7.65) Harris hip score (mean Group 1: 67.3 +15.5 (n= 71) experience with
M/F: 25/87 +SD) at 24 months Group 2: 70.6 +19.1 (n= 68) procedure, spinal
Other factors: P value(s): 0.26 anaesthesia, no.
Concurrent symptomatic medical receiving blood
Eq-5d index score (mean Group 1: 0.53 +0.29 (n= 79)
disease: 52 transfusion while
+SD) at 4 months Group 2: 0.61 +0.30 (n= 70)
Previously recognised cognitive failure: admitted,
P value(s): 0.06
40 postoperative
Ability to walk without any aid: 67 Eq-5d index score (mean Group 1: 0.56 +0.33 (n= 70) confusion, cognitive
Mean time from injury to admission: 8 +SD) at 12 months Group 2: 0.65 +0.30 (n= 62) failure at 4 months,
hours P value(s): 0.07 type of reoperation
Eq-5d index score (mean Group 1: 0.61 +0.31 (n= 52)
Group 2: hemiarthroplasty +SD) at 24 months Group 2: 0.72 +0.23 (n= 52) Notes:
No. randomised: 110 P value(s): 0.03
No. of dropouts: 0 Eq-5d visual analogue Group 1: 53 +18.5 (n= 69)
Mean age (SD): 82.5 (7.32) scale (mean +SD) at 4 Group 2: 62 +21.0 (n= 60)
M/F: 32/78 months P value(s): 0.01
Other factors:
Concurrent symptomatic medical Eq-5d visual analogue Group 1: 57 +21.6 (n= 59)
disease: 64 scale (mean +SD) at 12 Group 2: 63 +24.3 (n= 54)
Previously recognised cognitive failure: months P value(s): 0.16
29 Eq-5d visual analogue Group 1: 60 +18.0 (n= 45)
Ability to walk without any aid: 60 scale (mean +SD) at 24 Group 2: 60 +21.0 (n= 43)
Mean time from injury to admission: months P value(s): 0.84
5.5 hours No. patients with Barthel Group 1: 41/88
Index Score of 95 or 100 Group 2: 40/80
at 4 months P value(s): 0.66
No. patients with Barthel Group 1: 31/87
No. patients with Barthel Group 1: 24/69
APPENDIX E 321

Total number of Group 1: 70/111
complications at 24 Group 2: 16/108
months P value(s): <0.001
Total number of hips Group 1: 56/111
with any complication at Group 2: 16/108
24 months P value(s): <0.001
Total number of hips Group 1: 47/111
with major complication Group 2: 11/108
at 24 months P value(s): <0.001
Complications at 24 Group 1: 7/111
months – deep infection Group 2: 7/108
P value(s):
months – mechanical Group 2: 3/108
failure of internal P value(s):
fixation/non-union
months – dislocation of Group 2: 1/108
hemiarthroplasty P value(s):
months – avascular Group 2: 0/108
necrosis P value(s):
Median (range) time to Group 1: 137.5 (8-730) days (n= 111)
complication Group 2: 18 (6-730) days (n= 109)
P value(s): 0.01
322 APPENDIX E

Macauley et al., Patient group: Group 1 Mortality at 6 months Group 1: 5/23 Funding: American
2008197,197 Patients with a displaced intracapsular Hemiarthroplasty after surgery Group 2: 1/17 Association of Hip
proximal femoral fracture. (unipolar or bipolar, P value(s): 0.21 and Knee Surgeons,
Country of cemented or Orthopaedic
Mortality at mean follow Group 1: 9/23
study: uncemented stem). Research and
Inclusion criteria up of 34 months (29 to Group 2: 5/17
USA Education Foundation
42 months) P value(s): 0.53
- age >50 Group 2
Study design: - ability for independent ambulation Total hip Bodily pain at 12 months Group 1: 42.4 +11.5 (n= 23)
(SF-36 subscales 1-100) Group 2: 53.2 +10.2 (n= 17) Limitations:
RCT before fracture replacement with a
femoral head of (mean +SD) P value(s): 0.02
- displaced femoral neck fracture
List who was (Garden III or IV which the surgeon 28mm or more Pain on injured side at 12 Group 1: 88.5 +13.6 (n= 23) Outcomes not
masked to considered not amenable to (cemented or months (WOMAC 1-100) Group 2: 92.5 +14.6 (n= 17) reported:
interventions: treatment with open reduction uncemented stem). (mean +SD) P value(s): 0.50
Unclear if internal fixation (ORIF)) Additional outcomes
Bodily pain at 24 months Group 1: 44.7 +10.5 (n= 23)
anyone was reported: duration of
- ability to comprehend either (SF-36 subscales 1-100) Group 2: 54.8 +7.9 (n= 17)
masked to the operation
English or Spanish (mean +SD) P value(s): 0.03
intervention
after Pain on injured side at 24 Group 1: 77.8 +20.9 (n= 23)
randomisation. Exclusion criteria months (WOMAC 1-100) Group 2: 94.4 +6.8 (n= 17) Notes: study
- chronic severe dementia (defined (mean +SD) P value(s): 0.05 designed to
as <23 of 30 on Folstein Mini Physical function at 12 Group 1: 32.8 +10.0 (n= 23) demonstrate the
Duration of Mental State Examination (MMSE)) months (SF-36 subscales Group 2: 33.5 +12.0 (n= 17) feasibility of a large
follow-up: 1-100) (mean +SD) P value(s): 0.87 randomised,
- pathological fracture
24 months multicentre trial with
- other concomitant long bone Function at 12 months Group 1: 78.7 +16.8 (n= 23) multiple surgeons
fractures or fractures requiring (WOMAC 1-100) (mean Group 2: 75.9 +19.8 (n= 17) treating subjects with
surgical repair +SD) P value(s): 0.71 displaced
- preexisting arthritis of the Physical function at 24 Group 1: 35.1 +12.9 (n= 23) intracapsular femoral
ipsilateral hip months (SF-36 subscales Group 2: 38.6 +8.9 (n= 17) neck fractures.
1-100) (mean +SD) P value(s): 0.52
Setting: Hospital Function at 24 months Group 1: 65.1 +18.1 (n= 23)
(WOMAC 1-100) (mean Group 2: 81.8 +10.2 (n= 17)
APPENDIX E 323

+SD) P value(s): 0.66
All patients Physical component Group 1: 36.4 +9.2 (n= 23)
N: 41 summary score at 12 Group 2: 40.2 +9.9 (n= 17)
No. of dropouts: 1 (2.5%) months (SF-36 subscales P value(s): 0.35
1-100) (mean +SD)
Group 1: hemiarthroplasty
Physical component Group 1: 40.9 +12.3 (n= 23)
No. randomised: 23
summary score at 24 Group 2: 43.0 +7.5 (n= 17)
No. of dropouts: 0
months (SF-36 subscales P value(s): <0.68
Mean age (SD): 77 (9)
1-100) (mean +SD)
M/F: 9/14
Other factors: Harris Hip Score on Group 1: 80.6 +14.3 (n= 23)
Average no. comorbid conditions: 4.2 injured side at 12 months Group 2: 84.2 +12.0 (n= 17)
(1-11) (1-100) (mean +SD) P value(s): 0.55
Harris Hip Score on Group 1: 81.1 +11.7 (n= 23)
Group 2: total hip replacement injured side at 24 months Group 2: 84.0 +12.2 (n= 17)
No. randomised: 18 (1-100) (mean +SD) P value(s): 0.64
No. of dropouts: 1 TUG score (Take “Up and Group 1: 16.5 +10.1 (n= 23)
Mean age (SD): 82 (7) Go”score at 12 months Group 2: 17.2 +13.5 (n= 17)
M/F: 10/7 (mean +SD) P value(s): 0.89
Other factors:
Average no. comorbid conditions: 3.5 TUG score (Take “Up and Group 1: 16.9 +10.1 (n= 23)
(0-7) Go”)score at 24 Group 2: 14.7 +7.2 (n= 17)
months(mean +SD) P value(s): 0.64
Length of stay in hospital Group 1: 7.7 +5.5 (n= 23)
(mean +SD days) Group 2: 5.4 +2.8 (n= 17)
P value(s): 0.18
Length of stay in hospital Group 1: 7 (n= 23)
(median days) Group 2: 6 (n= 17)
324 APPENDIX E

Mouzopoulos et Patient group: Group 1 Mortality at 1 year Group 1: 6/43 Funding: not
al., 2008218,218 Patients with a displaced subcapital hip Internal fixation Group 2: 6/43 reported
fractures (Garden III or IV) (Richards plate Group 3: 5/43
Country of screw, Smith & P value(s):
Limitations: method
study: Nephew, Memphis, Mortality at 4 years Group 1: 15/43
Inclusion criteria of randomisation
Greece TN, USA) Group 2: 13/43
- displaced femoral neck fracture unclear study:
Group 3: 11/43 patients assigned in
Study design: (Garden III or IV) P value(s):
Group 2 order of type of
RCT Hemiarthroplasty Prefracture function Group 1: 85.2 +4.8 (n= 43) fixation:
Exclusion criteria (Merete, Berlin, according to the Barthel Group 2: 81.05 +8.95 (n= 43) hemiarthroplasty,
List who was Germany).
- previous hip surgery Index Score Group 3: 87.4 +17.4 (n= 43) total hip
masked to
- history of cancer or Paget’s disease Function according to the Group 1: 77.1 +7.1 (n= 32) replacement, internal
interventions: Group 3
- rheumatic arthritis Barthel Index Score at 1 Group 2: 76.8 +6.8 (n= 30) fixation. No
Total hip indication that
year Group 3: 84.8 +14.8 (n= 33)
replacement (Plus; anyone was masked
Duration of Setting: Hospital DePuy, Warsaw, IN, Function according to the Group 1: 80.1 +5.3 (n= 19)
to the intervention.
follow-up: USA). Barthel Index Score at 4 Group 2: 79.6 +6.3 (n= 20)
4 years All patients years Group 3: 85.3 +11.6 (n= 23)
N: 129 Harris Hip Score at 1 year Group 1: 71.3 +5.3 (n= 32) Outcomes not
No. of dropouts: 34 at 1 year, 67 at 4 Group 2: 77.81 +9.6 (n= 30) reported:
years Group 3: 83.7 +4.8 (n= 33)
P value <0.05 for comparison between Additional outcomes
Group 1: internal fixation group 1 and 3 reported: mentions
No. randomised: 43 but provides no
Harris Hip Score at 4 Group 1: 73.6 +6.7 (n= 19)
No. of dropouts: 11 at 1 year, 24 at 4 figures for range of
years Group 2: 79.5 +6.5 (n= 20)
years passive motion, and
Group 3: 83.7 +4.8 (n= 23)
Mean age (SD): 75.38 (4.62)* walking speed.
P value <0.05 for comparison between
M/F: 12/26* Barthel Index score
group 1 and 3
Other factors: prefracture
Average no. comorbid conditions: 4.2 Number of revisions Group 1: 12/43
(1-11) Group 2: 5/43
Group 3: 1/43 Notes:
APPENDIX E 325

P value(s):
Group 2: hemiarthroplasty
No. randomised: 43
No. of dropouts: 13 at 1 year, 23 at 4
years
Mean age (SD): 74.24 (3.77)*
M/F: 10/24*
Other factors:
Average no. comorbid conditions: 3.5
(0-7)
Group 3: total hip replacement

No. randomised: 43
No. of dropouts: 10 at 1 year, 20 at 4
years
Mean age (SD): 73.07 (4.93)*
M/F: 9/28*
Other factors:
Average no. comorbid conditions: 3.5
(0-7)
* data not provided for all patients

326 APPENDIX E
17.7 Evidence Table 7: Surgery – Cement versus no cement

Parker et al., Patient group: Group 1 Outcomes extracted for Results reported in forest plots for: Funding:
2010265 Skeletally mature patients with a Cemented older designs of - Mortality at up to 1 month, 1 to 3 Not reported
Country of proximal femoral fracture. prostheses hemiarthroplasty months, 1 year & 3 years
study: - Number of reoperations at 8 to 20
UK months Limitations:
Setting: Hospital Group 2
Uncemented - Failure to regain mobility at 12 to
Study design: prostheses 17 months Outcomes not
Systematic - Change in mobility score at 12 reported:
review months
including 19 Additional non- - Length of hospital stay Additional outcomes
RCTs, 6 relating comparative - Number of patients failing to reported: length of
to cemented prophylaxis: return home at 1.5 to 5 years surgery, hypotension
stems in old - Pain at 3 months and 1 to 2 years during surgery,
Not applicable
designs of - Pain score at 6 months operative blood loss,
hemi- - Number of reoperations at 8 to 20 postoperative blood
arthroplasty months transfusion, cost of
with 899 - Deep sepsis at 1 to 5 years treatment, leg
participants, 1 - Wound haematoma at 1 to 5 shortening, external
relating to new years rotation deformity
styles of stems - All medical complications
with 220 Outcomes extracted for Results reported in forest plots for:
participants Notes:
new designs of - Mortality at 30 days, 9 days, 1
Duration of hemiarthroplasty year & 2 years Review also
follow-up: - Number of reoperations at 12 compares: different
Average ranged months types of unipolar or
from 6 months - Need for pain medication at 12 bipolar
to 4 years months hemiarthroplasties,
- Unable to walk without aids at 12 unipolar vs. bipolar
to months hemiarthroplasty,
- Functional scores: Barthel Index, uncemented
APPENDIX E 327

Harris Hip Score and Eq-5d at 12 hemiarthroplasty vs.
months total hip
- Length of hospital stay replacement,
cemented
hemiarthroplasty vs.
total hip
replacement,
different types of
total hip
replacement.
328 APPENDIX E
17.8 Evidence Table 8: Extracapsular fixation

Ahrengart et al., Patient group: 96% of patients were Additional Group 1: 5 Funding:
20023,3 Patients with intertrochanteric operated on within 2 days. fissure/fracture Group 2: 2 The Karolinska
fractured femur. perioperatively Institute Foundation,
Country of Lund University,
Group 1 Gamma nail Other Group 1: 5
study: Setting: 5 hospitals. Skane County Council
The 12mm diameter technical/surgical Group 2: 2
Sweden and and Stryker-
Gamma nail was used in problems
Finland Howmedica.
Inclusion criteria: 73%, the 14mm nail in 20% Duration of hospital Group 1: 10 (1 – 100)
Study design: • Fracture types 1 to 5 of the and the 16mm nail in 7% of stay, mean (range) Group 2: 10 (1 – 100)
Evans’ classification of patients. The proximal Outcomes not
RCT
intertrochanteric fractures, as femur was reamed to a Wound infection Group 1: 10 (1 – 100) reported:
List who was modified by Jensen and 2mm larger diameter than Group 2: 10 (1 – 100) Place of residence
masked to Michaelsen. the diameter of the nail. In
interventions: Exclusion criteria: patients with stable
Cut out of lag screw Group 1: 14 Additional outcomes
Not reported. • Subtrochanteric and fractures, distal locking was
Group 2: 4 reported:
pathologic fractures, earlier used in 68% of patients, and
fractures or operations on the in unstable fractures 74% of Radiological
Duration of parameters,
same hip, or if the surgeon patients. Mortality of 6 Group 1: 41
follow-up: operation time, blood
was unfamiliar with the months Group 2: 37
6 months loss, % of fractures
Gamma nail technique. Group 2 Compression hip
All patients Healed fracture at 6 Group 1: 89% healed in
screw preoperative
N: 492 The Richard’s classic or the months Group 2: 88%
No. of dropouts: 66 (13%) position, hip rotation
Dynamic hip screw was
Group 1: Gamma nail used. 2 hole plates were Notes:
Lateral pain over the Group 1: 27%
No. randomised: 210 used in 5%, 4 hole in 67%, 5 femoral head screw Group 2: 26% Of the 5 hospitals
No. of dropouts: hole in 20%, 6 hole in 7%, at 6 months participating in the
Mean age (range): F: 82 (48-96) and 8 or 10 in 2% of study, 1 centre was
M: 77 (44-90) Pain at the top of Group 1: 20%
patients in whom a active for 3 years,
M/F: 63/147 the greater Group 2: 6%
compression screw was whereas the others
Other factors: trochanter at 6 p<0.001
used. participated for 2
months
APPENDIX E 329

ASA score: Needs walking aid Group 1: 72% years each.
1: 16% Group 2: 70%
2: 42% Additional non- In most cases patient
3: 34% comparative prophylaxis: Lives at home Group 1: 65% who were lost to
4: 8% 81% of patients received Group 2: 62% follow up were
5: 0 antibiotic prophylaxis. absent at the final
Group 2: Compression hip screw Prophylactic anticoagulants exam due to
Internal hip rotation Group 1: 15˚ (0 – 50˚)
No. randomised: 216 were given to 75% of the advanced age, other
of the fractured leg Group 2: 15˚ (0 – 45˚)
No. of dropouts: patients; 56% received physical illness, or
Age (mean +SD): F: 81 (54-99) dextran and 18% received dementia.
M: 74 (32-98) External hip rotation Group 1: 20˚ (0 – 70˚)
heparin preparations or
M/F: 60/156 of the fractured leg Group 2: 30˚ (0 –60˚)
warfarin. Spinal anaesthesia
Other factors: p = p<0.001
ASA score: used in 90% of
1: 20% patients.
2: 39%
3: 36%
4: 6%
5: 0
330 APPENDIX E
Evidence tables – extracapsular fractures

Aune et al., Patient group: Group 1 Requirement for Group 1: 13/177 Funding:
19949,9 Patients with hip fracture All the Gamma nails reoperation − Stable trochanteric =5 femoral Not reported
Setting: Orthopaedic hospitals, Norway (Howmedica) were shaft fractures and 2 cut out of
Country of modified to a 6 the lag screw
Limitations:
study: Inclusion criteria: degree valgus − Unstable trochanteric =4
Norway femoral shaft fractures and 1 Small study, little
• Trochanteric or subtrochanteric angle, 4 degrees
less than in the cut out of the lag screw detail about
femoral fractures randomization and
Study design:
Exclusion criteria: standard nail. The − Subtrochanteric = 1 femoral
Prospective shaft fracture few outcomes
• None stated slot for the lag
randomized screw had a 131 Group 2: 2/201 reported e.g.
study − Stable trochanteric = 1 cut out mortality etc.
All patients degree angle in
N: 378 relation to the of the lag screw
List who was shaft. The − Unstable trochanteric = 1 cut Outcomes not
No. of dropouts: 0
masked to diameters of the out of the lag screw reported:
interventions: nails used were 12 − Subtrochanteric = 0 Mortality, length of
Group 1: Gamma nail
Not reported. or 14mm. The P value(s): P < 0.003 stay in hospital, place
No. randomised: 177
Mean age (range): 82 (49-96) medullary canal of residence,
Duration of was over-reamed functional status.
M/F: 66/109
follow-up: 2mm. In 119 of 177
Other factors:
Median follow- nailings distal
Stable trochanteric = 84 Additional outcomes
up was 17 locking screws were
Unstable trochanteric = 76 reported: Further
months (10-27) inserted through a
Subtrochanteric = 14 details of the 15
jig.
patients requiring
Group 2: Hip compression screw (HCS) reoperation,
No. randomised: 201 Group 2
including time from
Age (mean +SD): Hip compression
operation to
M/F: 89/114 screw (Smith and
reoperation.
Other factors: Nephew)
Stable trochanteric = 89
Unstable trochanteric = 98 Notes: Fracture type
Subtrochanteric = 17 assessed by methods
of Jenson and Zickel.
APPENDIX E 331

Barton et al., Patient group: All surgeons Reoperation (screw cut- Group 1: 2 Funding:
201014,14 Patients with fracture of the proximal performing the out, implant failure, late Group 2: 3 No external funding
femur operations had fracture, and deep P value(s): 0.67
Country of experience with the infection) (all were screw cut out)
Limitations:
study: 2 implants. Mortality 30 days
Setting: Initial power
UK Following surgery, Group 1: 11
Dept Trauma and Orthopaedics, calculation produced
Frenchay Hospital, Bristol. patients were Group 2: 21
Study design: a sample
mobilized bearing P value(s): 0.13
RCT requirement of 220
Inclusion criteria: full weight under
patients.
Patients aged over 18 with AO/OTA 31- the supervision of a 1 year
Outcomes not
List who was physiotherapist. Group 1: 24
A2 fracture of the proximal femur. reported:
masked to Following Group 2: 32
interventions: discharge, patients P value(s): 0.26
No blinding of Exclusion criteria:
Pathological fractures, previous were evaluated Length of hospital stay Group 1: 31 (1 to 154) Additional outcomes
assessor or both clinically and reported:
proximal femoral fracture, reverse Group 2: 32 (1 to 164)
patients.
oblique fractures, and a decision by the radiographically at P value(s): 0.17 Requirement for
surgeon responsible for the patient’s 3, 6 and 12 months. transfusion,
Duration of Mobility (change in score Group 1: 1.49
care not to include the patient in the – points) Group 2: 1.86 demographic
follow-up:
study. Group 1 (1 – unaided, 2 – one P value(s): 0.26 characteristics (side
1 year
Sliding hip screw cane or crutch , 3 – two of fracture, mini-
All patients (Omega 2; Stryker, canes or crutches, 4 – mental score), tip-
N: 210 Newbury, UK) walker, 5 – wheelchair) apex distance >25mm
No. of dropouts: 2 A four-hole, 135˚ Notes:
Change in residence Group 1: 1.23
Mean age (range): 83.2 (42 to 99) plate was inserted.
(change in score – points) Group 2: 1.16
M/F: 44/166
(1 – own home, 2 – P value(s): 0.79
Group 2
sheltered housing, 3 –
Group 1: Long gamma nail
residential home, 4 –
No. randomised: 110 (Dyax; Stryker)
nursing home, 5 –
No. of dropouts: 0 The femur was hospital)
Mean age (range): 83.3 (56 to 97) reamed to 1mm
M/F: 25/85 EuroQol 5D QUALY
greater than the
Other factors: ASA score Group 1: 0.46
332 APPENDIX E

1: 2, 2: 46, 3: 59, 4: 3 diameter of the Group 2: 0.37
nail, and a 130˚ nail
Group 2: of the appropriate
No. randomised: 100 length was
No. of dropouts: 2 died before surgery inserted; all nails
Mean age (range): 83.1 (42 to 99) were locked distally
M/F: 19/81 with 2 screws.
Other factors: ASA score
1: 0, 2: 47, 3: 49, 4: 4
APPENDIX E 333

Bridle et al., Patient group: Group 1 Mortality Before discharge Funding:
199137,37 Patients with intertrochanteric The Gamma nail was Group 1: 10 Not reported
fractured femur. inserted using a ‘closed’ Group 2: 9
Country of technique under image 6 months post op
Limitations:
study: intensifier control. The Group 1: 15
Inclusion criteria: Allocation
UK, London patient is positioned on the Group 2: 19
• Patients diagnosed with concealment unclear.
traction table, and the Complications Group 1 2
Study design: intertrochanteric fractured
fracture is reduced with the CVA 4 0
Randomised femur
leg adducted. A 6 cm Bronchopneumonia 1 3 Outcomes not
prospective incision is made just Pulmonary embolism 1 0 reported:
comparison Exclusion criteria:
proximal to the greater Pressure score 4 1 Length of stay in
• Not reported
trochanter, which is entered Wound infection 1 2 hospital, reoperation.
List who was using a curved awl. The Wound haematoma 0 2
masked to All patients
entry point is just lateral to Accommodation Home
interventions: N: 100 Additional outcomes
the tip of the trochanter. A Before injury Group 1: 32
Not reported. No. lost to follow up: 6 reported:
guide wire is introduced Group 2: 24
into the femoral shaft, and Operative details
Group 1: Gamma nail Non-institution
Duration of flexible reamers are used to Notes: Treatment
No. randomised: 49 Group 1: 3
follow-up: the appropriate size. A nail, was randomised at
Age: 81.0 Group 2: 8
At least 6 1 to1.5 mm smaller than the the time of
M/F: 9/40 Non-hospital institution
months final reamer, is selected. No anaesthesia.
Other factors: Group 1: 9
attempt is made to ream Group 2: 13
ASA score:
the shaft to accept a large Hospital
I =2
nail. The angle of the nail Group 1: 5
II = 23
ranges from 125 to 140 Group 2: 6
III = 20
degrees.
IV = 4 Accommodation Home
Group 2 Latest review (at Group 1: 24
Fracture type: least 6 months post Group 2: 18
Dynamic hip screws were
Stable: 18 op) Non-institution
inserted using the standard
Unstable: 31 Group 1: 2
technique.
Group 2: 4
334 APPENDIX E

Anaesthesia: Non-hospital institution
Spinal = 6 Group 1: 3
General = 43 Group 2: 15
Hospital
Group 2: Dynamic hip screw (DHS) Group 1: 11
Mean age: 82.7 Mobility (before Unaided
M/F: 7/44 injury) Group 1: 31
ASA score: Sticks
I =2 Group 1: 6
II = 22 Group 2: 16
III = 16 Frame
IV = 11 Group 1: 17
Group 2: 9
Fracture type: Non-walker
Stable: 23 Group 1: 5
Mobility (final Unaided
Anaesthesia:
review) Group 1: 7
Spinal = 7
Group 2: 11
General = 44
Sticks
Group 1: 24
Group 2: 9
Frame
Group 1: 13
Group 2: 14
Non-walker
Group 1: 13
Group 2: 3
Cut out Group 1: 2
Group 2: 3
APPENDIX E 335

Femoral shaft Group 1: 4
fracture Group 2: 0
336 APPENDIX E

Ekstrom et al., Patient group: Spinal anaesthesia was Mortality at 1 year Trochanteric Funding:
200777,77 Unstable trochanteric and used, although 13 patients Group 1: 14 Not reported
subtrochanteric fractures had general anaesthesia and Group 2: 15
Country of 1 had a combination of Subtrochanteric
study: both. Group 1: 1
Inclusion criteria: Outcomes not
Sweden Group 2: 3
Unstable intertrochanteric proximal Group 1: Proximal femoral reported:
femoral fractures and Functional outcome- Trochanteric
Study design: nail (Stratec) able to walk the 15m Group 1: 86% Length of stay in
subtrochanteric fractures.
Randomised The nail used was a 240mm test at 6 weeks Group 2: 72% hospital
prospective long nail with a 130 degree Subtrochanteric
Exclusion criteria:
comparison shaft angle. The nail was Group 1: 94%
Stable trochanteric fractures, high Additional outcomes
inserted according to the Group 2: 77%
energy trauma, pathological reported:
List who was surgical technique
fractures, previous surgery to the Functional outcome Trochanteric Operative details,
masked to recommended by the
proximal femur, daily steroids of – rise from a chair Group 1: 25% ability to climb a
interventions: manufacturer
more than 10mg of prednisolone, without arm support Group 2: 19% curb, living
Not reported.
ongoing chemotherapy, irradiation (6 weeks) Subtrochanteric conditions, union,
Group 2: Medoff sliding
treatment, presence of Group 1: 35% minor complications.
Duration of plate (Medpac)
degenerative osteoarthritis of the Group 2: 31%
follow-up: Both 4 hole and 2 hole were Notes:
injured hip. Functional outcome Trochanteric
12 months used for trochanteric
fractures, whereas only the – rise from a chair Group 1: 46%
Setting: without arm support Group 2: 40%
6 hole plates were used for
2 orthopaedic hospitals, Sweden (4 months) Subtrochanteric
subtrochanteric fractures.
The locking screw set was Group 1: 56%
All patients used in all subtrochanteric Group 2: 23%
N: 210
fractures to prevent Functional outcome Trochanteric
No. of dropouts: 25% (7 exclusions
compression along the – rise from a chair Group 1: 50%
made: 5 wrong fracture and 2
femoral neck. No locking set without arm support Group 2: 53%
wrong treatment). screw was used in the (12 months) Subtrochanteric
trochanteric fractures. Group 1: 60%
Group 1: Proximal femoral nail
Group 2: 50%
No. randomised: 105
Pain while walking Trochanteric
APPENDIX E 337

No. of dropouts: 0 All patients received at 6 weeks (assesses Group 1: 30
Age (SD): 82 (48-96) preoperative iv antibiotics using a visual Group 2: 30
M/F: 24/76 with 2g of cloxacillin. analogue scale – VAS Subtrochanteric
Other factors: Subcutaneous low 0-100) Group 1: 30
Trochanteric =86 molecular weight heparin Group 2: 25
Jensen-Michaelsen (JM) was used as Pain while walking Trochanteric
JM 3: 16% thromboembolic at 4 months Group 1: 20
JM 4: 10% prophylaxis for 7 days. (assesses using a Group 2: 20
JM 5: 56% visual analogue scale Subtrochanteric
Subtrochanteric = 19 – VAS 0-100) Group 1: 0
Seinsheimer (S) Group 2: 20
S3: 1%
Pain while walking Trochanteric
S4: 8%
at 12 months Group 1: 0
S5: 9%
(assesses using a Group 2: 0
visual analogue scale Subtrochanteric
Group 2: Medoff sliding plate
– VAS 0-100) Group 1: 0
No. randomised: 98
Group 2: 0.5
No. of dropouts: 0
Mean age (SD): 82 (52-97) Complications: Trochanteric
M/F: 25/75 Femoral fracture Group 1: 1
Trochanteric = 85 Subtrochanteric
Jensen-Michaelsen (JM) Group 1: 0
JM 3: 11% Group 2: 0
JM 4: 19% Complications: Trochanteric
JM 5: 57% cut out Group 1: 5
Subtrochanteric = 13 Group 2: 1
Seinsheimer (S) Subtrochanteric
S3: 5% Group 1: 1
S4: 1% Group 2: 1
338 APPENDIX E

S5: 7% Complications: Trochanteric
femoral neck Group 1: 0
fracture Group 2: 0
Subtrochanteric
Group 1: 1
Group 2: 0
Complications: Trochanteric
Non union Group 1: 0
Group 2: 1
Subtrochanteric
Group 1: 0
Group 2: 1
reoperations Trochanteric
Group 1: 6
Group 2: 1
Subtrochanteric
Group 1: 3
Group 2: 0
APPENDIX E 339

Guyer et al., Patient group: All patients were Mortality (termed 30 days Funding:
1993A127,128 Pertrochanteric and intertrochanteric operated on within 24 lethality in the study) Group 1: 4 not reported
fractures hours where possible. Group 2: 2
Country of Late lethality (not defined)
study: Group 1: Gamma nail Group 1: 4 Limitations:
Exclusion criteria:
Switzerland The greater trochanter Group 2: 5 Allocation
Not reported
was exposed after Length of stay in hospital Group 1: 30.9 concealment
Study design: standard intramedullary (excluding those who Group 2: 30.9 unclear.
Setting:
Randomised Orthopaedic hospital, Switzerland technique and the entry died in hospital)
prospective point was holed with the Outcomes not
Reoperation Group 1: 5/50
comparison All patients awl. 12 mm diameter reported:
Group 2: 6/50
N: 100 nails used in 44 cases and
List who was 14mm in 6 cases. Complications Cranial screw perforation (cut out)
No. of dropouts: 0 Additional
masked to Group 1: 1
interventions: Group 2: Dynamic hip Group 2: 3 outcomes
Group 1: Gamma nail reported:
Not reported. screw Intra op femoral fragmentation
No. randomised: 50
135˚ 4 to 12 hole plates Group 1: 1 Operative details
No. of dropouts: 10 lost to follow up
Duration of were used. Group 2: 0 including blood loss
Age (SD): 79.5
follow-up: Wound haematoma and length of
M/F: 82% women
12 weeks Group 1: 2 surgery, leg
Other factors: All patients received Group 2: 2 shortening, social
Fracture stability: prophylactic Deep wound infection situation
Pertrochanteric: cephalosporin and low Group 1: 0
Stable: 23 dose heparin. Group 2: 1
Unstable: 24
Intertrochanteric: 3 Pain during walking (12 Group 1: 19/28
weeks) Group 2: 18/32
Group 2: Dynamic hip screw
Walking capacity (12 Full
No. randomised: 50
weeks) Group 1: 4/28
No. of dropouts: 14 lost to follow up
Group 2: 6/32
Mean age (SD): 80.3
More than 1 hr
M/F: 88% women
Group 1: 13/28
Other factors:
340 APPENDIX E

Fracture stability: Group 2: 16/32
Pertrochanteric: Less than 1 hr
Stable: 19 Group 1: 11/28
Unstable: 26 Group 2: 10/32
Intertrochanteric: 5
APPENDIX E 341

Hardy et al., Patient group: Group 1: Compression hip- Mobility score Pre op Funding:
1998134,134 Trochanteric proximal femoral screw (Parker and Palmer) Group 1: 2.3 (0.8) Smith and Nephew
fractures The compression hip-screw Ability to walk Group 2: 2.4 (0.8) Richards, Memphis,
Country of with a plate was inserted indoors (SD) Tennesse
study: with a standard technique 1 month
Inclusion criteria:
Belgium, by means of a straight Group 1: 0.9 (0.6)***
Brussels Group 2: 1.9 (0.7)*** Limitations:
Exclusion criteria: lateral incision on the lateral
Allocation
• Patients aged <60, aspect of the thigh, as
Study design: 6 month concealment unclear.
pathological fractures, described by Clawson*. The
Randomised barrel of the plate was at a Group 1: 1.5 (1.1)
prospective incorrect anatomy, history of Group 2: 1.9 (1.0)
fracture or operation 135 degree angle in each Outcomes not
comparison patient. reported:
involving same limb. 12 month
Reoperations, length
List who was Group 2: Intramedullary hip Group 1: 1.6 (1.2)
All patients of stay in hospital.
masked to screw Group 2: 1.9 (1.0)
interventions: N: 100 *** p<0.01
No. of dropouts: 0 A cannulated intramedullary
Not reported. Mobility score Pre op Additional outcomes
nail with a 4 degree
(Parker and Palmer) Group 1: 2.1 (2.3) reported:
Group 1: Compression hip-screw mediolateral bend to allow
Duration of insertion through the Ability to walk Group 2: 3.0 (2.6) Operative data e.g.
follow-up: No. randomised: 50
greater trochanter. The nail outdoors (SD) time, blood loss.
At least 6 No. of dropouts: 0
is 21 cm long and available 1 month Sliding of lag screw.
months Age (SD): 79.5 (±10.7)
M/F: 15/35 in 3 diameters (12, 14 and Group 1: 0.3 (0.7)**
Other factors: 16 mm). The opening for Group 2: 0.7 (0.9)** Notes:
Fracture stability: the lag-screw is available in The fractures healed
Stable: 16 2 angles (130 and 135 6 month
in all but one of the
Unstable: 34 degrees). It can be locked Group 1: 1.7 (2.2)*
seventy patients who
with one or 2 4.5 mm Group 2: 2.7 (2.1)*
were still alive at 12
ASA score: diameter interlocking months. The one
I =5 screws. A keyed centering 12 month
non-union was in a
II = 13 sleeve, which is held by a Group 1: 1.7 (2.2)*
patient who had a
III = 18 set-screw, passes through Group 2: 2.8 (2.2)*
compression hip
* p=0.05
342 APPENDIX E

IV = 13 the intramedullary nail and ** 0.01<p<0.05 screw.
V =1 over the lag-screw. The *** p<0.01
sleeve helps to prevent Perioperative Bronchpneumonia:
Jenson index rotation while allowing the complications Group 1: 6
1 = 10 lag-screw to slide freely. Group 2: 4
2=7 Cardiac failure
3=7 Piritramide was Group 1: 5
4 = 26 administered Group 2: 7
postoperatively and Mortality 3 months
Anaesthesia: paracetamol given in the Group 1: 12/50
Spinal: 36 recovery period. Group 2: 13/50
General: 14
Patients were permitted to 6 months
Group 2: Intramedullary hip screw
get out of bed and sit in a Group 1: 13/50
No. randomised: 50
chair on the second day Group 2: 13/50
No. of dropouts: 0
Mean age (SD): 81.7 (±11.8) after operation and bear full
weight on the fourth day. 1 year
M/F: 8/42
Group 1: 15/50
Other factors:
Group 2: 15/50
Fracture stability:
Stable: 13 Pain in hip whilst 3 months
Unstable: 37 walking Group 1: 7/40
(4 point scale, 1 = no Group 2: 4/37
ASA score: pain, 2 = slight pain
I =5 that does not effect 1 year
II = 12 ability tp walk, 3 = Group 1: 2/35
III = 23 moderate pain that Group 2: 2/35
IV = 10 that effects ability to
V =0 walk, 4 – severe
intractable pain
Jenson index even in bed)
1 = 11 Pain in mid portion 1 year
2 = 10 of thigh while Group 1: 2/35
APPENDIX E 343

3=5 walking, resulting in Group 2: 7/35
4 = 24 inability to walk
(4 point scale, 1 = no
Anaesthesia: pain, 2 = slight pain
Spinal: 36 that does not effect
General: 14 ability tp walk, 3 =
moderate pain that
that effects ability to
walk, 4 – severe
intractable pain
even in bed)
Cut-out Group 1: 0
Group 2: 1
*Clawson DK. Trochanteric fractures treated by the sliding screw plate fixation method. J. Trauma, 4:737-752, 1964.
344 APPENDIX E

Harrington et Patient group: Group 1: Post-op stay, days (SD) Group 1: 16.3 (7.5) Funding:
al., 2002137,137 Patients with hip fracture Compression hip Group 2:16.5 (8.8) Not reported
screw
Country of
Setting: Orthopaedic hospital, UK Mortality in hospital Group 1: 2/52 Limitations:
study: Group 2 Group 2: 4/50 Reference made to
UK Intramedullary hip
Inclusion criteria: some surgeons who
screw
Study design: • Unstable trochanteric proximal Ambulatory status at 1 Group 1: 22/33 had only used the
The nail is 21cm
Prospective femoral fractures year (retained pre injury Group 2: 19/30 IMHS on bone model
long with a 4
randomized living status) sessions.
degree valgus
study Exclusion criteria: Technical complications Group 1:
angulation and
• Patients aged <65 years, Screw cut out = 1
distal locking Outcomes not
List who was pathological fractures, previous Barrel-plate pulled off femur = 1
screws measuring reported:
masked to fractures, other fracture. Group 2:
4.5mm in diameter. Reoperation, length
interventions: • Patients with dementia who were Screw cut out = 1
A 12 mm diameter of stay in hospital,
Not reported. unable to give informed consent Intraoperative fracture propagation= 1
nail and 2 locking functional status,
were excluded Late fracture of femoral shaft = 1
screws were used pain.
Duration of for distal locking
follow-up: All patients
were used in all
1 year N: 102 Additional outcomes
patients.
No. lost to follow up: not reported reported:
Operative details,
Additional non-
ambulatory status
Group 1: Compression hip screw comparative
No. randomised:52 prophylaxis:
No. of dropouts: 0 n/a Notes:
Mean age (SD): 82.1 (8.6)
M/F: 11/41
Other factors:
ASA score
I: 4
II: 20
APPENDIX E 345

III: 17
IV: 11
V: 0
Anaesthesia:
Spinal: 34
General: 18
Group 2: Intramedullary hip screw

No. randomised: 50
No. of dropouts: 0
Mean age (SD): 83.8 (8.5)
M/F: 10/40
Other factors:
ASA score
I: 3
II: 22
III: 16
IV: 9
V: 0
Anaesthesia:
Spinal: 35
General: 15
346 APPENDIX E

Hoffman et al., Patient group: The selected device Delay to surgery (SD) Group 1: 1.9 (± 1.4) Funding:
1996147,147 Patients with hip fracture was inserted Group 2: 1.6 (± 1.1) Not reported
following a detailed Total hospital stay (SD) Group 1: 30.3 (±18.9)
Country of operative protocol Group 2: 31.4 (± 19.7)
Setting: Orthopaedic hospital, New Limitations:
study: based on the
Zealand Postoperative stay (SD) Group 1: 28.5 (±18.9) The manufacturer’s
New Zealand manufacturer’s Group 2: 29.8 (±20.1) guidelines were
Inclusion criteria: guidelines.
Study design: Postoperative CVA modified during the
Prospective • Trochanteric proximal femoral complications Group 1: 1 course of the study
randomized fractures Group 1: Ambi hip Group 2: 1 for the Gamma nail.
study • Patients aged >50 years screw
Cardiac Outcomes not
Exclusion criteria: Group 2 Gamma Group 1: 3
List who was reported:
masked to • Pathological fractures excluded nail Group 2: 2
The Gamma nail Reoperations,
interventions: functional status.
All patients was interlocked in Pressure areas
Not reported.
N: 69 all cases initially, as Group 1: 1
Duration of No. lost to follow up: none recommended, but Group 2: 0 Additional outcomes
Died before surgery: 2 after the first 5 reported:
follow-up:
Mean age: 81 years cases locking was Pneumonia Intraoperative
26 weeks
reserved for Group 1: 1 complications
Group 1: Ambi hip screw unstable fractures Group 2: 1
No. randomised:36 and in line with
Mean age (SD): 79.0 (10.4) manufacturer’s DVT
M/F: 12/24 updated Group 1: 0
Other factors: recommendation. Group 2: 1
ASA score: No cases were
Fracture union (% united) 6 weeks
II: 18 locked after patient
Group 1: 38
III: 15 number 50.
Group 2: 32
IV: 3
V: 0 Antibiotic 12 weeks
prophylaxis (IV Group 1: 79
APPENDIX E 347

Anaesthesia: cephradine – 1g) Group 2: 79
Spinal: 11 prior to induction of
General: 25 anaesthesia. 26 weeks
Group 1: 96
Fracture stability: Group 2: 96
Unstable: 12
Stable: 24 Resolution of hip pain (% 2 weeks
without pain) Group 1: 52
Group 2: Gamma nail Group 2: 48
No. randomised: 31
Mean age (SD): 83.2 (8.1) 6 weeks
M/F: 4/27 Group 1: 55
ASA score:
II: 10 12 weeks
III: 15 Group 1: 75
IV: 5 Group 2: 37
V: 1
26 weeks
Anaesthesia: Group 1: 71
Spinal: 6 Group 2: 60
General: 25
intra-operative fracture Group 1: 0
Fracture stability:
Group 2: 3
Unstable: 10
Stable: 21
348 APPENDIX E

Leung et al., Patient group: Group 1: Gamma nail Mortality 4 weeks Funding:
1992191,191 Trochanteric proximal femoral 2mm Kirschner wire passed Group 1: 7 Not reported
fractures percutaneously, anterior to Group 2: 5
Country of the femoral shaft and
study: parallel to the femoral neck. 6 months Limitations:
Inclusion criteria: Allocation
Hong Kong 6 to 8cm incision made Group 1: 13
Patients over 65 years with above the tip of the greater Group 2: 15 concealment unclear.
pertrochanteric fractures (including
Study design: trochanter and then the Mean duration of Group 1 n = 93 (30 stable, 63 unstable)
subtrochanteric extensions).
Randomised medullary canal is entered. hospital stay (acute Group 2 n = 93 (20 stable, 73 unstable) Outcomes not
prospective The cavity is reamed 1mm hospital) in days (SD) reported:
Exclusion criteria:
comparison larger than the diameter of Stable
• Pure subtrochanteric the intended nail. The nail is
fractures were excluded. Group 1: 9.2 (6.43)
passed into the canal, Additional outcomes
Setting: Group 2: 10.7 (6.27)
Duration of without hammering, and reported:
follow-up: Orthopaedic hospitals the corresponding device is Operative details
Unstable
7 months assembled on the nail Group 1: 9.5 (3.38) intra-operative
All patients mount and the lateral complications, mean
N: 225 (226 fractures) Group 2: 9.6 (4.46)
cortex of the femur is sliding of lag screws,
No. of dropouts: 0 Mean duration of Stable
perforated by the awl and shortening, external
hospital stay Group 1: 17.7 (11.97)
the lag screw guide wire is rotation.
Group 1: Gamma nail (convalescent Group 2: 15.4 (10.86)
inserted. Distal locking is Notes:
No. randomised: 93 hospital) in days (SD)
indicated for unstable
No. of dropouts: 0 Unstable
fractures.
Age (SD): 80.9 (±8.41) Group 1: 15.9 (8.2)
M/F: 25/68 Group 2: 19.1 (10.34)
Group 2: Dynamic hip
Other factors: screw mean time to full Stable
ASA grade Inserted using the standard weight bearing (SD) Group 1: 1.3 (0.88)
1:15 technique. Group 2: 1.9 (0.89)
2:47
3:23 Unstable
4:8 Group 1: 1.2 (0.64)
Fracture stability: Group 2: 1.7 (0.76)
APPENDIX E 349

Stable: 30 Postoperative Stable
Unstable: 63 mobility Independent
Group 1: 12 (40%)
Group 2: 8 (40%)
Group 2: Dynamic hip screw Aided
No. randomised: 93 Group 1: 11 (36.7%)
No. of dropouts: 0 Group 2: 11 (55%)
Mean age (SD): 78.3 (±9.46) Chair/bed bound
M/F: 30/63 Group 1: 7 (23.3%)
ASA grade Group 2: 1 (5%)
1:10
2:42 Unstable
3:38 Independent
4:3 Group 1: 22 (34.9%)
Fracture stability: Group 2: 23 (31.5%)
Stable: 20 Aided
Unstable: 73 Group 1: 36 (57.1%)
Group 2: 42 (57.5%)
Chair/bed bound
Group 1: 5 (8%)
Group 2: 8 (11%)
Pain in hip Stable
Group 1: 8 (26.7%)
Group 2: 5 (25%)
Unstable
Group 1: 14 (22.2%)
Group 2: 27 (40%)
Pain in thigh Stable
Group 1: 4 (13.4%)
Group 2: 5 (25%)
Unstable
350 APPENDIX E

Group 1: 7 (11.1%)
Group 2: 3 (4.1%)
Non union Stable
Group 1: 1
Group 2: 0
Unstable
Group 1: 0
Group 2: 0
Postoperative Infection
complications Group 1: 1
Group 2: 3
Superior cutting out

Group 1: 2
Group 2: 3
Fracture of shaft
Group 1: 2
Group 2: 0
APPENDIX E 351

Little et al., Patient group: A standard Mortality 30 day Funding:
2008195,195 Patients with hip fracture operative Group 1: 7/92 (7.6%) Not reported
Country of technique either Group 2: 6/98 (6.1%)
study: recommended by
Inclusion criteria: Limitations:
England manufacturer’s 1 year
Patients presenting to the Accident and
guidelines or as Group 1: 16/92 (17.4%)
Study design: Emergency department with an Group 2: 17/98 (17.3%) Outcomes not
extracapsular intertrochanteric fracture detailed in
Prospective previous studies Time to frame in days Group 1: 3.6 (3.3 to 3.9) reported:
randomized was used. (95% CI) Group 2: 4.23 (3.9 to 4.8) Reoperation, length
study Exclusion criteria:
Patients with subtrochanteric extensions of p = 0.012 of stay in hospital,
Patients with wound Group 1: 5 (5.4) pain
the fracture were excluded. Group 1: Holland
List who was
nail (long infections (%) None were Group 2: 10 (10.2)
masked to
All patients trochanteric-entry reopened and all healed p = 0.286 Additional outcomes
interventions:
N: 190 intramedullary within 6 weeks reported:
Not reported.
No. lost to follow up: 0 mail) Mobility at 1 year (95% Group 1: 5.9 (5.3 to 6.5) Intra-operative
Mean age: 83.4 (50 to 102) CI) Group 2: 3.8 (3.3 to 4.3) variables
Duration of
follow-up: p <0.001
Group 1: Holland nail Group 2 Gamma
1 year Patients with mobility Group 1: 49 (64) Notes:
No. randomised: 92 nail
restored at 1year (%) Group 2: 30 (37) 2 implant failures in
Mean age (range): 82.6 (54 to 102)
p <0.001 group II. The proximal
M/F: 8/84 Each patient was
ASA score: screws migrated
given a single-dose
1= 2 (2.2%), 2= 57 (62.0%) laterally in 4 patients
antibiotic
3= 33 (35.8%), 4= 0 in group I.
teicoplanin and
gentamicin
Group 2: Dynamic hip screw induction.
No. randomised: 98
Mean age (range): 84.2 (50 to 98)
M/F: 20/78
ASA score:
1= 3 (3.1%), 2= 55 (56.1%)
3= 37 (37.7%), 4= 3 (3.1%)
352 APPENDIX E

Miedel et al., Patient group: Group 1: Standard gamma Technical failures Trochanteric Grp1 Grp2 Funding:
2005214,214 Unstable trochanteric and nail No complication 87 91 Grants received from
subtrochanteric proximal femoral Penetration of 3 4 the Trygg-Hansa
Country of fractures Diameter 11mm, length lag screw Insurance company,
study: 200mm, valgus bend 10˚, Redisplacement/ 0 1 the Swedish
Sweden neck angle 125 or 130˚ medialisation Orthopaedic
Inclusion criteria: (Stryker Howmedica, intra-operative 3 0
• Acute unstable trochanteric Association and, in
Study design: Malmo, Sweden). Nails femoral fracture equal parts from
Randomised (J-M type 3-5) or were inserted by hand and Deep infection 0 1
subtrochanteric fractures Stryker Howmedica
prospective not by hammering and not and Swemac.
comparison after a simple fall. to use the awl before Subtrochanteric Grp1 Grp2
drilling for the distal locking No complication 16 10
List who was Exclusion criteria: screw. Penetration of 0 0 Outcomes not
masked to • Pathological fractures, lag screw reported:
interventions: rheumatoid arthritis or Group 2: Medoff sliding Redisplacement/ 0 2 Mortality, length of
Not reported. osteoarthritis were excluded. plate medialisation stay in hospital, place
• Fractures extending more Neck angle 135˚, 6 hole intra-operative 0 0 of residence, pain.
Duration of than 5cm distal to the lesser plate (Swemac, Linkoping, femoral fracture
follow-up: trochanter were excluded. Sweden). Used in the biaxial Deep infection 0 1 Additional outcomes
12 months dynamisation mode, which reported:
All patients allows sliding along both the Reoperation Trochanteric Some outcomes
N: 217 femoral neck and shaft. Group 1: 3 grouped together
Lost to follow up: 3
Group 2: 6 (e.g. not reported
Group 1: Standard gamma nail All patients were given low- separately for
molecular weight heparin Subtrochanteric trochanteric and
No. randomised: 109
before and for Group 1: 0 subtrochanteric) such
No. of dropouts: 0
approximately 10 to 14 days Group 2: 3 as length of stay in
Age (SEM): 84.6 (±0.6)
M/F: 17/92 before operation and one hospital, HRQOL
Other factors: dose of cefuroxim before (EQ0-5D), operative
Fracture type: operation. data, pain
Trochanteric 93 Notes:
APPENDIX E 353

J-M 3: 12
J-M 4: 28
J-M 5: 53
Subtrochanteric 16
S2B: 1
S2C: 11
S3A: 3
S3B: 1
S4: 0
S5: 0
Group 2: Medoff sliding plate

No. randomised: 108
No. of dropouts: 0
Mean age (SEM): 82.7 (±0.6)
M/F: 24/84
Other factors:
Fracture type:
Trochanteric 96
J-M 3: 11
J-M 4: 24
J-M 5: 61
Subtrochanteric 12
S2B: 0
S2C: 6
S3A: 2
S3B: 1
S4: 1
S5: 2
354 APPENDIX E

O’Brien et al., Patient group: The standard Length of hospital stay, Orthopaedic ward Funding:
1995244 Patients with hip fracture operative technique range (median), days Group 1: 4 – 102 (16) Not reported
for fracture fixation Group 2: 3 – 52 (14)
Country of was followed.
study: Total hospital stay
Canada Group 1: 4 – 108 (18) Mortality rate could
Inclusion criteria: Group 1: Dynamic Group 2: 3 – 92 (16) be higher as the
Study design: • Patients with intertrochanteric hip screw number of people
fractures of the femur The 135 degree Early (in hospital) local Superficial wound infection
Prospective lost to follow up is
four hole DHS was complications Group 1: 1
randomized unclear
Exclusion criteria: used more than Group 2: 0
study Outcomes not
• Fractures more than 1 week old 80% of the time in reported:
Wound haematoma
List who was • Pathological fractures this group.
Group 1: 0 Functional status,
masked to • Subtrochanteric fractures.
Group 2 Gamma Group 2: 1 place of residence,
interventions:
All patients nail
Not reported. Malalignment
N: 101 130 or 135 degree Additional outcomes
nails were used Group 1: 0 reported:
Duration of (102 fractures)
86% of the time. Group 2: 1
follow-up: No. lost to follow up: 18% Blood loss and fluid
88% of nails were replacement., length
52 weeks Early failure of fixation
Group 1: Dynamic hip screw distally locked. of surgery, early (in
Group 1: 0
No. randomised: 49 hospital) general
Group 2: 2
Mean age (range): 77 (39 to 94) All but 4 patients complications
M/F: 17/32 received
Intraoperative fracture
Other factors: prophylactic Notes:
Group 1: 0
Fracture stability: antibiotic coverage
Group 2: 2
Unstable: 21 with cefazolin
Stable: 28 intravenously.
Neuropraxia
Group 1: 2
Group 2: Gamma nail
Group 2: 0
No. randomised: 53
Mean age (range): 83 (57 to 95)) Late local complications Failure of fixation
Group 1: 1
APPENDIX E 355

M/F: 9/43 Group 2: 1
Other factors:
Fracture stability: Femoral shaft fracture
Stable: 30 Group 2: 1
Varus malunion
Group 1: 3
Group 2: 5
Complications requiring Varus collapse with pain
reoperation Group 1: 0
Group 2: 2
Varus collapse with malunion

Group 1: 1
Group 2: 0
Failure of fixation (cut-out)

Group 1: 1
Group 2: 2
Femoral shaft fracture

Group 1: 0
Group 2: 1
Mortality (early Group 1: 1
postoperative) Group 2: 6
356 APPENDIX E

Ovesen et al., Patient group: Group 1: Dynamic Mortality 4 months Funding:
1996251,251 Patients with hip fracture hip screw (DHS) Group 1: 3/66 Not reported
The use of a Group 2: 3/67
Country of trochanteric
study: stabilizing plate in 12 months
Orthopaedic hospital, Odense, Surgeon experience
Denmark combination with Group 1: 3/56
Denmark Group 2: 3/59 may cause bias as
the DHS was operations were by
Study design: allowed, but only Reoperation by 12 Group 1: 6
Inclusion criteria: surgical team on call
Prospective used in 2 patients. months Group 2: 12
randomized • Patients with Intertrochanteric – 49 surgeons
fractures having given informed Walking aids pre fracture Sticks, crutches or no walking aid participated in the
study Group 2 Gamma
consent. Group 1: 50 trial.
nail Group 2: 50
List who was Outcomes not
Exclusion criteria: The distal femur
masked to was reamed 13 mm reported:
interventions: • Subtrochanteric or pathological Walking frame or wheelchair
and the proximal Group 1: 22 Place of residence,
Not reported. fractures
femur to 18 mm. pain
• Secondary exclusions included Group 2: 22
wrong diagnosis and transfer to The use of a
Duration of hammer during
hospitals outside the inclusion Missing or deceased Additional outcomes
follow-up: insertion was
area. Group 1: 1 reported:
1 year avoided. Group 2: 1 Intraoperative details
All patients Additional non-
N: 150 comparative p = 0.41
Notes:
(101 fractures) prophylaxis: Walking aids at discharge Sticks, crutches or no walking aid
No. lost to follow up: 17% Prophylaxis against Group 1: 22
DVT and pulmonary Group 2: 13
Group 1: Dynamic hip screw embolism
No. randomised: 73 consisting of Walking frame or wheelchair
Mean age (sd): 78.5 (±11.7) Enoxaparine 40 mg Group 1: 47
M/F: 21/52 once daily starting Group 2: 59
Other factors: at admission until
lost to follow up = 4 mobilisation, Missing or deceased
APPENDIX E 357

ASA score: discharge or for 7 Group 1: 4
1 = 19 days. Antibiotic Group 2: 1
2 = 18 prophylaxis was
3 = 26 also given. p = 0.03
4 = 10 Walking aids at 4 months Sticks, crutches or no walking aid
Group 1: 43
Group 2: Trochanteric gamma nail Group 2: 37
No. randomised: 73
Mean age (sd): 79.9 (±10) Walking frame or wheelchair
M/F: 20/53 Group 1: 23
lost to follow up = 11
ASA score: Missing or deceased
1 = 20 Group 1: 7
2 = 21 Group 2: 6
3 = 25
4=7 p = 0.14
Complications requiring Group 1:
reoperation Cut- out = 2
Redislocation = 3
Femoral fracture = 0
Infection = 1
Haematoma = 0
Group 2:
Cut- out = 7
Redislocation = 0
Femoral fracture = 2
Infection = 2
Haematoma = 1
358 APPENDIX E

Pajarinen et al., Patient group: All operations were Mean hospitalisation Group 1: 5.4 (3) Funding:
2005 Patients with hip fracture performed within 2 time in days (sd) Group 2: 6.1 (3.3) Not reported
(Also Pajarinen days of admission,
2004)254,255 in most cases by a p = 0.251
Orthopaedic hospital, Helsinki, Finland senior orthopaedic Discharged to (%) Own home
Country of resident. Group 1: 4 (7.4)
study: Group 2: 6 (11.1) Outcomes not
Inclusion criteria: reported:
Finland
Study design: • Low energy extracapsular Standard operative
Nursing home Pain
Prospective pertrochanteric femoral fractures techniques, which
are recommended Group 1: 2 (3.7)
randomized Group 2: 1 (1.9)
Exclusion criteria: by the Additional outcomes
study
• Pathological fractures, multiple manufacturers and reported:
have been Rehabilitation hospital
List who was injuries, and those unable to give Intraoperative
informed consent were excluded. described in detail Group 1: 48 (88.0)
masked to details, radiographic
in instruction Group 2: 45 (83.3)
interventions: findings at 4 months
All patients manuals or earlier post-op.
Not reported. Died at our hospital
N: 108 studies were used.
No. lost to follow up: 15 (14%) Group 1: 0
Duration of Group 2: 3.7 (0.495) Notes:
follow-up: Group 1: Dynamic
4 months Group 1: Dynamic hip screw hip screw (DHS)
No. randomised: 54 Place of residence at 4 Own home
Mean age (sd): 80.3 (±10.8) months (%) Group 1: 22 (53.7)
Group 2 Proximal
M/F: 14/40 Group 2: 24 (57.1)
femoral nail
Other factors: p = 0.827
ASA score: Nursing home
2=8 Intravenous Group 1: 6 (14.6)
3 = 32 antibiotic Group 2: 10 (23.8)
4 = 14 prophylaxis was p = 0.405
given. Patients Institution
Anaesthetic: were also treated Group 1: 13 (31.7)
General = 2 with a low- Group 2: 8 (19.0)
molecular weight p = 0.214
APPENDIX E 359

Spinal = 52 heparin during their Recovery of abilities to Yes
stay in hospital pre-op status (%) Group 1: 32 (78)
Group 2: Proximal femoral nail Group 2: 34 (81)
No. randomised: 54
Mean age (sd): 80.9 (±9.1) No
M/F: 13/41 Group 1: 9 (22)
Other factors: Group 2: 8 (19)
ASA score: p = 0.791
2=6 Walking ability (%) No aids needed
3 = 28 Group 1: 12 (29.3)
4 = 20 Group 2: 15 (35.7)
p = 0.641
Anaesthetic: In need of aids, but independent
General = 3 Group 1: 22 (53.7)
Spinal = 51 Group 2: 24 (57.1)
p = 0.827
In need of assistance
Group 1: 7 (17.1)
Group 2: 3 (7.1)
p = 0.194
Recovery of walking Yes
ability to pre-op status Group 1: 22 (53.7)
(%) Group 2: 32 (76.2)
No
Group 1: 19 (46.3)
Group 2: 10 (23.8)
p = 0.040
Drop out patients Fracture redisplacement (reoperation)
Group 1: 2
Group 2: 2
p = 1.00
Died before follow up was complete
360 APPENDIX E

Group 1: 2
Group 2: 4
p= 0.678
Did not attend final review
Group 1: 9
Group 2: 6
p = 0.578
APPENDIX E 361

Park et al., Patient group: Mean time to union Systemic Funding:
1998258,258 Patients with hip fracture (weeks) Group 1: 14.3 Not reported
Group 1: Gamma Group 2: 15.1
Country of Asia Pacific nail p = 0.06
study: (GAPN)
University Hospital, Korea Unclear allocation
Korea These were Stable
Group 1: 14.28 concealment.
Inclusion criteria: inserted using a
Study design: closed technique Group 2: 14.55
Intertrochanteric fractures of the
Prospective under image p = 0.73 Outcomes not
randomized femur.
Patients aged 60 and over intensifier control. reported:
study Unstable Pain, place of
Group 2: Group 1: 14.31 residence
Exclusion criteria:
List who was Compression hip Group 2: 15.42
Not reported Additional outcomes
masked to screw (CHS) p = 0.03
reported:
interventions: CHS (135˚) were
All patients Operative details,
Not reported. inserted using the Mobility assessment Mean
N: 60 decrease of neck
No. lost to follow up: 0 standard technique. (Ceder et al) Group 1: 5.1
shaft angle, length of
Duration of Group 2: 4.7
follow-up: sliding of the lag
Group 1: Gamma Asia Pacific nail p >0.05
1 year screw.
(GAPN) Complications Fracture of the shaft of the femur
Group 2: 0 Notes:
Mean age: 73.7
M/F: 10/20
Other factors: Greater trochanter fracture
ASA score: Group 1: 1
1=3 Group 2: 0
2 = 19
3=8 Fracture displaced by nail insertion
4=0 Group 1: 2
Group 2: 0
Fracture pattern (Tronzo)
Stable (II): 14 (47%) Cut out
362 APPENDIX E

Unstable (III & IV): 16 (53%) Group 1: 1
Group 2: 1
Group 2: Compression hip screw (CHS)
No. randomised: 30 Deep infection
Mean age: 72.2 Group 1: 1
M/F: 14/16 Group 2: 1
Other factors:
ASA score: Non union
1=4 Group 1: 0
2 = 16 Group 2: 1
3=9
4=1
Fracture pattern (Tronzo)

Stable (II): 11 (37%)
Unstable (III & IV): 19 (63%)
APPENDIX E 363

Radford et al., Patient group: The operations Mortality 3 months Funding:
1993279,279 Patients with hip fracture were performed Group 1: 10 Not reported
using image Group 2: 12
Country of intensification. For Delayed wound healing Group 1: 8
Setting: Orthopaedic hospital, UK Limitations:
study: both implants they or persistent discharge Group 2: 3
England Includes diabetic
Inclusion criteria: aimed to have a leading to another course
central position of patients. Unclear
• Patients aged over 60, with a of antibiotics to be given
Study design: allocation
the screw in the Infection 3 months
Prospective pertrochanteric femoral concealment.
femoral head on (bacteriologically proven) Group 1: 4
randomized fracture
both Group 2: 0
study anteroposterior and Outcomes not
lateral views, with Thromboembolism Group 1: 6 reported:
List who was Exclusion criteria: during hospital stay Group 2: 8
its tip 5 to 10 mm Pain, place of
masked to • Not reported
from the Fixation failure requiring Group 1: 3 residence
interventions: subchondral bone. surgical revision Group 2: 2
Not reported. All patients
N: 200 Fracture of the femoral Group 1: 1
Additional outcomes
No. lost to follow up: not stated Group 1: Dynamic shaft Group 2: 11
Duration of reported:
hip screw Fracture of the femoral Group 1: 1
follow-up: 1 Prefracture mobility
Group 1: Dynamic hip screw 4 hole 135˚ plate shaft – requiring surgical Group 2: 3
year and housing score,
No. randomised: 100 with a screw of revision femoral shaft fracture
Mean age: 78 (60 to 90) appropriate length
Reoperation Group 1: 3 details of patients
M/F: 76/24 Group 2: 6 treated with gamma
Other factors: Group 2: Gamma
Cut-out Group 1: 3 nails, preoperative
Number with diabetes: 4 nail
Group 2: 2 blood loss.
Unstable: 43% A preoperative
radiograph was Non-union Group 1: 0
Group 2: Gamma nail taken of the other Group 2: 0 Notes:
No. randomised: 100 hip to compare Only surgeons of
Mean age: 72.2 with the implant registrar grade and
M/F: 14/16 template to decide above took part in
Other factors: the angle of the the trial and were
chosen nail. already experienced
364 APPENDIX E

Number with diabetes: 6 in the use of the DHS.
Unstable: 38% Distal locking of the The first 2 Gamma
nail in the femoral nail operations
shaft was performed by each
performed only surgeon were not
when indicated for included in the trial.
longitudinal
instability. Perioperative
fractures were
caused by too
forceful insertion of
the nail into the
femoral shaft – often
by hammer
APPENDIX E 365

Rahme et al., Patient group: Group 1: Proximal Length of stay in Group 1: 22 Funding:
2007281,281 Subtrochanteric femoral fractures femoral nail hospital Group 2: 25 not reported
Inclusion criteria: Treated with closed p=0.7
Country of All skeletally mature patients presenting reduction using a Non-union (absence of Group 1: 8 Limitations:
study: with acute subtrochanteric fractures traction table and bridging callus on 2 Group 2: 1
Australia percutaneous insertion Allocation
Exclusion criteria: radiographic views 9 p=0.025
Ipsilateral femoral shaft or neck fractures. of the nail (Synthes AG, concealment unclear.
months after injury)
Study design: Chur, Switzerland) Underpowered.
Randomised Revision Group 1: 8
All patients without anatomic
prospective Group 2: 0
N: 58 reduction. Outcomes not
comparison p=0.005
No. of dropouts: 0 reported:
Group 2: Intramedullary Mortality Group 1: 2
Pain, mobility,
List who was hip screw Group 2: 6
Group 1: Blade plate functional status
masked to Treated with open p=0.25
No. randomised: 29 Additional outcomes
interventions: No. of dropouts: 0 anatomic reduction, Infection Group 1: 1
Not reported. reported:
Mean age : 67 Internal fixation was Group 2: 3
achieved using a 95˚ p=0.6 mean operating time,
M/F: 12/17
Duration of angled blade plate blood transfusion,
Seinsheimer classification:
follow-up: (Synthes AG, Chur, infection (all,
Type 1: 0, Type 2: 8
9 months Switzerland). including whether an
organism was
Type 5: 9
confirmed as present
or not)
Group 2: Proximal femoral nail
No. randomised: 29 Notes:
No. of dropouts: 0 Intention to treat
Mean age: 73 analysis performed
M/F: 13/16
Seinsheimer classification:
Type 3: 10, Type 4: 1
Type 5: 10
366 APPENDIX E

Sadowski et al., Patient group: All procedures were Residence - Preoperative Home Funding:
2002294,294 Patients with hip fracture performed by staff (chi square with Yates Group 1: 15 Not reported
surgeons. correction) Group 2: 13
Country of
Setting: Orthopaedic hospital, Geneva, Limitations:
study: Group 1: Dynamic Nursing home
Switzerland. Includes diabetic
Switzerland hip screw Group 1: 4
Group 2: 7 patients
Inclusion criteria: Operative
Study design: technique p = 0.54
Prospective • Patients aged over 55, with
described by Blatter Postoperative data – Pneumonia Additional outcomes
randomized AO/OTA 31-A3 fractures
complications Group 1: 3 reported:
study (trochanteric proximal femoral and Janssen
fractures) (chi square) Group 2: 2 Operative time, blood
• Low energy fractures Group 2: Gamma transfusion, difficulty
nail Cardiac failure or infarction of operation, type of
Duration of Operative Group 1: 1 reduction, conversion
follow-up: technique as Group 2: 1 from static to
12 months Exclusion criteria:
• Patients with pathological described in dynamic construct,
Cerebrovascular accident consolidation time.
fractures, fractures associated Simmermacher et
al. The fracture was Group 1: 0 Notes:
with polytrauma, fractures
Group 2: 1
associated with polytrauma, a not exposed for
preexisting femoral deformity nailing unless it
could not be p = 0.83
preventing hip screw
osteosynthesis or reduced with closed Wound complications Group 1: 2
intramedullary nailing, techniques. A 10 or (chi square with Yates Group 2: 3
previous surgery on the 11mm diameter correction) p = 0.95
ipsilateral hip or femur, and a nail was used I 8/20
Hospital stay (days) Group 1: 18 ± 7
fractures extending 5cm distal patients and the lag (student t test) Group 2: 13 ± 4
to the inferior border of the screw measured
p = 0.01
lesser trochanter. 100 or 105mm in
10/20 patients. The Discharge to: Home
All patients proximal fragment (chi square) Group 1: 15
was reamed in all Group 2: 13
N: 39
APPENDIX E 367

No. lost to follow up: 1 20 patients, but Nursing home
distal reaming was Group 1: 4
Group 1: Dynamic condylar screw only performed on1 Group 2: 7
No. randomised: 19 patient. All of the
Mean age: 77 (±14) nails were Home
M/F: 5/14 interlocked distally Group 1: 15
Other factors: with 2 screws. Group 2: 13
ASA score:
1=1 All patients were Nursing home
2=9 given one dose of Group 1: 4
3=9 prophylactic Group 2: 7
4=0 intravenous p = 0.26
Anaesthesia: antibiotic. In Status of patient at 1 Mortality
General = 10 addition all patients year Group 1: 1
Regional = 9 were treated with (chi square) Group 2: 2
low-molecular
Group 2: Proximal femoral nail weight heparin Lost to follow-up
Mean age: 80 (±13) Group 2: 0
M/F: 7/13
Other factors: Available for review
Group 1: 17
1=0
Orthopaedic Implant failure
2=6
complications at 1 year Group 1: 6
3 = 11
(chi square) Group 2: 0
4=3
Anaesthesia:
Non-union
General = 11
Group 1: 1
Regional = 9
Group 2: 1
Infection
Group 1: 1
368 APPENDIX E

Group 2: 0
p = 0.007
cut-out
Group 1: 5
Group 2: 0
Major reoperations at 1 Group 1: 6 *

year Group 2: 0
(chi square) *(1 hip prosthesis, 1 change of implant,
4 change of implant and bone graft)
p = 0.008
Hip/thigh pain score at 1 Group 1: 1.77 ±0.73
year (student t test) Group 2: 1.44 ±0.86
p = 0.2
Jenson social-function Group 1: 2.5 ±1.3
score at 1 year Group 2: 2.6 ±1.0
(student t test) p = 0.9
Parker-and-palmer score Group 1: 6.0 ±3.5
at 1 year Group 2: 5.0 ±2.6
(student t test) p = 0.39
Residence at 1 year Home
(chi square) Group 1: 15
Group 2: 13
Nursing home
Group 1: 4
Group 2: 7
APPENDIX E 369

Saudan et al., Patient group: Group 1: Dynamic Postoperative data – Respiratory Funding:
2002300,300 Patients with hip fracture hip screw complications Group 1: 7 Not reported
In 50% of patients (chi square) Group 2: 7
Country of the length of the
Setting: Orthopaedic hospital, Geneva, Limitations:
study: screw was 90 or Cardiovascular
Switzerland.
Switzerland 95mm, and in Group 1: 9
almost all cases the Group 2: 5 Additional outcomes
Study design: side plate was 135˚
All fractures of the trochanteric region
Prospective with 4 holes Pulmonary embolism Intraoperative data.
randomized (in persons over the age of 55 years) Group 1: 1
caused by a low energy injury. Notes:
study Group 2: Proximal Group 2: 1
Included classifications were AO/OTA
femoral nail
Type 31-A1 or A2.
Operative Deep vein thrombosis
Duration of technique as Group 1: 1
Exclusion criteria:
follow-up: described by Group 2: 1
Pathologic fractures, fractures
12 months Simmermacher.
associated with polytrauma, a patient
Gastrointestinal
with previous ipsilateral hip or femur Group 1: 2
surgery, or any fractures with All patients were
given one dose of Group 2: 1
extension 5 cm distal to the inferior
border of the lesser trochanter. antibiotic
prophylaxis Neurologic
preoperatively, and Group 1: 1
All patients
treated with a low- Group 2: 2
N: 206
molecular weight p = 0.24
No. lost to follow up: 4%
heparin followed by Wound complications Group 1: 10
Group 1: Dynamic hip screw Coumadin as Group 2: 11
No. randomised: 106 prophylactic p = 0.71
Mean age: 83.7 (±10.1) anticoagulation, Hospital stay (days) Group 1: 14 ±10
M/F: 22/84 begun after surgery Group 2: 13 ±4
Other factors: and continued for 6 p = 0.71
ASA score: weeks.
Discharge to: Home
370 APPENDIX E

1=3 Group 1: 24
2 = 30 Group 2: 22
3 = 66
4=7 Nursing home/rehabilitation hospital
Anaesthesia: Group 1: 78
General = 37 Group 2: 74
Regional = 69
Died in hospital
Group 2: Proximal femoral nail Group 1: 4
Mean age: 83 (±9.7) p = 0.99
M/F: 24/76 Status of patient at 1 Died
Other factors: year Group 1: 13
1=1
2 = 30 Lost to follow up
3 = 63 Group 1: 4
4=6 Group 2: 5
Anaesthesia:
General = 38 Available for review
Regional = 62 Group 1: 89
Group 2: 79
Complications at 1 year Fixation failure (cut-out_
Group 1: 1
Group 2: 3
Non-union
Group 1: 0
Group 2: 0
Infection
Group 1: 1
Group 2: 3
APPENDIX E 371

p = 0.15
Reoperation at 1 year Hip prosthesis
Group 1: 1
Group 2: 3
Removal of implant and/or

debridement
Group 1: 1
Group 2: 3
p = 0.15
Habitation Home
Group 1: 50
Group 2: 37
Nursing home
Group 1: 39
Group 2: 42
p = 0.22
Pain (score) Group 1: 1.31 ±0.63
Group 2: 1.36 ±0.63
p = 0.59
Social function – Jensen Group 1: 2.65 ±1.14
(mean) Group 2: 2.88 ±1.16
p = 0.2
Mobility score – Group 1: 5.07 ±2.97
Palmer/Parker (mean) Group 2: 4.94 ±3.33
p = 0.8
372 APPENDIX E

Utrilla et al., Patient group: Fracture fixation Mortality 0 – 30 days Funding:
2005337,337 Patients with hip fracture was performed Group 1: 7 Not reported
within 4 days. Group 2: 10
Country of 4 surgeons
study: experienced in the 31 – 90 days
Orthopaedic hospital, Alicante, Spain
Spain standard gamma Group 1: 1
nail did all the Group 2: 5 Outcomes not
Study design: operations, but the
Patients aged over 65 years who
Prospective first 3 TGN 91 – 180 days List the outcomes in
randomized sustained a trochanteric fracture of Group 1: 3
the femur. operations which we are
study performed by each Group 2: 0 interested that are
Exclusion criteria:
not reported here
Patients with subtrochanteric fractures surgeon were not
181 – 365 days Additional outcomes
Duration of or subtrochanteric fracture extension, included in the Group 1: 8
pathologic fractures, history of a study. reported:
follow-up: Group 2: 6
previous injury involving the lower Perioperative data,
12 months Walking ability Total
limbs, and patients who had a severe Spinal anaesthesia leg shortening
concomitant medical condition (grade was performed in (Parker/Palmer score) at Group 1: 6.4 ±2.8 n= 82
V ASA score). 12 months Group 2: 6.2 ±2.8 n= 81
all but 3 patients. Notes:
p = 0.74
All patients Stable
Group 1: Group 1: 7.6 ±2.2
N: 210 Trochanteric
No. lost to follow up: 7 (3.3%) Group 2: 7.3 ±2.4
Gamma Nail (TGN) p = 0.92
This was a Unstable
Group 1: Trochanteric Gamma Nail modification of the
(TGN) Group 1: 7.0 ±2.1
standard implant: Group 2: 5.8 ±2.7
No. randomised: 106 shorter in length
Mean age: 80.6 (±7.5) p = 0.017
(180mm), with a
M/F: 38/66 Hip flexion (˚) Group 1: 97.9 ±10.3
lower mediolateral
Other factors: Group 2: 95.6 ± 9.5
curvature (4˚) and
ASA score: p = 0.15
available only in
1 = 13 Hip pain (no.) Group 1: 41
APPENDIX E 373

2 = 39 proximal and distal Group 2: 44
3 = 41 diameters of 17 and p = 0.75
4 = 11 11 mm. The neck Thigh pain (no.) Group 1: 50
shaft angle was Group 2: 45
Group 2: Compression hip screw 130˚ and was p = 0.52
No. randomised: 106 inserted by a
Postoperative Grp 1 Grp2
Mean age: 79.8 (±7.3) percutaneous
complications n 82 81
M/F: 28/78 technique. Distal
DVT 4 3
Other factors: locking with 1
Local wound 6 7
ASA score: screw only was
Deep infection 0 1
1 = 14 performed on those
Trochanter fracture 4 2
2 = 35 fractures with
Fixation failure 5 6
3 = 54 rotational instability
Cut out 1 2
4=3 of the diaphyseal
Reoperation 1 4
fragment.
Group 2:
Compression hip
screw (CHS)
The CHS was
inserted using the
standard technique,
the implant was a
135˚ plate with 4
holes.
All patients
received antibiotic
and
thromboembolic
prophylaxis.
374 APPENDIX E

Zou et al., Patient group: Surgery was Femoral shaft fracture Group 1: 0 Funding:
2009364,364 Consecutive patients with low-energy performed with the Group 2: 0 Not stated
trochanteric femoral fractures patient in the P value(s): not significant
Country of Setting: supine position on a Cut-out Group 1: 0 Limitations:
study: Dept orthopaedic surgery, The first fracture table, with Group 2: 0
China affiliated hospital of Soochow the injured P value(s): not significant
University, Suzhou, Jiangsu, China extremity slight Outcomes not
Study design: Non-union Group 1: 1 (unstable) reported:
adducted to
Group 2: 0
Inclusion criteria: facilitate insertion The Salvati and
RCT Patients with 31-A1 stable trochanteric of the implant. Wilson scoring
Breakage of implant Group 1: 0 system for hip
or 31-A2/31-A3 unstable trochanteric
Group 2: 2 (1 unstable, 1 stable, of function
fractures. After surgery the which 1 required reoperation)
Exclusion criteria: patients were P value(s): not significant
Patients with a pathological fracture or mobilised and given Additional outcomes
Duration of multiple injuries were excluded. Wound infection Group 1: 1 (stable)
standard reported: list
follow-up: Group 2: 1 (unstable)
rehabilitation additional outcomes
1 year All patients P value(s): not significant
instructions by a reported in the study
N: 121 physiotherapist. that we are not
interested in
Group 1
No. randomised: 63 Group 1
Dynamic hip screw Notes:
Stable: 52
Unstable: 11
Group 2
Age (mean +SD): 65 (34-89)
M/F: 24%/76% Proximal femoral
nail antirotation
Operative time: 93 +/- 13 mins
Group 2
No. randomised: 58
Stable: 42
Unstable: 16
Age (mean +SD): 65 (37-91)
M/F: 21%/79%
APPENDIX E 375

Operative time: 52 +/- 10 mins
376 APPENDIX E
17.9 Evidence Table 9: Surgical approach to hemiarthroplasty

Enocson et al., Patient group: Surgical approach Number of dislocations Group 1: 13/431 (3%) Funding:
200885,85 Consecutive patients who had a Group 2: 15/176 (9%) None reported
hemiarthroplasty for non-pathological Group 1 Group 2: 17/129 (13%)
Country of displaced femoral neck fracture 431 operations
Dislocation for posterior Logistic regression univariate analysis Limitations:
study: performed by an
lateral approach with Odds ratio: 3.0 (1.4, 6.4) Not stated how
Sweden anterolateral
Setting: posterior repair P=0.005 patients allocated to
approach.
Orthopaedics department compared to a surgeon. Surgical
Study design:
anterolateral approach. Logistic regression multivariate approach based on
Historical Group 2
Inclusion criteria: analysis adjusted for age, sex, surgeon’s own
cohort 176 operations
• Not reported performed by a
indication for surgery, surgeon seniority preference
and femoral head size
List who was posterolateral
Exclusion criteria: Odds ratio: 3.9 (1.6, 9.8)
masked to approach with Outcomes not
• None reported P=0.003
interventions: posterior repair. reported:
Not applicable Dislocation for posterior Logistic regression univariate analysis
All patients lateral approach without Odds ratio: 4.9 (2.3, 10) Mortality, length of
Group 3 stay in secondary
N: 739 hips in 720 patients 129 operations posterior repair P<0.001
No. of dropouts: not reported compared to care, requirement for
Duration of performed by a surgical revision,
follow-up: Age (mean +SD): women: 84 (54-103) , posterolateral anterolateral approach. Logistic regression multivariate
men 82 (55-97) years analysis adjusted for age, sex, wound infection.
Median 2.3 (0- approach without
10) years M/F: 147/592 posterior repair. indication for surgery, surgeon seniority
and femoral head size Operations
Odds ratio: 6.9 (2.6, 19) performed by
P<0.001 registrars or post-
registrars.
APPENDIX E 377
Evidence tables – surgical approach to hemiarthroplasty

Parker et al., Patient group: Surgical approach Outcomes extracted Results reported in forest plots for: Funding:
year269 Patients with displaced intracapsular -Number of dislocations None reported
hip fracture Group 1 -Pain at 1 month
Country of 57 cemented -Impairment of mobility at 6
Limitations:
study: Thompson hemi- months
Setting: Most operations
UK arthroplasties by an
Hospital anterolateral were performed by
Study design: approach. trainees with
All patients different levels of
Systematic
review N: 114 patients Group 2 experience. No
No. of dropouts: not reported blinding of anyone
including 1 RCT 57 cemented
Thompson hemi- reported. Unclear
arthroplasties by allocation
Duration of posterior approach concealment.
follow-up:
2 years Outcomes not
reported:
Mortality (only
presented in graphs),
length of stay in
secondary care,
reoperations (unable
to work out
numbers), quality of
life.
378 APPENDIX E
17.10 Evidence Table 10: Mobilisation strategies

Hauer et al., Patient group: Group 1 Barthel/Mahoney Group 1: 93.0 (8.2) Funding:
2002 139,140 Patients with hip fracture High intensity progressive activities of daily living Group 2: 96.1 (8.2) A grant received
resistance training of (ADL) p = 0.636 from the
Country of Setting: functionally relevant Lawton/Brody Group 1: 7.3 (1.4) Ministerium fur
study: muscle groups and a Instrumental activities of Group 2: 6.9 (1.3) Wissenscahft,
Germany Inclusion criteria: progressive functional daily living index p = 0.416 Forschung und
• Hip surgery, recent history of training for 3 days a week Maximal dynamic and Leg-press, fractured side 1RM (kg) Kunst Baden-
Study design: for 12 weeks. Intensity of Wuerttemberg and
injurious falls, age over 75 isometric muscle Group 1: 71 (35)
years, female, consent of strength training was strength, at 3 months Group 2: 50 (21) p = 0.021 the University of
RCT adjusted to 70-90% of the Heidelberg.
orthopaedic surgeon, patient mean, (+SD)
willingness to participate in the individual maximal Leg-press, non-affected side1RM
study. workload. Basic functions (kg) Limitations:
Duration of such as walking, stepping Group 1: 88 (39) Small study size
follow-up: or balancing were trained
Exclusion criteria: Group 2:67 (17) p = 0.018
3 month
• Acute neurological impairment, progressively with
severe cardio-vascular disease, increasing complexity. Leg-extensor, fractured side, Additional
unstable chronic or terminal Newton outcomes
illness, major depression, severe Group 2 Group 1: 68 (13) reported:
cognitive impairment or severe Patients in the control Group 2: 51 (22) p = 0.011 Further baseline
musculo-skeletal impairment. group met 3 times a week characteristics.
for 1 hour for motor Leg-extensor, non affected side, Balance score,
All patients placebo activities. Typical Newton functional reach,
N: 28 activities, which were not Group 1: 80 (11) total activity,
No. of dropouts: supposed to be relevant Group 2: 60 (20) p = 0.006 ‘sports’ activity.
Age (mean +SD): 81 (+3.9) for the study purpose, Household
M/F: All female were calisthenics, games Leg flexor, fractured side, Newton activities,
and memory tasks whilst Group 1: 37 (7) emotional state
Group 1 seated Group 2: 34 (13) p = 0.036 Notes:
No. randomised: 15
APPENDIX E 379

No. of dropouts: Leg flexor, non affected side,
Age (mean +SD): 81.7 (+7.6) Both groups received Newton
M/F: All female identical physiotherapy Group 1: 39 (11)
Group adherence: 93.1 (+13.5%) two times a week for 25 Group 2: 37 (13) p = 0.113
mins. Strength and Ankle plantar flexion, fractured
Group 2 balance training were side, Newton
No. randomised: 13 excluded during
No. of dropouts: physiotherapy and Group 1: 88 (30)
Age (mean +SD): 80.8 (+7.0) control group sessions. Group 2: 65 (33) p = 0.944
M/F: All female Physiotherapy consisted Ankle plantar flexion, non affected
Group adherence: 96.7 (+6.1%) of massage, stretching side, Newton
and application of heat or
ice. Group 1: 98 (32)
Group 2: 78 (32) p = 0.968
Handgrip strength, both Group 1: 121 (29)
hands, Kilopascal Group 2: 108 (28) p = 0.270
Maximal gait speed, Group 1: 0.72 (0.28)
m/sec Group 2: 0.49 (0.15) p = 0.121
Timed up and go, (sec) Group 1: 26.1 (17.8)
Group 2: 26.9 (9.8) p = 0.731
Tinetti’s performance Overall
oriented mobility Group 1: 23.5 (4.5)
assessment (POMA) Group 2: 20.5 (4) p = 0.505
Part 1
Group 1: 12.7 (2.2)
Group 2: 11.4 (2.4) p = 0.747
Part 2
Group 1: 10.8 (2.5)
Group 2: 9.1 (2.1) p = 0.249
380 APPENDIX E

Box step, cm Fractured leg
Group 1: 34.5 (6.4)
Group 2: 30.6 (9.8)
p = 0.482
Unaffected leg
Group 1: 38.5 (7.8)
Group 2: 34.4 (5.8)
p = 0.420
APPENDIX E 381
Evidence tables – mobilisation strategies

Karumo et al., Patient group: Group 1 – usual care Length of hospital Prosthesis Funding:
1977171,171 Patients with hip fracture Average of 30mins stay Group 1: 33.9 (+20.1) Not stated
physiotherapy per day. Group 2: 31.8 (+19.6)
Country of Inclusion criteria: Limitations:
study: Consecutive patients aged over 50 with Group 2 – Intensive Internal fixation
Finland Group 1: 36.0 (+23.2) Most data presented
dislocated fractures of the femoral neck. Physiotherapy
performed twice daily Group 2: 32.5 (+23.6) for overall trial
Study design: population or split by
Exclusion criteria: – average of 1 hour.
RCT Cochrane report: surgical treatment
Inadequate follow up examination. rather than rehab
Physiotherapy shame: Group 1: 35.01 (21.8)
List who was Group 2: 32.21 (22.03) type.
All patients Walking on crutches
masked to N: 100 on first postoperative Strength of the Prosthesis
interventions: Lost to follow up: 13 day with almost all adductor muscle (9 Group 1: 5.6 (3.3) Additional outcomes
allowed full weight weeks post op) – Group 2: 6.3 (5.7) reported:
Not reported Group 1 bearing from the operated leg Ability to move/sit
No. randomised: beginning. Internal fixation up/stand/walking
Duration of 23 treated with prosthesis From first post op day Group 1: 6.4 (4.0) ability/social
follow-up: 26 with internal fixation training in sitting in a Group 2: 4.5 (2.3) management – all
3 months No. of dropouts: chair with the hip and split by surgical
Age (mean +SD): knee joint in 90˚ Cochrane report: treatment. No
M/F: 13/26 flexion. Group 1: 5.26 (4.08) difference reported.
Subgroup category numbers: In second Group 2: 6.02 (3.69) Notes:
Other factors: postoperative week
training in walking up
Group 2 and down stairs.
No. randomised: Patients urged to
16 treated with prosthesis perform extension-
22 with internal fixation flexion movements of
the knee joint.
No. of dropouts:
Age (mean +SD):
M/F: 9/29
382 APPENDIX E

APPENDIX E 383

Moseley et al., Patient group: Group 1 Knee extensor strength 4 week Funding:
2009216,216 Patients with hip fracture High group. Weight (isometric knee extensor Group 1: 7.8 (3.9) Project grant from
bearing exercise twice strength at 90˚ measured Group 2: 7.7 (4.0) the National Health
Country of Setting: daily for a total of 60 using a spring and Medical Research
study: Inpatient rehab units of 3 teaching minutes per day for 16 balance.)kg, mean (SD) 16 week Council , Australia.
Australia hospitals in Sydney weeks. 5 weight bearing Group 1: 10.3 (5.0)
exercises were prescribed Group 2: 9.3 (4.4)
Limitations:
Study design: Inclusion criteria: in addition to walking on Walking speed 4 week
• Patients with surgical fixation a tread mill with partial (measured over a 6 m Group 1:0.53 (0.25)
RCT body weight support distance using a stop Group 2: 0.48 (0.22) Additional outcomes
for hip fracture admitted to
using a harness (for watch) m/sec. Mean (SD) reported: Fear of
inpatient rehab units who had
List who was inpatients) or a walking 16 week falling, balance, step
approval to weight bear or
masked to programme (after Group 1: 0.63 (0.32) test, body sway,
partial weight bear; able to
interventions: hospital discharge). The 5 Group 2: 0.60 (0.31) stability test, falls
tolerate the exercise
Assessor- efficiency scale.
programmes; able to take 4 plus weight bearing exercises Pain (7 item ordinal 4 week
blinded used for both legs Further participant
steps with a forearm support scale) – some. Moderate Group 1:44
included stepping in characteristics.
walking frame and the or severe. Group 2: 41
assistance of one person; no different directions,
Duration of
medical contraindications that standing up and sitting 16 week Notes:
follow-up: down, tapping the foot
would limit ability to exercise; Group 1: 30
living at home or low care and stepping onto and off Group 2: 29
16 weeks a block. Hand support
residential facility prior to the Quality of life (EQ 5D and 4 weeks
hip fracture, with the plan to could be used if
expressed as a utility Group 1: 0.53 (0.27)
return to this accommodation at necessary. The exercises score) Group 2: 0.53 (0.27)
discharge. were progressed by
• Subjects with cognitive reducing support from
the hands, increasing 16 week
impairment were included if a Group 1: 0.62 (0.30)
carer who was able to supervise block height, decreasing Group 2: 0.62 (0.26)
the exercise programme was chair height and
available. increasing the number of Length of stay in hospital 16 week
Group 1: 28 (15)
• Middle band of people with hip repetitions. This started
Group 2: 25 (14)
384 APPENDIX E

fracture as an inpatient Total exercise time with a Group 1: 545 (463)
programme, followed by physiotherapist or Group 2: 363 (318)
Exclusion criteria: home visits and a physiotherapy assistant P value(s): 0.001
• High functioning patients who structured home exercise as an inpatient, (min)
are discharged directly to home programme. mean (IQR)
and low functioning patients sit-to-stand (stand-ups 4 week
who are discharged to a Group 2 per sec) mean (SD) Group 1: 0.24 (0.15)
residential aged care facility Low group. Patients Group 2: 0.19 (0.09)
from the acute orthopaedic undertook 5 exercises in
ward were excluded. sitting or lying plus a 16 week
small amount of walking Group 1: 0.26 (0.14)
All patients using parallel bars or Group 2: 0.22 (0.11)
N: 160 walking aids for a total of
Barthel index 4 week
No. of dropouts: 30 mins each day for 4
Group 1: 93 (85-100)
weeks. The exercises
Group 2: 90 (85-95)
Group 1 were progressed by
No. randomised: 80 increasing the repetitions
16 week
No. of dropouts: 2 withdrew by 16 and resistance. This type
Group 1: 95 (90-100)
weeks of exercise programme is
Group 2: 95 (85-100)
Age (mean +SD): 84 (8) regarded as usual care.
M/F: 15:65
Subgroup category numbers: All patients received usual
Other factors: post-op mobilisation, and
the rehab programme
Group 2 usually provided by other
No. randomised: 80 health professionals and
No. of dropouts: 1 withdrew at 16 any gait aids were
weeks progressed as per usual
Age (mean +SD): 84 (7) protocols. No
M/F: 15:65 physiotherapy treatments
were administered during
the trial.
APPENDIX E 385

Oldmeadow et Patient group: Group 1 Function – Assistance independent Funding:
al., 2006247,247 Patients with hip fracture Early ambulation required to transfer from Group 1: 16 Not stated
(within 48 supine to sit, sit to stand Group 2: 4
Country of h/postoperative
study: assistance
The Alfred Hospital, Victoria, Australia day) with a
Australia physiotherapist Group 1: 10
during standard Group 2: 21
Inclusion criteria:
Study design: Additional outcomes
Consecutive patients admitted through working hours.
RCT P value(s): 0.009 reported: Further
the emergency department for surgical
fixation of an acute neck of femur Group 2 Function – Mean walking Group 1: 58.63 (0.05 – 400) baseline
Duration of characteristics,
fracture (by sliding screw, gamma nail delayed (longer metres Group 2: 29.71 (0 – 150)
follow-up: Troponin, subgroup
or a hemiarthoplasty) were considered than 48 P value(s): 0.03
1 week post h/postoperative analysis of true early
for inclusion in the study. Assistance required to Independent
surgery day 3 or 4) ambulation and failed
negotiate one step on Group 1: 10
day 7 post-surgery. Group 2: 23 early ambulation.
Exclusion criteria:
Pathological fractures, if postoperative All patients
orders were for non-weight bearing on received routine Failed/unable Notes:
the operated hip, the patient admitted postoperative Group 1: 13
from a nursing home or the patient was medical and nursing Group 2: 1
non-ambulant premorbidly. clinical care, as
currently practiced P value(s): 0.32
All patients at The Alfred.
Discharge destination Group 1:
N: 60 Home: 5
Mean age: 79.4 years (53-95) The physiotherapy
ambulation re- Fast stream rehab: 8
M/F: 68% women Slow stream rehab: 14
education program
was implemented Nursing home: 1
Group 1 Death: 1
No. randomised: 29 once per day over 7
No. of dropouts: 10 patients failed to days. This program
was the same for all Group 2:
achieve their first walk within the 48h. Home: 1
Age (mean +SD): 78.8 (2.14) and included
Fast stream rehab: 14
386 APPENDIX E

M/F: 8/21 walking re- Slow stream rehab: 16
Group 2 education, bed Nursing home: 0
No. randomised: 31 exercises and chest Death: 0
Age (mean +SD): 80.0 (2.08) physiotherapy as
M/F: 11/20 indicated. Only the P value(s): 0.19
time to first walk Length of stay, mean Group 1: 9.27 (4-33) – outlier removed
differed between (range) n =18.
groups 17.90 (5-33) – failed early ambulation n
= 10
Group 2: 11.39 (5-24)
P value(s): 0.59
APPENDIX E 387
17.11 Evidence Table 11: Multidisciplinary rehabilitation

Cameron Patient group: Group 1 Median length of Group 1: 13 (7-25) Funding:
199342,44 Patients with proximal femoral A nursing care plan starts hospital stay, days Group 2: 15 (8-44) Australian
hip fracture immediately post-op that (interquartile range) p=0.034 Department of
Country of supports early mobility and Health, Housing and
study: self-care. A physician sees the Mortality (obtained Group 1: 32 Community Services.
Setting:
Australia patient the same day or the from the Cochrane Group 2: 38
General hospital serving an
outer urban area of Sydney, next day of the operation to review- Handoll 2009) Limitations:
Study design: identify and treat concurrent 12 months
RCT Australia. Mean age
illness, review previous level Mean Barthel index 2 weeks after injury significantly lower in
Inclusion criteria: of disability and assess social Group 1: 32
Duration of accelerated rehab
follow-up: Patients aged over 50 with an support needs. The Physician Group 2: 38 group (p = 0.0042)
uncomplicated proximal also liaises with the
femoral fracture (non- orthopaedic surgeon 1 month after injury
4 months regarding likely complications No assessor blinding
pathological, no additional Group 1: 32
fractures), surgical or precautions (e.g. limitations Group 2: 38
intervention within 7 days of of weight bearing). The Outcomes not
injury and residence in the physician leads on planning reported:
district. the rehab according to the
patient’s pre fracture
Exclusion criteria: condition. Patients from
Additional outcomes
Fractures sustained whilst in nursing homes are returned
there as soon as feasible to reported:
hospital or who were
transferred to another undergo supervised Additional baseline
hospital for surgical mobilization and characteristics such
treatment. physiotherapy. Patients not as pre-injury
from nursing homes are situation, injury
All patients discharged once they can walk details.
N: 252 (with an aid) and go to the
Lost to follow up: toilet independently. The
Notes:
patient received
388 APPENDIX E

Age (mean +SD): 84 physiotherapy on each
M/F: 17% male weekday (ideally 2 sessions Patients stratified
Cognitively impaired: 122 per day). The orthopaedic into 3 groups:
surgeon and rehab physician
Nursing home, non
Group 1 Accelerated rehab review the patient 3 or 4 times
No.: 127 weekly. After discharge the nursing home +
moderate to severe
No. of dropouts: patient’s rehab continues
disability and non-
Age (mean): either at home
nursing home +
Nursing home: 84.2 (n = 48) (physiotherapist home visit) or
Non nursing home+moderate at a day hospital until they limited disability.
to severe disability: 87.2 (n = reach their pre-fracture level
21) of function or plateau at a Key difference in
Non-nursing home+limited lower level. accelerated rehab
disability: 79.2 (n = 58) was concentrated
M/F: Group 2 input of an
Conventional care experienced
Group 2 Usual care physician with
No. : 125 training in geriatric
No. of dropouts: and rehab medicine.
Age (mean):
Nursing home: 88.5 (n=46)
Non nursing home+moderate
to severe disability: 89.3 (n =
22)
Non-nursing home+limited
disability: 81.4 (n = 57)
M/F:
Other factors:
Living alone
APPENDIX E 389
Evidence tables – multidisciplinary rehabilitation

58,60
Crotty 2002 Patient group: Randomisation was Mortality at 12 months Group 1: 3 Funding:
Patients with hip fracture undertaken by the hospital Group 2: 4 Supported by the
Country of pharmacy department Moved to higher level of Group 1: 1 South Australian
study: (computer generated care Group 2: 2 Department of
Setting:
Australia allocation sequence in Human Services
2 Australian teaching hospitals Unable to walk Group 1: 0
in Adelaide (Flinders Medical sealed opaque envelopes). Group 2: 2
Study design:
centre, Repatriation General SF-36 physical Group 1: 38 (34.0-41.9) Limitations:
RCT
Hospital) Group 1 component score at one Group 2: 33.3 (27.6-39.1) Baseline data not
Duration of Patients were discharged year, mean (95% CI) given for
Inclusion criteria: within 48 hours of male/female ration
follow-up: SF-36 mental component Group 1: 53.8 (49.2-58.3)
Aged 65 or over, medically randomisation and were or mean age in each
score at one year, mean Group 2: 52.3 (47.3-57.3)
stable, needed a formal visited by physiotherapists, arm.
12 months (95% CI)
rehabilitation program, had occupational therapists, Outcomes not
adequate physical and mental speech pathologists, social reported:
capacity to participate in workers, and therapy aides, Length of hospital stay, Group 1: 7.8 (9.3)
rehabilitation, were expected who negotiated a set of mean (SD) – from Group 2: 14.3 (10.6)
to return home after discharge realistic, short-term, and Cochrane review, Handoll Additional outcomes
from the hospital, and had a measureable treatment 2009 reported:
home environment suitable for goals with both participants
rehabilitation. and their care-givers. Notes:
Exclusion criteria: Standard therapy services
If patients had inadequate podiatry, nursing care, and
social support in the assistance with light Length of rehab, mean Group 1: 28.3 (14.5)
community, no telephone at domestic tasks, were (SD) – from Cochrane Group 2: 14.3 (10.6)
home, or did not live in provided as required. review, Handoll 2009
Adelaide’s southern
metropolitan region. Group 2
Conventional care in routine
All patients hospital interdisciplinary
N: 66 rehabilitation.
390 APPENDIX E

Lost to follow up: 3 Hospital readmissions Group 1: 8
Age (mean +SD): 82.5 during 4 month follow up Group 2: 7
M/F: 33% male – from Cochrane review,
Handoll et al., 2009132,133
Group 1 Early discharge +
home rehab
No.: 34
Age (mean): not stated
M/F: not stated
Group 2 Usual care

No. : 32
Age (mean): not stated
M/F: not stated
APPENDIX E 391

Galvard Patient group: All patients were length of stay in hospital, Group 1: 53.3 (47.7) Funding:
1995107,107 Patients with hip fracture treated at the days (mean, SD) Group 2: 28 (24.2) Not stated
orthopaedic
Country of department and Median
study: then Group 1: 40
Vaernhem Hospital, Malmo, Sweden Higher number of
Sweden randomization Group 2: 21
took place subtrochanteric fractures and
Inclusion criteria: higher mean age of men in the
Study design: immediately after Mortality at 1 year Group 1: 45
Independently living hip fracture patients in geriatric MDR group.
RCT the operation, Group 2: 40
the municipality of Malmo Unclear allocation concealment.
Exclusion criteria: using a random
List who was number generator. Outcomes not reported:
People resident in nursing homes or waiting Total no. of patients Group 1: 36
masked to
for a nursing bed, or already in hospital readmitted to hospital Group 2: 57
interventions
Group 1 Additional outcomes reported:
:
All patients Patients were Baseline data – distribution of
N: 371 transferred on the fracture types. Destination at
Age (mean + range): 79 (52-102) second discharge from hospital. Causes
Duration of
follow-up: M/F: 26% male postoperative day, for hospital readmissions. Hip
and once weekly a pain and walking ability one year
1 year
Group 1 Geriatric (MDR) visiting postoperatively. Indoor walking
No.: 179 orthopaedic speed.
Age (mean +SD): surgeon would
men: 79.1 (8.6) decide on further
Notes:
women: 80.9 (9.2) treatment of the
M/F: 50/129 fracture Study states that longer length of
hospital stay in geriatric MDR
Group 2 Usual care Group 2 group may relate to lack of
No. : 192 Usual care – stayed experience in geriatric
Age (mean +SD): on the orthopaedic department at the time and that
men: 73.6 (10) ward. the orthopaedic (usual care)
women: 79.6 (8.2) group had over 25 years of
M/F: 45/147 experience with these patients.
392 APPENDIX E

Study details Patients Interventions Outcome Effect size Comments
measures
Gilchrist et Patient group: Patients were admitted to the Mortality Inpatient Funding:
al., 1988113,113 Patients with femoral neck fractures orthopaedic unit and had standard Group 1: 4 Not stated
preoperative medical assessment. After Group 2: 13
Country of surgery were transferred to orthopaedic
study: wards at Gartnavel General Hospital for 3 month
Glasgow, UK Orthopaedic unit, Western Infirmary Group 1: 10
rehab. Randomisation occurred at time of
transfer. Group 2: 18 Outcomes not
Study design:
RCT Women aged over65 6 month
Group 1
Group 1: 14
Exclusion criteria: Patients were under overall care of the
Duration of Group 2: 23
Patients referred from nearby orthopaedic surgical staff. A weekly Additional outcomes
follow-up:
hospitals, patients who made a rapid combined ward round was performed by reported:
recovery and were sent directly home. a geriatrician (consultant or senior Type of fracture,
6 months
registrar), an orthopaedic senior registrar, placement of patients
All patients and the senior ward nurse. A Length of Group 1: 44 (5.7)
admitted from home,
N: 222 physiotherapist, occupational therapist, stay in Group 2: 47.7 (7.7)
conditions in patients
Age (mean +SD): and a social worker participated in the hospital
at discharge
case conference that followed. Advice (mean, SE)
Group 1 Orthopaedic geriatric unit was given on medical problems that arose
Notes:
No.: 97 between ward rounds by consultation
Age (mean): 82 with the geriatrician. Patients were seen
Length of stay before transfer (days): on average, 4 times by a geriatrician.
10.2
Group 2
Group 2 Usual care (orthopaedic Similar nursing cover and paramedical
ward) services as group 1, but no case
No. : 125 conference. Referral for any medical
Age (mean): 80.6 problem to the geriatric service was made
Length of stay before transfer (days): by letter, and patients were seen by a
9.8 different geriatrician than in group 1.
APPENDIX E 393

Huusko et al., Patient group: Group 1 Mortality at 12 months Group 1: 18 Funding:
2002157,158 Patients with proximal femoral Intensive geriatric Group 2: 20 Study was supported
(Huusko et al., fracture rehab within hospital: by grants from
2000157,157 gives multidisciplinary Central Finland
subgroup data geriatric team Mortality at discharge Group 1: 5 Health Care District,
Setting:
for patients (geriatrician, specialist Group 2: 5 Kuopio University
with dementia) Specialist district hospital in
Jyvaskyla, Finland GP and nurses, Hospital, Emil
occupational therapist, Total days in hospital Group 1: 80 Aaltonen Foundation,
Country of physiotherapist, social Uulo Arthio
Inclusion criteria: (during 1 year) Group 2: 80
study: worker, Foundation and
Finland • Community-dwelling patients
neuropsychiatrist). Novartis Finland Ltd
with acute hip fractures over Length of hospital stay Group 1: 85 (13-365) N = 19
Twice daily
Study design: 64 years of age. (median + range) – severe Group 2: 67 (15-365) N = 9
physiotherapy; ADL Limitations:
RCT Exclusion criteria: dementia (mini mental
practice; daily
• Pathological fracture, multiple state examination score P=0.902 Imbalance of baseline
schedule; counselling;
List who was fractures, terminally ill, 0-11) characteristics.
information; discharge
masked to serious early complication, Intervention group
plan; home visits,
interventions: receiving calcitonin, unable to had a greater number
treatment at home Length of hospital stay Group 1: 47 (10-365) N = 24
No assessor communicate (median + range) – Group 2: 147 (18-365) N = 12 with Dementia
after discharge based in
blinding All patients moderate dementia (mini 32/120 vs. 20/123);
geriatric ward in same
N: 243 mental state examination fewer were
hospital as surgery.
Duration of Lost to follow up: score 12-17) functionally
Group 2
follow-up: Age (mean and range): 80 (66-97 independent in ADL
Discharge to local Place of residence and 1 year
M/F: 28% male before hip fracture
community hospitals, mortality – severe Independent living
12 months treatment by GP with (41 vs. 66)
Group 1 Geriatric rehab dementia (mini mental Group 1: 7
physiotherapists Outcomes not
No.: 84 state examination score Group 2: 3
usually available. reported:
Age (mean + range): 80 (67-92) 0-11) Nursing home
Transfer 2 to 5 days Group 1: 5
M/F: 36/84
after surgery. Group 2: 0 Additional outcomes
Living alone: 62
Dementia: 32 Hospital reported:
Group 1: 2 IADL and ADL change
Group 2: 3
394 APPENDIX E

Group 2 Usual care Dead from baseline.
No. : 90 Group 1: 5
Age (mean + range): 80 (66-97 Group 2: 3 Notes:
M/F: 33/90
Patients were
Living alone: 70 Group 1: n = 19
Dementia: 20 Group 2: n = 9 mobilised on the first
postoperative day.
Place of residence and 1 year

mortality – moderate Independent living
dementia (mini mental Group 1: 15
state examination score Group 2: 4
12-17) Nursing home
Group 1: 1
Group 2: 2
Hospital
Group 1: 4
Group 2: 4
Dead
Group 1: 4
Group 2: 2
Group 1: n = 24
Group 2: n = 12
APPENDIX E 395

Kennie et al., Patient group: Patients were Length of hospital stay Mean +/- SD(from Cochrane review, Funding:
1988176,176 Women with proximal femoral randomised to geriatric Handoll 2009) Forth Valley Health
fracture rehab or usual care Group 1: 37 (33) Board
Country of once the orthopaedic Group 2: 56 (54)
study: surgeon judged them
Stirling, UK fit to be moved to a Median
Orthopaedic ward and geriatric Group 1: 24 (8-197) No blinding of staff or
rehab ward, Stirling rehab ward.
Study design: Group 2: 41 (9-365) patients.
Both treatment and
RCT control groups received
Inclusion criteria:
Women aged over 65 physiotherapy, More dependent based Group 1: 22/43 Outcomes not
occupational therapy, on Katz score at 1 year Group 2: 28/35 reported:
Duration of and orthotic and other (from Cochrane review,
Exclusion criteria:
follow-up: services. Handoll 2009)
Mortality prior to randomisation,
pathological fractures, those likely Group 1 Type of residence after NHS or private nursing home Additional outcomes
1 year
to be discharged within 7 days of Transferred by discharge Group 1: 5 reported:
entering the trial, those remaining ambulance 5km to Group 2: 16
Additional baseline
unfit for transfer by ambulance to a orthopaedic beds in a data including
peripheral hospital. peripheral hospital. The Own home
residence,
median delay between Group 1: 31
independence and
All patients entry into the study Group 2: 19
mental state before
N: 108 and transfer was one Mortality (taken from At discharge admission, details of
Lost to follow up: day (range 0-7). A GP Reid 1989) Group 1: 5 fracture.
Age (mean +SD): provided day-to-day Group 2: 4
M/F: All female medical attention, and
a consultant physician Notes:
At 1 year
Group 1 Geriatric rehab in geriatric medicine Group 1: 10 Similar baseline
No.: 54 attended 2 ward round Group 2: 18 characteristics across
Age (median + range): 79 (65-94) and 1 conference of the groups, apart from
M/F: All female multidisciplinary team age and difference in
each week. mental state, with
Group 2 Usual care Orthopaedic advice was more moderate and
No. : 54 available on demand. severe impairment in
396 APPENDIX E

Age (median + range): 84 (66-94) the control group
M/F: All female Group 2 (p=0.06)
The control group
remained in the
orthopaedic admission
ward. A few of these
patients were moved
into other short stay
wards at the discretion
of the consultant
orthopaedic surgeon.
APPENDIX E 397

Naglie et al., Patient group: Within 48h of Mortality At discharge Funding:
2002222,222 Patients with hip fracture randomisation the research Group 1: 7 (5%) Supported by a grant
coordinator reviewed each Group 2: 13 (9.4%) from Ontario Ministry
Country of case for compliance with of Health and the
Setting:
study: the inclusion criteria and a 3 months Research Institute of
Teaching hospital in Toronto
Toronto, panel then reviewed Group 1: 10 (7.1%) the Queen Elizabeth
Canada eligibility. Group 2: 12 (8.7%) Hospital, Toronto.
Inclusion criteria:
• Patients aged over 70 from Separate staff provided
Study design: care in each group to 6 months
RCT the community and from Group 1:17 (12.1%) Limitations:
nursing homes prevent containment bias. Anticipated that the
Group 2: 21 (15.2%)
List who was intervention would
Exclusion criteria: Decline in ambulation- 3 months
masked to Group 1 increase length of
• Fractures occurring in an data missing for 3 patients Group 1: 73 (57%)
interventions: Protocols and standardized hospital stay.
acute care hospital, pathologic in group 1 and 8 patients Group 2: 72 (61%)
Assessors orders were used, early
fractures, multiple traumas, in group 2 at 3 months.
mobilisation, early Outcomes not
6 months
Duration of previous surgery on the participation in self care
Group 1:59 (47.6%) reported:
follow-up: fractured hip, expected and individualised discharge
survival less than 6 months, Group 2: 56 (47.9%)
6 months planning. All nursing staff
residence in a nursing home on the ward received Decline in transfers- data 3 months Additional outcomes
and dependence and at least specialised education about missing for 3 patients in Group 1: 57 (44.5%) reported:
one person for ambulation the care of elderly with hip group 1 and 8 patients in Group 2: 48 (40.7%) Baseline
before the fracture, or fracture. A physiotherapist, group 2 at 3 months. characteristics such
residence outside occupational therapist or a 6 months as: Functional and
metropolitan Toronto. clinical nurse specialist and Group 1: 45 (36.3%) cognitive scores,
Group 2: 44 (37.6%)
398 APPENDIX E

Patients were excluded social worker assigned to Change in residence 3 months medical indicators,
postoperatively if the surgery the ward routinely assessed Group 1: 31 (23.7%) surgical procedure.
failed for technical reasons, if all study patients within 72 Group 2: 32 (25.4%) Care by allied health
they required care in an hours. Daily medical care professional. Place of
intensive care unit or of there from a senior internal 6 months residence at
was no bed available on the medicine resident Group 1: 22 (17.7%) discharge,
interdisciplinary care ward. supervised by an internist- Group 2: 23 (19.7%)
geriatrician. Notes:
All patients
N: Group 2 Intervention group
received more
Lost to follow up:
physiotherapy hours
Age (mean +SD): Patients had access to allied than the control
M/F: health professionals if a p<0.001
consultation was requested,
Group 1 interdisciplinary care but had limited access to an Length of stay in hospital, Group 1: 29.2 (22.6)
No.: 141 occupational therapist or a days (SD) Group 2: 20.9 (18.8) A subgroup analysis
No. of dropouts: 0 clinical nurse specialist. in the paper shows a
Age (mean): 83.8 (6.9) trend towards benefit
M/F: 32/109 in patients with mild
Other factors: to moderate
Living alone: 23.4% cognitive
Mean time to surgery: 1.3 days impairment.
Subcapital fractures: 46.8%
NB Intensive
Group 2 Usual care intervention during
No. : 138 hospital stay.
Withdrawal: 1
Age (mean): 84.6 (7.3)
M/F: 24/114
Other factors:
Living alone: 23.2%
Mean time to surgery: 1.4 days
Subcapital fractures: 39.1%
APPENDIX E 399

Marcantonio et Patient group: Group 1: Intervention Delirium: Total Group 1: 20 Funding:
al., 2001203,203 Patients with proximal hip fracture Geriatric consultation cumulative incidence Group 2: 32 Part funded by a pilot
preoperatively or within 24h during acute project grant from
Country of postoperatively. A geriatrician hospitalisation the Older Americans
Inclusion criteria:
study: performed daily visits for the Independence Centre
All patients aged 65 and older
USA duration of hospitalisation Severe delirium: Group 1: 7 and a grant from the
admitted to an academic tertiary
medical center for primary surgical and made targeted cumulative incidence Group 2: 18 Charles Farnsworth
Study design: recommendations based on a during acute Trust.
repair of hip fracture.
RCT structured protocol. The hospitalisation
Exclusion criteria:
Presence of metastatic cancer or protocol included 10 modules Hospital days of delirium Group 1: 2.9+2 Limitations:
other comorbid comorbid illnesses each containing 2 to 5 specific per episode (mean +SD) Group 2: 3.1+2.3 Not MDR rehab,
Duration of recommendations. Detailed
likely to reduce life expectancy to focus on impact of
follow-up: fully in the paper, includes
less than 6 months, or inability to Hospital length of stay Group 1: 5+2 geriatric consultation.
obtain informed consent within 24h adequate CNS oxygen (median +IQR) Group 2: 5+2
of surgery or 48h of admission delivery, fluid/electrolyte
balance, treatment of severe Notes:
All patients pain, elimination of Recommendations
N: 126 unnecessary medications, made, and adherence
Lost to follow up: regulation of bowel/bladder to them varied. Full
Discharged to Group 1: 92%
Age (mean +SD): function, adequate nutritional data given in paper.
institutional setting Group 2: 88%
M/F: intake, early mobilization and
(nursing home, rehab
rehab, management of postop
hospital)
Group 1 Geriatric consultation complications, appropriate
No.: 62 environmental stimuli,
No. of dropouts: treatment of agitated Delirium at hospital Group 1: 8
Age (mean): 78+8 delirium. discharge Group 2: 12
M/F: 79% female
Other factors: Group 2:
Pre fracture dementia: (Blessed usual care
score ≥4) :21 Management by orthopaedic
Prefracture ADL impairment (Katz team, including internal
ADL score <5: 11 medicine or geriatric consults
400 APPENDIX E

on a reactive rather than
Group 2 Usual care proactive basis.
No. : 64
No. of dropouts:
Age (mean): 80+8
M/F: 78% female
Other factors:
Pre fracture dementia: (Blessed
score ≥4) :29
Prefracture ADL impairment (Katz
ADL score <5: 18
APPENDIX E 401

Shyu et al., Patient group: Patients recruited from the Length of hospital Group 1: 10.1 (3.7) Funding:
2008305,306 Patients with hip fracture emergency room by research stay, mean days (SD) Group 2: 9.72 (4.96) Supported by grants
assistants. from the National
Country of Health Research
Setting: Recovery of walking at 6 months
study: Institute, Taiwan.
Teaching hospital in Taiwan Group 1 ability Group 1: 62
Taiwan
Interdisciplinary programme Group 2: 44
Inclusion criteria: of geriatric consultation, Limitations:
Study design:
Aged 60 or over, admitted to continuous rehabilitation and at 12 months
RCT
hospital for an accidental single- discharge planning. Group 1: 61
Outcomes not
side hip fracture, receiving hip Geriatrician and geriatric Group 2: 49
Duration of reported:
arthroplasty or internal fixation, nurses provided geriatric Mortality at 6 months
follow-up:
able to perform full range of assessment/consultation; Group 1: 6
1 year motion against gravity and physiotherapist, geriatric Group 2: 8
against some or full resistance nurses and rehab physician Additional outcomes
and had a prefracture Chinese were responsible for rehab at 12 months reported:
Barthel Index score >70, and programme, Early Group 1: 13 Marital status,
living in northern Taiwan mobilisation, home visit and Group 2: 15 educational
follow-up services provided.4x Non- Group 1: 59 background.
Exclusion criteria: 30min physical therapy recovery/decline in Group 2: 56 Occurrence of falls,
Severely cognitively impaired, sessions per patient, 2 walking ability – self-care ability,
making them unable to follow assessments from a physical long-term at 12 depressive symptoms
orders or terminally ill. therapist and one visit from months (additional
rehab physician. 4 home visits info from Cochrane Notes:
All patients during first month and 4 review Handoll 2009)
N: 162 during second and third Includes early
Age (mean +SD): 78 month from a geriatric nurse. mobilisation and
M/F: 31.5% male intensive rehab.
Group 2
Group 1 Intervention On trauma or orthopaedic Intervention
No.: 80 ward. Occasional consultation resembles a geriatric
Age (mean): 77.36 (8.19) with other disciplines hip fracture rehab
M/F: 25/55 depending on patient’s
402 APPENDIX E

Independent walking ability: 68 condition. Exercises taught by programme and early
nurses in first 2 to 3 days. supportive discharge.
Group 2 Usual care Physical therapy sessions
No. : 82 varied according to insurance
Age (mean): 78.94 (7.28) policy.
M/F: 26/56
Independent walking ability: 69
APPENDIX E 403

Stenvall et al., Patient group: Living independently At 4 months Funding:
2007320,320 Patients with femoral neck fracture Group 1: 54 Supported by the
Group 2: 46 Vardal Foundation,
Group 1
Country of the Joint Committee
Setting: Geriatric unit
study: At 12 months of the Northern
Umea University Hospital, Sweden specializing in geriatric
Sweden Group 1: 47 Health Region of
orthopaedic patients. Group 2: 36
Inclusion criteria: Active prevention, Sweden, the JC
Study design: Independent walking At 4 months Kempe Memorial
RCT • Patients aged 70 years or detection and
treatment of post op ability Group 1: 59 Foundation, the
older
complications Group 2: 52 Dementia Fund, and
List who was the Foundation of the
Exclusion criteria: implemented daily.
masked to At 12 months Medical Faculty, the
interventions: • Patients with severe Early mobilisation, with
daily training was Group 1: 55 Borgerskapet of
rheumatoid arthritis, severe
provided by Group 2: 45 Ulmea Research
hip osteoarthritis or a
physiotherapists, Independent walking At 4 months Foundation, the Erik
Duration of pathological fracture, or
without walking aid Group 1: 31 and Anne-Marie
follow-up: severe renal failure. Patients occupational therapists
indoors Group 2: 19 Detlof’s Foundation,
12 month who were bed bound prior to and care staff during
hospital stay. University of Ulmea
the fracture.
Assessment at 4 At 12 months and the County
months by geriatric Group 1: 35 Council of
All patients
team. Group 2: 22 Vasterbotten and the
N: 199
Swedish Research
Lost to follow up: Independent in P-ADL At 4 months
Group 2 Council.
Age (mean +SD): (poorer personal activities Group 1: 35
M/F: 26% male Specialist orthopaedic of daily living) Group 2: 23
unit following Limitations:
Group 1 Intervention conventional At 12 months Not blinded, but
No.: 102 postoperative routines. Group 1: 33 independent
No. of dropouts: A geriatric unit was Group 2: 17 assessors.
Age (mean): 82.3 (6.6) used for those needing Intensity and quality
Length of stay in hospital At 12 months
M/F: 28/74 longer rehab n = 40, of outpatient rehab is
Group 1: 30 (18.1)
Other factors: but this was not the unknown.
Group 2: 40 (40.6)
404 APPENDIX E

Living alone: same ward as the p=0.028
intervention. Mortality At discharge Outcomes not
Group 2 Usual care Group 1: 6 reported:
No. : 97 Group 2: 7 Baseline data such as
No. of dropouts:
Age (mean): 82 (5.9) health and medical
At 12 months problems, functional
M/F: 23/74 Group 1: 16 performance prior to
Other factors: Group 2: 18 fracture.
Living alone
Hospital readmissions At 12 months
Group 1: 38
Additional outcomes
Group 2: 30
reported:
More dependent based Group 1: 35 Notes:
on Katz index at 1 year Group 2: 49
Paper contains a
Non recovery in ADL at 1 Group 1: 51 detailed description
year Group 2: 59 of the intervention
and control group.
APPENDIX E 405

Swanson et al., Patient group: Patients were identified Length of stay (discharge Mean Funding:
1998325,325 Patients with femoral fractures. by the trial coordinator criteria used e.g. when Group 1: 21 (17.2-24.4) Medicare Incentives
in the Accident and medically stable and able Group 2: 32.5 (24.2-41.1) Hospital Access
Country of Emergency Department to transfer and walk p<0.01 Program.
Setting:
study: independently with or
Royal Brisbane Hospital, teaching
Australia without aids) Median
hospital. Group 1 Group 1: 17 Limitations:
Multidisciplinary team: Underpowered –
Study design: Group 2: 24
Inclusion criteria: full time initial power analysis
RCT p<0.01
Patients aged 55 or over; non- physiotherapist, determined that 120
pathological fractures; residing at occupational therapist, patients (60 in each
Duration of Mortality In hospital
home or in a hostel; independently clinical nurse arm) would have the
follow-up: Group 1:2
mobile (with or without a walking consultant, half time power to detect a
aid); able to give informed consent; social worker, Group 2: 2
12months reduction in mean
accessible for follow up (i.e., geriatrician, length of stay of 7
residing in the Brisbane area); and orthopaedic surgeon. 12 months
days at 0.05 level of
public patients. Early mobilisation (1st Group 1:5
significance. However
day after surgery if Group 2: 6
the difference in
Exclusion criteria: possible), twice daily length of stay was
Patients with dementia, with intense sessions by Modified Barthel Index at 6 months larger than
inadequate English to give informed physiotherapist, daily discharge (95% CI) Group 1: 92.8 (90.0-95.6) anticipated.
consent or residing in a nursing assessment, treatment Group 2: 85.6 (81.3-89.8)
home. or counselling by the
12 months No assessor blinding
occupational therapist
All patients and social worker. Group 1: 95.3 (SD 9.8) Outcomes not
N: 71 Review by geriatrician Group 2: 89 (SD 15.8) reported:
406 APPENDIX E

Lost to follow up: 0 on next working day Complications (additional Chest infection, cardiac problem
Age (mean +SD): after surgery, 2 from Cochrane review) bedsore Additional outcomes
M/F: 22% male additional ward rounds Group 1: 6 reported:
attended by all staff, Group 2: 13
Group 1 Early intervention weekly case conference
No.: 38 attended by all staff, Stroke emboli Notes:
Age (mean): 78.5 (75.3-81.7) coordination of care by Group 1: 4 Surgery was carried
M/F: 11/27 trial coordinator, home Group 2: 1 out within 48 hours
Living at home: 35 (92.1%) assessment visit before of admission for 90%
discharge. of intervention and
Group 2 Usual care 80% of standard care.
No. : 33 Group 2 12 month data from
No. of dropouts: Standard orthopaedic Day 2001.
Age (mean): 77.8 (74.0-81.6) management including
M/F: 5/28 daily visits from a
Living at home: 29 (87.9%) physiotherapist, and
social worker or
occupational therapist
visits as requested by
hospital staff. Weekly
discharge planning,
home visits as
requested by social
worker.
APPENDIX E 407

Vidan et al., Patient group: All patients had an Median total length of Group 1: 18 (13 – 24) Funding:
2005344,344 Patients with hip fracture orthopaedic surgeon hospital stay (25th to 75th Group 2: 16 (13 – 19) Not stated
and a nurse assigned percentile) p = 0.06 Limitations:
Country of
Setting: Hospital General Universitario when they were In hospital mortality Group 1: 9 (5.5%) Usual care group
study: admitted to hospital.
“Gregorio Maranon”. Group 2: 1 (0.6%) have a higher
Spain The intervention and p = 0.03 percentage of
Inclusion criteria: control group shared coexisting conditions
Study design: Mortality – end of Group 1: 39
the same orthopaedic
• Consecutive patients aged 65 and scheduled follow up Group 2: 28
wards and used the
RCT older between February 1 and (from Cochrane review) Outcomes not
same hospital-wide
December 15, 1997 for acute hip Major medical Confusion reported:
support services,
Duration of fracture surgery. complications Group 1: 67 (44.1%)
including physical
follow-up: therapy and social work. Group 2: 53 (34.2%)
Exclusion criteria: Additional outcomes
p = 0.07
• Inability to walk before the The orthopaedic reported:
12 months surgeon made the
fracture and dependency in all Pressure sores Additional baseline
basic activities of daily living; decision of discharge data: coexisting
Group 1: 27 (16.9%)
pathological hip fracture; known moment in both groups. Group 2: 8 (5.2%) conditions, type of
terminal illnesses, defined as p = 0.001 fracture, type of
those associated with a life Group 1 surgery. Also medical
expectancy of less than 12 The surgeon and Pneumonia complications: heart
months. orthopaedic nurses Group 1: 6 (3.7%) failure, DVT,
managed patients, with Group 2: 6 (3.9%) myocardial infarction,
All patients counselling from p = 0.95 arrhythmia.
N: 319 different specialists as Notes:
Lost to follow up: needed. Heart failure ADL = activities of
Age (mean +SD): Group 1: 5 daily living. (Bathing,
M/F: 18.5% male Group 2 Group 2: 12 dressing, using the
A geriatrician visited the Time from surgery to Group 1: 10.2 (6) toilet, getting from
Group 1 Usual care patients daily and was bed to chair, and
rehabilitation, days, Group 2: 8.3 (3.9)
No.: 164 responsible for medical continence)
mean (SD) p = 0.007
No. of dropouts: not stated care. After initial
Recovery of ADL or FAC Group 1: 3 (2%) FAC = Functional
408 APPENDIX E

Age (mean): 82.6 (±7.4) assessment and within at time of hospital Group 2: 5 (3%) Ambulation
M/F: 35/129 72 hours after discharge Classification. This
Living at home before admission: 134 admission, there was an Recovery of ADL or FAC Group 1: 59/134 (44%) consists of 6 different
(82%) interdisciplinary at time of 3 months Group 2: 82/144 (57%) functional levels.
Type of surgery: meeting, including the p = 0.03
Internal fixation: 101 (61.6%) orthopaedic and
Incomplete recovery of Group 1: 75
Prosthetic replacement: 53 (32.3%) geriatric teams, to
ADL and mobility at 1 Group 2: 67
Others: 10 (6.1%) discuss the patient’s
year (from Cochrane
Mean time to surgery, hours (SD): 78.5 medical, functional, and
review)
±53.2 social problems and to
elaborate a
Group 2 Intervention comprehensive
No. : 155 therapeutic plan. The
No. of dropouts: not stated meeting was repeated
Age (mean): 81.1 (±7.8) weekly.
M/F: 24/131
Living at home before admission: 135
(87%)
Internal fixation: 91 (58.7%)
Prosthetic replacement: 58 (37.4%)
Others: 6 (3.9%)
Mean time to surgery, hours (SD): 75.8
±43.2
APPENDIX E 409

Ziden et al., Patient group: A geriatric nurse who Balance confidence – Swedish 1 month Funding:
2008 and Ziden Community-dwelling patients performed the randomisation version of the Falls Efficacy Group 1: 117.4 (12.0) Supported by the
et al., with hip fracture using sealed envelopes. Scale – 1 month (SD) . 0-10 Group 2: 85.5 (30.5) Vardal Institute, the
2010360,361 scale where 0 indicates very p<0.0001 Hjalmar Svensson’s
confident, no fear of falling, 10 Foundation and the
Setting: Patients with hip fracture
Country of is not confident, very afraid of Geriatric Section of
study: Patients admitted to the were referred from the falling. Swedish. Includes 13
emergency unit at the emergency unit to a geriatric the Swedish
Sweden items covering activities of Association of
Sahlgrenska University ward with home rehab (group
daily living. Registered
Study design: Hospital 1) or with conventional care.
Physiotherapists.
RCT Activities of daily living and 1 month
Inclusion criteria: Both groups performed early
Acute hip fracture surgery, leisure activities – degree of Self-care Limitations:
Duration of mobilization, preferably independence assessed by Group 1: 38.4 (2.9)
follow-up: medically approved by the within 48 h. When needed an No length of hospital
responsible geriatric doctor Functional Independent Group 2: 33.5 (7.2)
occupational therapist or stay or total length of
as being in need of geriatric Measure (FIM) motor scale
1 year physiotherapist made a home rehab in control
care and rehab, aged 65 or (mean, SD). 13 items with a 7 Mobility
visit with the patient to assess group.
over and able to speak and point grading scale (0 = totally Group 1: 18.3 (1.5)
if they could manage and dependent and 7 = totally Group 2: 16.3 (3.3)
understand Swedish. what aids they needed. independent) max score 91 Outcomes not
points Locomotion reported:
Exclusion criteria:
Severe mental illness with Group 1 Group 1: 10.4 (2.5)
expected survival of less than Group 2: 7.6 (3.6)
one year, severe drug or Conventional care and rehab Additional outcomes
alcohol abuse, mental illness as in group 2, plus supported 6 months – median with range
reported:
or documented severe discharge. An initial meeting Self-care
with the patient aimed to Group 1: 40 (33-42) Other baseline
cognitive impairment. characteristics such
establish individual goals. Group 2: 37 (6-42)
as walking ability,
All patients Close contact with social
Locomotion number of medical
N: 102 home services and relatives to
plan discharge and Group 1: 31 (15-34) diagnosis, functional
Total of 212 randomised: independence and
Excluded: 99 cooperation during 1st few Group 2: 30 (5-35)
instrumental activity.
410 APPENDIX E

Declined to participate: 11 weeks at home. Home rehab 1 year – median with range Subsequent falls,
Lost to follow up: consisted of a 3 week Self-care frequency of
Age (mean +SD): 81.9 (6.8) intervention period. Group 1: 40 (23-42) activities, balance
M/F: Group 2: 38 (12-42) confidence.
Group 2
Group 1 Home rehab Participation in standard Locomotion
Notes:
No.: 48 rehab including daily training Group 1: 32 (11-35)
No. of dropouts: in basic activities: transfer Group 2: 29 (9-35)
Age (mean): 81.2 (5.9) techniques, technical aids,
M/F: 19/29 indoor and stair walking. Also Basic physical mobility – 1 month
Other factors: physiotherapy and timed “up and go” test. Assess Group 1: 24.9 (15.4)
Living alone: 26 occupational therapy group total time for standing up from Group 2: 30.8 (16.0)
sessions. Prior to discharge a chair, walking 3 m, turning
Group 2 Usual care the home service officer and 180o returning and sitting
No. : 54 patient’s next of kin was down (performed twice, one
No. of dropouts: contacted to make plans for trial and one timed)
Age (mean): 82.5 (7.6) the future. All rehab
Functional lower extremity 1 month
M/F: 12/42 measures were adapted to
muscle strength. Ability to rise Group 1: 1.8 (0.8)
Other factors: the patient’s individual
from a chair was measured by Group 2: 3.3 (3.6)
Living alone 39 medical and functional status
sit-to-stand. Best of 3 trials
and personal goals.
was recorded. Made in
participant’s home with
ordinary chairs, preferable
with armrests, and ordinary
walking aids were used, if
needed (secs)
Length of hospital stay (mean Group 1: 18.4 (8.4)

+/- SD) Group 2: 20.0 (6.8)
APPENDIX E 411
17.12 Evidence Table 12: Patient views

Study Archibald 20038. Country: UK. Setting: community hospital in Bradford
Aim To explore experiences of individuals who had suffered a hip fracture. Not to produce generalisable findings but to generate "rich description" of the
experience of incurring and recovering from hip fracture to inform nursing practice.
Population 5 patients with hip fracture
Age >65; 4 women and 1 man; all were cognitively intact
Method of In depth audio-recorded interviews with open ended questions, ranging between 25 and 50 minutes duration were conducted during stay in the
gaining views hospital.
Data analysis “Colaizzi's analysis framework. 6 step methodological interpretation
• Interviews transcribed verbatim and read to get a feel for responses
• Significant statements and phrases extracted
• Meanings formulated from significant statements
• Organised into clusters of themes
• Themes used to provide full description of experience
• Researcher returns description to participants for confirmation of validity
Findings 4 main themes: injury experience, pain experience, recovery experience, disability experience.
Injury – relates to falling and breaking their hip
Pain - Most participants described the pain they had. One mentioned being in a lot of pain in orthopaedic unit despite pain killers. Another mentioned
they thought the pain went with rest after a while, but not completely. Only 1 person was still having pain at time of interview. One said "…I have not
suffered, not what I call real pain, at all",
Recovery - operation: varied comments - some did not remember anything or much, one had a “horrendous” recollection of operating theatre: “The
operation was pretty horrendous. I had the injection in the spinal cord, [an] epidural… There was no pain, but the noises [laughs] – it was like being in
an engineering shop or something. The noise was terrible. I thought ‘What are they doing me?’ Anyway, it came to an en (it took quite a long
time)..and before I knew it I was back on the ward.”
Recovery - beginning struggle: 3 patients discussed this, 1: not being able to do anything, 2: struggling to get to toilet & into the chair, 3: hated using
bed pan.
Recovery - regaining independence: Motivation found to be key factor in recovery, all comments in study positive comments about regaining
independence during their rehabilitation.
Disability: comments about reduced functional status, dependence on others, being house bound.
Comments Not stated how patients were selected for the study. No baseline data provided about patients. The role of the researcher is not described.
412 APPENDIX E
Evidence tables – patient views

Study Borkan 1991 & 199228,29. Country: USA. Setting: 4 hospitals (no more detail)
Aim Two research questions addressed:
• What are the meanings present in the narratives of elderly hip fracture patients?
• What is the importance of narrative elements as prognostic indicators or 'risk factors' for predicting rehabilitation outcomes?
Population 80 patients with hip fracture (from a pool of 174) "functionally hardy elderly, intact mental status, independent or lightly-supervised residence outside
long-term care facilities, full pre-fracture ambulation; >65 years; 65 women and 15 men; diagnosed within 48 hours of fracture; treated surgically
within 1 week.
Excluded open pathological or multiple fractures.
Method of Interviewed during first week after hip fracture, generally 1 or 2 days after surgery, in participant’s hospital room. In depth initial interviews included
gaining views demographics, open ended questions and standardised scales. Combination of open-ended and multiple choice questions. Interview content validated
through pretesting with 10 subjects, and reviewed by a panel of experts. Inconsistencies and ambiguities revised or deleted from study. Follow up
interviews at 3 and 6 months post-fracture generally conducted in participants’ current residence, except where movement to distant states or
particular patient preferences precluded face to face contact. These attempted to match some of the patient’s perceptions to what actually happened.
In addition, observations carried out on main orthopaedic floors over the course of 2 years in order to familiarise research team with the treatment
and rehabilitation as well as to confirm information drawn from interviews and uncover unexpected associations.
Data analysis Quantitative analysis & qualitative narrative. Names coded and interview transcripts sent to independent expert panel to identify emergent or
recurrent themes. 13 dimensions identified and grouped into 3 composite. Subjects’ narrative accounts rated on a 7 point bipolar scale.
Findings Gives themes around the patient perception of hip fracture, how it happened, how they perceive their injury, what the future holds, their subsequent
level of ability and their future. Categories derived from narratives are rated on a bipolar scale and presented in 3 groups. The remaining percentage
not given for each category relates to patients either not giving a view or indicating an equal rating for both polar elements.
1. Explanation of fracture: described as disease (1%) or fracture (49%); fall as secondary (4%) or primary (82%); etiology, internal degeneration –
primary (6%) or secondary (11%); broke and fell (10%) or fell and broke (64%); course of rehabilitation described as chronic (19%) or acute
(49%); functional severity – total impairment (14%) or complete recovery (70%); range of severity – whole body (11%) or affected leg or hip
(15%).
2. Perception of disability: vulnerable (41%) or not vulnerable (34%); dependency increased (21%) or not increased (30%); sense of alienation
from the world – alienated (20%) or integrated (29%); objectification of body part – alienation (4%) or wholeness (7%).
3. Futurity: hopefulness (54%) or hopelessness (19%).
Expectations of recovery during initial hospitalisation: 43 (53.7%) expected full recovery; 14 (17.5%) partial recovery; the rest did not know or did not
give an answer. Narrative responses varied "from stubborn optimism to despair".
Expectations of living situation: 61% predicted going home, 15% predicted going into a nursing home (none came from nursing home), 9% going to
children's house, 15% did not know or did not respond. Actual figures: 34 (43%) discharged to long-term care institutions, 13 (38%) of these remained
in institution at 1 year, 18 (53%) returned home, 3 (9%) died.
Comments The role of the researcher is not well described.
APPENDIX E 413

Study Bowman 199733. Country: Canada. Setting: hospital
Aim To describe sleep satisfaction, pain perceptions & psychological concerns of patients undergoing planned & emergency hip operations. Two additional
questions on perceptions of how they would manage.
Population 43 out of 50 consecutively admitted patients: 17 with hip fracture & 26 undergoing elective hip replacement. Gender for overall study 29 women and
14 men. Characteristics of hip fracture patients: mean age 80 (+7.5); 8/17 had delirium; 11/17 patients claimed to be active or very active prior to
fracture
Method of Pain assessment was conducted using a visual analogue scale. Sleep satisfaction was conducted using a ‘Likert’ scale.
gaining views
Not much detail on methods for qualitative part of study. Interviewed on day of admission. Two structured questions but no details on how or by
whom they were delivered. 1. What are your biggest concerns at this time as a result of this injury and your upcoming surgery? 2. Do you have any
concerns about your ability to recover fully and quickly?
Data analysis Numerical analysis of responses to two questions.
Findings 6/17 feared being unable to walk again; additional 3/17 concerned about recovery and managing on their own; 5/17 put their trust in God.
Comments Little detail about methods used for the qualitative part of this review. Little baseline data provided about patients. The role of the researcher is not
described.
414 APPENDIX E

Study Furstenberg 1986105. Country: USA. Setting: Large urban teaching hospital
Aim 2 parts to study: 1. community residents without hip fracture, 2. hospitalised patients with hip fracture. "The purpose of hospital study was to
construct a natural history of the hip fracture, from the events surrounding the fracture through the hospitalisation period.
Population 11 patients hospitalised for hip fracture. Patient characteristics: age 59 to 85 years; 4 men & 7 women; cognitively intact, fracture that had not
resulted from malignancy or its treatment.
Method of Interviewed at one or more points during their hospital stay. "Ethnographic interviews" recorded and transcribed in full. Interviews took place in
gaining views physical therapy hospital rooms or in rehabilitation centre for 3 who were transferred. During interview informants requested to talk about the
fracture, their reactions to it, their pre-fracture functioning, their experiences during hospitalisation and the process for planning for discharge.
Data analysis "Analysis consisted of identifying salient and recurrent issues and themes and grouping the portions of the interviews dealing with each theme. The
variations on each theme were described, and correlates of these variations were identified".
Findings Split into two main sets: (1) immediate expectations about recovery explicitly or implicitly expressed by patients; (2) contextual factors to the evolving
expectations about recovery.
1. Immediate or early expectations of recovery - most expressions of despair and discouragement. Only 1 patient feared "it was over". First
reactions "varied from shock to a focus on immediate problems, and for some, immediate concern about the consequences of their way of
life. As the situation progressed, patients' concerns focused more exclusively on limitations on their functioning and the implications these
would have". Most expressed worry about the degree to which they would recover, and when. Several talked repeatedly about the slowness
of the process of recovery of physical function. Some worried about being burdens on their caretakers, some worried about further falls.
Those who went temporarily to a home of an adult child worried about being able to return to independent living. Summary - hip fracture was
going to result in extended period of slow recovery of function, with attendant dependency, postponement or relinquishment of cherished
plans and changed living situation with the threat of permanent loss of independent living. Also suffered uncertainty about timing &
completeness of return to full recovery.
2. Contextual factors - as time progressed. Only positive points, not negative ones, came out in this section. Patients observing their own
progress sometime after surgery commented that although progress slow they could see improvement. Participants also took encouragement
from others progress. The study notes that while patients could focus on positive and negative points, the informants only focused on
encouraging examples.
3. Contextual factors - health professionals influence on patients' perceptions. Healthcare professionals’ cues, encouragement and feedback
guided the informants’ perceptions about their own progress. Quotes of the healthcare professionals were scattered throughout participants’
responses. Some patients “referred to the elusiveness of the doctors and their own unanswered questions.”
4. Contextual factors - other health issues. Also reports a few comments by patients on other health issues.
Comments Little baseline data provided about patients. The role of the researcher is not described.
APPENDIX E 415

Study Olsson 2007249. Country: Sweden. Setting: geriatric/orthopaedic ward
Aim To describe patients' own perceptions of their situation and views of their responsibility in the rehabilitation process.
Population 13 hip fracture patients from a geriatric/orthopaedic ward, non-institutional residence pre-fracture, median age 81 (range 71 to 93) years, 2 men & 11
women. Excluded patients with severe illness, cognitive impairment, dementia or pathological fracture.
Method of 30-45 minute interviews conducted in informant’s room or in a secluded area of the ward as soon after the operation as the informants felt strong
gaining views enough. Semi-structured questions were used "such that the main questions, related to the informant's perception of the transitional properties, were
included in all interviews." Deliberate efforts were made to encourage informants to reveal and comment freely on their personal experiences of and
reflections on their situation, without imposing the interviewer's own values on what was being said. The interviewees all talked freely and appeared
to be grateful for the attention and for having someone to listen to their reflections. All interviews were recorded and transcribed.
Data analysis Transcripts read several times. 5 transitional properties & 542 meaning units identified & pooled. A “saturation” was observed when 9 interviews had
been conducted "...meaning units describing qualitatively similar conceptions were grouped together and the nature of this similarity was articulated."
Categories were labelled and exemplified with representative quotations from interviews. To test the reliability of the categories the second author
evaluated the categories in relation to the interviews.
Findings Participant’s responses were categorised into different conceptions:
• autonomous – appeared confident and accustomed to managing for themselves and being in control of their lives. Willing to listen to staff,
but made their own decisions. Even if they appeared strong they felt just as vulnerable as the other groups. However, they were aware of the
importance of information, personal support and their own responsibility.
One informant commented that more information given preoperatively could have made a great difference:
o “Of course, if someone had come and sat down for a little while and talked. If they had said something like, this is what it will be like
and so on and after a while you will be able to walk and maybe manage on your own again. That would have been reassuring, it really
would. Because I really must say, at moments like that, you get a feeling of being small and insignificant.”
• modest – gave the impression of being vulnerable and dependent on others and they expressed themselves cautiously. Instead of demanding
community aftercare like the Autonomous, they were willing to go along with what was offered. These informants appreciated information
offered to them but for some reason they did not request more, even though they seemed to want to. They worried more about their future
ability to walk and maintain their former lifestyles than the other two groups. They feared being discharged, saw only problems and appeared
unaware of the progress they had made. They were reluctant to talk about their hopes for the future and did not see their responsibility as
clearly as the autonomous.
• heedless – “appeared to view their situation with some detachment, almost as if it did not concern them. The Heedless did not doubt they
would recover and they were confident that people around them would care for them.” “The Heedless were characterized predominantly by
a reluctance to reflect on their own situation, by a refusal to accept responsibility and by their need for information. ….They did not appear
to have reached a stage where planning for the future was relevant.”
Also identifies some common traits:
• [lack of] awareness - most lacked adequate awareness about their condition, what to do and how to act and needed more information. Only
416 APPENDIX E
1 patient knew someone who had undergone rehabilitation for hip fracture.
• shocking event - although several suspected they had a fracture all were distressed by the diagnosis. Period before surgery was mostly
blurred and filled with fear and pain. They worried about how they would function postoperatively;
• zest for life - all expressed a strong desire to recuperate. While confined to bed they were worried remembering the pain and inability to
move their leg. The suffering experienced in anticipation and preparation for the operation led them to believe they might not be able to
walk.
Comments Not stated how patients were 'strategically selected' for the study. Little baseline data provided about patients. The role of the researcher is not
described.
APPENDIX E 417

Study Pownall 2004274. Country: UK. Setting: trauma and orthopaedic ward
Aim Critical appraisal of an individual patient narrative of their experience with hip fracture. Undertaken in an effort to understand further the nature of
personal experience. Narrative was acquired as part of a routine nursing evaluation and helped to illuminate nursing care issues through they eyes of
the patient.
Population A 60 year old woman with an intracapsular fracture in Nottingham hospital. She stated she was fully independent prior to fracture.
Method of Interviewed prior to discharge from acute trauma and orthopaedic ward, exact time point unclear. A list of structured questions were devised but not
gaining views rigidly adhered to:
• What did you feel about requiring hospitalisation?
• What were the good aspects of your hospitalisation?
• What were the bad aspects of your hospitalisation?
• What do you feel could be improved?
Data analysis Narrative assessment of patient’s views
Findings A few areas for potential improvement for the hospital/department were identified:
• communication skills
• time management for staff so time spent with patient is used effectively
• pain management
Ann's comments that were included in the study:
• I could not understand why I had to wait so long in A & E, they had done the X-ray, it was broken the X-ray person told me that. So why did I
have to wait?
• The pain was unbearable; I didn’t care what happened or what was said I just wanted to get rid of the pain.
• The staff were so kind, they could not do enough for me.
• Initially, I could not understand why they (the staff) wanted to keep checking my bottom, I was comfortable why keep moving me?
• It was terrible to be kept nil by mouth the first day, I didn’t feel like eating but I really wanted a drink.
• It was such a disappointment to be told my operation was cancelled; I just wanted to be fixed.
• When I came back from theatre I really needed a drink, but I could not reach my glass. I didn’t want to bother the staff they looked so busy.
• It was a relief to come back from theatre and be able to press a button and get pain relief, but it was taken away the next day when the
physiotherapist came. So I had to keep asking for pain killers.
• The staff are so busy no one has time to sit and explain things to you.
• I could hear the nurse explaining the operation to my son, but what about me I needed to know.
• It was frightening to wake up from the operation and see that I was having a blood transfusion, no-one said that I might need a blood
transfusion. It makes you feel something has gone terribly wrong.
• I couldn’t believe it when they wanted to mobilise me the day after the operation, even my son was shocked to see me out of bed.
418 APPENDIX E
Comments Almost no methodology described so results could be unreliable. It is unclear how this patient was chosen. The role of the researcher is not described.
APPENDIX E 419

Study Slauenwhite 1998314. Country: Canada. Setting: interviewed at home after discharge from hospital
Aim Purpose of this study was to investigate the impact of enhanced early discharge on families experiencing repaired hip fracture in an older adult.
Population Convenience sample of 23 caregivers for 23 patients who had experienced hip fracture. Patient characteristics: age 75.9 (range 56-97) years; 19
women 4 men. Care giver characteristics: 16 women, 7 men.
Method of Interviewed 4 to 6 weeks after discharge with questions adapted from Canadian Patient Centered Hospital Care Survey, a “validated tool”. The
gaining views developers of the tool determined the adaptations would not affect the validity or reliability of the questionnaire. Questionnaires mailed to caregiver 1
week before interview (questions reported in study). "Data were considered qualitative if the interviewee elaborated on an answer, expressing strong
beliefs about a topic. At the end of the interview, specific questions were asked to elicit how the patient and family experienced the illness episode."
One interviewer was used. Interviews were taped then transcribed.
Data analysis 2 investigators separately analysed the transcripts and developed themes that emerged from the data. Themes were then compared, contrasted and
collapsed until only main themes remained.
Findings • Length of stay not a major issue for 15/23 families, care-recipient thought too long while patient-carer thought too short for 3 families, 4
families said people heal better in own homes.
• 20/23 families stated pain management not a problem in hospital or at home
• several families thought transition from house to home a problem as took several hours to days for all info to be relayed to home care
system. This went hand in hand for those with comorbidities.
• “Instrumental functioning” not a concern when patients were allowed to manipulate their own resources in their own home.
• Older people and men more capable of role flexibility while younger people and women talked more about role strain.
• Many caregivers had stories of dissatisfaction which was suggested to be related to health care system and mismatched care. Mismatched
care not well defined.
Comments No description of ‘early supported discharge’. No baseline description of patients and no indication of how patients were selected.
420 APPENDIX E

Study Williams 1994354. Country: USA. Setting: At home after discharge from 4 hospitals
Aim Aim to gain information on: (1) the recovery pattern in functional status & mood in first 14 weeks after hospital discharge; (2) factors most associated
with the extent of assistance required in specific mobility activities & patient assessment of their problems; (3) problems patients identified as most
important; (4) advice those patients would give to others.
Population 120 consecutive patients meeting inclusion criteria with hip fracture. Older patients who were relatively healthy & home dwelling before fracture.
Mean 79.9 (+9.7) range 60 to 100). Included intracapsular (68), extracapsular (52), internal fixation (76), femoral head replacement (44). Sample
included only white women as a result of low number in region of study.
Method of Interviewed before hospital discharge and followed up at 2, 8 and 14 weeks. At 14 weeks participants asked what advice they would give to other
gaining views persons who fractured their hip. Also assessed functional status, perceived return to normal mobility, mood states, and other factors including urinary
problems according to scales.
Data analysis Coding of responses to advice to give to other hip fracture patients done by the two "co-principal investigators" with recategorisation occurring until
100% agreement was reached.
Findings Advice to patients with newly fractured hips from women with a personal experience of hip fracture
Number of comments by category:
• 94 importance of mental attitude - maintain hope & look to the future
• 76 follow experts advice
• 34 mobility – keep mobile, rest before getting up to walk, use walker to help get up
• 15 maintain healthy lifestyle
• 7 use caution & be careful not to fall
• 3 limit stay in institution and get help to be at home if possible;
• 6 gave no specific advice as they commented that everyone is different.
Comments
APPENDIX E 421

Study Wykes 2009355. Country: Australia. Setting: rehabilitation hospital
Aim Pilot study to explore "the impact of fractured neck of femur on the lives of previously independent women and identifies their concerns when
participating in inpatient rehabilitation".
Population 5 patients undergoing inpatient rehabilitation for hip fracture at 2 rehabilitation hospitals, aged 60-85 years, living alone and independently before
fracture, cognitively intact and able to converse fluently in English
Method of Interviewed in a private room during stay in rehabilitation hospital. Interviews were shared by two researchers previously unknown to the patients.
gaining views Interviews taped and transcribed verbatim. Each woman invited to tell her story with as few interruptions as possible. Main questions were "Can you
tell me how you came to be here in this rehabilitation ward?" and "What do you think about while you are in hospital?" To ensure in depth coverage
patients were frequently asked "Could you please tell me more about that?”
Data analysis Thematic analysis using stages set out by Burnard 1991. Two researchers independently made notes of themes apparent in the data as a whole.
Transcript lines were coded. Similar codes combined into higher order categories. Same two researchers carried out the analysis. Researchers engaged
in "reflexive self-awareness". This included a conscious awareness of previous experiences of and with patients who had fractured a neck of the femur.
Findings Two major findings:
1. Impact of fracture for previously independent women was an issue for all. Primarily, others had to assume responsibility for things they had
done previously.
2. Concerns following fracture listed in 4 sections:
a. behaviour of others (22 instances identified) - these included:
what others do - things staff said or did (1 women upset when she overheard staff talking about the possibility of limb shortening if
they stayed in a wheelchair too long, 1 women upset about being put on a ward with people “completely of the planet” as a result of
dementia; 1 patient commented that staff don’t understand because they encouraged her to walk when she felt she could not
because of Parkinson’s Disease interfering with her mobility), friends & family doing things without consulting her
what others do not do -; family not told by staff when patient moved hospital; not enough information about complications
what others expect – 1 women concerned by staff expecting her daughter to look after her before rehabilitation started; family and
friends expectations upset participants.
b. what was happening to them - possible accommodation changes after discharge; possible loss of independence; money issues
c. impact of their injury on others - inconveniencing and upsetting others
d. other health issues – 2 women had pre-existing conditions that overshadowed their concerns about hip fracture and had adverse
effects on their rehabilitation outcomes. – 1 severely disabled with Parkinson’s Disease; 1 had recent cardiac surgery and a long-
standing vertebral disc prolapse.
Comments Study notes: only 5 patients included so it only reveals some of the concerns of older women with hip fractures; not enough data to explore the
differences between hospitals; analysis only at 1 point in time.
422 APPENDIX E

Study Young 2009358, Country: USA. Setting: rehabilitation programme
Aim To explore the perceptions of older adults regarding their functional recovery 1 year after hip fracture.
Population 62 hip fracture patients (‘convenience sample’ from a longitudinal study of rehabilitation and functional recovery after hip fracture involving 280
patients). Age 65 or older (average: 78, range 65-91), 47 women, 15 men, cognitively intact, community dwelling, admitted to one of the five
predetermined rehabilitation sites with a primary diagnosis of acute hip fracture, receiving a surgical procedure, non pathological fracture, no evidence
of metastatic cancer.
Method of Participants invited and completed an exit interview immediately after the 12-month post hip fracture follow up data collection. The exit interview
gaining views was a thematic survey with open-ended questions that explored areas of functional recovery and participants’ willingness to engage in rehabilitation
activities. Questions:
1. Have you been satisfied with your functional recovery since your hip fracture surgery? – YES, NO
a. If “YES” what do you think has helped the most with regards to your recovery process?
b. If “NO” what do you think has hindered your recovery process most?
c. If “NO” what things would you have liked to see differently regarding you recovery process?
2. What do you think needs to be done to help improve the functional recovery process for future hip fracture patients?
3. What one piece of advice would you give a hip fracture patient to help them with their recovery?
Responses were transcribed verbatim by a physical therapist and a physician assistant, both of whom were familiar with hip fracture care and received
three sessions of interview training at the Center on Aging and Health at John Hopkins University in Baltimore.
Data analysis Data analysis conducted using basic content analysis. “Although the interview guide used in this study contained specific themes and directed
participants to address things that facilitated their recovery process, response analysis was conducted using participants’ own words to capture their
particular responses and ideas about thematic areas.” A list including a definition of each code was developed and continually revised as new codes
were added. “
“Confirmability” Data were initially coded by first reviewer, a geriatric nurse practitioner, and researcher familiar with the hip fracture trajectory. The
coded data were then given to a second researcher, an epidemiologist and gerontologist who had studied patients post-hip fracture across the entire
recovery period. The second interviewer independently coded the transcripts, compared her coding to the coding of the first reviewer, and then
discussed the findings with the first reviewer. As the discrepancies were identified, the reviewers went back to the data to clarify their interpretations.
This process repeated until consensus was reached. Codes were then grouped based on similarities and differences.
Data credibility was addressed by presenting the findings to an interdisciplinary group of clinicians and researchers (one physician, four
epidemiologists, three exercise trainers, one physical therapist, and one occupational therapist) familiar with the hip fracture trajectory to establish if
the findings made sense and were consistent with the current understanding of the recovery process post hip-fracture. The findings were presented in
a small group and one on one in the clinical setting. Participants were asked to verbally confirm or refute the findings.
Findings 53 participants were satisfied with their functional recovery, 9 were not satisfied. 25 codes were identified and collapsed into four main themes.
1. Facilitators of recovery (identified by 53 participants satisfied with their recovery):
• professionals (40) – comments covered being buoyed by seeing physician frequently, having good doctors or surgeons, getting
“correct” or “professional” care. “They evaluated professionals as a team and did not single out one provider over another in terms
APPENDIX E 423
of help and support received”. Communication and a positive attitude by professionals also important;
• social support (13) – from family and friends essential to their recovery. Specifically mentioned verbal encouragement helped them
maintain a positive attitude
• determination (12) – own determination to exercise and be involved
• lifestyle factors (4) & environment (1) – eating healthy food, taking appropriate medications and vitamins, and engaging in physical
activity. “an environment that encouraged healthy behaviors (i.e. facilitated physical activity) was important to promote exercise”
• individualised care – verbal encouragement (4);
• spirituality (4) – spirituality and belief in a supreme being helped them maintain their optimism throughout the process
• identifying goals (3) – returning home, regaining independence and being able to walk like they could prefracture
2. Factors that hinder recovery (identified by 9 participants dissatisfied with their recovery):
• medical complications/comorbidities (4)
• unpleasant sensations (3) – pain reported as a limiting factor
• age (1)
3. System recommendations to facilitate recovery:
• more care (26) – more direct physical & occupational therapy and more education about the recovery process and ways to optimise
physical function
• better care (9) – follow up and care in the home setting after discharge from rehabilitation
• spirituality (3), social support (2) – some participants said they would have like exposure to spiritual support options throughout the
course of their rehabilitation programme. Some participants also felt that additional social and spiritual supports were needed from
family and friends.
• additional information (8)
• elimination of unpleasant sensations (4)
• policy (1)
4. Peer advice to facilitate recovery:
• participate (48) & listen to providers (19) – listen to healthcare instructions and participate as much as possible in rehabilitation
activities. Comments included “listen to the advice from medical staff such as doctors, therapists, and nurses” and “Do a lot of
physical and occupational therapy even if it’s painful
• positive attitude (20) & determination (13) – participants strongly recommended that older adults who sustain hip fractures
maintain a positive attitude, avoid worry and remain determined throughout the recovery experience
• be careful (8) – avoid subsequent trauma, prevent anything that would impede recovery, prevent falls
• push through pain (6), relieve pain – “do your physical therapy even though it may hurt” & “use all offered medications that could
alleviate pain and relax muscles”
• don’t worry (4).
Numbers in brackets relate to the number of times noted
424 APPENDIX E
Comments Paper reports the study used to as the basis to recruit participants for this paper had stringent eligibility criteria because it was designed to evaluate
rehabilitation. Therefore, the findings of this study may only be applicable to a similar patient group. Although the findings were found to be credible
with rehabilitation clinicians and researchers they were not verified with patients who had sustained hip fracture. Themes were determined by the
interview guide.
APPENDIX E 425

Study Ziden 2008 & Ziden 2010362,363. Country: Sweden. Setting: hospital
Aim Aim to explore & describe the experienced consequences of an acute hip fracture among home dwelling elderly people shortly after discharge. "The
ambition was to let the subjects concretize their experiences, for instance by describing in as great details as possible their ordinary daily activities
before and after the fracture."
Population Patients selected from a larger sample of 102 participants (ZIDEN2008 RCT) with acute hip fracture, >65 years old, living in own home, no cognitive
impairment, and able to understand Swedish. Participants asked if they were willing to participate a few days after surgery.
At 1 month: 18 participants, 16 women and 2 men
At 1 year: 15 participants, 13 women and 2 men
Method of Semi-structured interviews using the phenomenographic method. Interviews held in patients own homes 1 month & 1 year after hospital discharge.
gaining views Interviews conducted in a conversational manner that allowed interviewees to speak freely and to express their own experiences of the consequences
of the hip fracture. As an introduction, the subject was asked to narrate what had happened when he or she broke their hip. Follow up questions and
prompts were used, such as "Tell me more about it", What does this mean to you?" and "Can you clarify?" Interviews were taped and were transcribed
verbatim.
Data analysis Phenomenographic method described by Dahlgren & Fallsberg: interviews read through repeatedly to obtain a total concurrent overview; statements
extracted that dealt with consequences of hip fracture to achieve a concentrated and representative version of entire dialogues; quotes from previous
step were compared in order to uncover sources of variation or agreement; similar quotes were grouped together, an attempt was made to "describe
the essence of similarity within each group" (stage called articulating); these groups were then labelled/categorised and compared to ensure
categories did not overlap. The grouping and describing stages were revised several times before the analysis was judged to be satisfactory. Sequence
of steps in the analysis made separately by authors before joint discussions leading finally to consensus.
Findings At 1 month 8 categories in 3 focused areas were identified:
In relation to your body and yourself:
• You are limited to move and have lost confidence in your body (18 people)
• You become humble and grateful (7 people)
• You respect yourself and your own needs (2 people)
In relation to others:
• You become more dependent on others (12 people)
• You gain more human contact and are treated in a friendly way by others (2 people)
In relation to the life situation:
• You are secluded and trapped at home (4 people)
• You are old, closer to death and have lost your zest for life (4 people)
• You take one day at a time and are uncertain about the future (7 people)
At 1 year 6 categories in 2 focused areas were identified:
Experienced consequences of a hip fracture 1 year after discharge
426 APPENDIX E
• Isolated life with more restricted activity and fewer social contacts
a. more insecure and afraid (11 patients)
b. more limited ability to move (12 patients)
• Disappointed and sad that identity and life have changed (8 patients)
• Satisfied with the situation or feeling even better than before fracture (5 patients)
Conceptions of what influences hip fracture recovery
• Own mind and actions influence recovery (10 patients)
• Treatment and actions from others influences recovery (4 patients)
• You cannot influence recovery (6 patients)
Comments
APPENDIX F 427
18 Appendix F: Evidence tables - Economic

studies
Abbreviations
CI Confidence interval
ITT Intention to treat analysis
Int Intervention
LOS Length Of Stay
LR+ Positive likelihood ratio
M/F Male/female
N Total number of patients randomised
NA Not Applicable
NR Not reported
QALY Quality-Adjusted Life Years
QoL Quality of life
RR Relative risk
SA Sensitivity analysis
SD Standard Deviation
SE Standard Error
Sig Statistically significant at 5%
428 APPENDIX F
18.1 Evidence Table 13: General versus regional anaesthesia

details
Chakladar 2010 Patient group: Group 1: Mean (SD) anaesthetic Group 1: 31 (15) Funding/conflict of interest:
UK Hypothetical patients Spinal anaesthesia time (minutes) Group 2: 27 (16) The authors declared there
undergoing uncomplicated p value: p<0.0001 were no competing interests or
Economic analysis: anaesthetic for hip fracture Mean (SD) cost of Group 1: £66.73 (30.05) external funding.
Cost analysis repair. Group 2: anaesthesia equipment Group 2: £108.15 (38.53)
General anaesthesia per patient (2010 GBP) p value: NR Limitations:
Study design Partial economic evaluation.
Mean (SD) cost of Group 1: £1.81 (0)
Survey Survey on hypothetical
airway equipment per Group 2: £25.68 (2.28)
patients, not on real cohorts.
patient (2010 GBP) p value: NR
Duration of follow- Spinal anaesthesia after failure
up: Mean (SD) cost of Group 1: £105.90 (0) of regional was not included in
NA personnel per patient Group 2: £106.76 (0) the analysis. Anaesthetists
(2010 GBP) p value: NR from one hospital only were
Perspective: UK Mean (SD) cost of drugs Group 1: £19.03 (11.00) interviewed.
NHS per patient (2010 GBP) Group 2: £25.17 (11.04)
p value: NR Overall quality and
Discount rates: Mean (SD) cost of Group 1: £0.43 (0.13) applicability
Costs: NA gases/inhalational Group 2: £6.26 (3.94) Potentially serious limitations
Effects: NA agents per patient p value: NR and partial applicability.
(2010 GBP)
Mean total cost per Group 1: £193.81 (37.49) Data sources:
patient (SD) Group 2: £270.58 (44.68) Anaesthetic time from Brighton
2010 GBP, sum of p value: p<0.0001 Hip Fracture Database.
previous categories of
costs. Notes:
Cost-effectiveness NR * 20 anaesthetic consultants
APPENDIX F 429

details
Sensitivity analysis NR
Abbreviations: NR=not reported, NA=not applicable

430 APPENDIX F
18.2 Evidence Table 14: Displaced intracapsular fractures

details
Johansson2006163 Patient group: Group 1: Number of hips that required Group 1: 34 (44%) Funding/conflict of interest:
Sweden Patients 75 years or older Internal fixation reoperation (%) Group 2: 11 (16%) NR
who were admitted to the performed with two p value: NR
Economic analysis: Linkoping University Hospital parallel and Number of patients with a Group 1: 6/48*** Limitations:
cost-consequences with displaced femoral neck percutaneously Harris hip score excellent or Group 2: 24/24 Costs derived only from one hospital.
analysis fractures with walking ability inserted screws good/fair or poor at 1 year** p value: <0.0001
prior to the trauma, no after closed Overall quality and applicability
Number of patients with a Group 1: 6/42***
Study design contraindications to major reduction. Potentially serious limitations and partial
Harris hip score excellent or Group 2: 20/21
RCT surgery, no rheumatic joint applicability.
good/fair or poor at 2 p value: <0.001
disease.
years**
Group 2: Additional outcomes:
Duration of follow- All patients Total hip Mean cost per patient Group 1: 13,100 There was no difference in the change of
up: N: 143* replacement 2000 Euros, cost of surgical (£11,575) average cost of community services/place
2 years Mean age (range): 84 (75– performed with a procedures, hospital stay, Group 2: 12,800 of residency between the two groups.
101) cemented radiographic examination, (£11,310) Pain was significantly higher in Group 1.
Perspective: M/F: 34/109 prosthesis using a home rehabilitation, p value: NR
Hospital Drop outs: 16 patients poster-lateral emergency and outpatient Notes:
approach. visits, hospital overheads, *143 patients were followed up but two
Discount rates: Group 1 complications and patients in Group 1 and one patient in
Costs: NR N: 78* reoperations. Group 2 were randomised twice in the
Effects: NA Age (mean): All patients had Cost-effectiveness NR same group because they had bilateral
M/F: postoperative fractures.
Drop outs: 9 patients physiotherapy. ** Data for 7 patients in Group 1 and 4 in
Sensitivity analysis NR Group 2 were missing at 1 year, and data
Group 2 for 9 patients in Group 1 and 7 in Group 2
N: 68* were missing at 2 years.
Age (mean): *** Once a patient scored as poor due to a
M/F: failure they remained in this group despite
Drop outs: 7 patients reoperation.
APPENDIX F 431
Abbreviations: NR=not reported, NA=not applicable, M/F=male/female, Sig=statistically significant at 5%, N=total number of patients randomised, Int=intervention, SA=sensitivity analysis
432 APPENDIX F
Evidence table: displaced intracapsular fractures

details
Keating 2005 173 Patient group: previously fit Group 1: Number of deaths Group 1: 3 (3%) Funding/conflict of
UK patients of 60 years or older Internal fixation within 4 months of Group 2: 6 (5%) interest:
with displaced subcapital hip operation (%) Group 3: 2 (4%) grant from the National
Economic analysis: fractures. p value: Not sig Health Service Health
Cost-utility analysis Group 2: Number of patients Group 1: 26 (22%) Technology Assessment
All patients Bipolar hemiarthroplasty with further surgery Group 2: 6 (5%) programme.
N: 298 within 4 months of first Group 3: 5 (7%)
Study design Age (range): 60 - 93 operation (%) p value: NR Limitations:
RCT M/F: 65/233 Group 3: Small number of
Number of deaths Group 1: 10 (8%)
Drop outs: Total hip replacement patients.
within 12 months of Group 2: 11 (10%)
operation (%) Group 3: 4 (6%)
Duration of follow- Group 1 Overall quality and
p value: Not sig
up: N: 118 applicability
2 years Age (mean): 74.9 Number of patients Group 1: 37 (31%) Minor limitations and
M/F: 29/89 with further surgery Group 2: 6 (5%) partial applicability.
Perspective: Drop outs: 19 (18/19 died) within 12 months of Group 3: 6 (9%)
NHS first operation (%) p value: NR Additional outcomes:
Group 2 Number of deaths Group 1: 18 (15%) Place of discharge,
Discount rates: N: 111 within 24 months of Group 2: 18 (16%) adverse events
Costs: 0%* Age (mean): 75.4 operation (%) Group 3: 6 (9%)
Effects: 0% M/F: 19/92 p value: Not sig Notes:
Drop outs: 20 (19/20 died) Number of patients Group 1: 46 (39%) * Costs were not
with further surgery Group 2: 6 (5%) discounted because
Group 3 within 24 months of Group 3: 6 (9%) most of the costs were
N: 69 first operation (%) p value: <0.001 (Group 1 vs 2 and 3) incurred within 1 year
Age (mean): 75.2 Not sig (Group 2 vs 3) of injury.
M/F: 17/52 ** Group1 vs 3 was sig
Drop outs: 7 (7/7 died) EQ-5D utility scores at 4 Group 1: 0.56 (0.29) after adjusting for age
months – mean (SD) Group 2: 0.61 (0.29) and gender
Group 3: 0.68 (0.24)
p value: Not sig**
APPENDIX F 433

details
EQ-5D utility scores at Group 1: 0.58 (0.34)
12 months – mean (SD) Group 2: 0.64 (0.33)
Group 3: 0.70 (0.29)
p value: 0.04 (Group 1 vs 3)
Other groups not sig
EQ-5D utility scores at Group 1: 0.55 (0.38)
24 months – mean (SD) Group 2: 0.53 (0.35)
Group 3: 0.69 (0.32)
p value: 0.008 (Group 2 vs 3)
Other groups not sig
Mean cost per patient Group 1: 12,623 (10,768 – 14,478)
over 2 years (95% CI) Group 2: 9,897 (8,062 – 11,732)
2001 GBP, cost of Group 3: 9,399 (8,265 – 10,532)
hospital admission p value: Sig (Group 1 vs 3)
(inpatient and day Other groups not sig
case), theatre costs,
prosthesis and profile
of hardware, excluding
non-hip-related
admissions.
Cost-effectiveness Total Hip Replacement is dominant.
Cost per utility gained
Sensitivity analysis Results did not change when cost of

Two-way SA prostheses and cost of readmission
were varied over a range from -50% to
+100% around the baseline values.
Abbreviations: NR=not reported, M/F=male/female, Sig=statistically significant at 5%, N=total number of patients randomised, SA=sensitivity analysis
434 APPENDIX F
18.3 Evidence Table 15: Cemented arthroplasties

details
Santini 2005298 Patient group: at least 65 Group 1: VELCA functional score Group 1: 9.13 Funding:
Italy years old, with life Cemented bipolar Group 2: 8.95 The authors declared no
expectancy of at least 3 hemiarthroplasty p value: Not sig conflict of interest.
Economic analysis: months, low-energy trauma. Peri-operative mortality Group 1: 3 (24.5%)
Cost-consequences – number of patients Group 2: 2 (26.4%) Limitations:
analysis All patients Group 2: (%) p value: Not sig Surgical time not included in
N: 106 Uncemented bipolar cost calculation although it
Mortality at 1year – Group 1: 13 (24.5%)
Study design Age (mean): NR hemiarthroplasty was significantly different
number of patients (%) Group 2: 14 (26.4%)
RCT* M/F: 24/82 (group 2 had shorter
p value: Not sig
Drop outs: 0 operating time). The only
Number of patients Group 1: 21 difference considered was
Duration of follow- Group 1 with complications Group 2: 21 the cost of prostheses.
up: N: 53 p value: NR
One year Age (mean): 82 Mean cost per Group 1: 3,093 (£2,400) Overall quality and
M/F: 13/40 patient** 2001 Euros, Group 2: 4,008 (£3,110) applicability:
Perspective: Drop outs: 0 cost of medical and p value: NR Potentially serious limitations
Provider nursing staff, drugs, and partial applicability.
Group 2 diagnostic procedures,
Discount rates: N: 53 prostheses, blood Additional outcomes:
Costs: NA Age (mean): 80 transfusion and hospital Social environment at 1 year
Effects: NA M/F: 11/42 stay. was similar in the two
Drop outs: 0 Cost-effectiveness NR groups.
Notes:
Sensitivity analysis NR * included in our clinical
review
**only cost of prostheses
was different between the
two groups.
APPENDIX F 435
Abbreviations: NR=not reported, NA=not applicable, M/F=male/female, N=total number of patients randomised, VELCA=Verona Elderly Care, Sig=statistically significant at 5%
436 APPENDIX F
18.4 Evidence Table 16: Multidisciplinary rehabilitation

details
Cameron Patient group: Group 1: Accelerated Median length of stay, days Group 1: 13 (7-25) Funding
199445 Patients with proximal femoral hip rehab (involving: (interquartile range) Group 2: 15 (8-44) Australian Department of Health,
fracture early mobilization p value = 0.034 Housing, and Community
Country: after surgery, Mean Barthel index: Group 1: 63 (49.6%) Services.
Australia All patients comprehensive No. of patients recovered
N: 252 rehabilitation at 4 months from surgery Group 2: 52 (41.6%) Conflict of interest: NR
Economic Age (mean): 84 program, early
analysis: M/F: 14/70 discharge from 95% CI (-3% to 21%) Limitations
CEA Drop outs: 0 hospital, community- A longer follow up could have
based rehabilitation). p value = Not significant better reflected differences in
Group 1: Accelerated Rehab costs and outcomes.
Study design N: 127 Group 2:
Mean Barthel index: Group 1: 31 (24%)
RCT Age (mean): Conventional care Health-related QoL were not
No. of patients worse at 4 Group 2: 39 (31%)
Nursing home: 84.2 (n=48) calculated.
months from surgery
Non-nursing home + moderate to
p value= NR
Duration of severe disability: 87.2 (n=21)
follow-up Non-nursing home +limited Overall quality and applicability
4 months disability: 79.2 (n=58) Mean Barthel index: Group 1: 19 The study has potentially serious
No. of patients death at 4 Group 2: 20 limitations and partial
Perspective: M/F: NR months from surgery applicability.
Health care p value = NR
provider Notes:
Group 2 (1)Calculated using the Power
Discount N: 125 Mean cost per patient Group 1: A$ 10,620 Purchasing Parity (PPP) of 1990
rates Age (mean): Year: 1990 (£ 4678.9 – 1990 PPP)(1)
NA Nursing home: 88.5 (n=46) Currency: Australian dollars
Non-nursing home +moderate to Group 2: A$ 12,790
severe disability: 89.3 (n=22) Cost components: inpatient (£ 5635.01 – 1990 PPP)(1)
Non-nursing home +limited hospital (surgical, post
APPENDIX F 437

details
disability: 81.4 (n=57) surgical), readmissions, p value = 0.186
community support
M/F: NR services, institutional care.
Cost-effectiveness The accelerated rehab
Incremental cost per program is the dominant
additional recovered strategy (more effective, less
patient costly)
Sensitivity analysis Accelerated rehab is more
Threshold sensitivity costly than usual care when:
analysis (1) The difference in LOS
between the 2
strategies is less than
1.5 – 2 days
(2) Cost of treatment is
more than 40% per
bed day compared to
conventional care.
These results were not
sensitive to the % of patients
recovering nor to the
definition of recovery.
438 APPENDIX F
Evidence table - Multidisciplinary rehabilitation

details
Farnworth 1994 91 Patient group: Group 1: Fractured Hip In-hospital mortality at 1 Group 1: 16 (24%) Funding/conflict of interest:
Patients with hip Management Program year Group 2: 19 (27%) NR
Australia fracture (FHMP) comprising: p value = NR
orthopaedic surgeon, Length of Stay – days Group 1: 7.3 Limitations:
All patients geriatric physician, nurses, (nursing home patients) Group 2: 10.2 The year at which cost data
Economic N: 138 occupational therapist, p value = NR refer is not clear.
analysis: Age (mean): NR physiotherapist.
Length of Stay – days (non- Group 1: 21.5
CCA M/F: 23/115 Rehabilitation took place in The duration of follow up is not
nursing home patient) Group 2: 28.2
Drop outs: 0 the patient’s normal clear.
p value = NR
environment.
Study design Group 1 Readmission within 1 year Group 1: 4 (6%) No sensitivity analysis was
Case study with N: 67 Group 2: Usual care Group 2: 6 (8%) conducted.
historical control Age (mean), (SD): 78.4 p value = NR
(8.8) Mean cost per patient Group 1: $Aus11 060 (£4872) Health related QoL outcomes
M/F 10/57 Year: 1990 (1) were not calculated.
Duration of Group 2: $Aus 9280 (£4088) (1)
follow-up Currency: $Aus p value = NR No incremental analysis was
6 months Group 2 conducted.
N: 71 Cost components:
Age (mean): 79.8 (10.7) -Hospital costs Overall quality and
Perspective: M/F 13/58 - FHMP costs (staff time, applicability
Health care use of medical goods, office The study has potentially
provider space and travel time for serious limitations and partial
home visits). applicability
Discount rates Cost-effectiveness NR
NA Additional outcomes:
Changes in living arrangements
Sensitivity analysis NR at discharge from hospital and
1 year after hip fracture.
Notes:
APPENDIX F 439

details
(1) The costs were expressed in
GBP using the Power
Purchasing Parity for 1990.
440 APPENDIX F

details
Galvard 1995107 Patient group: Patients Group 1: Rehabilitation in Readmissions to Group 1: 36 Funding/ Conflict of interest: NS
with hip fracture geriatric department (Patients hospital Group 2: 57
Sweden transferred on second p value = NR Limitations:
All patients postoperative day. Orthopaedic Mortality at 1 year Group 1: 45 No sensitivity analysis was
Economic analysis: N: 371 surgeon would visit them once Group 2: 40 performed.
CCA Age (mean): NR weekly) p value = NR
M/F: 95/276 Health related QoL outcomes are not
Mean length of stay in Group 1: 53.3 (47.7)
Drop outs: 0 calculated.
hospital, days (SD) Group 2: 28 (24.2)
Study design Group 2: Usual care
p value = NR
RCT Group 1 (rehabilitation in orthopaedic No incremental analysis was
N: 192 department) Mean cost per patient Group 1: SEK 94,026.05 conducted.
Age (mean) - female: Year: 1989 (£6590.82) (1)
Duration of follow- 79.6 years (SD 8.2) Group 2: SEK 84,536.81 The source used to estimate the unit
up: Age (mean) - male: Currency: Swedish (£5925.67) (1) cost of resources was unclear.
1 year 73.6 years (SD 10.0) Krona (SEK) p value = NR
M/F 45/147 Overall quality and applicability
Drop outs: 0 Cost components: The study has potentially serious
Perspective: Technical aids, home limitations and partial applicability
NHS and PPS Group 2 adjustment costs, stay
N: 179 at convalescent home, Additional outcomes:
Age (mean) - male: new hospital admission, Destination at discharge: 72.4% of
79.1 (SD 8.6) daily costs at patients from group 1 and 72.0% of
Discount rates Age (mean) - female: orthopaedic and patients in group 2
NA 80.9 (SD 9.2) geriatric department. returned to their previous living
arrangements (NS).
M/F 50/129
Drop outs: 0 Notes:
Cost-effectiveness NR (1) Values in GBP obtained using the
Power Purchasing Parity (PPP) for
APPENDIX F 441

details
Sensitivity analysis NR 1989.
442 APPENDIX F

details
Hollingworth Patient group: Group 1: Community LOS (mean inpatients Group 1: 32.5 Funding/conflict of interest: NR
1993148 Hip fracture patients. rehabilitation - Hospital at days) Group 2: 41.7
home (HAH) scheme. The p value: <0.001 Limitations:
UK scheme provides care from Unclear follow up time
All patients trained nurses, nursing Readmission rates at 1 Group 1: 53 (6.8%)
Economic N: 1080 auxiliaries, year (for patients with Group 2: 8 (2.7%) Parameters’ uncertainty has not
analysis: Age (mean): NR physiotherapists, and access to HAH scheme p value =0.008 been subjected to appropriate
Cost analysis M/F: 198/882 occupational therapists in and for usual care probabilistic sensitivity analysis.
Drop outs: NA the patient’s home for up patients)
to 24 hours a day under the No incremental analysis was
Mean cost per patient Group 1: £4884
Study design Group 1 medical supervision of the conducted.
Year: 1992 Group 2: £5606
Case series N=779 (2) general practitioner. The
Currency: UK sterling p value = 0.048
Age (mean): 78.7 (SD 11.2) scheme lasts for up to two Health-related QoL were not
M/F: 143/636 weeks – after then, other determined.
Cost components: Ward,
Duration of Drop outs: NA community services take
Hospital at home, hotel,
follow-up: over. Information on costs obtained from
overheads, medical,
Until discharge (1) Group 2 the hospital finance department,
theatre, other treatment.
N: 301 not from official statistics.
Age (mean): 79.8 (SD 10.9) Group 2: Usual care Cost-effectiveness NR
Perspective: M/F: 55/246 Overall quality and applicability
NHS Drop outs: NA The study has potentially serious
Sensitivity analysis The costs in the HAH limitations and partial applicability
Discount rates One-way deterministic scheme would still be lower
NR sensitivity analysis. than in the usual care case Notes:
even if inpatients costs (1) The duration of follow up
were 50% lower than was unclear from the
predicted and the HAH paper
costs were 50% higher. (2) These were patients with
access to the HAH scheme.
Of these 779 patients, 292
patients were actually
APPENDIX F 443

details
discharged to the scheme.
444 APPENDIX F

details
Huusko 2002158 Patient group: Group 1: 2 weeks Length of stay, days Group 1: 34 (95% CI 28-38) Funding/conflict of
Patients with acute hip intensive rehabilitation Group 2: 42 (95% CI 35-48) interest:
fracture on the geriatric ward p value: 0.05 Study was supported by
Finland Mortality (at discharge) Group 1: 5 (4%) grants from Central
Group 2: 5 (4%) Finland Health Care
Economic analysis: All patients District, Kuopio
CCA N: 243 Group 2: Standard care in University Hospital,
Age (mean): 80 a local hospital Emil Aaltonen
Study design M: 69 Foundation, Uulo
Mortality (at 1 year) Group 1: 18 (15%)
RCT F: 174 Arthio Foundation and
Group 2: 20 (16%)
Drop outs: Novartis Finland Ltd
Duration of follow- Group 1 Patients regaining their Group 1: 5 Limitations:

up: N: 120 independency in ADL – Group 2: 6 No sensitivity analysis
One year Age (mean, range): 80 (66- median, baseline to 3 p value: 0.004 No HRQoL
97) months
M: 36
Perspective: F: 84 Patients regaining their Group 1: 5 Overall quality and
NHS Drop outs: independency in ADL – Group 2: 6 applicability
median, baseline to 1 p value: 0.008 The study has limited
Discount rates: Group 2 year applicability and
Costs: NA N: 123 potentially serious
Effects: NA Age (mean, range): 80 (67- limitations
92) Mean cost per patient Group 1: € 17,900
M: 33 (includes hospital care, (£11,723) Additional outcomes:
F: 90 nursing home care, and Group 2: € 15,900 Pre-fracture
Drop outs: outpatient services) (£10,414) instrumental activities
PPP = 0.667223 (of p value: NR of daily living – IADL
2002) (median) – baseline to 3
months and baseline to
APPENDIX F 445

details
Cost-effectiveness NR 1 year
Data sources:
Notes:
446 APPENDIX F

details
O’Cathain 1994 Patient group: Group 1: Hospital at home Health-related QoL: Group 1: 14 Funding: The study was funded by Trent
245
Patients with fracture neck of scheme (patients discharged Emotional reaction at Group 2: 24 Regional Health Authority, Southern
femur to their own homes and discharge (from the p value <0.05 Derbyshire Community Health Services and
UK cared for by a community Nottingham Health Southern Derbyshire Department of Public
HAH team under the clinical Profile Health.
Economic All patients responsibility of the GP for a questionnaire)(1)
analysis: N: 110 maximum of 12 days. The Mortality Group 1: 5.3% Conflict of interest: NR
CCA Age (mean): NR HAH team consisted of Group 2: 5.9%
M/F: 16/94 district nurses, community p value: NR Limitations:
Drop outs: 14 physiotherapists, The length of period during which costs are
Readmission rate (at Group 1: 15.8%
Study design occupational therapists and calculated is unclear.
three months) Group 2: 8.8%
Non Group 1 generic workers.)
p value: 0.187 (NS)
randomised N: 76 A longer follow up would have better
trial with Age (mean): 76.4 (SD 10.0) Hospital LOS, median Group 1: 10 reflected differences in costs and
concurrent M/F: 11/65 Group 2: Usual care number of days Group 2: 17 outcomes.
controls Drop outs: 8 (interquartile range) p value: <0.001
No sensitivity analysis was conducted.
Group 2 Mean cost per Group 1: £1500
Duration of N: 34 patient Group 2: £1870 No incremental analysis was conducted.
follow-up: Age (mean): 77.6 (SD 9.7) p value: NR
3 months M/F: 5/29 Year: 1992 Overall quality and applicability
Drop outs: 6 Currency: UK sterling The study has potentially serious
Perspective: limitations and limited applicability
NHS Cost components:
staff costs, Notes:
Discount rates orthopaedic bed cost. (1) The other dimensions of the NHP
NA Cost-effectiveness NR (Physical mobility, pain, sleep, energy and
social isolation) were not statistically
significant.
APPENDIX F 447
448 APPENDIX F

details
Parker 1991270 Patient group: Group 1: early supported LOS (mean, days) Group 1: 29 Funding/conflict of interest:
Patients with acute hip discharge scheme – Group 2: 38
Economic analysis: fracture hospital at home scheme p value: 0.035
CCA Mortality (at 90 days) Group 1: 40 (14%) Limitations:
Group 2: 14 (11%) No sensitivity analysis
All patients Costs were not discounted
Study design N: 410 Group 2: usual inpatient
Prospective Age (mean): 77 rehabilitation
observational study F: 80% Overall quality and
Mean cost per patient Group 1: £1165.30
Drop outs: applicability
Group 2: £365.50*
The study has limited
p value: NR
Duration of follow- Group 1 applicability and potentially
up: N: 284 Cost-effectiveness NR serious limitations
3 years Age (mean, range): 77
F: 79% Additional outcomes:
Drop outs: 113 Sensitivity analysis NR
Perspective:
NHS Group 2 Data sources:
N: 126 Hospital records
Discount rates: Age (mean, range): 77
Costs: NR F: 83% Notes:
Effects: NR Drop outs: NA *HAH cost saving (-£799.80).
Only 171 patients (60% of
284) were discharged using
the HAH scheme, and the
mean cost of the scheme
refers to this group only.
APPENDIX G 449
19 Appendix G: Forest plots
Figure 3-1: Illustration of precise and imprecise outcomes based on the confidence interval of ..........................................26
Figure G-2. Sensitivity and specificity: Sonography and isotope scanning (reference standard: MRI) .............................. 452
Figure G-3. Mortality: Early (≤24 hours) vs. late surgery................................................................................................................ 453
Figure G-4. Return to independent living: late (>24 hours) vs. early surgery ........................................................................... 453
Figure G-5. Pressure ulcers: late (>24 hours) vs. early surgery..................................................................................................... 453
Figure G-6. Major complications: late (>24 hours) vs. early surgery........................................................................................... 454
Figure G-7. Mortality – in hospital: late (24-48 hours) vs. early surgery .................................................................................... 454
Figure G-8. Complications: late (24-48 hours) vs. early surgery................................................................................................... 454
Figure G-9. Pressure ulcers: late (24-48 hours) vs. early surgery ................................................................................................. 454
Figure G-10. Mortality – at 4 months: late (>36 hours) vs. early surgery .................................................................................. 455
Figure G-11. Pressure ulcers: late (>36 hours) vs. early surgery .................................................................................................. 455
Figure G-12. Return to independent living: late (>36 hours) vs. early surgery ......................................................................... 455
Figure G-13. Mortality: late (>48 hours) vs. early surgery ............................................................................................................ 455
Figure G-14. Return to independent living: late (>48 hours) vs. early surgery ......................................................................... 456
Figure G-15. Pressure ulcers: late (>48 hours) vs. early surgery .................................................................................................. 456
Figure G-16. Major complications: late (>48 hours) vs. early surgery ........................................................................................ 456
Figure G-17. Minor complications: late (>48 hours) vs. early surgery ........................................................................................ 456
Figure G-18. Mortality – 30 days: late (>24 hours) vs. early surgery with the exclusion of patients unfit for surgery .. 457
Figure G-19. Combined mortality and needing total assistance in locomotion at 6 months: late (>24 hours) vs. early
surgery with the exclusion of patients unfit for surgery ................................................................................................................ 457
Figure G-20. Major post operative complications: late (>24 hours) vs. early surgery with the exclusion of patients unfit
for surgery .................................................................................................................................................................................................. 457
Figure G-21. Mortality: late (>48 hours) vs. early surgery with the exclusion of patients unfit for surgery ..................... 457
Figure G-22. Change in residence (more dependent): late (>48 hours) vs. early surgery with the exclusion of patients
unfit for surgery ........................................................................................................................................................................................ 458
Figure G-23. Return to original residence: late (>48 hours) vs. early surgery with the exclusion of patients unfit for
surgery ........................................................................................................................................................................................................ 458
Figure G-24. Pain: Nerve blocks vs. no block (systemic drugs) ..................................................................................................... 459
Figure G-25. Unsatisfactory pain control pre-operatively or ‘need for breakthrough analgesia’: Nerve blocks vs. no
block (systemic drugs) ............................................................................................................................................................................. 459
Figure G-26. Unsatisfactory pain control postoperatively: Nerve blocks vs. (systemic drugs) ............................................. 460
Figure G-27. Nausea and/ or vomiting: Nerve blocks vs. (systemic drugs) ............................................................................... 460
Figure G-28. Need for anti-emetics: Nerve blocks vs. (systemic drugs) ..................................................................................... 460
Figure G-29. Wound infection: Nerve blocks vs. (systemic drugs) .............................................................................................. 461
Figure G-30. Pneumonia: Nerve blocks vs. (systemic drugs) ......................................................................................................... 461
Figure G-31. Any cardiac complication: Nerve blocks vs. no block (systemic drugs) .............................................................. 462
Figure G-32. Myocardial infarction: Nerve blocks vs. no block (systemic drugs) ..................................................................... 462
Figure G-33. Puritis: Nerve blocks vs. no block (systemic drugs) ................................................................................................. 462
Figure G-34. Pulmonary embolism: Nerve blocks vs. no block (systemic drugs) ..................................................................... 463
Figure G-35. Deep vein thrombosis: Nerve blocks vs. no block (systemic drugs) .................................................................... 463
Figure G-36. Mortality: Nerve blocks vs. no block (systemic drugs)............................................................................................ 464
Figure G-37. Pressure sores: Nerve blocks vs. no block (systemic drugs) .................................................................................. 465
Figure G-38. Confusional state: Nerve blocks vs. no block (systemic drugs) ............................................................................ 465
450 APPENDIX G
Figure G-40. Mortality at 1 month (random effects model): Regional (spinal or epidural) versus general anaesthesia
...................................................................................................................................................................................................................... 465
Figure G-41. Mortality- early up to 1 month: Regional (spinal or epidural) versus general anaesthesia .......................... 466
Figure G-42. Length of stay in hospital: Regional (spinal or epidural) versus general anaesthesia .................................... 467
Figure G-43. Vomiting: Regional (spinal or epidural) versus general anaesthesia .................................................................. 467
Figure G-44. Acute confusional state: Regional (spinal or epidural) versus general anaesthesia ....................................... 467
Figure G-45. Pneumonia: Regional (spinal or epidural) versus general anaesthesia .............................................................. 468
Figure G-46. Myocardial infarction: Regional (spinal or epidural) versus general anaesthesia ........................................... 468
Figure G-47. Pulmonary embolism (Peto odds ratio): Regional (spinal or epidural) versus general anaesthesia ........... 469
Figure G-48. Pulmonary embolism (random effects model): Regional (spinal or epidural) versus general anaesthesia
...................................................................................................................................................................................................................... 469
Figure G-49. Pulmonary embolism (fatal and non fatal): Regional (spinal or epidural) versus general anaesthesia ..... 470
Figure G-50. Deep vein thrombosis: Regional (spinal or epidural) versus general anaesthesia .......................................... 470
Figure G-51. Reoperation rate for technically demanding hip fractures at 6 months: Senior/higher grade surgeon
versus junior/lower grade surgeon...................................................................................................................................................... 471
Figure G-52. Dislocation rate for arthroplasty: Senior/higher grade surgeon versus junior/lower grade surgeon ........ 471
Figure G-53. Perioperative mortality - older designs of arthroplasty: cemented vs. uncemented. ................................... 472
Figure G-54. Mortality – at up to 1 month - older designs of arthroplasty: cemented vs. uncemented. ......................... 472
Figure G-55. Mortality at between 1 and 3 months - older designs of arthroplasty: cemented vs. uncemented. ......... 473
Figure G-56. Mortality at 1 year - older designs of arthroplasty: cemented vs. uncemented. ............................................ 474
Figure G-57. Mortality at 3 years - older designs of arthroplasty: cemented vs. uncemented. .......................................... 474
Figure G-58. Number of patients failing to regain mobility - older designs of arthroplasty: cemented vs. uncemented.
...................................................................................................................................................................................................................... 475
Figure G-59. Change in mobility score - older designs of arthroplasty: cemented vs. uncemented. ................................. 475
Figure G-60. Length of hospital stay - older designs of arthroplasty: cemented vs. uncemented. ..................................... 476
Figure G-61. Number of patients failing to return home - older designs of arthroplasty: cemented vs. uncemented. 476
Figure G-62. Number of patients reporting pain at 3 months - older designs of arthroplasty: cemented vs.
uncemented. ............................................................................................................................................................................................. 477
Figure G-63. Number of patients reporting pain at 1 to 2 years - older designs of arthroplasty: cemented vs.
uncemented. ............................................................................................................................................................................................. 477
Figure G-64. Pain score at 6 months - older designs of arthroplasty: cemented vs. uncemented. .................................... 478
Figure G-65. Reoperations - older designs of arthroplasty: cemented vs. uncemented. ...................................................... 478
Figure G-66. Deep sepsis - older designs of arthroplasty: cemented vs. uncemented. ......................................................... 479
Figure G-67. Wound haematoma - older designs of arthroplasty: cemented vs. uncemented........................................... 479
Figure G-68. Mortality - newer designs of arthroplasty: cemented vs. uncemented. ........................................................... 480
Figure G-69. Reoperations - newer designs of arthroplasty: cemented vs. uncemented. .................................................... 480
Figure G-70. Pain – need for pain medication - newer designs of arthroplasty: cemented vs. uncemented. ................. 480
Figure G-71. Unable to walk without aids at 12 months –newer designs of arthroplasty: cemented vs. uncemented.
...................................................................................................................................................................................................................... 480
Figure G-72. Barthel Index –newer designs of arthroplasty: cemented vs. uncemented. .................................................... 481
Figure G-73. Harris Hip Score and Eq-5d scores –newer designs of arthroplasty: cemented vs. uncemented. .............. 481
Figure G-74. Length of hospital stay –newer designs of arthroplasty: cemented vs. uncemented. ................................... 481
Figure G-75. Mortality: Internal fixation versus hemiarthroplasty.............................................................................................. 482
Figure G-76. Reoperations: Internal fixation versus hemiarthroplasty ...................................................................................... 483
Figure G-77. Failure to return to same residence by final follow up: Internal fixation versus hemiarthroplasty ............ 484
Figure G-78. Failure to regain mobility: Internal fixation versus hemiarthroplasty ................................................................ 484
Figure G-79. Patients reporting pain at 1 year: Internal fixation versus hemiarthroplasty .................................................. 485
Figure G-80. Harris Hip Score: Internal fixation versus hemiarthroplasty ................................................................................. 485
Figure G-81. Number of patients with Barthel Index Score of 95 or 100: Internal fixation versus hemiarthroplasty ... 486
Figure G-82. Euroquol Eq-5d score: Internal fixation versus hemiarthroplasty ....................................................................... 486
Figure G-83. Length of hospital stay: Internal fixation versus hemiarthroplasty..................................................................... 486
Figure G-84. Mortality: Internal fixation versus total hip replacement ..................................................................................... 487
Figure G-85. Reoperations – all – at final follow up of study: Internal fixation versus total hip replacement ................. 488
Figure G-86. Number of patients reporting pain at 1 year: Internal fixation versus total hip replacement ..................... 489
Figure G-87. Length of hospital stay: Internal fixation versus total hip replacement ............................................................ 489
Figure G-88. Mortality: Hemiarthroplasty versus total hip replacement .................................................................................. 490
Figure G-89. Reoperations - all: Hemiarthroplasty versus total hip replacement ................................................................... 491
APPENDIX G 451
Figure G-90. Number of patients reporting pain at 1 year: Hemiarthroplasty versus total hip replacement .................. 491
Figure G-91. Pain scores: Hemiarthroplasty versus total hip replacement ............................................................................... 492
Figure G-92. Failure to regain mobility at end of study: Hemiarthroplasty versus total hip replacement ....................... 492
Figure G-93. Functional scores (lower scores advantageous): Hemiarthroplasty versus total hip replacement ............ 492
Figure G-94. Functional status (higher scores advantageous): Hemiarthroplasty versus total hip replacement............ 493
Figure G-95. Quality of life scores: Hemiarthroplasty versus total hip replacement .............................................................. 494
Figure G-96. Length of hospital stay: Hemiarthroplasty versus total hip replacement ......................................................... 494
Figure G-97. 30 days mortality: Intramedullary implants versus extramedullary implants .................................................. 495
Figure G-98. 3 months mortality: Intramedullary implants versus extramedullary implants .............................................. 495
Figure G-99. 12 months mortality: Intramedullary implants versus extramedullary implants ............................................ 496
Figure G-100. Reoperation – within the follow up period of the study: Intramedullary implants versus extramedullary
implants ...................................................................................................................................................................................................... 497
Figure G-101. Operative or postoperative fracture of femur - within the follow up period of the study: Intramedullary
implants versus extramedullary implants .......................................................................................................................................... 498
Figure G-102. Cut-out (at latest follow up): Intramedullary implants versus extramedullary implants ............................ 499
Figure G-103. Infection (deep infection or requires reoperation – at latest follow up): Intramedullary implants versus
extramedullary implants ........................................................................................................................................................................ 500
Figure G-104. Non-union (at latest follow-up): Intramedullary implants versus extramedullary implants ...................... 501
Figure G-105. Pain – patient reported outcomes: Intramedullary implants versus extramedullary implants ................. 501
Figure G-106. Length of stay in hospital (in days): Intramedullary implants versus extramedullary implants ................ 502
Figure G-107. Mean mobility score (Parker Palmer score): Intramedullary implants versus extramedullary implants 502
Figure G-96. 30 days mortality: Intramedullary implants versus extramedullary implants .................................................. 503
Figure G-97. 3 months mortality: Intramedullary implants versus extramedullary implants .............................................. 503
Figure G-98. 12 months mortality: Intramedullary implants versus extramedullary implants ............................................ 503
Figure G-99. Reoperation – within the follow up period of the study: Intramedullary implants versus extramedullary
implants ...................................................................................................................................................................................................... 504
Figure G-100. Operative or postoperative fracture of femur - within the follow up period of the study: Intramedullary
implants versus extramedullary implants .......................................................................................................................................... 504
Figure G-103. Non-union (at latest follow-up): Intramedullary implants versus extramedullary implants ...................... 507
Figure G-104. Pain – patient reported outcomes: Intramedullary implants versus extramedullary implants ................. 507
Figure G-105. Length of stay in hospital (in days): Intramedullary implants versus extramedullary implants ................ 507
Figure G-106. Mean mobility score (Parker Palmer score): Intramedullary implants versus extramedullary implants 508
Figure G-108. Mortality at 12 months: Intramedullary implants versus extramedullary implants ..................................... 508
Figure G-109. Reoperation within follow up period of the study: Intramedullary implants versus extramedullary
implants ...................................................................................................................................................................................................... 508
Figure G-112. Non-union (at latest follow up): Intramedullary implants versus extramedullary implants ...................... 509
Figure G-113. Independent to transfer at day 7: Early versus delayed mobilisation .............................................................. 510
Figure G-114. Independent to step at day 7: Early versus delayed mobilisation .................................................................... 510
Figure G-115. Discharge to home or rehabilitation programme: Early versus delayed mobilisation ................................. 510
Figure G-116. Discharge to nursing home or died: Early versus delayed mobilisation .......................................................... 511
Figure G-117. Strength measures: intensive physiotherapy versus usual care ........................................................................ 512
Figure G-118. Tinetti's POMA (Performance Orientated Mobility Assessment): intensive physiotherapy versus usual
care .............................................................................................................................................................................................................. 512
Figure G-119. Functional performance measures: intensive physiotherapy versus usual care ........................................... 513
Figure G-120. Functional performance tests: intensive physiotherapy versus usual care .................................................... 513
Figure G-121. Walking speed: intensive physiotherapy versus usual care ................................................................................ 513
Figure G-122. Knee extensor strength: intensive physiotherapy versus usual care ............................................................... 514
Figure G-123. Functional performance tests: intensive physiotherapy versus usual care .................................................... 514
Figure G-124. Quality of life: intensive physiotherapy versus usual care .................................................................................. 515
Figure G-125. Walking speed: intensive physiotherapy versus usual care ................................................................................ 515
Figure G-126. Pain: intensive physiotherapy versus usual care ................................................................................................... 515
452 APPENDIX G
Figure G-127. Length of hospital stay: intensive physiotherapy versus usual care ................................................................. 516
Figure G-128. Adductor muscle strength (kp) at 9 weeks: intensive physiotherapy versus usual care ............................. 516
Figure G-129. Length of hospital stay: intensive physiotherapy versus usual care ................................................................. 516
Figure G-130. Mortality at 6 months: hospital MDR versus usual care ...................................................................................... 517
Figure G-131. Mortality at 12 months: hospital MDR versus usual care ................................................................................... 517
Figure G-132. Mortality (at discharge): hospital MDR versus usual care................................................................................... 518
Figure G-133. Functional outcomes at 6 months: orthogeriatric hospital MDR versus usual care ..................................... 518
Figure G-134. Functional outcomes at 1 year: orthogeriatric hospital MDR versus usual care ........................................... 519
Figure G-135. Functional outcomes at 1 year: hip fracture programme versus usual care .................................................. 519
Figure G-136. : Functional outcomes: Barthel scores at long-term follow-up: hip fracture programme versus usual care
...................................................................................................................................................................................................................... 519
Figure G-137. Complications: hospital MDR versus usual care .................................................................................................... 520
Figure G-138. Length of hospital stay: hospital MDR versus usual care .................................................................................... 521
Figure G-139. Readmitted to hospital during follow up: hospital MDR versus usual care .................................................... 521
Figure G-140. Mortality: Home-based MDR versus usual care .................................................................................................... 522
Figure G-141. “Poor outcome” – institutional care and unable to walk: Home-based MDR versus usual care .............. 522
Figure G-142. SF-36 scores at 12 months (0: worst to 100: best): Home-based MDR versus usual care .......................... 522
Figure G-143. Lengths of hospital or rehabilitation stays (days): Home-based MDR versus usual care ............................ 523
Figure G-144. Readmission to hospital during 4 month follow-up: Home-based MDR versus usual care ........................ 523
Figure G-145. Degree of independence (Functional Independent Measure): Home-based MDR versus usual care ..... 523
Figure G-146. Mobility and strength tests: Home-based MDR versus usual care ................................................................... 524
Figure 147: Decision tree with Markov states - investment for early surgery vs. no hospital investment for early surgery
...................................................................................................................................................................................................................... 538
Figure 148: Cycle 0 Markov model ...................................................................................................................................................... 554
Figure 149: Cycles 1 to 3 of the Markov model ................................................................................................................................ 556
Figure 150: Cycle 4 - onwards of the Markov model ...................................................................................................................... 557
Figure 151: Place of residence at discharge ...................................................................................................................................... 558
Figure 152: Incremental cost-effectiveness scatter plot: HFP vs GORU/MARU ....................................................................... 582
Figure 153: Incremental cost-effectiveness scatter plot: GORU/MARU vs. usual care .......................................................... 583
Figure 154: Incremental cost-effectiveness scatter plot - HFP vs usual care ............................................................................ 584
Figure 155: Model structure - community MDR vs usual care ..................................................................................................... 587
Figure 156: Two-way sensitivity analysis on length of stay at home and in hospital ............................................................. 593
Figure 157: Incremental cost-effectiveness scatter plot - Community MDR vs usual care ................................................... 597
19.1 Radiology
Figure G-2. Sensitivity and specificity: Sonography and isotope scanning (reference standard: MRI)
APPENDIX G 453
19.2 Timing of surgery
Figure G-3. Mortality: Early (≤24 hours) vs. late surgery

Odds Ratio Odds Ratio
Study or Subgroup log[Odds Ratio] SE IV, Fixed, 95% CI IV, Fixed, 95% CI
1.1.1 In hospital
Bergeron 2006 -0.1278334 0.24016 0.88 [0.55, 1.41]
Weller 2005 0.15700375 0.039324 1.17 [1.08, 1.26]
1.1.2 30 days
Majumdar 2006 -0.1053605 0.217003 0.90 [0.59, 1.38]
Bottle 2006 0.22314355 0.024509 1.25 [1.19, 1.31]
1.1.3 3 months
Weller 2005 0.10436002 0.027606 1.11 [1.05, 1.17]
1.1.4 4 months
Alani 2008 0.06765865 0.237527 1.07 [0.67, 1.70]
1.1.5 1 year
Weller 2005 0.12221763 0.038281 1.13 [1.05, 1.22]
0.2 0.5 1 2 5
Favours late surgery Favours early surgery
Figure G-4. Return to independent living: late (>24 hours) vs. early surgery
Study or Subgroup log[Odds Ratio] SE Weight IV, Fixed, 95% CI IV, Fixed, 95% CI
Alani 2008 -0.1508229 0.33145 100.0% 0.86 [0.45, 1.65]
Total (95% CI) 100.0% 0.86 [0.45, 1.65]

Heterogeneity: Not applicable
0.01 0.1 1 10 100
Test for overall effect: Z = 0.46 (P = 0.65) Favours early surgery Favours late surgery
Figure G-5. Pressure ulcers: late (>24 hours) vs. early surgery

Alani 2008 0.78390154 0.302455 100.0% 2.19 [1.21, 3.96]
Total (95% CI) 100.0% 2.19 [1.21, 3.96]

0.01 0.1 1 10 100
Test for overall effect: Z = 2.59 (P = 0.010) Favours late surgery Favours early surgery
454 APPENDIX G
Figure G-6. Major complications: late (>24 hours) vs. early surgery
Bergeron 2006 -0.1392621 0.203921 100.0% 0.87 [0.58, 1.30]
Total (95% CI) 100.0% 0.87 [0.58, 1.30]

0.01 0.1 1 10 100
Figure G-7. Mortality – in hospital: late (24-48 hours) vs. early surgery
Lefaivre 2009 -0.1984509 0.344369 100.0% 0.82 [0.42, 1.61]
Total (95% CI) 100.0% 0.82 [0.42, 1.61]

0.01 0.1 1 10 100
Figure G-8. Complications: late (24-48 hours) vs. early surgery

2.2.1 Minor
Lefaivre 2009 0.42526774 0.190999 100.0% 1.53 [1.05, 2.22]
Subtotal (95% CI) 100.0% 1.53 [1.05, 2.22]
Test for overall effect: Z = 2.23 (P = 0.03)
2.2.2 Major
Lefaivre 2009 -0.040822 0.309577 100.0% 0.96 [0.52, 1.76]
Subtotal (95% CI) 100.0% 0.96 [0.52, 1.76]
0.01 0.1 1 10 100

Figure G-9. Pressure ulcers: late (24-48 hours) vs. early surgery
Lefaivre 2009 0.20701417 0.279058 100.0% 1.23 [0.71, 2.13]
Total (95% CI) 100.0% 1.23 [0.71, 2.13]

0.01 0.1 1 10 100
APPENDIX G 455
Figure G-10. Mortality – at 4 months: late (>36 hours) vs. early surgery
Alani 2008 0.04879016 0.259163 100.0% 1.05 [0.63, 1.74]
Total (95% CI) 100.0% 1.05 [0.63, 1.74]

0.01 0.1 1 10 100
Alani 2008 1.22964055 0.289723 100.0% 3.42 [1.94, 6.03]
Total (95% CI) 100.0% 3.42 [1.94, 6.03]

0.01 0.1 1 10 100
Test for overall effect: Z = 4.24 (P < 0.0001) Favours late surgery Favours early surgery
Alani 2008 -0.8209806 0.371247 100.0% 0.44 [0.21, 0.91]
Total (95% CI) 100.0% 0.44 [0.21, 0.91]

0.01 0.1 1 10 100
Figure G-13. Mortality: late (>48 hours) vs. early surgery

4.1.1 In hospital
Bergeron 2006 0.14842001 0.306737 1.16 [0.64, 2.12]
Lefaivre 2009 -0.0725707 0.462615 0.93 [0.38, 2.30]
Weller 2005 0.47000363 0.060492 1.60 [1.42, 1.80]
4.1.2 30 days
Bottle 2006 0.3074847 0.026284 1.36 [1.29, 1.43]
Grimes 2002A -0.3424903 0.228015 0.71 [0.45, 1.11]
4.1.3 3 months
Weller 2005 0.33647224 0.047513 1.40 [1.28, 1.54]
4.1.4 4 months
Alani 2008 -0.1508229 0.343293 0.86 [0.44, 1.69]
4.1.5 1 year
Weller 2005 0.45742485 0.116587 1.58 [1.26, 1.99]
0.01 0.1 1 10 100

456 APPENDIX G

Alani 2008 -1.1086626 0.438176 100.0% 0.33 [0.14, 0.78]
Total (95% CI) 100.0% 0.33 [0.14, 0.78]

0.01 0.1 1 10 100

Alani 2008 1.46787435 0.314867 4.34 [2.34, 8.04]
Grimes 2002A 0.18232156 0.146777 1.20 [0.90, 1.60]
Lefaivre 2009 0.82855182 0.333584 2.29 [1.19, 4.40]
0.01 0.1 1 10 100

Figure G-16. Major complications: late (>48 hours) vs. early surgery

Bergeron 2006 0.27763174 0.26127 1.32 [0.79, 2.20]
Lefaivre 2009 0.79299252 0.371919 2.21 [1.07, 4.58]
0.01 0.1 1 10 100

Figure G-17. Minor complications: late (>48 hours) vs. early surgery
Lefaivre 2009 0.81977983 0.252969 100.0% 2.27 [1.38, 3.73]
Total (95% CI) 100.0% 2.27 [1.38, 3.73]

0.01 0.1 1 10 100
APPENDIX G 457
Figure G-18. Mortality – 30 days: late (>24 hours) vs. early surgery with the exclusion of patients
unfit for surgery
Experimental Control Risk Ratio Risk Ratio
Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI
Moran 2005 85 982 85 1166 100.0% 1.19 [0.89, 1.58]
Total (95% CI) 982 1166 100.0% 1.19 [0.89, 1.58]

Total events 85 85
0.01 0.1 1 10 100
Figure G-19. Combined mortality and needing total assistance in locomotion at 6 months: late (>24
hours) vs. early surgery with the exclusion of patients unfit for surgery
Orosz 2004 -0.4780358 0.287445 100.0% 0.62 [0.35, 1.09]
Total (95% CI) 100.0% 0.62 [0.35, 1.09]

0.01 0.1 1 10 100
Figure G-20. Major postoperative complications: late (>24 hours) vs. early surgery with the
exclusion of patients unfit for surgery

Orosz 2004 -1.3470736 0.665298 100.0% 0.26 [0.07, 0.96]
Total (95% CI) 100.0% 0.26 [0.07, 0.96]

0.005 0.1 1 10 200
Figure G-21. Mortality: late (>48 hours) vs. early surgery with the exclusion of patients unfit for
surgery

6.1.1 Mortality - 30 days
Moran 2005 36 497 134 1651 100.0% 0.89 [0.63, 1.27]
Subtotal (95% CI) 497 1651 100.0% 0.89 [0.63, 1.27]
Total events 36 134
6.1.2 Mortality - 1 year

Siegmeth 2005A 238 3454 24 174 100.0% 0.50 [0.34, 0.74]
Subtotal (95% CI) 3454 174 100.0% 0.50 [0.34, 0.74]
Total events 238 24
0.01 0.1 1 10 100

Favours early surgery Favours late surgery
458 APPENDIX G
Figure G-22. Change in residence (more dependent): late (>48 hours) vs. early surgery with the
exclusion of patients unfit for surgery
Siegmeth 2005A 240 3454 22 174 100.0% 0.55 [0.37, 0.83]
Total (95% CI) 3454 174 100.0% 0.55 [0.37, 0.83]

Total events 240 22
0.01 0.1 1 10 100
Figure G-23. Return to original residence: late (>48 hours) vs. early surgery with the exclusion of
patients unfit for surgery

Siegmeth 2005A 240 3454 22 174 100.0% 0.55 [0.37, 0.83]
Total (95% CI) 3454 174 100.0% 0.55 [0.37, 0.83]

Total events 240 22
0.01 0.1 1 10 100
APPENDIX G 459
19.3 Analgesia
Figure G-24. Pain: Nerve blocks vs. no block (systemic drugs)
Nerve block Control (no block) Std. Mean Difference Std. Mean Difference
Study or Subgroup Mean SD Total Mean SD Total Weight IV, Fixed, 95% CI IV, Fixed, 95% CI
1.1.1 Three in one block (on admission)
Gille 2006 1.22 0.43 50 1.58 0.73 50 47.2% -0.60 [-1.00, -0.20]
Kullenberg 2004 1.9 0.9 40 2.3 0.7 40 38.3% -0.49 [-0.94, -0.05]
Murgue 2006 2.1 8.4 16 5.7 10.5 14 14.5% -0.37 [-1.10, 0.35]
Subtotal (95% CI) 106 104 100.0% -0.52 [-0.80, -0.25]
Heterogeneity: Chi² = 0.32, df = 2 (P = 0.85); I² = 0%
Total (95% CI) 106 104 100.0% -0.52 [-0.80, -0.25]

-4 -2 0 2 4
Test for overall effect: Z = 3.72 (P = 0.0002) Favours block Favours no block
Test for subgroup differences: Not applicable
Figure G-25. Unsatisfactory pain control preoperatively or ‘need for breakthrough analgesia’:
Nerve blocks vs. no block (systemic drugs)
Nerve block Control (no block) Risk Ratio Risk Ratio
Foss 2007 3 24 3 24 6.3% 1.00 [0.22, 4.47]
Gille 2006 5 50 12 50 25.2% 0.42 [0.16, 1.10]
Kullenberg 2004 4 40 12 40 25.2% 0.33 [0.12, 0.95]
Murgue 2006 3 16 8 14 18.0% 0.33 [0.11, 1.00]
Subtotal (95% CI) 130 128 74.8% 0.42 [0.24, 0.72]
Total events 15 35
1.3.2 Psoas block (on admission)

Chudinov 1999 3 20 12 20 25.2% 0.25 [0.08, 0.75]
Subtotal (95% CI) 20 20 25.2% 0.25 [0.08, 0.75]
Total events 3 12
Total (95% CI) 150 148 100.0% 0.37 [0.23, 0.61]

Total events 18 47
0.01 0.1 1 10 100
Test for overall effect: Z = 3.94 (P < 0.0001) Favours block Favours no block
460 APPENDIX G
Figure G-26. Unsatisfactory pain control postoperatively: Nerve blocks vs. no block (systemic
drugs)

Study or Subgroup Events Total Events Total M-H, Random, 95% CI M-H, Random, 95% CI
1.4.1 Psoas block (on admission)
Chudinov 1999 1 20 10 20 0.10 [0.01, 0.71]
1.4.2 Continous femoral nerve block (after surgery)

Cuvillon 2007 15 21 15 21 1.00 [0.68, 1.47]
0.01 0.1 1 10 100

Favours block Favours no block
Figure G-27. Nausea and/ or vomiting: Nerve blocks vs. no block (systemic drugs)

Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI
Foss 2007 3 24 3 24 11.5% 1.00 [0.22, 4.47]
Gille 2006 0 50 1 50 2.6% 0.33 [0.01, 7.99]
Murgue 2006 0 16 1 14 2.7% 0.29 [0.01, 6.69]
Subtotal (95% CI) 90 88 16.8% 0.70 [0.20, 2.41]
Total events 3 5
Heterogeneity: Tau² = 0.00; Chi² = 0.73, df = 2 (P = 0.69); I² = 0%
1.8.2 Psoas block (at surgery)

Spansberg 1996 2 10 2 10 8.4% 1.00 [0.17, 5.77]
Subtotal (95% CI) 10 10 8.4% 1.00 [0.17, 5.77]
Total events 2 2
1.8.3 Continous femoral nerve block after surgery

Cuvillon 2007 9 21 11 41 51.8% 1.60 [0.79, 3.24]
Tuncer 2003 4 20 7 20 23.0% 0.57 [0.20, 1.65]
Subtotal (95% CI) 41 61 74.8% 1.03 [0.38, 2.82]
Total events 13 18
Total (95% CI) 141 159 100.0% 1.05 [0.63, 1.75]

Total events 18 25
0.01 0.1 1 10 100
Figure G-28. Need for anti-emetics: Nerve blocks vs. no nerve block(systemic drugs)

Study or Subgroup Events Total Events Total M-H, Fixed, 95% CI M-H, Fixed, 95% CI
Tuncer 2003 0 20 5 20 0.09 [0.01, 1.54]
0.001 0.1 1 10 1000

APPENDIX G 461
Figure G-29. Wound infection: Nerve blocks vs. no nerve block (systemic drugs)

1.24.1 Epidural block (for 4 days after surgery)
Foss 2005 0 28 2 27 0.19 [0.01, 3.85]
0.01 0.1 1 10 100

Figure G-30. Pneumonia: Nerve blocks vs. no nerve block (systemic drugs)

Fletcher 2003 2 26 4 24 16.9% 0.46 [0.09, 2.30]
Haddad 1995 2 25 11 25 44.6% 0.18 [0.04, 0.74]
Subtotal (95% CI) 51 49 61.5% 0.26 [0.09, 0.73]
Total events 4 15
1.15.2 Continuous epidural block (on admission)

Matot 2003 2 34 2 34 8.1% 1.00 [0.15, 6.70]
Subtotal (95% CI) 34 34 8.1% 1.00 [0.15, 6.70]
Total events 2 2

White 1980 3 16 5 20 18.0% 0.75 [0.21, 2.67]
Subtotal (95% CI) 16 20 18.0% 0.75 [0.21, 2.67]
Total events 3 5

Foss 2005 3 28 3 27 12.4% 0.96 [0.21, 4.37]
Subtotal (95% CI) 28 27 12.4% 0.96 [0.21, 4.37]
Total events 3 3
Total (95% CI) 129 130 100.0% 0.49 [0.26, 0.94]

Total events 12 25
0.01 0.1 1 10 100
462 APPENDIX G
Figure G-31. Any cardiac complication: Nerve blocks vs. no nerve block (systemic drugs)

Matot 2003 2 34 11 34 91.7% 0.18 [0.04, 0.76]
Subtotal (95% CI) 34 34 91.7% 0.18 [0.04, 0.76]
Total events 2 11

Foss 2005 1 28 1 28 8.3% 1.00 [0.07, 15.21]
Subtotal (95% CI) 28 28 8.3% 1.00 [0.07, 15.21]
Total events 1 1
Total (95% CI) 62 62 100.0% 0.25 [0.07, 0.84]

Total events 3 12
0.01 0.1 1 10 100
Figure G-32. Myocardial infarction: Nerve blocks vs. no nerve block (systemic drugs)

Matot 2003 1 34 4 34 0.25 [0.03, 2.12]
0.01 0.1 1 10 100

Figure G-33. Puritis: Nerve blocks vs. no nerve block (systemic drugs)

Tuncer 2003 0 20 5 20 0.09 [0.01, 1.54]
0.001 0.1 1 10 1000

APPENDIX G 463
Figure G-34. Pulmonary embolism: Nerve blocks vs. no nerve block (systemic drugs)

Haddad 1995 0 25 2 25 83.1% 0.20 [0.01, 3.97]
Subtotal (95% CI) 25 25 83.1% 0.20 [0.01, 3.97]
Total events 0 2

Foss 2005 1 28 0 27 16.9% 2.90 [0.12, 68.15]
Subtotal (95% CI) 28 27 16.9% 2.90 [0.12, 68.15]
Total events 1 0
Total (95% CI) 53 52 100.0% 0.66 [0.11, 3.86]

Total events 1 2
0.01 0.1 1 10 100
Figure G-35. Deep vein thrombosis: Nerve blocks vs. no nerve block (systemic drugs)

Fletcher 2003 1 26 1 24 14.5% 0.92 [0.06, 13.95]
Haddad 1995 3 25 2 25 28.0% 1.50 [0.27, 8.22]
Subtotal (95% CI) 51 49 42.5% 1.30 [0.31, 5.46]
Total events 4 3

White 1980 2 16 1 20 12.4% 2.50 [0.25, 25.15]
Subtotal (95% CI) 16 20 12.4% 2.50 [0.25, 25.15]
Total events 2 1

Cuvillon 2007 1 21 1 41 9.5% 1.95 [0.13, 29.68]
Subtotal (95% CI) 21 41 9.5% 1.95 [0.13, 29.68]
Total events 1 1

Foss 2005 0 28 2 27 35.6% 0.19 [0.01, 3.85]
Subtotal (95% CI) 28 27 35.6% 0.19 [0.01, 3.85]
Total events 0 2
Total (95% CI) 116 137 100.0% 1.12 [0.43, 2.93]

Total events 7 7
0.01 0.1 1 10 100
464 APPENDIX G
Figure G-36. Mortality: Nerve blocks vs. no nerve block (systemic drugs)

Fletcher 2003 3 26 3 24 16.2% 0.92 [0.21, 4.14]
Haddad 1995 1 25 4 25 20.8% 0.25 [0.03, 2.08]
Subtotal (95% CI) 51 49 37.0% 0.54 [0.17, 1.77]
Total events 4 7

Matot 2003 0 34 4 34 23.4% 0.11 [0.01, 1.99]
Subtotal (95% CI) 34 34 23.4% 0.11 [0.01, 1.99]
Total events 0 4
1.27.3 Lateral cutaneous block (at surgery)

Jones 1985 1 10 0 9 2.7% 2.73 [0.12, 59.57]
Subtotal (95% CI) 10 9 2.7% 2.73 [0.12, 59.57]
Total events 1 0
1.27.4 Three in one block and subcostal block (at surgery)

Hood 1991 0 25 1 25 7.8% 0.33 [0.01, 7.81]
Subtotal (95% CI) 25 25 7.8% 0.33 [0.01, 7.81]
Total events 0 1

White 1980 1 20 0 20 2.6% 3.00 [0.13, 69.52]
Subtotal (95% CI) 20 20 2.6% 3.00 [0.13, 69.52]
Total events 1 0

Cuvillon 2007 2 21 6 41 21.1% 0.65 [0.14, 2.95]
Subtotal (95% CI) 21 41 21.1% 0.65 [0.14, 2.95]
Total events 2 6

Foss 2005 1 28 1 27 5.3% 0.96 [0.06, 14.65]
Subtotal (95% CI) 28 27 5.3% 0.96 [0.06, 14.65]
Total events 1 1
Total (95% CI) 189 205 100.0% 0.59 [0.29, 1.21]

Total events 9 19
0.005 0.1 1 10 200
APPENDIX G 465
Figure G-37. Pressure sores: Nerve blocks vs. no nerve block (systemic drugs)
Nerve block Control (No block) Risk Ratio Risk Ratio

Haddad 1995 1 25 5 25 41.4% 0.20 [0.03, 1.59]
Kullenberg 2004 0 40 0 40 Not estimable
Subtotal (95% CI) 65 65 41.4% 0.20 [0.03, 1.59]
Total events 1 5

Cuvillon 2007 2 21 4 41 58.6% 0.98 [0.19, 4.90]
Subtotal (95% CI) 21 41 58.6% 0.98 [0.19, 4.90]
Total events 2 4
Total (95% CI) 86 106 100.0% 0.51 [0.11, 2.39]

Total events 3 9
0.01 0.1 1 10 100
Figure G-38. Confusional state: Nerve blocks vs. no nerve block (systemic drugs)

Kullenberg 2004 6 40 12 40 43.6% 0.50 [0.21, 1.20]
Subtotal (95% CI) 40 40 43.6% 0.50 [0.21, 1.20]
Total events 6 12

White 1980 3 16 3 20 9.7% 1.25 [0.29, 5.38]
Subtotal (95% CI) 16 20 9.7% 1.25 [0.29, 5.38]
Total events 3 3

Cuvillon 2007 6 21 19 41 46.7% 0.62 [0.29, 1.31]
Subtotal (95% CI) 21 41 46.7% 0.62 [0.29, 1.31]
Total events 6 19
Total (95% CI) 77 101 100.0% 0.63 [0.37, 1.06]

Total events 15 34
0.01 0.1 1 10 100
19.4 Anaesthesia
Figure G-39. Mortality at 1 month (random effects model): Regional (spinal or epidural)
versus general anaesthesia
466 APPENDIX G
Regional General Risk Ratio Risk Ratio

Berggren 1987 1 28 0 29 1.8% 3.10 [0.13, 73.12]
Davis 1981 3 64 9 68 9.3% 0.35 [0.10, 1.25]
Davis 1987 17 259 16 279 21.8% 1.14 [0.59, 2.22]
Juelsgaard 1998 4 15 2 14 6.8% 1.87 [0.40, 8.65]
McKenzie 1984 8 73 13 75 17.1% 0.63 [0.28, 1.44]
McLaren 1978 4 56 17 60 12.7% 0.25 [0.09, 0.70]
Racle 1986 2 35 5 35 6.5% 0.40 [0.08, 1.93]
Valentin 1986 17 281 24 297 23.9% 0.75 [0.41, 1.36]
Total (95% CI) 811 857 100.0% 0.68 [0.44, 1.05]

Total events 56 86
0.01 0.1 1 10 100
Test for overall effect: Z = 1.75 (P = 0.08) Favours regional Favours general
Figure G-40. Mortality- early up to 1 month: Regional (spinal or epidural) versus general
anaesthesia
Additional analysis: The authors pooled mortality data from Adams 1990 and Bigler 1985 which reported
early mortality during hospital stay and Ungemach 1987 which reported mortality at 2 weeks with data
from the mortality at one month analysis.

Adams 1990 4 24 3 32 2.8% 1.78 [0.44, 7.21]
Berggren 1987 1 28 0 29 0.5% 3.10 [0.13, 73.12]
Bigler 1985 1 20 1 20 1.1% 1.00 [0.07, 14.90]
Davis 1981 3 64 9 68 9.6% 0.35 [0.10, 1.25]
Davis 1987 17 259 16 279 17.0% 1.14 [0.59, 2.22]
Juelsgaard 1998 4 15 2 14 2.3% 1.87 [0.40, 8.65]
McKenzie 1984 8 73 13 75 14.1% 0.63 [0.28, 1.44]
McLaren 1978 4 56 17 60 18.1% 0.25 [0.09, 0.70]
Racle 1986 2 35 5 35 5.5% 0.40 [0.08, 1.93]
Ungemach 1993 3 57 3 57 3.3% 1.00 [0.21, 4.75]
Valentin 1986 17 281 24 297 25.7% 0.75 [0.41, 1.36]
Total (95% CI) 912 966 100.0% 0.73 [0.54, 0.99]

Total events 64 93
0.01 0.1 1 10 100
APPENDIX G 467
Figure G-41. Length of stay in hospital: Regional (spinal or epidural) versus general
anaesthesia
Regional General Mean Difference Mean Difference
McKenzie 1984 38.8 55.5 73 42.9 69.3 75 6.1% -4.10 [-24.30, 16.10]
Racle 1986 20.09 10.6 35 20.05 11.4 35 93.9% 0.04 [-5.12, 5.20]
Total (95% CI) 108 110 100.0% -0.21 [-5.21, 4.78]

-100 -50 0 50 100
Figure G-42. Vomiting: Regional (spinal or epidural) versus general anaesthesia

Bigler 1985 1 20 2 20 67.9% 0.50 [0.05, 5.08]
McLaren 1978 1 26 1 29 32.1% 1.12 [0.07, 16.95]
Total (95% CI) 46 49 100.0% 0.70 [0.12, 3.94]

Total events 2 3
0.01 0.1 1 10 100
Figure G-43. Acute confusional state: Regional (spinal or epidural) versus general
anaesthesia
Regional Control Risk Ratio Risk Ratio
Berggren 1987 4 28 7 29 29.5% 0.59 [0.19, 1.80]
Bigler 1985 1 20 1 20 4.3% 1.00 [0.07, 14.90]
Casati 2003 1 15 3 15 12.9% 0.33 [0.04, 2.85]
Kamitani 2003 0 19 1 21 6.1% 0.37 [0.02, 8.50]
Racle 1986 5 35 11 35 47.2% 0.45 [0.18, 1.17]
Total (95% CI) 117 120 100.0% 0.50 [0.26, 0.95]

Total events 11 23
0.01 0.1 1 10 100
468 APPENDIX G
Figure G-44. Pneumonia: Regional (spinal or epidural) versus general anaesthesia

1.15.1 Fatal (reason for death only)
Adams 1990 1 24 1 32 3.0% 1.33 [0.09, 20.26]
Davis 1981 2 64 4 68 13.4% 0.53 [0.10, 2.80]
Davis 1987 5 259 4 279 13.3% 1.35 [0.37, 4.96]
Juelsgaard 1998 2 15 0 14 1.8% 4.69 [0.24, 89.88]
McKenzie 1984 5 73 3 75 10.3% 1.71 [0.42, 6.91]
McLaren 1978 1 56 5 60 16.7% 0.21 [0.03, 1.78]
Subtotal (95% CI) 491 528 58.5% 1.00 [0.52, 1.94]
Total events 16 17
1.15.2 Other (non fatal or fatal)

Berggren 1987 1 28 2 29 6.8% 0.52 [0.05, 5.40]
Bigler 1985 1 20 2 20 6.9% 0.50 [0.05, 5.08]
Racle 1986 3 35 8 35 27.7% 0.38 [0.11, 1.30]
Subtotal (95% CI) 83 84 41.5% 0.42 [0.16, 1.13]
Total events 5 12
Total (95% CI) 574 612 100.0% 0.76 [0.44, 1.30]

Total events 21 29
0.01 0.1 1 10 100
Figure G-45. Myocardial infarction: Regional (spinal or epidural) versus general

anaesthesia
Davis 1981 0 64 1 68 11.4% 0.35 [0.01, 8.53]
Davis 1987 2 259 1 279 7.5% 2.15 [0.20, 23.62]
McKenzie 1984 0 73 2 75 19.3% 0.21 [0.01, 4.21]
McLaren 1978 0 56 3 60 26.5% 0.15 [0.01, 2.90]
Subtotal (95% CI) 452 482 64.7% 0.44 [0.13, 1.50]
Total events 2 7
1.16.2 Other (non fatal or fatal)

Juelsgaard 1998 1 15 0 14 4.0% 2.81 [0.12, 63.83]
Racle 1986 2 35 4 35 31.3% 0.50 [0.10, 2.56]
Subtotal (95% CI) 50 49 35.3% 0.76 [0.20, 2.96]
Total events 3 4
Total (95% CI) 502 531 100.0% 0.55 [0.22, 1.37]

Total events 5 11
0.01 0.1 1 10 100
APPENDIX G 469
Figure G-46. Pulmonary embolism (Peto odds ratio): Regional (spinal or epidural) versus
general anaesthesia
Regional General Peto Odds Ratio Peto Odds Ratio

Study or Subgroup Events Total Events Total Weight Peto, Fixed, 95% CI Peto, Fixed, 95% CI
Adams 1990 1 24 0 32 4.6% 10.31 [0.20, 541.25]
Berggren 1987 2 28 0 29 9.2% 7.95 [0.48, 130.33]
Bigler 1985 2 20 0 20 9.1% 7.79 [0.47, 129.11]
Brichant 1995 1 46 0 42 4.7% 6.77 [0.13, 342.76]
Davis 1981 0 64 4 68 18.2% 0.14 [0.02, 1.00]
Davis 1987 0 259 1 279 4.7% 0.15 [0.00, 7.35]
McKenzie 1984 1 73 3 75 18.3% 0.37 [0.05, 2.68]
McLaren 1978 1 56 5 60 26.7% 0.27 [0.05, 1.37]
Racle 1986 1 35 0 35 4.7% 7.39 [0.15, 372.38]
Total (95% CI) 605 640 100.0% 0.72 [0.31, 1.69]

Total events 9 13
0.001 0.1 1 10 1000
Figure G-47. Pulmonary embolism (random effects model): Regional (spinal or epidural)

Adams 1990 1 24 0 32 9.0% 3.96 [0.17, 93.17]
Berggren 1987 2 28 0 29 9.9% 5.17 [0.26, 103.18]
Bigler 1985 2 20 0 20 10.0% 5.00 [0.26, 98.00]
Brichant 1995 1 46 0 42 8.9% 2.74 [0.11, 65.59]
Davis 1981 0 64 4 68 10.4% 0.12 [0.01, 2.15]
Davis 1987 0 259 1 279 8.8% 0.36 [0.01, 8.77]
McKenzie 1984 1 73 3 75 16.2% 0.34 [0.04, 3.22]
McLaren 1978 1 56 5 60 17.8% 0.21 [0.03, 1.78]
Racle 1986 1 35 0 35 8.9% 3.00 [0.13, 71.22]
Total (95% CI) 605 640 100.0% 0.88 [0.32, 2.39]

Total events 9 13
0.001 0.1 1 10 1000
470 APPENDIX G
Figure G-48. Pulmonary embolism (fatal and non fatal): Regional (spinal or epidural)

Adams 1990 1 24 0 32 3.0% 3.96 [0.17, 93.17]
Bigler 1985 1 20 0 20 3.4% 3.00 [0.13, 69.52]
Davis 1981 0 64 4 68 30.0% 0.12 [0.01, 2.15]
Davis 1987 0 259 1 279 9.9% 0.36 [0.01, 8.77]
McKenzie 1984 1 73 3 75 20.4% 0.34 [0.04, 3.22]
McLaren 1978 1 56 5 60 33.2% 0.21 [0.03, 1.78]
Subtotal (95% CI) 496 534 100.0% 0.43 [0.17, 1.10]
Total events 4 13
1.26.2 Non fatal

Berggren 1987 2 28 0 29 24.4% 5.17 [0.26, 103.18]
Bigler 1985 1 20 0 20 24.8% 3.00 [0.13, 69.52]
Brichant 1995 1 46 0 42 25.9% 2.74 [0.11, 65.59]
Racle 1986 1 35 0 35 24.8% 3.00 [0.13, 71.22]
Subtotal (95% CI) 129 126 100.0% 3.46 [0.74, 16.29]
Total events 5 0
0.001 0.1 1 10 1000

Favours regional Favours general
Figure G-49. Deep vein thrombosis: Regional (spinal or epidural) versus general
anaesthesia

1.23.1 Fatal (underlying reason for death only)
McLaren 1978 0 26 2 29 3.9% 0.22 [0.01, 4.43]
Subtotal (95% CI) 26 29 3.9% 0.22 [0.01, 4.43]
Total events 0 2
1.23.2 Other: venography diagnosis

Brichant 1995 14 46 13 42 22.2% 0.98 [0.52, 1.84]
McKenzie 1984 8 20 16 20 26.2% 0.50 [0.28, 0.89]
Subtotal (95% CI) 66 62 48.4% 0.72 [0.47, 1.11]
Total events 22 29
1.23.3 Other: fibrinogen scan diagnosis

Davis 1981 17 37 30 39 47.8% 0.60 [0.40, 0.88]
Subtotal (95% CI) 37 39 47.8% 0.60 [0.40, 0.88]
Total events 17 30
Total (95% CI) 129 130 100.0% 0.64 [0.48, 0.86]

Total events 39 61
0.01 0.1 1 10 100
APPENDIX G 471
19.5 Surgical interventions
19.5.1 Surgeon seniority
Figure G-50. Reoperation rate for technically demanding hip fractures at 6 months:
Senior/higher grade surgeon versus junior/lower grade surgeon

PALM2007 0.69813472 0.3523805 100.0% 2.01 [1.01, 4.01]
Total (95% CI) 100.0% 2.01 [1.01, 4.01]

0.01 0.1 1 10 100
Test for overall effect: Z = 1.98 (P = 0.05) Favours experimental Favours control
Figure G-51. Dislocation rate for arthroplasty: Senior/higher grade surgeon versus
junior/lower grade surgeon
1.3.1 Hemiarthroplasty
ENOCSON2008 0.26236426 0.4105709 1.30 [0.58, 2.91]
1.3.2 Total hip replacement

ENOCSON2009 -0.1053605 0.56979393 0.90 [0.29, 2.75]
0.01 0.1 1 10 100

Favours experimental Favours control
472 APPENDIX G
19.5.2 Cement in older designs of arthroplasty
Figure G-52. Perioperative mortality - older designs of arthroplasty: cemented vs.

uncemented.
Cemented Uncemented Risk Ratio Risk Ratio
1.1.1 Cemented Thompson versus uncemented Moore
Parker 2010 0 200 0 200 Not estimable
Subtotal (95% CI) 200 200 Not estimable
Total events 0 0
Test for overall effect: Not applicable
1.1.2 Cemented Thompson versus uncemented Thompson

Harper 1994 1 77 0 66 100.0% 2.58 [0.11, 62.21]
Subtotal (95% CI) 77 66 100.0% 2.58 [0.11, 62.21]
Total events 1 0
Total (95% CI) 277 266 100.0% 2.58 [0.11, 62.21]

Total events 1 0
0.01 0.1 1 10 100
Test for overall effect: Z = 0.58 (P = 0.56) Favours cement Favours no cement
Figure G-53. Mortality – at up to 1 month - older designs of arthroplasty: cemented vs.

uncemented.
Cement Uncemented Risk Ratio Risk Ratio
Parker 2010 9 200 11 200 84.4% 0.82 [0.35, 1.93]
Subtotal (95% CI) 200 200 84.4% 0.82 [0.35, 1.93]
Total events 9 11
1.12.2 Cemented Thompson bipolar versus uncemented Moore bipolar

Emery 1991 2 27 2 26 15.6% 0.96 [0.15, 6.34]
Subtotal (95% CI) 27 26 15.6% 0.96 [0.15, 6.34]
Total events 2 2
Total (95% CI) 227 226 100.0% 0.84 [0.38, 1.84]

Total events 11 13
0.1 0.2 0.5 1 2 5 10
APPENDIX G 473
Figure G-54. Mortality at between 1 and 3 months - older designs of arthroplasty:

cemented vs. uncemented.
Parker 2009 22 200 29 200 58.8% 0.76 [0.45, 1.27]
Subtotal (95% CI) 200 200 58.8% 0.76 [0.45, 1.27]
Total events 22 29

Harper 1994 12 77 6 66 13.1% 1.71 [0.68, 4.32]
Subtotal (95% CI) 77 66 13.1% 1.71 [0.68, 4.32]
Total events 12 6
1.13.3 Cemented Moore versus uncemented Moore

Sonne-Holm 1982 11 55 11 57 21.9% 1.04 [0.49, 2.19]
Subtotal (95% CI) 55 57 21.9% 1.04 [0.49, 2.19]
Total events 11 11

Emery 1991 4 27 3 26 6.2% 1.28 [0.32, 5.19]
Subtotal (95% CI) 27 26 6.2% 1.28 [0.32, 5.19]
Total events 4 3
Total (95% CI) 359 349 100.0% 0.98 [0.68, 1.41]

Total events 49 49
0.1 0.2 0.5 1 2 5 10
474 APPENDIX G
Figure G-55. Mortality at 1 year - older designs of arthroplasty: cemented vs.

uncemented.
Parker 2009 53 200 62 200 55.3% 0.85 [0.63, 1.17]
Subtotal (95% CI) 200 200 55.3% 0.85 [0.63, 1.17]
Total events 53 62

Branfoot 2000 7 38 14 53 10.4% 0.70 [0.31, 1.56]
Harper 1994 20 77 17 66 16.3% 1.01 [0.58, 1.76]
Subtotal (95% CI) 115 119 26.8% 0.89 [0.56, 1.40]
Total events 27 31

Emery 1991 8 27 6 26 5.5% 1.28 [0.52, 3.19]
Subtotal (95% CI) 27 26 5.5% 1.28 [0.52, 3.19]
Total events 8 6
1.14.4 Cemented bipolar versus uncemented bipolar hemiarthroplasty

Santini 2005 13 53 14 53 12.5% 0.93 [0.48, 1.78]
Subtotal (95% CI) 53 53 12.5% 0.93 [0.48, 1.78]
Total events 13 14
Total (95% CI) 395 398 100.0% 0.90 [0.71, 1.13]

Total events 101 113
0.1 0.2 0.5 1 2 5 10
Figure G-56. Mortality at 3 years - older designs of arthroplasty: cemented vs.

uncemented.

Parker 2009 102 200 96 200 100.0% 1.06 [0.87, 1.30]
Subtotal (95% CI) 200 200 100.0% 1.06 [0.87, 1.30]
Total events 102 96
0.1 0.2 0.5 1 2 5 10

Favours cement Favours no cement
APPENDIX G 475
Figure G-57. Number of patients failing to regain mobility - older designs of arthroplasty:

Parker 2009 90 144 93 137 56.1% 0.92 [0.78, 1.09]
Subtotal (95% CI) 144 137 56.1% 0.92 [0.78, 1.09]
Total events 90 93

Sonne-Holm 1982 19 33 15 25 26.1% 0.96 [0.62, 1.48]
Subtotal (95% CI) 33 25 26.1% 0.96 [0.62, 1.48]
Total events 19 15

Emery 1991 8 19 16 20 17.8% 0.53 [0.30, 0.93]
Subtotal (95% CI) 19 20 17.8% 0.53 [0.30, 0.93]
Total events 8 16
Total (95% CI) 196 182 100.0% 0.84 [0.64, 1.11]

0.1 0.2 0.5 1 2 5 10
Figure G-58. Change in mobility score - older designs of arthroplasty: cemented vs.
uncemented.
Cemented Uncemented Mean Difference Mean Difference
Parker 2009 1.4 1.9 150 2.2 1.9 144 100.0% -0.80 [-1.23, -0.37]
Subtotal (95% CI) 150 144 100.0% -0.80 [-1.23, -0.37]
-2 -1 0 1 2
Favours Cement Favours no cement
476 APPENDIX G
Figure G-59. Length of hospital stay - older designs of arthroplasty: cemented vs.
uncemented.
Cement Uncemented Mean Difference Mean Difference

Parker 2009 20.3 22.3 200 24.7 25.8 200 13.5% -4.40 [-9.13, 0.33]
Subtotal (95% CI) 200 200 13.5% -4.40 [-9.13, 0.33]

Harper 1994 14.38 9.54 77 16.56 6.34 66 43.7% -2.18 [-4.80, 0.44]
Subtotal (95% CI) 77 66 43.7% -2.18 [-4.80, 0.44]

Emery 1991 21.8 11.7 24 19.5 8.4 23 8.9% 2.30 [-3.50, 8.10]
Subtotal (95% CI) 24 23 8.9% 2.30 [-3.50, 8.10]

Santini 2005 17.23 9.1 53 17.46 6.29 53 33.9% -0.23 [-3.21, 2.75]
Subtotal (95% CI) 53 53 33.9% -0.23 [-3.21, 2.75]
Total (95% CI) 354 342 100.0% -1.42 [-3.15, 0.32]

-10 -5 0 5 10
Test for subgroup differences: Chi² = 4.04, df = 3 (P = 0.26), I² = 25.8%
Figure G-60. Number of patients failing to return home - older designs of arthroplasty:
Parker 2009 13 200 21 200 81.2% 0.62 [0.32, 1.20]
Subtotal (95% CI) 200 200 81.2% 0.62 [0.32, 1.20]
Total events 13 21

Emery 1991 3 19 5 20 18.8% 0.63 [0.17, 2.29]
Subtotal (95% CI) 19 20 18.8% 0.63 [0.17, 2.29]
Total events 3 5
Total (95% CI) 219 220 100.0% 0.62 [0.34, 1.12]

Total events 16 26
0.1 0.2 0.5 1 2 5 10
APPENDIX G 477
Figure G-61. Number of patients reporting pain at 3 months - older designs of

arthroplasty: cemented vs. uncemented.

Parker 2009 60 163 74 161 86.7% 0.80 [0.62, 1.04]
Subtotal (95% CI) 163 161 86.7% 0.80 [0.62, 1.04]
Total events 60 74

Sonne-Holm 1982 7 29 10 22 13.3% 0.53 [0.24, 1.17]
Subtotal (95% CI) 29 22 13.3% 0.53 [0.24, 1.17]
Total events 7 10
Total (95% CI) 192 183 100.0% 0.77 [0.60, 0.98]

Total events 67 84
0.02 0.1 1 10 50
Figure G-62. Number of patients reporting pain at 1 to 2 years - older designs of

arthroplasty: cemented vs. uncemented.
Parker 2009 28 141 45 131 61.5% 0.58 [0.38, 0.87]
Subtotal (95% CI) 141 131 61.5% 0.58 [0.38, 0.87]
Total events 28 45

Sonne-Holm 1982 10 33 12 25 18.0% 0.63 [0.33, 1.22]
Subtotal (95% CI) 33 25 18.0% 0.63 [0.33, 1.22]
Total events 10 12

Emery 1991 6 19 16 20 20.5% 0.39 [0.20, 0.79]
Subtotal (95% CI) 19 20 20.5% 0.39 [0.20, 0.79]
Total events 6 16
Total (95% CI) 193 176 100.0% 0.55 [0.40, 0.75]

Total events 44 73
0.1 0.2 0.5 1 2 5 10
478 APPENDIX G
Figure G-63. Pain score at 6 months - older designs of arthroplasty: cemented vs.
uncemented.

Study or Subgroup Mean SD Total Mean SD Total IV, Fixed, 95% CI IV, Fixed, 95% CI
Parker 2009 1.8 1.2 147 2.4 1.4 142 -0.60 [-0.90, -0.30]
-1 -0.5 0 0.5 1
Favours cement Favours no cement
Figure G-64. Reoperations - older designs of arthroplasty: cemented vs. uncemented.

Parker 2009 10 200 18 200 93.5% 0.56 [0.26, 1.17]
Subtotal (95% CI) 200 200 93.5% 0.56 [0.26, 1.17]
Total events 10 18

Branfoot 2000 0 38 1 53 6.5% 0.46 [0.02, 11.03]
Subtotal (95% CI) 38 53 6.5% 0.46 [0.02, 11.03]
Total events 0 1
Total (95% CI) 238 253 100.0% 0.55 [0.27, 1.14]

Total events 10 19
0.05 0.2 1 5 20
APPENDIX G 479
Figure G-65. Deep sepsis - older designs of arthroplasty: cemented vs. uncemented.
Parker 2010 6 200 5 200 66.6% 1.20 [0.37, 3.87]
Subtotal (95% CI) 200 200 66.6% 1.20 [0.37, 3.87]
Total events 6 5

Harper 1994 1 77 0 66 7.2% 2.58 [0.11, 62.21]
Subtotal (95% CI) 77 66 7.2% 2.58 [0.11, 62.21]
Total events 1 0

Sonne-Holm 1982 0 55 1 57 19.6% 0.35 [0.01, 8.30]
Subtotal (95% CI) 55 57 19.6% 0.35 [0.01, 8.30]
Total events 0 1

Santini 2005 1 53 0 53 6.7% 3.00 [0.12, 72.02]
Subtotal (95% CI) 53 53 6.7% 3.00 [0.12, 72.02]
Total events 1 0
Total (95% CI) 385 376 100.0% 1.25 [0.48, 3.24]

Total events 8 6
0.01 0.1 1 10 100
Figure G-66. Wound haematoma - older designs of arthroplasty: cemented vs.

uncemented.
Cemented Uncemented Odds Ratio Odds Ratio
Parker 2010 2 200 1 200 100.0% 2.01 [0.18, 22.35]
Subtotal (95% CI) 200 200 100.0% 2.01 [0.18, 22.35]
Total events 2 1
0.01 0.1 1 10 100

Favours cemen Favours no cement
480 APPENDIX G
19.5.3 Cement in newer designs of arthroplasty
Figure G-67. Mortality - newer designs of arthroplasty: cemented vs. uncemented.

2.1.1 30 days
Figved 2009 4 108 8 105 100.0% 0.49 [0.15, 1.57]
Subtotal (95% CI) 108 105 100.0% 0.49 [0.15, 1.57]
Total events 4 8
2.1.2 90 days
Figved 2009 13 108 15 105 100.0% 0.84 [0.42, 1.68]
Subtotal (95% CI) 108 105 100.0% 0.84 [0.42, 1.68]
Total events 13 15
2.1.3 12 months
Figved 2009 20 108 30 105 100.0% 0.65 [0.39, 1.07]
Subtotal (95% CI) 108 105 100.0% 0.65 [0.39, 1.07]
Total events 20 30
2.1.4 24 months
Figved 2009 32 108 36 105 100.0% 0.86 [0.58, 1.28]
Subtotal (95% CI) 108 105 100.0% 0.86 [0.58, 1.28]
Total events 32 36
0.1 0.2 0.5 1 2 5 10

Favours cemented Favours uncemented
Figure G-68. Reoperations - newer designs of arthroplasty: cemented vs. uncemented.

Figved 2009 7 112 8 108 100.0% 0.84 [0.32, 2.25]
Total (95% CI) 112 108 100.0% 0.84 [0.32, 2.25]

Total events 7 8
0.1 0.2 0.5 1 2 5 10
Test for overall effect: Z = 0.34 (P = 0.73) Favours hemiarthroplasty Favours uncemented
Figure G-69. Pain – need for pain medication - newer designs of arthroplasty: cemented
vs. uncemented.
2.7.2 Need for pain medication at 12 months
Figved 2009 23 91 14 77 100.0% 1.39 [0.77, 2.51]
Subtotal (95% CI) 91 77 100.0% 1.39 [0.77, 2.51]
Total events 23 14
0.1 0.2 0.5 1 2 5 10

Figure G-70. Unable to walk without aids at 12 months –newer designs of arthroplasty:
APPENDIX G 481

Figved 2009 4 91 6 77 100.0% 0.56 [0.17, 1.93]
Total (95% CI) 91 77 100.0% 0.56 [0.17, 1.93]

Total events 4 6
0.1 0.2 0.5 1 2 5 10
Test for overall effect: Z = 0.91 (P = 0.36) Favours cemented Favours uncemented
Figure G-71. Barthel Index –newer designs of arthroplasty: cemented vs. uncemented.
2.8.2 Barthel score less than 19 at 12 months
Figved 2009 46 91 29 77 100.0% 1.34 [0.94, 1.91]
Subtotal (95% CI) 91 77 100.0% 1.34 [0.94, 1.91]
Total events 46 29
0.1 0.2 0.5 1 2 5 10

Figure G-72. Harris Hip Score and Eq-5d scores –newer designs of arthroplasty: cemented
vs. uncemented.
2.10.2 Harris hip score at 12 months
Figved 2009 78.9 15.7 90 79.8 17.6 77 100.0% -0.90 [-6.00, 4.20]
Subtotal (95% CI) 90 77 100.0% -0.90 [-6.00, 4.20]

2.10.4 ED-5D index score at 12 months
Figved 2009 0.68 0.23 56 0.61 0.32 57 100.0% 0.07 [-0.03, 0.17]
Subtotal (95% CI) 56 57 100.0% 0.07 [-0.03, 0.17]

2.10.6 ED-5D visual analogue score at 12 months
Figved 2009 61 17.7 61 65 20.1 60 100.0% -4.00 [-10.75, 2.75]

Subtotal (95% CI) 61 60 100.0% -4.00 [-10.75, 2.75]

-10 -5 0 5 10
Favours uncemented Favours cemented
Figure G-73. Length of hospital stay –newer designs of arthroplasty: cemented vs.
uncemented.
Figved 2009 7.8 4.11 109 8.4 9.02 106 100.0% -0.60 [-2.48, 1.28]
Total (95% CI) 109 106 100.0% -0.60 [-2.48, 1.28]

-4 -2 0 2 4
Test for overall effect: Z = 0.62 (P = 0.53) Favours cemented Favours uncemented
482 APPENDIX G
19.5.4 Internal fixation versus hemiarthroplasty
Figure G-74. Mortality: Internal fixation versus hemiarthroplasty
Fixation Hemiarthroplasty Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI Year M-H, Fixed, 95% CI
2.1.1 Up to 1 month
Frihagen 2007 7 112 10 110 100.0% 0.69 [0.27, 1.74] 2007
Subtotal (95% CI) 112 110 100.0% 0.69 [0.27, 1.74]
Total events 7 10
2.1.3 Up to 3-6 months

Soreide 1979 6 51 5 53 3.9% 1.25 [0.41, 3.83] 1979
Jensen 1984 5 50 14 52 11.0% 0.37 [0.14, 0.96] 1984
Svenningsen 1985 18 110 9 59 9.4% 1.07 [0.51, 2.24] 1985
van Vugt 1993 1 21 4 21 3.2% 0.25 [0.03, 2.05] 1993
van Dortmont 2000 11 31 10 29 8.3% 1.03 [0.52, 2.05] 2000
Puolakka 2001 1 16 1 15 0.8% 0.94 [0.06, 13.68] 2001
Parker 2002 41 226 49 229 39.0% 0.85 [0.58, 1.23] 2002
Blomfeldt 2005 5 30 4 30 3.2% 1.25 [0.37, 4.21] 2005
Keating 2006 3 118 6 111 5.0% 0.47 [0.12, 1.84] 2006
Frihagen 2007 16 112 20 110 16.2% 0.79 [0.43, 1.44] 2007
Subtotal (95% CI) 765 709 100.0% 0.81 [0.64, 1.03]
2.1.4 Up to 12 months
Soreide 1979 9 51 11 53 6.1% 0.85 [0.38, 1.88] 1979
Jensen 1984 11 50 19 52 10.6% 0.60 [0.32, 1.13] 1984
Svenningsen 1985 25 110 13 59 9.6% 1.03 [0.57, 1.86] 1985
van Vugt 1993 2 20 5 21 2.8% 0.42 [0.09, 1.92] 1993
van Dortmont 2000 20 31 14 29 8.2% 1.34 [0.85, 2.11] 2000
Parker 2002 61 226 63 229 35.6% 0.98 [0.73, 1.32] 2002
Blomfeldt 2005 10 30 7 30 4.0% 1.43 [0.63, 3.25] 2005
Keating 2006 10 118 11 111 6.4% 0.86 [0.38, 1.93] 2006
Frihagen 2007 24 112 29 110 16.6% 0.81 [0.51, 1.30] 2007
Subtotal (95% CI) 748 694 100.0% 0.93 [0.78, 1.12]
2.1.5 Up to 24 to 36 months
Jensen 1984 20 50 28 52 10.5% 0.74 [0.49, 1.13] 1984
Svenningsen 1985 40 110 21 59 10.4% 1.02 [0.67, 1.56] 1985
van Vugt 1993 5 21 6 21 2.3% 0.83 [0.30, 2.31] 1993
van Dortmont 2000 28 31 22 29 8.7% 1.19 [0.94, 1.51] 2000
Puolakka 2001 8 16 7 15 2.8% 1.07 [0.52, 2.22] 2001
Parker 2002 87 209 97 209 37.0% 0.90 [0.72, 1.11] 2002
Roden 2003 7 53 4 47 1.6% 1.55 [0.48, 4.97] 2003
Blomfeldt 2005 13 30 12 30 4.6% 1.08 [0.59, 1.97] 2005
Keating 2006 18 118 18 111 7.1% 0.94 [0.52, 1.71] 2006
Frihagen 2007 39 112 39 110 15.0% 0.98 [0.69, 1.40] 2007
Subtotal (95% CI) 750 683 100.0% 0.96 [0.84, 1.09]
0.01 0.1 1 10 100

Favours internal fixation Favours hemiarthroplasty
APPENDIX G 483
Figure G-75. Reoperations: Internal fixation versus hemiarthroplasty
Favours fixation Hemiarthroplasty Risk Ratio Risk Ratio

2.14.1 Screws versus Thompson
Puolakka 2001 7 16 1 15 1.0% 6.56 [0.91, 47.21]
van Dortmont 2000 4 31 1 29 1.0% 3.74 [0.44, 31.55]
Subtotal (95% CI) 47 44 2.1% 5.15 [1.22, 21.68]
Total events 11 2
2.14.2 Screws versus Moore

Blomfeldt 2005 10 30 4 30 4.0% 2.50 [0.88, 7.10]
Jensen 1984 8 50 2 52 2.0% 4.16 [0.93, 18.65]
Parker 2002 90 226 15 229 15.0% 6.08 [3.63, 10.17]
Subtotal (95% CI) 306 311 21.0% 5.21 [3.36, 8.09]
Total events 108 21
Test for overall effect: Z = 7.37 (P < 0.00001)
2.14.3 SHS versus Moore

Skinner 1989 30 91 22 91 22.2% 1.36 [0.85, 2.18]
Subtotal (95% CI) 91 91 22.2% 1.36 [0.85, 2.18]
Total events 30 22
2.14.4 SHS versus Stanmore bipolar

van Vugt 1993 6 21 7 22 6.9% 0.90 [0.36, 2.23]
Subtotal (95% CI) 21 22 6.9% 0.90 [0.36, 2.23]
Total events 6 7
2.14.5 Screws versus Charnley-Hastings bipolar cemented hemiarthroplasty

Frihagen 2007 70 111 13 108 13.3% 5.24 [3.09, 8.89]
Subtotal (95% CI) 111 108 13.3% 5.24 [3.09, 8.89]
Total events 70 13
2.14.6 Screws, SHS or nail & plate versus Christiansen bipolar

Soreide 1979 9 51 5 53 4.9% 1.87 [0.67, 5.21]
Svenningsen 1985 16 110 8 59 10.5% 1.07 [0.49, 2.36]
Subtotal (95% CI) 161 112 15.4% 1.33 [0.72, 2.47]
Total events 25 13
2.14.7 SHS versus Thompson or Monk bipolar

Davison 2001 28 93 8 187 5.4% 7.04 [3.34, 14.83]
Subtotal (95% CI) 93 187 5.4% 7.04 [3.34, 14.83]
Total events 28 8
2.14.8 Screws or SHS versus bipolar hemiarthroplasty

Keating 2006 46 118 6 111 6.2% 7.21 [3.21, 16.22]
Subtotal (95% CI) 118 111 6.2% 7.21 [3.21, 16.22]
Total events 46 6
2.14.9 2 von Bahr screws versus Variokopf bipolar hemiarthroplasty

Roden 2003 31 53 7 47 7.5% 3.93 [1.91, 8.07]
Subtotal (95% CI) 53 47 7.5% 3.93 [1.91, 8.07]
Total events 31 7
Total (95% CI) 1001 1033 100.0% 3.59 [2.93, 4.39]

Total events 355 99
Heterogeneity: Chi² = 48.84, df = 12 (P < 0.00001); I² = 75%
0.01 0.1 1 10 100
Test for overall effect: Z = 12.41 (P < 0.00001) Favours fixation Favours hemiarthroplasty
484 APPENDIX G
Figure G-76. Failure to return to same residence by final follow up: Internal fixation versus
hemiarthroplasty

Jensen 1984 4 23 7 23 20.5% 0.57 [0.19, 1.69]
Parker 2002 25 164 27 162 79.5% 0.91 [0.56, 1.51]
Total (95% CI) 187 185 100.0% 0.84 [0.54, 1.33]

Total events 29 34
0.1 0.2 0.5 1 2 5 10
Test for overall effect: Z = 0.73 (P = 0.46) Favours fixation Favours hemiarthropl
Figure G-77. Failure to regain mobility: Internal fixation versus hemiarthroplasty

2.14.1 Screws versus Thompson
van Dortmont 2000 4 10 3 14 5.3% 1.87 [0.53, 6.57]
Subtotal (95% CI) 10 14 5.3% 1.87 [0.53, 6.57]
Total events 4 3

Blomfeldt 2005 6 20 10 23 10.5% 0.69 [0.31, 1.56]
Jensen 1984 5 23 11 23 9.3% 0.45 [0.19, 1.10]
Parker 2002 98 164 105 166 32.7% 0.94 [0.80, 1.12]
Subtotal (95% CI) 207 212 52.6% 0.79 [0.53, 1.18]
2.14.3 2 von Bahr screws versus Variokopf bipolar hemiarthroplasty

Roden 2003 23 40 13 44 17.9% 1.95 [1.15, 3.30]
Subtotal (95% CI) 40 44 17.9% 1.95 [1.15, 3.30]
Total events 23 13
2.14.5 Screws v Christiansen bipolar

Soreide 1979 19 30 23 36 24.2% 0.99 [0.69, 1.43]
Subtotal (95% CI) 30 36 24.2% 0.99 [0.69, 1.43]
Total events 19 23
Total (95% CI) 287 306 100.0% 1.02 [0.74, 1.39]

0.1 0.2 0.5 1 2 5 10
Test for overall effect: Z = 0.12 (P = 0.91) Favours fixation Favours hemiarthroplasty
APPENDIX G 485
Figure G-78. Patients reporting pain at 1 year: Internal fixation versus hemiarthroplasty
Fixation hemiarthroplasty Risk Ratio Risk Ratio

Blomfeldt 2005 3 20 6 23 8.5% 0.57 [0.16, 2.01]
Parker 2002 58 160 72 163 44.9% 0.82 [0.63, 1.07]
Subtotal (95% CI) 180 186 53.3% 0.81 [0.62, 1.05]
Total events 61 78
2.28.7 Screws or SHS versus bipolar hemiarthroplasty

Keating 2006 65 100 49 95 46.7% 1.26 [0.99, 1.61]
Subtotal (95% CI) 100 95 46.7% 1.26 [0.99, 1.61]
Total events 65 49
Total (95% CI) 280 281 100.0% 0.97 [0.66, 1.44]

0.1 0.2 0.5 1 2 5 10
Test for overall effect: Z = 0.14 (P = 0.89) Favours fixation Favours hemiarthroplasty
Figure G-79. Harris Hip Score: Internal fixation versus hemiarthroplasty

Fixation Hemiarthroplasty Mean Difference Mean Difference
2.30.1 at 4 months
Frihagen 2007 59.6 19.5 89 67.7 15.8 84 100.0% -8.10 [-13.38, -2.82]
Subtotal (95% CI) 89 84 100.0% -8.10 [-13.38, -2.82]
2.30.2 at 12 months
Frihagen 2007 65.8 15.9 87 72.6 17.5 74 100.0% -6.80 [-12.00, -1.60]
Subtotal (95% CI) 87 74 100.0% -6.80 [-12.00, -1.60]
2.30.3 at 24 months
Frihagen 2007 67.3 15.5 71 70.6 19.1 68 100.0% -3.30 [-9.10, 2.50]
Subtotal (95% CI) 71 68 100.0% -3.30 [-9.10, 2.50]
-100 -50 0 50 100

Favours hemiarthroplasty Favours fixation
486 APPENDIX G
Figure G-80. Number of patients with Barthel Index Score of 95 or 100: Internal fixation
versus hemiarthroplasty
Internal fixation Hemiarthroplasty Risk Ratio Risk Ratio

2.19.1 at 4 months
Frihagen 2007 41 88 40 80 100.0% 0.93 [0.68, 1.27]
Subtotal (95% CI) 88 80 100.0% 0.93 [0.68, 1.27]
Total events 41 40
2.19.2 at 12 months
Frihagen 2007 31 87 39 73 100.0% 0.67 [0.47, 0.95]
Subtotal (95% CI) 87 73 100.0% 0.67 [0.47, 0.95]
Total events 31 39
2.19.3 at 24 months
Frihagen 2007 24 69 26 68 100.0% 0.91 [0.58, 1.42]
Subtotal (95% CI) 69 68 100.0% 0.91 [0.58, 1.42]
Total events 24 26
0.01 0.1 1 10 100

Favours fixation Favours hemiarthroplasty
Figure G-81. Euroquol Eq-5d score: Internal fixation versus hemiarthroplasty

2.31.1 at 4 months
Frihagen 2007 0.53 0.29 79 0.61 0.3 70 100.0% -0.08 [-0.18, 0.02]
Subtotal (95% CI) 79 70 100.0% -0.08 [-0.18, 0.02]
2.31.2 at 12 months
Frihagen 2007 0.56 0.33 70 0.65 0.3 62 100.0% -0.09 [-0.20, 0.02]
Subtotal (95% CI) 70 62 100.0% -0.09 [-0.20, 0.02]
2.31.3 at 24 months
Frihagen 2007 0.61 0.31 52 0.72 0.23 52 100.0% -0.11 [-0.21, -0.01]
Subtotal (95% CI) 52 52 100.0% -0.11 [-0.21, -0.01]
-100 -50 0 50 100

Favours hemiarthroplasty Favours fixation
Figure G-82. Length of hospital stay: Internal fixation versus hemiarthroplasty

Frihagen 2007 8.2 7.35 111 10.2 11.95 109 29.8% -2.00 [-4.63, 0.63]
Keating 2006 10.7 7 118 10.8 7 111 62.5% -0.10 [-1.91, 1.71]
Parker 2002 20.8 32.6 226 20.5 27 229 6.8% 0.30 [-5.20, 5.80]
van Dortmont 2000 24 33 31 19.5 29 29 0.8% 4.50 [-11.20, 20.20]
Total (95% CI) 486 478 100.0% -0.60 [-2.04, 0.83]

-10 -5 0 5 10
Test for overall effect: Z = 0.82 (P = 0.41) Favours fixation Favours hemiarthtoplasty
APPENDIX G 487
19.5.5 Internal fixation versus total hip replacement
Figure G-83. Mortality: Internal fixation versus total hip replacement

Fixation Total hip replacement Risk Ratio Risk Ratio
3.1.1 At 2-4 months
Johansson 2002 7 78 3 68 47.7% 2.03 [0.55, 7.56]
Keating 2006 3 69 2 69 29.7% 1.50 [0.26, 8.70]
Neander 1997 2 10 1 10 14.9% 2.00 [0.21, 18.69]
Tidermark 2003 B 3 53 0 49 7.7% 6.48 [0.34, 122.37]
Subtotal (95% CI) 210 196 100.0% 2.21 [0.91, 5.40]
Total events 15 6
3.1.2 At 12-18 months

Johansson 2002 17 78 16 68 77.4% 0.93 [0.51, 1.69]
Keating 2006 6 69 4 69 18.1% 1.50 [0.44, 5.08]
Neander 1997 2 10 1 10 4.5% 2.00 [0.21, 18.69]
Subtotal (95% CI) 157 147 100.0% 1.08 [0.64, 1.82]
Total events 25 21
3.1.3 At 24 months
Johansson 2002 23 78 20 68 60.0% 1.00 [0.61, 1.66]
Jonsson 1996 2 24 3 23 8.6% 0.64 [0.12, 3.48]
Keating 2006 9 69 6 69 16.8% 1.50 [0.56, 3.99]
Tidermark 2003 B 10 53 5 49 14.6% 1.85 [0.68, 5.03]
Subtotal (95% CI) 224 209 100.0% 1.18 [0.79, 1.75]
Total events 44 34
0.01 0.1 1 10 100

Favours fixation Favours THR
488 APPENDIX G
Figure G-84. Reoperations – all – at final follow up of study: Internal fixation versus total
hip replacement
Fixation T HR Risk Ratio Risk Ratio
3.7.1 at 1 to 2 years
Johansson 2002 36 78 13 68 30.7% 2.41 [1.40, 4.16]
Jonsson 1996 7 24 2 23 4.5% 3.35 [0.78, 14.50]
Keating 2006 27 69 6 69 13.3% 4.50 [1.98, 10.21]
Neander 1997 1 10 1 10 2.2% 1.00 [0.07, 13.87]
Subtotal (95% CI) 181 170 50.7% 2.98 [1.95, 4.56]
Total events 71 22
3.7.2 at 4 years
Tidermark 2003 B 25 53 2 49 4.6% 11.56 [2.89, 46.25]
Subtotal (95% CI) 53 49 4.6% 11.56 [2.89, 46.25]
Total events 25 2
3.7.3 at 13 years
Skinner 1989 30 91 20 89 44.7% 1.47 [0.90, 2.38]
Subtotal (95% CI) 91 89 44.7% 1.47 [0.90, 2.38]
Total events 30 20
Total (95% CI) 325 308 100.0% 2.70 [1.99, 3.67]

Total events 126 44
0.01 0.1 1 10 100
Test for overall effect: Z = 6.34 (P < 0.00001) Favours fixation Favours THR
APPENDIX G 489
Figure G-85. Number of patients reporting pain at 1 year: Internal fixation versus total hip
replacement
Internal fixation THR Risk Ratio Risk Ratio
Jonsson 1996 9 17 5 18 14.3% 1.91 [0.80, 4.55]
Keating 2006 38 61 29 61 85.7% 1.31 [0.94, 1.82]
Total (95% CI) 78 79 100.0% 1.40 [1.02, 1.90]

Total events 47 34
0.1 0.2 0.5 1 2 5 10
Test for overall effect: Z = 2.12 (P = 0.03) Favours fixation Favours THR
Figure G-86. Length of hospital stay: Internal fixation versus total hip replacement
Internal fixation THR Mean Difference Mean Difference
Keating 2006 10.6 6 69 12.3 10 69 100.0% -1.70 [-4.45, 1.05]
Total (95% CI) 69 69 100.0% -1.70 [-4.45, 1.05]

-10 -5 0 5 10
Test for overall effect: Z = 1.21 (P = 0.23) Favours fixation Favours THR
490 APPENDIX G
19.5.6 Hemiarthroplasty versus total hip replacement
Figure G-87. Mortality: Hemiarthroplasty versus total hip replacement

Hemiarthroplasty Total hip replacement Risk Ratio Risk Ratio
7.1.1 at 3-6 months
Keating 2006 5 69 2 69 16.6% 2.50 [0.50, 12.45]
Macaulay 2008 5 23 1 17 9.6% 3.70 [0.47, 28.81]
Skinner 1989 15 100 8 80 73.8% 1.50 [0.67, 3.36]
Subtotal (95% CI) 192 166 100.0% 1.88 [0.96, 3.68]
Total events 25 11
7.1.2 at 1 year
Blomfeldt 2007 3 60 4 60 11.8% 0.75 [0.18, 3.21]
Keating 2006 6 69 4 69 11.8% 1.50 [0.44, 5.08]
Mouzopoulos 2008 6 43 6 43 17.6% 1.00 [0.35, 2.86]
Skinner 1989 27 100 18 80 58.8% 1.20 [0.71, 2.02]
Subtotal (95% CI) 272 252 100.0% 1.15 [0.76, 1.74]
Total events 42 32
7.1.3 at 2 to 4 years
Baker 2006 7 41 3 40 10.2% 2.28 [0.63, 8.19]
Keating 2006 9 69 6 69 20.1% 1.50 [0.56, 3.99]
Macaulay 2008 9 23 5 17 19.3% 1.33 [0.54, 3.26]
Mouzopoulos 2008 13 43 15 43 50.4% 0.87 [0.47, 1.60]
Subtotal (95% CI) 176 169 100.0% 1.23 [0.80, 1.87]
Total events 38 29
0.01 0.1 1 10 100

Favours hemiarthroplasty Favours THR
APPENDIX G 491
Figure G-88. Reoperations - all: Hemiarthroplasty versus total hip replacement

Hemiarthroplasty Total hip Risk Ratio Risk Ratio
Baker 2006 3 41 5 40 13.4% 0.59 [0.15, 2.29]
Blomfeldt 2007 0 60 2 60 6.6% 0.20 [0.01, 4.08]
Dorr 1986 5 37 9 39 23.2% 0.59 [0.22, 1.59]
Keating 2006 5 69 6 69 15.9% 0.83 [0.27, 2.60]
Mouzopoulos 2008 5 43 1 43 2.6% 5.00 [0.61, 41.04]
Skinner 1989 24 100 13 80 38.2% 1.48 [0.80, 2.71]
Total (95% CI) 350 331 100.0% 1.06 [0.70, 1.60]

Total events 42 36
0.01 0.1 1 10 100
Test for overall effect: Z = 0.26 (P = 0.79) Favours hemi. Favours THR
Figure G-89. Number of patients reporting pain at 1 year: Hemiarthroplasty versus total
hip replacement
Hemiarthroplasty Total hip Risk Ratio Risk Ratio
Keating 2006 30 60 29 61 54.5% 1.05 [0.73, 1.52]
Skinner 1989 20 73 0 62 45.5% 34.91 [2.15, 565.58]
Total (95% CI) 133 123 100.0% 5.18 [0.05, 515.13]

Total events 50 29
0.01 0.1 1 10 100
492 APPENDIX G
Figure G-90. Pain scores: Hemiarthroplasty versus total hip replacement

Hemiarthroplasty Total hip Mean Difference Mean Difference
7.15.6 Harris Hip Score for pain - 12 months
Blomfeldt 2007 39.1 6.26 55 43.1 6.26 56 100.0% -4.00 [-6.33, -1.67]
Subtotal (95% CI) 55 56 100.0% -4.00 [-6.33, -1.67]
-4 -2 0 2 4
Favours THR Favours hemi.
Figure G-91. Failure to regain mobility at end of study: Hemiarthroplasty versus total hip
replacement
Hemiarthroplasty Total hip (THR) Risk Ratio Risk Ratio
Dorr 1986 6 37 7 39 32.7% 0.90 [0.33, 2.44]
Skinner 1989 11 73 13 62 67.3% 0.72 [0.35, 1.49]
Total (95% CI) 110 101 100.0% 0.78 [0.43, 1.40]

Total events 17 20
0.1 0.2 0.5 1 2 5 10
Figure G-92. Functional scores (lower scores advantageous): Hemiarthroplasty versus

total hip replacement
Hemiarthroplaty Total hip Mean Difference Mean Difference
7.14.1 Oxford Hip Score - mean of 40 months
Baker 2006 22.3 6.7 33 18.8 6.7 36 100.0% 3.50 [0.34, 6.66]
Subtotal (95% CI) 33 36 100.0% 3.50 [0.34, 6.66]
Total (95% CI) 33 36 100.0% 3.50 [0.34, 6.66]

-10 -5 0 5 10
APPENDIX G 493
Figure G-93. Functional status (higher scores advantageous): Hemiarthroplasty versus

total hip replacement
7.19.1 Barthel score - one year
Mouzopoulos 2008 76.8 6.8 30 84.8 14.8 33 100.0% -8.00 [-13.61, -2.39]
Subtotal (95% CI) 30 33 100.0% -8.00 [-13.61, -2.39]
7.19.2 Barthel score - four years

Mouzopoulos 2008 79.6 6.3 20 85.3 11.6 23 100.0% -5.70 [-11.19, -0.21]
Subtotal (95% CI) 20 23 100.0% -5.70 [-11.19, -0.21]
7.19.3 Hip rating questionnaire - 24 months

Keating 2006 73.8 16 50 79.9 17 56 100.0% -6.10 [-12.38, 0.18]
Subtotal (95% CI) 50 56 100.0% -6.10 [-12.38, 0.18]
7.19.4 Harris Hip Score - total score - 12 months

Blomfeldt 2007 79.4 12.14 55 87.2 12.14 56 41.7% -7.80 [-12.32, -3.28]
Mouzopoulos 2008 77.8 9.6 30 81.6 4.9 33 58.3% -3.80 [-7.62, 0.02]
Subtotal (95% CI) 85 89 100.0% -5.47 [-8.39, -2.55]
7.19.7 Harris Hip Score - total score - four years

Mouzopoulos 2008 79.5 6.5 20 83.7 4.8 23 100.0% -4.20 [-7.66, -0.74]
Subtotal (95% CI) 20 23 100.0% -4.20 [-7.66, -0.74]
7.19.8 Harris Hip Score for function - 12 months

Blomfeldt 2007 31.6 9.23 55 35.3 9.23 56 100.0% -3.70 [-7.13, -0.27]
Subtotal (95% CI) 55 56 100.0% -3.70 [-7.13, -0.27]
7.19.9 Short form 36 physical score - mean of 40 months

Baker 2006 38.1 10.85 33 40.53 10.85 36 100.0% -2.43 [-7.56, 2.70]
Subtotal (95% CI) 33 36 100.0% -2.43 [-7.56, 2.70]
7.19.11 Self reported walking distance (kilometres) - mean of 40 months

Baker 2006 1.9 3.35 33 3.6 3.35 36 100.0% -1.70 [-3.28, -0.12]
Subtotal (95% CI) 33 36 100.0% -1.70 [-3.28, -0.12]
-4 -2 0 2 4
494 APPENDIX G
Figure G-94. Quality of life scores: Hemiarthroplasty versus total hip replacement
7.22.5 EuroQol (EQ-5d) questionnaire - 24 months
Keating 2006 0.53 0.36 65 0.69 0.32 66 100.0% -0.16 [-0.28, -0.04]
Subtotal (95% CI) 65 66 100.0% -0.16 [-0.28, -0.04]
-4 -2 0 2 4
Figure G-95. Length of hospital stay: Hemiarthroplasty versus total hip replacement
Keating 2006 11.5 8 69 12.3 10 69 100.0% -0.80 [-3.82, 2.22]
Total (95% CI) 69 69 100.0% -0.80 [-3.82, 2.22]

-10 -5 0 5 10
APPENDIX G 495
19.5.7 Trochanteric extracapsular fracture – all studies
Figure G-96. 30 days mortality: Intramedullary implants versus extramedullary implants

Intramedullary Extramedullary Risk Ratio Risk Ratio
1.1.1 All
Barton 2010 21 100 11 110 19.0% 2.10 [1.07, 4.13]
Bridle 1991 10 49 9 51 16.0% 1.16 [0.51, 2.60]
Guyer 1993A 4 50 2 50 3.6% 2.00 [0.38, 10.43]
Harrington 2002 4 50 2 52 3.6% 2.08 [0.40, 10.86]
Leung 1992 7 113 5 113 9.1% 1.40 [0.46, 4.28]
Little 2008 7 92 6 98 10.5% 1.24 [0.43, 3.56]
O'Brien 1995 6 52 1 49 1.9% 5.65 [0.71, 45.29]
Radford 1993 12 100 10 100 18.1% 1.20 [0.54, 2.65]
Utrilla 2005 7 106 10 106 18.1% 0.70 [0.28, 1.77]
Subtotal (95% CI) 712 729 100.0% 1.44 [1.04, 1.99]
Total events 78 56
1.1.2 Unstable
Barton 2010 21 100 11 110 84.2% 2.10 [1.07, 4.13]
Harrington 2002 4 50 2 52 15.8% 2.08 [0.40, 10.86]
Subtotal (95% CI) 150 162 100.0% 2.10 [1.12, 3.93]
Total events 25 13
0.05 0.2 1 5 20
Favours intramedullary Favours extramedullary
Figure G-97. 3 months mortality: Intramedullary implants versus extramedullary implants

Guyer 1993A 4 50 5 50 23.8% 0.80 [0.23, 2.81]
Hardy 1998 12 50 13 50 61.9% 0.92 [0.47, 1.82]
Ovesen 2006 3 73 3 73 14.3% 1.00 [0.21, 4.79]
Total (95% CI) 173 173 100.0% 0.90 [0.52, 1.59]

Total events 19 21
0.01 0.1 1 10 100
Test for overall effect: Z = 0.35 (P = 0.73) Favours intramedullary Favours extramedullary
496 APPENDIX G
Figure G-98. 12 months mortality: Intramedullary implants versus extramedullary

implants
1.3.1 All
Ahrengart 2002 41 210 37 216 21.1% 1.14 [0.76, 1.70]
Barton 2010 32 100 24 110 13.2% 1.47 [0.93, 2.31]
Bridle 1991 15 49 19 51 10.8% 0.82 [0.47, 1.43]
Ekstrom 2007 14 86 15 85 8.7% 0.92 [0.47, 1.79]
Hardy 1998 15 50 15 50 8.7% 1.00 [0.55, 1.82]
Leung 1992 13 113 15 113 8.7% 0.87 [0.43, 1.74]
Little 2008 16 92 17 98 9.5% 1.00 [0.54, 1.86]
Ovesen 2006 3 73 3 73 1.7% 1.00 [0.21, 4.79]
Sadowski 2002 2 20 1 19 0.6% 1.90 [0.19, 19.27]
Saudan 2002 16 106 13 100 7.7% 1.16 [0.59, 2.29]
Utrilla 2005 19 106 16 106 9.3% 1.19 [0.65, 2.18]
Subtotal (95% CI) 1005 1021 100.0% 1.09 [0.91, 1.31]
1.3.3 Unstable
Barton 2010 32 100 24 110 58.7% 1.47 [0.93, 2.31]
Ekstrom 2007 14 86 15 85 38.7% 0.92 [0.47, 1.79]
Sadowski 2002 2 20 1 19 2.6% 1.90 [0.19, 19.27]
Subtotal (95% CI) 206 214 100.0% 1.27 [0.88, 1.83]
Total events 48 40
0.05 0.2 1 5 20
APPENDIX G 497
Figure G-99. Reoperation – within the follow up period of the study: Intramedullary
implants versus extramedullary implants
1.4.1 All
Aune 1994 12 160 2 187 7.3% 7.01 [1.59, 30.87]
Barton 2010 3 100 2 110 5.6% 1.65 [0.28, 9.67]
Ekstrom 2007 6 86 1 85 4.2% 5.93 [0.73, 48.22]
Guyer 1993A 5 50 6 50 10.6% 0.83 [0.27, 2.55]
Hardy 1998 3 50 4 50 7.6% 0.75 [0.18, 3.18]
Hoffman 1996 1 31 1 36 2.7% 1.16 [0.08, 17.80]
Leung 1992 4 93 2 93 6.1% 2.00 [0.38, 10.65]
Little 2008 0 92 1 98 2.0% 0.35 [0.01, 8.60]
Miedel 2005 3 93 6 96 8.3% 0.52 [0.13, 2.00]
O'Brien 1995 5 53 2 49 6.5% 2.31 [0.47, 11.37]
Ovesen 2006 12 73 6 73 13.1% 2.00 [0.79, 5.04]
Pajarinen 2005 2 54 2 54 4.9% 1.00 [0.15, 6.84]
Radford 1993 6 100 3 100 8.2% 2.00 [0.51, 7.78]
Sadowski 2002 0 20 6 19 2.5% 0.07 [0.00, 1.22]
Saudan 2002 6 100 2 106 6.6% 3.18 [0.66, 15.39]
Utrilla 2005 1 106 4 106 4.0% 0.25 [0.03, 2.20]
Subtotal (95% CI) 1261 1312 100.0% 1.39 [0.87, 2.23]
Total events 69 50
1.4.2 Stable
Aune 1994 7 84 1 89 100.0% 7.42 [0.93, 59.01]
Subtotal (95% CI) 84 89 100.0% 7.42 [0.93, 59.01]
Total events 7 1
1.4.3 Unstable
Aune 1994 5 76 1 98 19.0% 6.45 [0.77, 54.04]
Barton 2010 3 100 2 110 21.7% 1.65 [0.28, 9.67]
Ekstrom 2007 9 86 1 85 19.6% 8.90 [1.15, 68.69]
Miedel 2005 3 93 6 96 24.9% 0.52 [0.13, 2.00]
Sadowski 2002 0 20 6 19 14.7% 0.07 [0.00, 1.22]
Subtotal (95% CI) 375 408 100.0% 1.41 [0.32, 6.14]
Total events 20 16
0.005 0.1 1 10 200

498 APPENDIX G
Figure G-100. Operative or postoperative fracture of femur - within the follow up period
of the study: Intramedullary implants versus extramedullary implants

1.5.1 All
Ahrengart 2002 5 210 2 216 14.8% 2.57 [0.50, 13.11]
Aune 1994 9 160 0 187 3.5% 22.19 [1.30, 378.23]
Bridle 1991 4 49 0 51 3.7% 9.36 [0.52, 169.40]
Ekstrom 2007 1 86 0 85 3.8% 2.97 [0.12, 71.79]
Guyer 1993A 1 50 0 50 3.8% 3.00 [0.13, 71.92]
Hardy 1998 3 50 0 50 3.8% 7.00 [0.37, 132.10]
Harrington 2002 1 50 0 52 3.7% 3.12 [0.13, 74.78]
Hoffman 1996 3 31 0 36 3.5% 8.09 [0.43, 150.85]
Leung 1992 2 93 0 93 3.8% 5.00 [0.24, 102.75]
Miedel 2005 3 93 0 96 3.7% 7.22 [0.38, 137.95]
O'Brien 1995 2 53 0 49 3.9% 4.63 [0.23, 94.10]
Ovesen 2006 2 73 0 73 3.8% 5.00 [0.24, 102.38]
Park 1998 3 30 0 30 3.8% 7.00 [0.38, 129.93]
Radford 1993 11 100 1 100 7.5% 11.00 [1.45, 83.61]
Saudan 2002 0 100 0 106 Not estimable
Utrilla 2005 4 106 2 106 15.0% 2.00 [0.37, 10.69]
Zou 2009 0 58 0 63 Not estimable
Subtotal (95% CI) 1392 1443 81.9% 5.61 [2.98, 10.59]
Total events 54 5
1.5.2 Stable
Aune 1994 5 84 0 89 3.7% 11.65 [0.65, 207.45]
Subtotal (95% CI) 84 89 3.7% 11.65 [0.65, 207.45]
Total events 5 0
1.5.3 Unstable
Aune 1994 4 76 0 98 3.3% 11.57 [0.63, 211.68]
Ekstrom 2007 1 86 0 85 3.8% 2.97 [0.12, 71.79]
Harrington 2002 1 50 0 52 3.7% 3.12 [0.13, 74.78]
Miedel 2005 3 93 0 96 3.7% 7.22 [0.38, 137.95]
Subtotal (95% CI) 305 331 14.5% 6.05 [1.38, 26.63]
Total events 9 0
Total (95% CI) 1781 1863 100.0% 5.90 [3.33, 10.44]

Total events 68 5
0.01 0.1 1 10 100
Test for overall effect: Z = 6.09 (P < 0.00001) Favours intramedullary Favours extramedullary
APPENDIX G 499
Figure G-101. Cut-out (at latest follow up): Intramedullary implants versus extramedullary
implants
1.6.1 All
Aune 1994 3 160 2 187 4.3% 1.75 [0.30, 10.36]
Barton 2010 3 100 2 110 4.4% 1.65 [0.28, 9.67]
Bridle 1991 2 49 3 51 6.8% 0.69 [0.12, 3.98]
Ekstrom 2007 5 86 1 85 2.3% 4.94 [0.59, 41.42]
Guyer 1993A 1 50 3 50 7.0% 0.33 [0.04, 3.10]
Hardy 1998 0 50 1 50 3.5% 0.33 [0.01, 7.99]
Harrington 2002 1 50 1 52 2.3% 1.04 [0.07, 16.18]
Hoffman 1996 1 31 1 36 2.1% 1.16 [0.08, 17.80]
Leung 1992 2 93 3 93 7.0% 0.67 [0.11, 3.90]
Little 2008 0 92 2 98 5.6% 0.21 [0.01, 4.38]
Miedel 2005 3 93 4 96 9.1% 0.77 [0.18, 3.37]
O'Brien 1995 3 53 1 49 2.4% 2.77 [0.30, 25.78]
Ovesen 2006 7 73 5 73 11.6% 1.40 [0.47, 4.21]
Pajarinen 2005 1 54 1 54 2.3% 1.00 [0.06, 15.58]
Park 1998 1 30 1 30 2.3% 1.00 [0.07, 15.26]
Radford 1993 2 100 3 100 7.0% 0.67 [0.11, 3.90]
Sadowski 2002 0 20 5 19 13.1% 0.09 [0.01, 1.47]
Saudan 2002 3 100 1 106 2.3% 3.18 [0.34, 30.07]
Utrilla 2005 1 104 2 106 4.6% 0.51 [0.05, 5.53]
Subtotal (95% CI) 1446 1508 100.0% 0.95 [0.63, 1.45]
Total events 39 42
1.6.2 Stable
Aune 1994 2 84 1 76 100.0% 1.81 [0.17, 19.56]
Subtotal (95% CI) 84 76 100.0% 1.81 [0.17, 19.56]
Total events 2 1
1.6.3 Unstable
Aune 1994 1 89 1 98 9.6% 1.10 [0.07, 17.34]
Barton 2010 3 100 2 110 19.2% 1.65 [0.28, 9.67]
Ekstrom 2007 6 105 2 98 20.8% 2.80 [0.58, 13.55]
Harrington 2002 1 50 1 52 9.9% 1.04 [0.07, 16.18]
Miedel 2005 3 109 4 108 40.5% 0.74 [0.17, 3.24]
Subtotal (95% CI) 453 466 100.0% 1.41 [0.63, 3.15]
Total events 14 10
0.01 0.1 1 10 100

500 APPENDIX G
Figure G-102. Infection (deep infection or requires reoperation – at latest follow up):
Intramedullary implants versus extramedullary implants
1.7.1 All
Guyer 1993A 0 50 1 50 9.7% 0.33 [0.01, 7.99]
Hardy 1998 0 50 0 50 Not estimable
Hoffman 1996 0 31 0 36 Not estimable
Leung 1992 1 93 3 93 19.4% 0.33 [0.04, 3.15]
Little 2008 0 92 0 98 Not estimable
Miedel 2005 0 93 1 96 9.6% 0.34 [0.01, 8.34]
O'Brien 1995 0 53 0 49 Not estimable
Ovesen 2006 2 73 1 73 6.5% 2.00 [0.19, 21.58]
Pajarinen 2005 0 54 0 54 Not estimable
Park 1998 1 30 1 30 6.5% 1.00 [0.07, 15.26]
Radford 1993 1 100 0 100 3.2% 3.00 [0.12, 72.77]
Sadowski 2002 0 20 1 19 9.9% 0.32 [0.01, 7.35]
Saudan 2002 3 79 1 89 6.1% 3.38 [0.36, 31.84]
Utrilla 2005 0 104 1 106 9.6% 0.34 [0.01, 8.24]
Subtotal (95% CI) 922 943 80.5% 0.86 [0.38, 1.93]
Total events 8 10
1.7.2 Unstable
Miedel 2005 0 93 1 96 9.6% 0.34 [0.01, 8.34]
Sadowski 2002 0 20 1 19 9.9% 0.32 [0.01, 7.35]
Subtotal (95% CI) 113 115 19.5% 0.33 [0.04, 3.10]
Total events 0 2
0.01 0.1 1 10 100

APPENDIX G 501
Figure G-103. Non-union (at latest follow-up): Intramedullary implants versus

extramedullary implants
1.8.1 All
Ekstrom 2007 0 86 0 85 Not estimable
Harrington 2002 1 50 0 52 9.9% 3.12 [0.13, 74.78]
Leung 1992 1 93 0 93 10.1% 3.00 [0.12, 72.71]
Ovesen 2006 0 73 0 73 Not estimable
Park 1998 0 30 1 30 30.3% 0.33 [0.01, 7.87]
Radford 1993 0 100 0 100 Not estimable
Sadowski 2002 1 20 1 19 20.7% 0.95 [0.06, 14.13]
Zou 2009 0 58 1 63 29.0% 0.36 [0.02, 8.70]
Subtotal (95% CI) 610 621 100.0% 1.01 [0.30, 3.46]
Total events 3 3
1.8.2 Stable
Leung 1992 1 93 0 93 100.0% 3.00 [0.12, 72.71]
Subtotal (95% CI) 93 93 100.0% 3.00 [0.12, 72.71]
Total events 1 0
1.8.3 Unstable
Harrington 2002 1 50 0 52 25.4% 3.12 [0.13, 74.78]
Leung 1992 0 98 0 98 Not estimable
Sadowski 2002 1 18 1 17 53.2% 0.94 [0.06, 13.93]
Zou 2009 1 11 0 16 21.4% 4.25 [0.19, 95.68]
Subtotal (95% CI) 263 268 100.0% 2.20 [0.43, 11.24]
Total events 3 1
0.005 0.1 1 10 200

Figure G-104. Pain – patient reported outcomes: Intramedullary implants versus

Hardy 1998 9 50 4 50 4.5% 2.25 [0.74, 6.83]
Hoffman 1996 9 31 9 36 9.4% 1.16 [0.53, 2.56]
Leung 1992 22 93 32 93 36.0% 0.69 [0.43, 1.09]
Utrilla 2005 50 104 45 106 50.1% 1.13 [0.84, 1.53]
Total (95% CI) 278 285 100.0% 1.03 [0.81, 1.30]

Total events 90 90
0.01 0.1 1 10 100
502 APPENDIX G
Figure G-105. Length of stay in hospital (in days): Intramedullary implants versus
Intramedullary Extramedullary Mean Difference Mean Difference
Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
1.11.1 All
Harrington 2002 16.5 8.8 50 16.3 7.5 52 7.2% 0.20 [-2.98, 3.38]
Hoffman 1996 29.8 20.1 31 28.5 18.9 36 1.2% 1.30 [-8.09, 10.69]
Leung 1992 26.9 8.2 93 28.3 4.5 93 12.5% -1.40 [-3.30, 0.50]
O'Brien 1995 23.7 19 53 27.6 26.8 49 1.3% -3.90 [-12.98, 5.18]
Ovesen 2006 16.4 8.4 73 14.4 9.4 73 8.2% 2.00 [-0.89, 4.89]
Pajarinen 2005 6.1 3.3 54 5.4 3 54 16.6% 0.70 [-0.49, 1.89]
Sadowski 2002 13 4 20 18 7 19 6.1% -5.00 [-8.60, -1.40]
Saudan 2002 13 4 100 14 10 106 11.7% -1.00 [-3.06, 1.06]
Subtotal (95% CI) 474 482 64.7% -0.54 [-1.93, 0.84]
1.11.2 Stable
Leung 1992 9.2 6.43 30 10.7 6.27 20 6.2% -1.50 [-5.08, 2.08]
Subtotal (95% CI) 30 20 6.2% -1.50 [-5.08, 2.08]
1.11.3 Unstable
Harrington 2002 16.5 8.8 50 16.3 7.5 52 7.2% 0.20 [-2.98, 3.38]
Leung 1992 9.5 3.38 63 9.6 4.46 73 15.8% -0.10 [-1.42, 1.22]
Sadowski 2002 13 4 20 18 7 19 6.1% -5.00 [-8.60, -1.40]
Subtotal (95% CI) 133 144 29.1% -1.31 [-4.07, 1.44]
-10 -5 0 5 10
Figure G-106. Mean mobility score (Parker Palmer score): Intramedullary implants versus
Hardy 1998 1.9 1 50 1.6 1.2 50 61.4% 0.30 [-0.13, 0.73]
Sadowski 2002 5 2.6 20 6 3.5 19 3.1% -1.00 [-2.94, 0.94]
Saudan 2002 4.94 3.33 100 5.07 2.97 106 15.4% -0.13 [-0.99, 0.73]
Utrilla 2005 6.4 2.8 104 6.2 2.8 106 20.1% 0.20 [-0.56, 0.96]
Total (95% CI) 274 281 100.0% 0.17 [-0.17, 0.51]

-2 -1 0 1 2
Test for overall effect: Z = 1.00 (P = 0.32) Favours extramedullary Favours intramedullary
APPENDIX G 503
19.5.8 Trochanteric extracapsular fracture – studies from 2000
Figure G-107. 30 days mortality: Intramedullary implants versus extramedullary implants

17.1.1 All
Barton 2010 21 100 11 110 37.1% 2.10 [1.07, 4.13]
Harrington 2002 4 50 2 52 6.9% 2.08 [0.40, 10.86]
Little 2008 7 92 6 98 20.6% 1.24 [0.43, 3.56]
Utrilla 2005 7 106 10 106 35.4% 0.70 [0.28, 1.77]
Subtotal (95% CI) 348 366 100.0% 1.43 [0.91, 2.25]
Total events 39 29
17.1.2 Unstable
Barton 2010 21 100 11 110 84.2% 2.10 [1.07, 4.13]
Harrington 2002 4 50 2 52 15.8% 2.08 [0.40, 10.86]
Subtotal (95% CI) 150 162 100.0% 2.10 [1.12, 3.93]
Total events 25 13
0.05 0.2 1 5 20

implants
Ovesen 2006 3 73 3 73 100.0% 1.00 [0.21, 4.79]
Total (95% CI) 73 73 100.0% 1.00 [0.21, 4.79]

Total events 3 3
0.01 0.1 1 10 100

implants
504 APPENDIX G

17.3.1 All
Ahrengart 2002 41 210 37 216 29.3% 1.14 [0.76, 1.70]
Barton 2010 32 100 24 110 18.4% 1.47 [0.93, 2.31]
Ekstrom 2007 14 86 15 85 12.1% 0.92 [0.47, 1.79]
Little 2008 16 92 17 98 13.2% 1.00 [0.54, 1.86]
Ovesen 2006 3 73 3 73 2.4% 1.00 [0.21, 4.79]
Sadowski 2002 2 20 1 19 0.8% 1.90 [0.19, 19.27]
Saudan 2002 16 106 13 100 10.8% 1.16 [0.59, 2.29]
Utrilla 2005 19 106 16 106 12.9% 1.19 [0.65, 2.18]
Subtotal (95% CI) 793 807 100.0% 1.17 [0.94, 1.45]
17.3.3 Unstable
Barton 2010 32 100 24 110 60.2% 1.47 [0.93, 2.31]
Ekstrom 2007 14 86 15 85 39.8% 0.92 [0.47, 1.79]
Subtotal (95% CI) 186 195 100.0% 1.25 [0.86, 1.82]
Total events 46 39
0.05 0.2 1 5 20
Figure G-110. Reoperation – within the follow up period of the study: Intramedullary
implants versus extramedullary implants
17.4.1 All
Barton 2010 3 100 2 110 11.5% 1.65 [0.28, 9.67]
Ekstrom 2007 6 86 1 85 9.1% 5.93 [0.73, 48.22]
Little 2008 0 92 1 98 4.7% 0.35 [0.01, 8.60]
Miedel 2005 3 93 6 96 15.5% 0.52 [0.13, 2.00]
Ovesen 2006 12 73 6 73 21.2% 2.00 [0.79, 5.04]
Pajarinen 2005 2 54 2 54 10.3% 1.00 [0.15, 6.84]
Sadowski 2002 0 20 6 19 5.9% 0.07 [0.00, 1.22]
Saudan 2002 6 100 2 106 13.2% 3.18 [0.66, 15.39]
Utrilla 2005 1 106 4 106 8.7% 0.25 [0.03, 2.20]
Subtotal (95% CI) 724 747 100.0% 1.10 [0.52, 2.34]
Total events 33 30
17.4.3 Unstable
Barton 2010 3 100 2 110 26.8% 1.65 [0.28, 9.67]
Ekstrom 2007 9 86 1 85 24.3% 8.90 [1.15, 68.69]
Miedel 2005 3 93 6 96 30.7% 0.52 [0.13, 2.00]
Sadowski 2002 0 20 6 19 18.2% 0.07 [0.00, 1.22]
Subtotal (95% CI) 299 310 100.0% 0.99 [0.19, 5.13]
Total events 15 15
0.005 0.1 1 10 200

Figure G-111. Operative or postoperative fracture of femur - within the follow up period
of the study: Intramedullary implants versus extramedullary implants
APPENDIX G 505

17.5.1 All
Ahrengart 2002 5 210 2 216 26.5% 2.57 [0.50, 13.11]
Ekstrom 2007 1 86 0 85 6.8% 2.97 [0.12, 71.79]
Harrington 2002 1 50 0 52 6.6% 3.12 [0.13, 74.78]
Miedel 2005 3 93 0 96 6.6% 7.22 [0.38, 137.95]
Ovesen 2006 2 73 0 73 6.7% 5.00 [0.24, 102.38]
Utrilla 2005 4 106 2 106 26.9% 2.00 [0.37, 10.69]
Subtotal (95% CI) 776 797 80.0% 3.05 [1.22, 7.62]
Total events 16 4
17.5.3 Unstable
Ekstrom 2007 1 86 0 85 6.8% 2.97 [0.12, 71.79]
Harrington 2002 1 50 0 52 6.6% 3.12 [0.13, 74.78]
Miedel 2005 3 93 0 96 6.6% 7.22 [0.38, 137.95]
Subtotal (95% CI) 229 233 20.0% 4.43 [0.76, 25.84]
Total events 5 0
Total (95% CI) 1005 1030 100.0% 3.32 [1.48, 7.48]

Total events 21 4
0.01 0.1 1 10 100
506 APPENDIX G
implants
17.6.1 All
Barton 2010 3 100 2 110 7.7% 1.65 [0.28, 9.67]
Ekstrom 2007 5 86 1 85 4.1% 4.94 [0.59, 41.42]
Harrington 2002 1 50 1 52 3.9% 1.04 [0.07, 16.18]
Little 2008 0 92 2 98 9.8% 0.21 [0.01, 4.38]
Miedel 2005 3 93 4 96 15.9% 0.77 [0.18, 3.37]
Ovesen 2006 7 73 5 73 20.1% 1.40 [0.47, 4.21]
Pajarinen 2005 1 54 1 54 4.0% 1.00 [0.06, 15.58]
Sadowski 2002 0 20 5 19 22.7% 0.09 [0.01, 1.47]
Saudan 2002 3 100 1 106 3.9% 3.18 [0.34, 30.07]
Utrilla 2005 1 104 2 106 8.0% 0.51 [0.05, 5.53]
Subtotal (95% CI) 830 862 100.0% 1.02 [0.59, 1.75]
Total events 24 24
17.6.3 Unstable
Barton 2010 3 100 2 110 21.2% 1.65 [0.28, 9.67]
Ekstrom 2007 6 105 2 98 23.1% 2.80 [0.58, 13.55]
Harrington 2002 1 50 1 52 10.9% 1.04 [0.07, 16.18]
Miedel 2005 3 109 4 108 44.8% 0.74 [0.17, 3.24]
Subtotal (95% CI) 364 368 100.0% 1.44 [0.62, 3.34]
Total events 13 9
0.01 0.1 1 10 100


17.7.1 All
Little 2008 0 92 0 98 Not estimable
Miedel 2005 0 93 1 96 15.6% 0.34 [0.01, 8.34]
Ovesen 2006 2 73 1 73 10.6% 2.00 [0.19, 21.58]
Pajarinen 2005 0 54 0 54 Not estimable
Sadowski 2002 0 20 1 19 16.3% 0.32 [0.01, 7.35]
Saudan 2002 3 79 1 89 9.9% 3.38 [0.36, 31.84]
Utrilla 2005 0 104 1 106 15.7% 0.34 [0.01, 8.24]
Subtotal (95% CI) 515 535 68.1% 1.04 [0.36, 3.02]
Total events 5 5
17.7.2 Unstable
Miedel 2005 0 93 1 96 15.6% 0.34 [0.01, 8.34]
Sadowski 2002 0 20 1 19 16.3% 0.32 [0.01, 7.35]
Subtotal (95% CI) 113 115 31.9% 0.33 [0.04, 3.10]
Total events 0 2
Total (95% CI) 628 650 100.0% 0.81 [0.32, 2.08]

Total events 5 7
0.01 0.1 1 10 100
APPENDIX G 507
Figure G-114. Non-union (at latest follow-up): Intramedullary implants versus

17.8.1 All
Harrington 2002 1 50 0 52 16.6% 3.12 [0.13, 74.78]
Ovesen 2006 0 73 0 73 Not estimable
Sadowski 2002 1 20 1 19 34.7% 0.95 [0.06, 14.13]
Zou 2009 0 58 1 63 48.7% 0.36 [0.02, 8.70]
Subtotal (95% CI) 387 398 100.0% 1.02 [0.21, 4.99]
Total events 2 2
17.8.3 Unstable
Harrington 2002 1 50 0 52 25.4% 3.12 [0.13, 74.78]
Sadowski 2002 1 18 1 17 53.2% 0.94 [0.06, 13.93]
Zou 2009 1 11 0 16 21.4% 4.25 [0.19, 95.68]
Subtotal (95% CI) 165 170 100.0% 2.20 [0.43, 11.24]
Total events 3 1
0.005 0.1 1 10 200

Figure G-115. Pain – patient reported outcomes: Intramedullary implants versus

Utrilla 2005 50 104 45 106 100.0% 1.13 [0.84, 1.53]
Total (95% CI) 104 106 100.0% 1.13 [0.84, 1.53]

Total events 50 45
0.01 0.1 1 10 100
Figure G-116. Length of stay in hospital (in days): Intramedullary implants versus
Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
17.11.1 All
Harrington 2002 16.5 8.8 50 16.3 7.5 52 12.8% 0.20 [-2.98, 3.38]
Ovesen 2006 16.4 8.4 73 14.4 9.4 73 13.9% 2.00 [-0.89, 4.89]
Pajarinen 2005 6.1 3.3 54 5.4 3 54 20.6% 0.70 [-0.49, 1.89]
Sadowski 2002 13 4 20 18 7 19 11.4% -5.00 [-8.60, -1.40]
Saudan 2002 13 4 100 14 10 106 17.2% -1.00 [-3.06, 1.06]
Subtotal (95% CI) 297 304 75.8% -0.35 [-2.14, 1.44]
17.11.3 Unstable
Harrington 2002 16.5 8.8 50 16.3 7.5 52 12.8% 0.20 [-2.98, 3.38]
Sadowski 2002 13 4 20 18 7 19 11.4% -5.00 [-8.60, -1.40]
Subtotal (95% CI) 70 71 24.2% -2.33 [-7.42, 2.77]
Total (95% CI) 367 375 100.0% -0.84 [-2.55, 0.88]

-10 -5 0 5 10
508 APPENDIX G
Figure G-117. Mean mobility score (Parker Palmer score): Intramedullary implants versus
Sadowski 2002 5 2.6 20 6 3.5 19 7.9% -1.00 [-2.94, 0.94]
Saudan 2002 4.94 3.33 100 5.07 2.97 106 40.0% -0.13 [-0.99, 0.73]
Utrilla 2005 6.4 2.8 104 6.2 2.8 106 52.0% 0.20 [-0.56, 0.96]
Total (95% CI) 224 231 100.0% -0.03 [-0.57, 0.52]

-2 -1 0 1 2
Test for overall effect: Z = 0.10 (P = 0.92) Favours extramedullary Favours intramedullary
19.5.9 Subtrochanteric extracapsular fracture.
Figure G-118. Mortality at 12 months: Intramedullary implants versus extramedullary

implants
2.1.3 12 months
Ekstrom 2007 1 19 3 13 45.4% 0.23 [0.03, 1.96]
Rahme 2007 6 29 2 29 54.6% 3.00 [0.66, 13.65]
Subtotal (95% CI) 48 42 100.0% 0.93 [0.08, 11.52]
Total events 7 5
Total (95% CI) 48 42 100.0% 0.93 [0.08, 11.52]

Total events 7 5
0.01 0.1 1 10 100
Figure G-119. Reoperation within follow up period of the study: Intramedullary implants
versus extramedullary implants
Aune 1994 1 14 0 17 23.6% 3.60 [0.16, 82.05]
Ekstrom 2007 3 19 0 13 25.3% 4.90 [0.27, 87.59]
Miedel 2005 0 16 3 12 25.3% 0.11 [0.01, 1.93]
Rahme 2007 0 29 8 29 25.8% 0.06 [0.00, 0.97]
Total (95% CI) 78 71 100.0% 0.56 [0.06, 5.47]

Total events 4 11
0.01 0.1 1 10 100
Miedel 2005 0 16 1 12 63.0% 0.25 [0.01, 5.76]
Rahme 2007 3 29 1 29 37.0% 3.00 [0.33, 27.18]
Total (95% CI) 45 41 100.0% 1.27 [0.28, 5.88]

Total events 3 2
0.01 0.1 1 10 100
APPENDIX G 509
implants
Ekstrom 2007 1 19 1 13 100.0% 0.68 [0.05, 9.98]
Total (95% CI) 19 13 100.0% 0.68 [0.05, 9.98]

Total events 1 1
0.01 0.1 1 10 100
Figure G-122. Non-union (at latest follow up): Intramedullary implants versus
Ekstrom 2007 0 19 1 13 29.3% 0.23 [0.01, 5.32]
Rahme 2007 1 29 8 29 70.7% 0.13 [0.02, 0.94]
Total (95% CI) 48 42 100.0% 0.15 [0.03, 0.82]

Total events 1 9
0.01 0.1 1 10 100
510 APPENDIX G
19.6 Mobilisation strategies
19.6.1 Timing of mobilisation
Figure G-123. Independent to transfer at day 7: Early versus delayed mobilisation

Oldmeadow 2006 16 29 4 31 100.0% 4.28 [1.62, 11.30]
Total (95% CI) 29 31 100.0% 4.28 [1.62, 11.30]

Total events 16 4
0.01 0.1 1 10 100
Test for overall effect: Z = 2.93 (P = 0.003) Favours late ambulation Favours early ambulation
Figure G-124. Independent to step at day 7: Early versus delayed mobilisation

Oldmeadow 2006 10 29 23 31 100.0% 0.46 [0.27, 0.80]
Total (95% CI) 29 31 100.0% 0.46 [0.27, 0.80]

Total events 10 23
0.01 0.1 1 10 100
Test for overall effect: Z = 2.77 (P = 0.006) Favours late ambulation Favours early ambulation
Figure G-125. Discharge to home or rehabilitation programme: Early versus delayed

mobilisation

2.3.1 Home
Oldmeadow 2006 5 29 1 31 100.0% 5.34 [0.66, 43.06]
Subtotal (95% CI) 29 31 100.0% 5.34 [0.66, 43.06]
Total events 5 1
2.3.2 Fast stream rehab

Oldmeadow 2006 8 29 14 31 100.0% 0.61 [0.30, 1.24]
Subtotal (95% CI) 29 31 100.0% 0.61 [0.30, 1.24]
Total events 8 14
2.3.3 Slow stream rehab

Oldmeadow 2006 14 29 16 31 100.0% 0.94 [0.56, 1.55]
Subtotal (95% CI) 29 31 100.0% 0.94 [0.56, 1.55]
Total events 14 16
0.01 0.1 1 10 100

Favours late ambulation Favours early ambulation
APPENDIX G 511
Figure G-126. Discharge to nursing home or died: Early versus delayed mobilisation

2.4.4 Nursing home
Oldmeadow 2006 1 29 0 31 100.0% 3.20 [0.14, 75.55]
Subtotal (95% CI) 29 31 100.0% 3.20 [0.14, 75.55]
Total events 1 0
2.4.5 Death
Oldmeadow 2006 1 29 0 31 100.0% 3.20 [0.14, 75.55]
Subtotal (95% CI) 29 31 100.0% 3.20 [0.14, 75.55]
Total events 1 0
0.01 0.1 1 10 100

Favours early ambulation Favours late ambulation
512 APPENDIX G
19.7 Intensive exercise or physiotherapy vs. usual care
19.7.1 Intensive physiotherapy (Strength training)
Figure G-127. Strength measures: intensive physiotherapy versus usual care

Intensive Control Mean Difference Mean Difference
3.3.1 Leg-press fractured side (kg)
Hauer 2002 71 35 12 50 21 12 100.0% 21.00 [-2.09, 44.09]
Subtotal (95% CI) 12 12 100.0% 21.00 [-2.09, 44.09]
3.3.2 Leg extensor fractured side (Newtons)

Hauer 2002 68 13 12 51 22 12 100.0% 17.00 [2.54, 31.46]
Subtotal (95% CI) 12 12 100.0% 17.00 [2.54, 31.46]
3.3.4 Ankle plantar flexion fractured side (Newtons)

Hauer 2002 88 30 12 65 33 12 100.0% 23.00 [-2.23, 48.23]
Subtotal (95% CI) 12 12 100.0% 23.00 [-2.23, 48.23]
-50 -25 0 25 50
Favours control Favours intensive
Figure G-128. Tinetti's POMA (Performance Orientated Mobility Assessment): intensive

physiotherapy versus usual care

3.7.1 Overall POMA (0 to 30. higher = better)
Hauer 2002 23.5 4.5 12 20.5 4 12 100.0% 3.00 [-0.41, 6.41]
Subtotal (95% CI) 12 12 100.0% 3.00 [-0.41, 6.41]
3.7.2 POMA part 1 (balance: 0 to 15)

Hauer 2002 12.7 2.2 12 11.4 2.4 12 100.0% 1.30 [-0.54, 3.14]
Subtotal (95% CI) 12 12 100.0% 1.30 [-0.54, 3.14]
3.7.3 POMA part 2 (gait: 0 to 15)

Hauer 2002 10.8 2.5 12 9.1 2.1 12 100.0% 1.70 [-0.15, 3.55]
Subtotal (95% CI) 12 12 100.0% 1.70 [-0.15, 3.55]
-10 -5 0 5 10
APPENDIX G 513
Figure G-129. Functional performance measures: intensive physiotherapy versus usual

care

3.10.1 Barthel's ADL (activities of daily living) (0 to 100: fully independent)
Hauer 2002 93 8.2 12 96.1 8.2 12 100.0% -3.10 [-9.66, 3.46]
Subtotal (95% CI) 12 12 100.0% -3.10 [-9.66, 3.46]
3.10.2 Lawton's IADL (instrumental activities of daily living) (0 to 8: fully competent)

Hauer 2002 7.3 1.4 12 6.9 1.3 12 100.0% 0.40 [-0.68, 1.48]
Subtotal (95% CI) 12 12 100.0% 0.40 [-0.68, 1.48]
-10 -5 0 5 10
Figure G-130. Functional performance tests: intensive physiotherapy versus usual care
3.8.1 Timed up-and-go (seconds)
Hauer 2002 26.1 17.8 12 26.9 9.8 12 100.0% -0.80 [-12.30, 10.70]
Subtotal (95% CI) 12 12 100.0% -0.80 [-12.30, 10.70]
3.8.2 Chair rise (seconds)

Hauer 2002 16.9 5.7 12 18.7 6.3 12 100.0% -1.80 [-6.61, 3.01]
Subtotal (95% CI) 12 12 100.0% -1.80 [-6.61, 3.01]
-100 -50 0 50 100

Favours intensive Favours control
Figure G-131. Walking speed: intensive physiotherapy versus usual care
Experimental Control Mean Difference Mean Difference

3.6.3 3 months
Hauer 2002 0.72 0.28 12 0.49 0.15 12 100.0% 0.23 [0.05, 0.41]
Subtotal (95% CI) 12 12 100.0% 0.23 [0.05, 0.41]
-1 -0.5 0 0.5 1
Favours control Favours experimental
514 APPENDIX G
19.7.2 Intensive physiotherapy (treadmill training)
Figure G-132. Knee extensor strength: intensive physiotherapy versus usual care

3.2.1 4 weeks
Moseley 2009 7.8 3.9 80 7.7 4 80 100.0% 0.10 [-1.12, 1.32]
Subtotal (95% CI) 80 80 100.0% 0.10 [-1.12, 1.32]
3.2.2 16 weeks
Moseley 2009 10.3 5 80 9.3 4.4 80 100.0% 1.00 [-0.46, 2.46]
Subtotal (95% CI) 80 80 100.0% 1.00 [-0.46, 2.46]
-100 -50 0 50 100

Figure G-133. Functional performance tests: intensive physiotherapy versus usual care
3.9.4 Sit-tostand test at 4 weeks
Moseley 2009 0.24 0.15 80 0.19 0.09 80 100.0% 0.05 [0.01, 0.09]
Subtotal (95% CI) 80 80 100.0% 0.05 [0.01, 0.09]
3.9.5 Sit-to-stand test at 16 weeks

Moseley 2009 0.26 0.14 80 0.22 0.11 80 100.0% 0.04 [0.00, 0.08]
Subtotal (95% CI) 80 80 100.0% 0.04 [0.00, 0.08]
-4 -2 0 2 4
Favours intensive Favours control
APPENDIX G 515
Figure G-134. Quality of life: intensive physiotherapy versus usual care

3.12.1 4 weeks
Moseley 2009 0.53 0.27 80 0.53 0.27 80 100.0% 0.00 [-0.08, 0.08]
Subtotal (95% CI) 80 80 100.0% 0.00 [-0.08, 0.08]
3.12.2 16 weeks
Moseley 2009 0.62 0.3 80 0.62 0.26 80 100.0% 0.00 [-0.09, 0.09]
Subtotal (95% CI) 80 80 100.0% 0.00 [-0.09, 0.09]
-100 -50 0 50 100

Figure G-135. Walking speed: intensive physiotherapy versus usual care

3.5.1 4 weeks
Moseley 2009 0.53 0.25 80 0.48 0.22 80 100.0% 0.05 [-0.02, 0.12]
Subtotal (95% CI) 80 80 100.0% 0.05 [-0.02, 0.12]
3.5.2 8 weeks
Moseley 2009 0.63 0.32 80 0.6 0.31 80 100.0% 0.03 [-0.07, 0.13]
Subtotal (95% CI) 80 80 100.0% 0.03 [-0.07, 0.13]
-1 -0.5 0 0.5 1
Favours control Favours experimental
Figure G-136. Pain: intensive physiotherapy versus usual care

3.11.1 4 weeks
Moseley 2009 44 80 41 80 100.0% 1.07 [0.80, 1.44]
Subtotal (95% CI) 80 80 100.0% 1.07 [0.80, 1.44]
Total events 44 41
3.11.2 16 weeks
Moseley 2009 30 80 29 80 100.0% 1.03 [0.69, 1.55]
Subtotal (95% CI) 80 80 100.0% 1.03 [0.69, 1.55]
Total events 30 29
0.01 0.1 1 10 100

516 APPENDIX G
Figure G-137. Length of hospital stay: intensive physiotherapy versus usual care
Moseley 2009 28 15 80 25 14 80 100.0% 3.00 [-1.50, 7.50]
Total (95% CI) 80 80 100.0% 3.00 [-1.50, 7.50]

-100 -50 0 50 100
Test for overall effect: Z = 1.31 (P = 0.19) Favours intensive Favours control
19.7.3 Intensive (more frequent) physiotherapy
Figure G-138. Adductor muscle strength (kp) at 9 weeks: intensive physiotherapy versus
usual care
Control Intensive Mean Difference Mean Difference

Karumo 1977 5.26 4.08 38 6.02 3.69 49 100.0% -0.76 [-2.42, 0.90]
Total (95% CI) 38 49 100.0% -0.76 [-2.42, 0.90]

-10 -5 0 5 10
Figure G-139. Length of hospital stay: intensive physiotherapy versus usual care

Karumo 1977 32.21 22.03 38 35.01 21.8 49 100.0% -2.80 [-12.09, 6.49]
Total (95% CI) 38 49 100.0% -2.80 [-12.09, 6.49]

-100 -50 0 50 100
APPENDIX G 517
19.8 Multidisciplinary rehabilitation
19.8.1 Hospital-based MDR
Hospital based MDR has been split into orthogeriatric hospital MDR (including GORU and
MARU) and hip fracture programmes.
Figure G-140. Mortality at 6 months: hospital MDR versus usual care
Intervention Control Risk Ratio Risk Ratio

9.1.1 Orthogeriatric hospital MDR
Gilchrist 1988 14 97 23 125 48.6% 0.78 [0.43, 1.44]
Naglie 2002 17 141 21 138 51.4% 0.79 [0.44, 1.44]
Subtotal (95% CI) 238 263 100.0% 0.79 [0.52, 1.21]
Total events 31 44
Total (95% CI) 238 263 100.0% 0.79 [0.52, 1.21]

Total events 31 44
0.01 0.1 1 10 100
Test for overall effect: Z = 1.09 (P = 0.27) Favours hospital MDR Favours usual care
Figure G-141. Mortality at 12 months: hospital MDR versus usual care

Galvard 1995 45 179 40 192 20.9% 1.21 [0.83, 1.75]
Huusko 2002 18 120 20 123 10.7% 0.92 [0.51, 1.66]
Kennie 1988 10 54 18 54 9.8% 0.56 [0.28, 1.09]
Stenvall 2007a 16 102 18 97 10.0% 0.85 [0.46, 1.56]
Subtotal (95% CI) 455 466 51.4% 0.95 [0.74, 1.23]
Total events 89 96
9.2.2 Hip fracture programme

Cameron 1993 32 127 38 125 20.8% 0.83 [0.56, 1.24]
Shyu 2008 7 80 7 82 3.8% 1.02 [0.38, 2.79]
Swanson 1998 5 38 6 33 3.5% 0.72 [0.24, 2.16]
Vidan 2005 28 155 39 164 20.6% 0.76 [0.49, 1.17]
Subtotal (95% CI) 400 404 48.6% 0.81 [0.61, 1.06]
Total events 72 90
Total (95% CI) 855 870 100.0% 0.88 [0.73, 1.06]

0.01 0.1 1 10 100
518 APPENDIX G
Figure G-142. Mortality (at discharge): hospital MDR versus usual care

Galvard 1995 19 179 20 192 27.3% 1.02 [0.56, 1.85]
Gilchrist 1988 4 97 13 125 16.0% 0.40 [0.13, 1.18]
Huusko 2002 5 120 5 123 7.0% 1.02 [0.30, 3.45]
Kennie 1988 5 54 4 54 5.6% 1.25 [0.35, 4.40]
Naglie 2002 7 141 13 138 18.6% 0.53 [0.22, 1.28]
Stenvall 2007a 6 102 7 97 10.1% 0.82 [0.28, 2.34]
Subtotal (95% CI) 693 729 84.6% 0.78 [0.54, 1.13]
Total events 46 62

Swanson 1998 2 38 2 33 3.0% 0.87 [0.13, 5.83]
Vidan 2005 1 155 9 164 12.4% 0.12 [0.02, 0.92]
Subtotal (95% CI) 193 197 15.4% 0.27 [0.07, 0.96]
Total events 3 11
Total (95% CI) 886 926 100.0% 0.70 [0.50, 1.00]

Total events 49 73
0.01 0.1 1 10 100
Figure G-143. Functional outcomes at 6 months: orthogeriatric hospital MDR versus usual
care

9.4.1 Non-recovery/decline in walking at long-term follow-up
Naglie 2002 59 124 56 117 100.0% 0.99 [0.76, 1.29]
Subtotal (95% CI) 124 117 100.0% 0.99 [0.76, 1.29]
Total events 59 56
9.4.2 Decline in transfers (bed to chair etc) at long-term follow-up

Naglie 2002 45 124 44 117 100.0% 0.96 [0.69, 1.34]
Subtotal (95% CI) 124 117 100.0% 0.96 [0.69, 1.34]
Total events 45 44
0.5 0.7 1 1.5 2

Favours hospital MDR Favours usual care
APPENDIX G 519
Figure G-144. Functional outcomes at 1 year: orthogeriatric hospital MDR versus usual
care

5.9.1 More dependent (based on Katz index) at 1 year
Kennie 1988 22 43 28 35 37.5% 0.64 [0.46, 0.89]
Stenvall 2007a 35 84 49 76 62.5% 0.65 [0.48, 0.88]
Subtotal (95% CI) 127 111 100.0% 0.64 [0.51, 0.81]
Total events 57 77
5.9.7 Non-recovery in activities of daily living (ADL) at 1 year

Stenvall 2007a 51 84 59 76 100.0% 0.78 [0.63, 0.96]
Subtotal (95% CI) 84 76 100.0% 0.78 [0.63, 0.96]
Total events 51 59
0.5 0.7 1 1.5 2

Figure G-145. Functional outcomes at 1 year: hip fracture programme versus usual care

6.10.3 Non-recovery in ADL/decline in walking at 1 year
Shyu 2008 19 80 33 82 30.1% 0.59 [0.37, 0.95]
Vidan 2005 67 127 75 125 69.9% 0.88 [0.71, 1.09]
Subtotal (95% CI) 207 207 100.0% 0.79 [0.65, 0.97]
Total events 86 108
0.5 0.7 1 1.5 2

Figure G-146. : Functional outcomes: Barthel scores at long-term follow-up: hip fracture
programme versus usual care
Intervention Control Mean Difference Mean Difference

9.7.1 Chinese Barthel Index at 6 months
Shyu 2008 90.53 19.4 73 84.36 24.02 75 100.0% 6.17 [-0.86, 13.20]
Subtotal (95% CI) 73 75 100.0% 6.17 [-0.86, 13.20]
9.7.2 Modified Barthel Index at 6 months

Swanson 1998 95.3 9.8 33 89 15.8 27 100.0% 6.30 [-0.53, 13.13]
Subtotal (95% CI) 33 27 100.0% 6.30 [-0.53, 13.13]
-100 -50 0 50 100

Favours usual care Favours hospital MDR
520 APPENDIX G
Figure G-147. Complications: hospital MDR versus usual care

9.10.1 pressure sores
Vidan 2005 8 155 27 164 100.0% 0.31 [0.15, 0.67]
Subtotal (95% CI) 155 164 100.0% 0.31 [0.15, 0.67]
Total events 8 27
9.10.2 heart failure

Vidan 2005 12 155 5 164 100.0% 2.54 [0.92, 7.04]
Subtotal (95% CI) 155 164 100.0% 2.54 [0.92, 7.04]
Total events 12 5
9.10.3 pneumonia
Vidan 2005 6 155 6 164 100.0% 1.06 [0.35, 3.21]
Subtotal (95% CI) 155 164 100.0% 1.06 [0.35, 3.21]
Total events 6 6
9.10.4 confusion
Vidan 2005 53 155 67 164 100.0% 0.84 [0.63, 1.11]
Subtotal (95% CI) 155 164 100.0% 0.84 [0.63, 1.11]
Total events 53 67
9.10.5 chest infection, cardiac problem, bedsore

Swanson 1998 6 38 13 33 100.0% 0.40 [0.17, 0.94]
Subtotal (95% CI) 38 33 100.0% 0.40 [0.17, 0.94]
Total events 6 13
9.10.6 stroke, emboli

Swanson 1998 4 38 1 33 100.0% 3.47 [0.41, 29.56]
Subtotal (95% CI) 38 33 100.0% 3.47 [0.41, 29.56]
Total events 4 1
9.10.7 Delirium
Marcantonio 2001 20 62 32 64 100.0% 0.65 [0.42, 1.00]
Subtotal (95% CI) 62 64 100.0% 0.65 [0.42, 1.00]
Total events 20 32
9.10.8 Severe delirium

Marcantonio 2001 7 62 18 64 100.0% 0.40 [0.18, 0.89]
Subtotal (95% CI) 62 64 100.0% 0.40 [0.18, 0.89]
Total events 7 18
0.1 0.2 0.5 1 2 5 10

APPENDIX G 521
Figure G-148. Length of hospital stay: hospital MDR versus usual care
Favours Intervention Control Mean Difference Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI Year IV, Random, 95% CI
Kennie 1988 37 33 54 56 54 54 8.5% -19.00 [-35.88, -2.12] 1988
Gilchrist 1988 44 56.14 97 47.7 86.09 125 7.7% -3.70 [-22.48, 15.08] 1988
Galvard 1995 53.3 47.7 179 28 24.2 192 13.4% 25.30 [17.52, 33.08] 1989
Naglie 2002 29.2 22.6 141 20.9 18.8 138 14.8% 8.30 [3.43, 13.17] 2002
Stenvall 2007a 30 18.1 101 40 40.6 97 12.8% -10.00 [-18.82, -1.18] 2007
Subtotal (95% CI) 572 606 57.3% 1.32 [-12.83, 15.47]
Heterogeneity: Tau² = 223.20; Chi² = 46.15, df = 4 (P < 0.00001); I² = 91%

Cameron 1993 19.5 20.5 127 28.5 30.3 125 14.1% -9.00 [-15.40, -2.60] 1993
Swanson 1998 20.8 11 38 32.6 23.8 33 12.8% -11.80 [-20.64, -2.96] 1998
Shyu 2008 10.1 3.7 80 9.72 4.96 82 15.8% 0.38 [-0.97, 1.73] 2005
Subtotal (95% CI) 245 240 42.7% -6.06 [-14.50, 2.38]
Total (95% CI) 817 846 100.0% -1.30 [-8.56, 5.97]

Heterogeneity: Tau² = 86.70; Chi² = 75.54, df = 7 (P < 0.00001); I² = 91%
-100 -50 0 50 100
Test for overall effect: Z = 0.35 (P = 0.73) Favours Intervention Favours Control
Figure G-149. Readmitted to hospital during follow up: hospital MDR versus usual care

Galvard 1995 36 160 57 172 33.8% 0.68 [0.47, 0.97]
Stenvall 2007a 38 96 30 90 19.0% 1.19 [0.81, 1.74]
Subtotal (95% CI) 256 262 52.8% 0.86 [0.67, 1.12]
Total events 74 87

Cameron 1993 16 103 11 101 6.8% 1.43 [0.70, 2.92]
Shyu 2008 23 80 19 82 11.5% 1.24 [0.73, 2.09]
Swanson 1998 3 35 2 31 1.3% 1.33 [0.24, 7.44]
Vidan 2005 44 155 46 164 27.5% 1.01 [0.71, 1.44]
Subtotal (95% CI) 373 378 47.2% 1.14 [0.87, 1.48]
Total events 86 78
Total (95% CI) 629 640 100.0% 0.99 [0.82, 1.19]

0.05 0.2 1 5 20
522 APPENDIX G
19.9 Home-based MDR versus usual inpatient rehabilitation

Figure G-150. Mortality: Home-based MDR versus usual care
Home MDR Usual care Risk Ratio Risk Ratio
7.1.1 Mortality at 12 months
Crotty 2003 3 34 4 32 100.0% 0.71 [0.17, 2.91]
Subtotal (95% CI) 34 32 100.0% 0.71 [0.17, 2.91]
Total events 3 4
0.01 0.1 1 10 100

Favours home MDR Favours usual care
Figure G-151. “Poor outcome” – institutional care and unable to walk: Home-based MDR
versus usual care
Home MDR Usual care Risk Ratio Risk Ratio

7.2.3 Moved to higher level of care
Crotty 2003 1 34 2 32 100.0% 0.47 [0.04, 4.94]
Subtotal (95% CI) 34 32 100.0% 0.47 [0.04, 4.94]
Total events 1 2
7.2.4 Unable to walk

Crotty 2003 0 34 2 32 100.0% 0.19 [0.01, 3.78]
Subtotal (95% CI) 34 32 100.0% 0.19 [0.01, 3.78]
Total events 0 2
0.01 0.1 1 10 100

Favours home MDR Favours usual care
Figure G-152. SF-36 scores at 12 months (0: worst to 100: best): Home-based MDR versus
usual care
Favours usual care Usual care Mean Difference Mean Difference

7.3.1 Physical component summary scores
Crotty 2003 38 8.8 34 33.3 10.7 32 100.0% 4.70 [-0.04, 9.44]
Subtotal (95% CI) 34 32 100.0% 4.70 [-0.04, 9.44]
7.3.2 Mental component summary scores

Crotty 2003 53.8 7.9 34 52.3 8.8 32 100.0% 1.50 [-2.54, 5.54]
Subtotal (95% CI) 34 32 100.0% 1.50 [-2.54, 5.54]
-20 -10 0 10 20
Favours usual care Favours home MDR
APPENDIX G 523
Figure G-153. Lengths of hospital or rehabilitation stays (days): Home-based MDR versus
usual care
Home-based MDR Usual care Mean Difference Mean Difference

7.4.1 Length of hospital stay
Crotty 2003 7.8 9.3 34 14.3 10.6 32 27.8% -6.50 [-11.32, -1.68]
Ziden 2008 18.4 8.4 48 20 6.8 54 72.2% -1.60 [-4.59, 1.39]
Subtotal (95% CI) 82 86 100.0% -2.96 [-5.50, -0.42]
7.4.2 Length of rehabilitation (hospital + home)

Crotty 2003 28.3 14.5 34 14.3 10.6 32 100.0% 14.00 [7.90, 20.10]
Subtotal (95% CI) 34 32 100.0% 14.00 [7.90, 20.10]
-100 -50 0 50 100

Favours Home-based MDR Favours Usual care
Figure G-154. Readmission to hospital during 4 month follow-up: Home-based MDR

versus usual care
Home-based MDR Usual care Risk Ratio Risk Ratio

Crotty 2003 8 34 7 32 1.08 [0.44, 2.62]
0.01 0.1 1 10 100
Favours Home-based MDR Favours Usual care
Figure G-155. Degree of independence (Functional Independent Measure): Home-based

MDR versus usual care
Home-based MDR Usual care Mean Difference Mean Difference

7.6.2 FIM Self-care
Ziden 2008 38.4 2.9 48 33.5 7.2 54 100.0% 4.90 [2.81, 6.99]
Subtotal (95% CI) 48 54 100.0% 4.90 [2.81, 6.99]
7.6.3 FIM Mobility

Ziden 2008 18.3 1.5 48 16.3 3.3 54 100.0% 2.00 [1.02, 2.98]
Subtotal (95% CI) 48 54 100.0% 2.00 [1.02, 2.98]
7.6.4 FIM Locomotion

Ziden 2008 10.4 2.5 48 7.6 3.6 54 100.0% 2.80 [1.61, 3.99]
Subtotal (95% CI) 48 54 100.0% 2.80 [1.61, 3.99]
-10 -5 0 5 10
Favours Usual care Favours Home-based MDR
524 APPENDIX G
Figure G-156. Mobility and strength tests: Home-based MDR versus usual care
Experimental Usual care Mean Difference Mean Difference

7.7.5 Up and go test
Ziden 2008 24.9 15.4 48 30.8 16 54 100.0% -5.90 [-12.00, 0.20]
Subtotal (95% CI) 48 54 100.0% -5.90 [-12.00, 0.20]
7.7.6 Sit-to-stand test

Ziden 2008 1.8 0.8 48 3.3 3.6 54 100.0% -1.50 [-2.49, -0.51]
Subtotal (95% CI) 48 54 100.0% -1.50 [-2.49, -0.51]
-100 -50 0 50 100

Favours usual care Favours home rehab
20 Appendix H: Health economic analysis
20.1 Cost analysis of nerve blocks, non-opioid analgesics and

opioid analgesics
20.1.1 Nerve block cost analysis
No studies were identified on the cost-effectiveness of nerve blocks compared to systemic

analgesia in providing adequate pain relief and reducing side effects and mortality.
As a consequence, we conducted a cost analysis where the different types and level of resources
used to administer a nerve block to a patient with a suspected hip fracture are based on the
GDG’s opinion, summarised in Table 73 below.
Table 73: Cost analysis for nerve block

Resources Unit price Source of unit price
Spinal pack (gown and drape) * £4.50 NHS hospital***

Biogel glove £1.07 NHS hospital***
Chlorhexidine** £1.08 NHS hospital***
Vial with Lidocaine 1% £0.38 (10-mL am) BNF 58
Vial of 0.5% Levobupivacaine £3.88 (5mg/mL) BNF 58
APPENDIX H 525
Syringes (10ml) £0.06 NHS hospital***

Filter needle £0.23 NHS hospital***
Regional block needle £5.78 NHS hospital***
Hypodermic needle £1.35 NHS hospital***
Personnel costs (consultant £36.00 PSSRU 2009; GDG estimate
anaesthetist) (£1.8 per minute*20
minutes)
Total cost £54.33
* Most anaesthetists use full aseptic precautions, with a gown and gloves plus a dressing pack
** Chlorhexidine built into swabs are standard practice.
*** Peterborough and Stamford Hospital NHS Foundation Trust
The personnel costs can vary depending on the time required to administer a nerve block,
which in turn depends on the technique used (nerve stimulator, ultrasound- guided,
landmark only) and the block used (3-in-1, femoral nerve only or fascia iliaca block). If a
fascia iliaca block is administered using a landmark technique only, then the following
sequence would be observed:
- Obtaining equipment (needle, disinfectant, gloves, local anaesthetic etc)
- Estimating patient’s weight
- Obtaining patient’s consent
- Identifying landmark
- Disinfecting skin
- Anaesthetising skin
- Passing needle
- Injecting local anaesthetic
- Maintaining manual pressure distal to injecting site for a minute after injection
The GDG estimates that the whole process would require about 15 - 20min, and that the time
required would not change substantially if the block is administered by a consultant anaesthetist
or a SAS (staff and associate specialist).
At present, in most emergency departments that do advocate nerve blocks for hip fracture
patients, the block would be performed by 'middle grade doctors', i.e. specialist registrars (SpR),
senior specialist trainees (ST3-6) or senior clinical fellows. In some departments, junior doctors
can also administer the procedure. In operating departments and if asked to do elsewhere
anaesthetists will always have a trained assistant with them, usually an ODP, which would
increase the total cost for a nerve block to £63.33 (assuming an ODP wage of £27 per hour as that
of a senior nurse)
The GDG recognises that there is likely to be a wide variation in practice as far as the
administration of nerve blocks is concerned.
1) The nerve block may be administered with a ultrasound-guided technique, which would
require the use of ultrasound anaesthetic machines. An average cost of these machines
has been estimated at around £34,000 from hospital records supplied by the
Peterborough and Stamford Hospital NHS Foundation Trust. The equivalent annual cost
would be £5,313, assuming a life expectancy of 7 years and discount rate of 3.5%.
526 APPENDIX H
If we assume that the ultrasound machine would be used solely for nerve blocks in the
anaesthetic department and that it would be used 7 hours per day every day, including
weekends with 4 scans per hour, then the machine costs 52p per scan.
2) Bupivacaine can be used as local anaesthetic instead of Levobupivacaine, but the

difference in price would be minimal.
3) A nerve locator could be used when performing the nerve block, but its cost would be
minimal (GDG expert’s opinion)
20.1.2 Non-opioid analgesics
We assume that patients will take a simple analgesic, such as paracetamol, continuously
throughout their inpatient stay. The GDG noted that aspirin would not generally be used as an
analgesic for our population, unless it is used as a low dose to prevent strokes. The average cost
of these drugs is less than £0.1p per dose (BNF 58).
20.1.3 Opioid analgesics
Table 74: Opioids controlled drugs

Category Dose cost
(source: BNF 58)
Diamorphine hydrochloride £2.69
Morphine salts £0.36
Oxycodone hydrochloride £1.60
Buprenorphine £0.72
Average cost £1.34
The opioids reported in table Table 74 are non-controlled drugs and can be administered within
existing nurse drug rounds, and therefore there is little extra cost associated with their
administration.
Table 75Table 74 summarises the opioids controlled drugs that could be administered to
hip fracture patients. This category of analgesics requires an additional round of two
trained nurses to administer. The GDG estimates that this would involve approximately 15
minutes per dose, with an extra cost of £10.50 (considering that the cost per hour of a
staff nurse is £21 (PSSRU 2009)). Hence, the cost of administering these controlled drugs
is £11.84 (nurse time plus drug cost).
APPENDIX H 527
Table 74: Opioids controlled drugs

Category Dose cost
(source: BNF 58)
Diamorphine hydrochloride £2.69
Morphine salts £0.36
Oxycodone hydrochloride £1.60
Buprenorphine £0.72
Average cost £1.34
The opioids reported in table Table 74 are non-controlled drugs and can be administered within
existing nurse drug rounds, and therefore there is little extra cost associated with their
administration.
Table 75: Opioids non-controlled drugs

Category Dose cost
(source: BNF 58)
Codeine phosphate £1.83
Dihydrocodeine Tartrate £2.58
Tramadol Hydrochoride £1.47
Average cost £1.96
The remaining opioid drugs (dipipanone hydrochloride, hydromorphone hydrochloride,

meptazinol, methadone hydrochloride, paperetum, pentazocine, pethidine hydrochloride) are
very rarely used in our population, as they are highly specialist analgesics for palliative care.
Fentanyl is rarely used in acute care, and is therefore not included in the dose cost.
20.2 Hourly wage costs for a planned trauma list

The GDG suggested to consider a general emergency theatre as a likely alternative to a
planned trauma list.
A general emergency theatre is one to which multiple specialities have access for
unplanned operations. Under these circumstances there will be necessary discussions
between the various specialties as to whose patient should go first. With an emergency
theatre, there is no start and finish time that can be forecasted in advance and great
variation in the professional grade of the personnel involved.
When the hip fracture patient does go to theatre, he will clearly need the same supporting
staff of surgeon, anaesthetist, nursing staff, radiographer etc. as for a planned trauma list.
Thus, some costs will be common across the two types of lists with the exception that an
emergency trauma list is more likely involve more junior staff.
528 APPENDIX H
Overall, the GDG has identified the following differences between an emergency and a
planned trauma list:
a) Senior responsible staff involved
With a general emergency theatre the involvement of senior staff may be regarded as a
covering on-call commitment. With a trauma list it becomes a regular work commitment to
which there needs to be programmed activities allocated for both senior responsible
anaesthetic and surgical staff. Since the nature of the work is known appropriate scrub
staff can be allocated
b) Where necessary a new operating room
Providing trauma cases with the same level of care enjoyed by elective cases may require
extra operating theatre space. There have been attempts in many hospitals to use
operating theatres for a greater proportion of the 24 hour day to better use that resource.
This has in general proved to be difficult; largely because trained staff prefer to have their
regular commitments in what would be regarded as normal working hours. Genuine
emergency procedures are a small proportion of any theatre workload and these need to
be carried out at the necessary time whenever that may be. However, the bulk of
procedures are urgent or elective, these should all be given the same advantages of a
properly staffed theatre. Should it be necessary for best use of theatre space to utilise
evening operating lists it may be preferable that these are occupied by the well prepared
elective patients rather than the rapidly prepared often unwell urgent patient. Since this is
unlikely to occur more operating space may well be required for daytime lists.
The advantage of a general emergency list is it uses the resources already available, and
may run from early in the morning till late in the evening (therefore many operations can
be performed sequentially). On the other hand, a planned trauma list needs to be run in
parallel with other lists, preferably in the morning. It may be difficult to find a physical
space for a planned trauma list to be carried out, in which case a new operating room may
be required.
c) Ad-hoc technical resources
A planned trauma list needs a dedicated image intensifier, so it depends upon the other
lists running as to whether its availability may be a problem.
d) Type of patients operated
A planned trauma list would only operate trauma patients whereas in a general
emergency theatre there would be operations on different types of patients
The table below estimates the cost of one hour of personnel input for a planned trauma
list during weekly normal working hours (that is, excluding weekends and public holidays
personnel costs).
APPENDIX H 529
Personnel input cost for a planned trauma list – weekly normal working hours
Categories of personnel Cost of hourly
wage (source:
PSSRU 2009)
Consultant surgeon £108
Consultant anaesthetist £108
Scrub nurse (senior staff £27
nurse)
Unscrub nurse (runner – staff £21
nurse)
Radiographer £25
Anaesthesia assistant [ODP] £27
(as senior staff nurse)
Recovery nurse (staff nurse) £21
Total personnel costs £337
As for the personnel costs of a general emergency theatre, we assume that it mainly relies
on registrars (both surgeons and anaesthetists) rather than consultants, and use a hourly
cost for registrars of £38 (per 48 hour week; source: PSSRU 200961). Any emergency
theatre also relies on having consultant surgeons and anaesthetist on call, and this cost
would also have to be considered in the overall costs for an emergency theatre. Once
again we consider the personnel costs during weekly normal working hours, and thus
exclude weekends and public holidays personnel costs nor additional personnel costs for
out-of-hours operations, which are quite common with a general emergency theatre.
Personnel input cost for an general emergency theatre – weekly normal working hours
Categories of personnel Cost of
hourly wage
(source:
PSSRU
2009)
Registrar surgeon £38
Registrar anaesthetist £38
Consultant surgeon on call* £23
Consultant anaesthetist on call* £23
Scrub nurse (senior staff nurse) £27
Unscrub nurse (runner – staff £21
nurse)
530 APPENDIX H
Radiographer £25
Anaesthesia assistant [ODP] (as £27
senior staff nurse)
Recovery nurse (staff nurse) £21
Manufacturer Price for Sliding Hip Price for Short Price for Long
Screw (for intramedullary nail intramedullary nail
extramedullary (for intramedullary (for intramedullary
fixation) fixation) fixation)
IMP
Stryker £357 £854 £1384
Biomet £260.70 £745 £1,090
Zimmer (1) £175 £826 £1,177
Synthes £260.35 £796.05 £1,142.85
Smith & Nephew (2) £245 £823.45 £1,083.16
DePuy £217 £516 NA
Total personnel costs £243

*= Assumes that the average emergency work undertaken per week for on-call duty is 3
hours. If the amount of this emergency on-call work raises to 6 hours per week, the hourly
rate paid to the consultant would be £39. Source: hourly on call salary costs provided by
the NICE costing implementation team.
Thus, a planned trauma list has additional personnel cost compared to a general
emergency theatre of £94 per hour. It is very important to stress that this estimate does
not consider the additional salary costs linked with operations taking place during
weekends or public holidays and outside normal working hours.
20.3 Prices for sliding hip screws and short and long
intramedullary nails
In the table above we report the prices for sliding hip screws, short intramedullary and long intramedullary
nails from quotations received by some of the major manufacturers of implants. All quotations are 2010
prices. All prices include VAT.
APPENDIX H 531
Average price £252.51 £760.08 £1,175.40
Manufacturer Price for Sliding Hip Price for Short Price for Long
Screw (for intramedullary nail intramedullary nail
extramedullary (for intramedullary (for intramedullary
fixation) fixation) fixation)
IMP
Stryker £357 £854 £1384
Biomet £260.70 £745 £1,090
Zimmer (1) £175 £826 £1,177
Synthes £260.35 £796.05 £1,142.85
Smith & Nephew (2) £245 £823.45 £1,083.16
DePuy £217 £516 NA
Average price £252.51 £760.08 £1,175.40

532 APPENDIX H
20.4 Cost analysis of the interventions for intensive mobilisation strategies.

Cost analysis of the interventions for intensive mobilisation strategies.
Study Intervention Control Other resources Unit costs Incremental cost of
intervention over usual care
Hauer et al 1 hour of 1 hour of Using data provided from a £23 per hour for physiotherapist input £12
2002 140 physiotherapist for 3 physiotherapist for 3 GDG member, the cost of
weeks weeks the equipment that would Other costs (for stepping and strength
be used in the intervention training) are considered as negligible
group was estimated at and have not been included in the cost
£49.00 per patient. analysis
This estimate is based on a
study currently under way,
where the costs per person
for the exercise equipment
was estimated to be £49.00.
This cost assumes no re-use
of equipment and does not
include overhead costs.
When appropriately cleand,
the equipment could be re-
used, in which case,
assuming that it is re-used
up to four times, the
relevant cost per person
would be approximately
£12.
Karumo Physiotherapy Average of 30mins Crutches £23 per hour for physiotherapist input £180.18
1977A 171 performed twice daily physiotherapy per
– average of 1 hour for day for 14 days (£161 control; £322 intervention)
14 days
APPENDIX H 533
Cost of crutches: £19.18 (a)

Moseley et Weight bearing Exercise for 30 mins For inpatients: additional £23 per hour for physiotherapist input £827.62 (for the inpatient part
al., 2009216 exercise twice daily for each day for 4 weeks. inpatients costs of the rehabilitation
a total of 60 minutes £13,029 for Biodex treadmill with programme)
per day for 16 weeks. Treadmill with partial body-weight support.
weight-support The costs of the outpatient
Walking on a tread LOS in hospital: 25 Cost per day of a bed (elderly person part of the programme was
mill with partial body days. care: £152 (b)) not calculated as it was not
weight support using a clear from the study what
harness (for Assumption: a physiotherapist is types of resources where used
inpatients) or a present for all the duration of in that part of the rehab
walking programme treatment when inpatient programme.
(after hospital
discharge). Treadmill costs – assumptions:
- Treadmill live is 5 years
LOS in hospital: 28 - Treadmill overall use: 4
days (4 weeks) hours per day for 5 days of
the week
For 84 days: walking - Discount rate: 3.5%
programme with - Treadmill used for 20 minutes
home visits and per session
exercise programme - Cost one session of treadmill
imputable to the
This started as an intervention: £0.54.
inpatient programme, - Cost of treadmill sessions
followed by home over 4 weeks: 7*4*£0.54=
visits and a structured £15.12
home exercise
programme. INTERVENTION COSTS:
Bed days cost: £152*28=£4256
Attributable treadmill costs: £15.12

per patient
534 APPENDIX H
Physiotherapist costs (intervention):

£23*28=£644
Total inpatient costs of intervention:

£4915.12
CONTROL COSTS:
Physiotherapist costs:
£23*0.5*25=£287.5
Bed day costs:

£152*25=£3800
Total cost for control: £4087.5
(a) Average cost obtained from the NHS Supply Catalogue 2010 for the following manufacturers: Sunrise Medical Ltd, NHS Supply Chain and Days
Healthcare UK Limited
(b): We have estimated the hospital stay using the unit cost per excess day associated with complex elderly patients (that is, the unit cost per day for days
exceeding the trim point). Using all the HRG unit costs reported for all Complex Elderly patients (Hospital Episode Statistics for England, Inpatient Statistics,
2007-08) we found a weighted mean of £152.
APPENDIX H 535
20.5 Cost-effectiveness analysis of hospital investment versus no

hospital investment for early surgery
20.5.1 Introduction
The GDG assigned a high priority in the economic plan for an original economic analysis to the
question:
“ In patients with hip fractures what is the clinical and cost effectiveness of early surgery (within
24, 36 or 48 hours) on the incidence of complications such as mortality, pneumonia, pressure
sores, cognitive dysfunction and increased length of hospital stay?”
A review of the literature was conducted. The literature search and review methods can be found
in Chapter 3. No cost-effectiveness analysis was found which addressed our clinical question. As a
consequence, the GDG felt that an original decision model was essential in order to inform their
recommendations.
The following general principles were adhered to:
• The GDG was consulted during the construction and interpretation of the model.
• When published data was not available, we used hospital records and experts’ opinion to
populate the model.
• Model assumptions were reported fully and transparently.
• The results were subject to sensitivity analysis and limitations were discussed.
• We followed the methods of the NICE reference case. Therefore costs were calculated
from the NHS and PSS perspective. Health gain was measured in terms of quality-
adjusted life-years (QALYs) gained. Both future costs and QALYs were discounted at 3.5%.
• The model employed a cost-effectiveness threshold of £20,000 per QALY gained.
• The model was peer-reviewed by another health economist at the NCGC.
20.5.2 Background
There are fundamentally two reasons why a patient with a diagnosed hip fracture is delayed in
receiving surgery. First, the patient may be considered to be unfit for surgery for medical reasons,
and therefore made to wait until the medical team optimises her status. Alternatively, a patient
may be deemed to be fit for surgery at the time of admission, but will still incur delays linked with
administrative reasons, such as lack of space on theatre lists and/or problems with theatre,
surgical and anaesthetic staff cover.
In our economic analysis, we focus exclusively on the administrative reasons for surgical delay.
This is because, albeit all studies in the clinical review were initially considered for inclusion in the
economic model, the GDG concluded that only the subgroup of papers with a population that
excluded patients unfit for surgery was appropriate for basing the economic model upon.
In particular, the GDG considered that by removing patients unfit for surgery (defined as those for
whom: ‘any medical reason when orthopaedic or anaesthetic staff felt that operation should be
delayed in order to improve the patient’s fitness for surgery’308) from our model, we would be
excluding confounding factors from the decision model, thus allowing more confidence in the
cost-effectiveness findings.
536 APPENDIX H
Those studies that had not excluded patients unfit for surgery from their population would
potentially have an imbalance in baseline characteristics which could result in skewing the data in
favour of the early surgery group. Even though these studies had used logistic regression to adjust
for confounding factors (such as ASA score, sex, age and comorbidities like cardiac problems), the
GDG still felt that the subgroup of papers that excluded patients unfit for surgery were more
robust.
Overall, three studies which excluded patients unfit for surgery from their population were
included in our clinical review: Moran (2005), Siegmeth (2005) and Orosz (2004)215,250,308. Of these,
only Siegmeth308 reports data regarding whether patients returned to their original place of
residence or whether they changed residence (at 1 year follow up) and this was considered
essential information for modelling the different health states in our analysis.
Siegmeth (2005)308 excluded patients who were delayed for any medical reason when orthopaedic
or anaesthetic staff felt that operation should have been delayed in order to improve the patient’s
fitness for surgery. Reasons for delays included anaemia requiring transfusion, correction of
electrolyte imbalance, uncontrolled diabetes and untreated heart failure. The GDG agreed that
the study adopted a set of diagnostically objective criteria in deciding which patients were
considered fit for surgery, and that no selection bias had been introduced in this process.
Furthermore, Siegmeth308 is a study set in the UK, and as such was considered to be more
applicable to our question than studies set in different countries. As the paper interprets “early
surgery” as surgery that took place within 48 hours from admission, we adopt this specific cut-off
point in our model.
20.5.3 Population and time horizon
The population for the cost-effectiveness analysis consists of hip fracture patients (male and
female) hospitalised for surgery and considered to be fit for surgery. The model spans over a life-
time horizon.
20.5.4 Software
The cost-effectiveness analyses were conducted using TreeAge Pro 2008.
20.5.5 Methods
We built a decision tree with Markov states where the expected costs and effectives of two
alternatives are evaluated and compared: “investment for early surgery” vs. “no investment for
early surgery”. As discussed in section 20.5.8, this investment consists of the addition of extra
operating lists to the existing weekly number of theatre lists.
As mentioned in section 20.5.2, the health states of the model reflect the outcomes of Siegmeth
(2005)308: at one year after surgery, patients can be “living in their own home”, “living in a
residential home”, “living in a nursing home”, or “dead”.
Since patients were followed at 1 year from surgery in Siegmeth (2005)308, the cycle length of the
Markov model is supposed to last one year. At the end of each cycle, patients can either stay in
the same health state or can transit to the “dead” state (the “absorbing” health state in the
model). This is because no data were available from Siegmeth (2005)308 over the possible
transitions of patients between the other health state (“living in own home”, “living in residential
home” or “living in nursing home”). Hence, we assume that patients’ place of residence at 1 year
stays the same for the rest of their lifetime. Although this is obviously a simplification, it is unlikely
APPENDIX H 537
that the impact of the intervention (“investment for early surgery”) will have an effect after 1 year
from surgery.
The model starts with a simple decision node, which represent the decision to invest or not in
providing extra operating theatre lists. Following the investment, surgery takes place. However,
whether surgery will indeed take place “early” (within 48 hours from admission) or “late” is an
uncertain event. As a consequence, in our decision model we are able to address the question of
whether it is cost-effective to invest in extra operating lists (and therefore in extra personnel and
all the required resources) in order to increase the probability that those patients deemed “fit for
surgery” at admission are indeed operated within a certain time target. The probabilities of a
patient being in one of the four possible health states in the first cycle depend on whether they
have been operated within 48 hours or after 48 hours.
20.5.6 Treatment effects
The proportion of patients in each health state depends on the effectiveness of the treatment
(that is, of investment for early surgery), and on the proportion of patients still alive, which falls as
the number of cycles and therefore age increases.
Primary data were obtained from a GDG expert advisor regarding the proportion of patients in
each health state at 1 year follow up. These data (reported in Table 76 below) have been
extracted from the same database used in the Siegmeth308 study included in our clinical review,
and therefore refer to patients who were delayed for surgery not for medical reason but only for
administrative reasons.
538 APPENDIX H
Hip fracture patients
“Investment” OR
“No investment”
for early surgery
Surgery > 48 hours

Surgery ≤ 48 hours
Living own home

Living own home
PNH late
Living residential
Living residential
home
home
Living nursing Living nursing

home Dead Dead home
POH early = probability of “living in own home” after early surgery POH late = probability of “living in own home” after late surgery
PRH early = probability of “living in residential home” after early surgery PRH late = probability of “living in residential home” after late surgery
PNH early = probability of “living in nursing home” after early surgery PNH late = probability of “living in nursing home” after late surgery
P D early = probability of being “dead” after early surgery P D late = probability of being “dead” after late surgery
Figure 157: Decision tree with Markov states - investment for early surgery vs. no hospital investment for early surgery
APPENDIX H 539
Table 76: Place of residence and mortality at 1 year

Patients who had Patients who had RR (surgery ≤ 48 hours
surgery ≤ 48 hours surgery > 48 hours vs. surgery > 48 hours)
Total number of admissions 3445 (0.952%) 175 (0.048%)
No. patients living in own 1734 (0.503%) 76 (0.434%) 1.16
home at 1 year
No. patients living in 489 (0.142%) 22 (0.126%) 1.13
residential home at 1 year
No. patients living in nursing 307 (0.089%) 16 (0.091%) 0.97
home at 1 year
No. patients dead at 1 year 915 (0.266%) 61 (0.349%) 0.76
It is important to point out that, for the first cycle in our model, the mortality data are based on
the information obtained from the database reported in Table 76.
For the long-term mortality, we considered a mean age of 81 for our cohort of patient, as this was
the mean age of patients in Siegmeth308. Following Parker(1992)268, the life expectancy after the
first cycle was assumed to be the same as that of the general population, and was obtained from
the Life Tables for the general population of England and Wales in the year 2005-2007 from the
Government Actuary Department:
(http://www.gad.gov.uk/Documents/Demography/EOL/ILT%202005-07/wltewm0507.xls).
This value was then adjusted for the ratio male/female corresponding to the patients
characteristics in the study as follows:
Total LE = LEfemale * %female + LEmale * %male
20.5.7 Quality of life
The EQ-5D utility weights for patients living in their own home, in a residential or nursing home
used in our model are based on the findings of the paper by Tidermark (2002)328 and are
summarised in Table 77 below.
Table 77: EQ-5D scores for health states

Health state Utility score
Living in own home (at 1 year from the fracture) 0.64
Living in own home (after 1 year from the fracture) 0.56
Living in an institution 0.35

540 APPENDIX H
(a) Source: Tidermark (2002)
We have assumed that patients living in their own home correspond to those “living
independently” in Tidermark (2002)328.
For each strategy, the expected QALYs in each cycle are calculated as follows:
Expected QALYs = Σ (Ui x Pi )
Where:
Ui = the utility score for health state i
Pi = the proportion of patients in health state i
and where health state i could be any of the health states reported in table 1.
The overall lifetime expected QALYs are given by the sum of QALYs calculated for each cycle. The
incremental QALYs gained associated with a treatment strategy (“investment for early surgery” in
our case) are calculated as the difference between the expected QALYs with that strategy and the
expected QALYs with the comparator (that is, “no investment for early surgery”).
20.5.8 Cost analysis
20.5.8.1 Early surgery implementation

costs
The “investments for early surgery” in our model consists of adding extra operating lists aimed at
increasing the theatre capacity as a way of reducing the time hip fracture patients have to wait
before they receive surgery. The evidence for this strategy refers to hospital records supplied by a
GDG member. In 2008, the John Radcliffe hospital in Oxford implemented a policy aimed at
increasing the number of patients operated with 48 hours from admissions. This was achieved by
adding an extra five half-day operating lists to the weekly number of lists. All the extra lists were
added during a normal working week, not during the weekend. Each extra theatre list consisted of
four hours of operating time. Table 78 below describes the extra personnel that had to be
employed to run these extra lists and the associated costs incurred by the hospital.
Table 78: Personnel costs for extra operating lists

Categories of personnel Hours per Cost of hourly wage Additional personnel
additional list (source: PSSRU 2009) costs for the 5 extra
lists
Consultant surgeon 4 £108 £2,160
Consultant anaesthetist 4 £108 £2,160
Orthogeriatrician 1 £108 £540
Scrub nurse (as senior staff nurse) 4 £27 £540
Unscrub nurse (runner) 4 £21 £420
Radiographer 4 £25 £500
APPENDIX H 541
Anaesthesia assistant [ODP] (as 4 £27 £540

senior staff nurse)
Recovery nurse (as staff nurse) 4 £21 £420
Total personnel cost for 5 additional weekly lists £7,280
Total personnel costs for 5 additional lists over 1 year £378,560
In addition to the extra personnel costs, we have to consider the overhead costs involved with
running the operating theatre for the extra five half-day lists. These costs have been estimated on
the basis of hospital records obtained from the Peterborough and Stamford District Hospital, and
are summarised in the Table 79 below.
Table 79: Overhead costs for the additional operating lists

Resource Cost per minute (£)
Energy 0.18
Premises maintenance 0.09
Staff uniforms and clothing 0.01
Medical and surgical equipment (including instruments) 0.82
Dressings 0.06
Total overhead costs per minute £1.16
Total overhead costs for 5 additional weekly lists £1,392
Total overhead costs for 5 additional over 1 year £72,384
It follows that the overall total implementation cost for early surgery amounts to £450,944.
Probability of early surgery after hospital investment

The following table summarises the number of patients operated within 48 hours from
admissions before the extra operating lists were added (i.e. at baseline, year 2007-08) and for the
years following the investment in extra operating lists. These data are also based on hospital
records supplied by the John Radcliffe Hospital in Oxford.
Table 80: Patients operated within and after 48 hours from admission - before and after
investments in extra operating lists
2007-8 2008-9 2009-10 2010-2011*
(baseline) (intervention)
Total cases operated during the 431 434 441 123
year
Number of patients fit for 363 347 374 114
542 APPENDIX H
surgery within 48 hours during

the year
Number of patients delayed over 68 87 67 9
48hrs because unfit for surgery
Number of patients fit for 192 233 316 109
surgery and operated within 48 (52.89%) (67.15%) (84.49%) (95.61%)
hrs (%)
Number of patients fit for 171 (47.11%) 114 (32.85%) 58 (15.51%) 5 (4.39%)
surgery but delayed >48hrs
*data collected up to July 2010.
As Table 80 shows, the addition of the extra operating lists affected the probability that patients
fit for surgery are operated “early” (in our case, within 48 hours from admission). However, even
following this investment, early surgery is still a random event which is affected by many other
factors beyond the number of operating sessions available. Still, the data in Table 80 shows that
there is a clear trend in the increase in the number of patients fit for surgery that are operated
within 48 hours. There are several possible reasons for this trend, but they can mostly be seen as
the result of a learning process (by all the health care professionals involved in the care of the
patients) that produced positive spillover effects and efficiency gains in the years following the
implementation of the extra operating lists.
We use the data for 2008-09 as our intervention in the base case analysis. Data referring to other
years (2009-10 and 2010-11) are used in a sensitivity analysis.
Incremental cost per patient of implementation costs for extra theatre lists
The extra cost per patient of implementing an early surgery strategy for the first year following
the investment (that is, for 2008-09) correspond to £450,944/434 = £1039.04 (where 434 is the
total number of patients operated for hip fracture – whether within or after 48 hours from
admission – in the intervention year).
20.5.8.2 Costing hospital length of stay
In addition to the costs linked with the extra operating lists, we have consider the costs for the
length of hospital stay. We assume that the daily cost of a hospital bed in an orthopaedic ward
corresponds to £241.69 (which is obtained from a weighted average of the costs of the excess bed
days for hip all hip fracture procedures (major, intermediate and minor) with all types of
complications). This cost is then multiplied by the length of stay for each group of patients,
summarised in Table 81 below and based on the findings of 308
Table 81: Mean length of hospital stay

Surgery ≤ 48 hours Surgery > 48 hours CI
Mean hospital stay in 21.6 36.5 (5.7 – 16)
days (95% CI) p<0.0001
20.5.8.3 Health state costs

APPENDIX H 543
Health and social care costs for patients in the “living at own home” health state
We acknowledge that even if a patient is discharged to his own home and returns to an
independent living status, he will still incur in a higher level of use of health and domiciliary social
care compared to his pre-fracture status, as it is unlikely that he will completely regain his pre-
fracture level of independence. The PSSRU (2009)61 describes five possible “community care
packages” for individuals who live in their own home and consume a level of health and
domiciliary social care resources that varies according to their specific level of independence in
functional status. For our model, we assume that the health and domiciliary social care costs for
the patients in the “living in their own home” health state is an average of the cost of the “very
low”, “low” and “medium” community care packages stated in the report. It follows that the
weekly average health care costs for patients living in their own home after the fracture amounts
to £9.9, and the weekly domiciliary social care costs to £98.1. While the health care costs are fully
funded by the NHS, the domiciliary social care costs will only be partially met by the local
authority. We found no published evidence regarding a national average of the percentage of
domiciliary social care funded by local authorities71, 348, 72, 144. In our base case analysis, we
assume that 60% of these costs would be funded by the local authorities, and then test this
assumption in a sensitivity analysis.
Health and social care costs for patients in the “living in residential home” and “living in nursing
home” health states
For patients living in a residential or in a nursing home, we need to consider the cost of long term
care. This is estimated from the unit cost of stay in private nursing homes and in private
residential care reported in the PSSRU 2009. The health care costs and fees per permanent
residential week are described in Table 82.
Table 82: Weekly health and social care costs for patients living in residential or nursing homes
Place of residence Weekly health care costs Weekly fees
Private nursing home £30.80, of which: £ 678

• £30 (GP weekly home visit)
• £0.80 (community nursing)
Private residential care £26.3, of which: £467
• £19.30 (GP weekly home visit)
• £7.00 (community nursing)
Once again, while the NHS fully funds the health care costs, it does not pay towards long-term
care for all patients. Moreover, only a proportion of the weekly fees will be met by the local
authorities. We found no published evidence regarding a national average of the percentage of
long-term care costs funded by local authorities, and as a consequence we assumed that the
proportion of the costs of long-term care borne by the NHS and PSS is equal to 60% in the base
case analysis, and changed it afterwards in a sensitivity analysis.
544 APPENDIX H
20.5.9 Cost-effectiveness analysis
Table 83 below summarised the findings of the cost-effectiveness analysis for the determinist
case. We found that, for the first year following the investment in extra operating lists, the
strategy “investment for early surgery” is not cost-effective at a willingness to pay of £20k per
QALYs gained.
Table 83: Cost-effectiveness results - deterministic analysis – first year following investment in
extra lists
Strategy Cost Incremental Effectiveness Incremental Incremental
Cost Effectiveness cost-
effectiveness
(ICER)
No hospital investment for early surgery £46.4K 2.32
Hospital investment for early surgery £47.4K £1.0K 2.3622 0.0421 £/QALY 22776
(with probability of early surgery =67.15%)
Table 84: Costs breakdown for "investment" and "no investment" in early surgery reports a
breakdown of all the cost categories included in the model for the first year in which the extra
operating lists were introduced.
Table 84: Costs breakdown for "investment" and "no investment" in early surgery
Resource item Investment in extra No investment in
operating lists extra operating lists
Rehab cost NA NA
Hospital-related costs (for length of stay and 7442 6917
investment in extra operating lists)
Readmission NA NA
Community health care (own home) 1664 1630
Community social care (own home) 9892 9690
Community health care (residential and nursing home) 2224 2206
Community social care (residential and nursing home) 26200 26000
Total cost £47422 £46443
In order to ascertain how robust the findings of Table 83 are, we ran a series of sensitivity
analyses. Deterministic sensitivity analysis showed that the findings of our model are not sensitive
to the hospital bed day cost. However, threshold sensitivity analyses found that “investing for
early surgery” is the strategy with the highest net benefit in correspondence to a range of values
for different variables of the model, as summarised in Table 85 below.
Table 85: Threshold sensitivity analyses

Variable Threshold values Strategy with highest net benefit
Probability of being operated >0.68 Investment for early surgery
within 48 hours when investing
for early surgery
APPENDIX H 545
Probability of living at home at 1 >0.53 Investment for early surgery

year for early surgery
Probability of living in nursing <0.10 Investment for early surgery
home at 1 year – early surgery
Probability of living in residential <0.15 Investment for early surgery
home at 1 year – early surgery
Mean length of hospital stay for <18.47 days Investment for early surgery
early surgery patients
Number of extra operating lists >4.38 No investment for early surgery
(of 4 hours each)
Proportion of social care costs >0.43 No investment for early surgery
paid by the NHS and local
authorities
Cost per day in hospital >£292.50 Investment in early surgery
20.5.9.1 Probabilistic sensitivity analysis
A probabilistic sensitivity analysis was performed to assess the robustness of the model results to
plausible variations in the model parameters. Probability distributions were assigned to each
model parameter, where there was some measure of parameter variability. We then re-calculated
the main results 10000 times, and each time all the model parameters were set simultaneously,
selecting from the respective parameter distribution at random. Table 86 summarises the type
and properties of distributions used in the probabilistic sensitivity analysis.
Table 86: Description of the type and properties of distributions used in the probabilistic
sensitivity analysis
Parameter Type of distribution Properties of distribution
Baseline risk Beta Bounded on 0 – 1 interval. Derived from sample
size, number of patients experiencing events
Cost Gamma Bounded at 0, positively skewed. Derived from
mean and standard error
Utility Beta Bounded on 0 – 1 interval. Derived from mean and
sample size
Risk ratio Lognormal Bounded at 0. Derived from log (RR) and standard
error of log (RR)
Table 87 reports the distribution, parameters and expected values for each variable of the model.
546 APPENDIX H
Table 87: Distributions, parameters and expected values for probabilistic sensitivity analysis
Name Baseline value Distributions and parameters Expected
(deterministic value
analysis)
EQ- 5D “living own home” 0.64 Beta, Real-numbered parameters, 0.64
alpha = 37.12, beta = 20.88
EQ - 5D – “living in nursing home” 0.35 Beta, Real-numbered parameters, 0.35

and “living in residential home” alpha = 2.45, beta = 4.55
EQ- 5D “living in own home” after 1 0.56 Beta, Real-numbered parameters, 0.56
year alpha = 31.92, beta = 25.08
Cost per hour – consultant (surgeon 108 Gamma, alpha = 15.36583528, 108
and anaesthetist) lambda = 0.142276253
Cost per hour (staff nurse) 21 Gamma, alpha = 15.36583528, 21
lambda = 0.731706442
Cost per hour - ODP 27 Gamma, alpha = 15.36583528, 27
lambda = 0.56910501
Cost per hour -radiographer 25 Gamma, alpha = 15.36583528, 25
lambda = 0.614633411
Cost per hour – senior nurse 27 Gamma, alpha = 15.36583528, 27
lambda = 0.56910501
Operating time per each extra list 4 Triangular, Min = 1, Likeliest = 4, 4
(hours) Max = 7
Initial age 81 None
Length of hospital stay – early 21.6 Log-Normal, u (mean of logs) = 21.6
surgery 3.038030773, sigma (std dev of
logs) = 0.2632965680
Length of hospital stay – late 36.5 Log-Normal, u (mean of logs) = 36.5
surgery 3.562649719, sigma (std dev of
logs) = 0.263296568
No of patients operated in the 434 Poisson, lambda = 434 434
intervention year (2008-09)
No of weekly extra operating lists 5 Triangular, Min = 3, Likeliest = 5, 5
added Max = 7
Overhead cost per minute 1.16 Gamma, alpha = 15.36583528, 1.16
lambda = 13.24640973
APPENDIX H 547
Probability of surgery within 48 0.5289 Beta, Integer parameters only, n = 0.5289

hours without investments in extra 363, r = 192
lists
Probability of surgery within 48 0.6715 Beta, Integer parameters only, n = 0.6715
hours after investments in extra 347, r = 233
lists
Proportion of social care costs 0.6 Triangular, Min = 0.20, Likeliest = 0.6
borne by local authorities 0.60, Max = 1; Expected value: 0.6
Probability of dead – late surgery 0.349 Beta, Integer parameters only, n = 0.349
175, r = 61
Probability of living in own home – 0.434 Beta, Integer parameters only, n = 0.434
late surgery 175, r = 76
Probability of living in nursing home 0.092 Beta, Integer parameters only, n = 0.092
– late surgery 175, r = 16
Probability of living in residential 0.125714 Beta, Integer parameters only, n = 0.12571428
home – late surgery 175, r = 22 6
Relative risk of living in nursing 0.97 Log-Normal, u (mean of logs) = - 0.97
home 0.060565609, sigma (std dev of
logs) = 0.24538297
Relative risk of living in own home 1.16 Log-Normal, u (mean of logs) = 1.16
0.144607796, sigma (std dev of
logs) = 0.08731791
Relative risk of living in residential 1.13 Log-Normal, u (mean of logs) = 1.13
home 0.101743909, sigma (std dev of
logs) = 0.202354755
Relative risk mortality 0.76 Log-Normal, u (mean of logs) = - 0.76
0.280072176, sigma (std dev of
logs) = 0.106163367
Weekly health care costs for 30.8 Gamma, alpha = 15.36583528, 30.8
patients living in a nursing home lambda = 0.498890756
patients living in their own home lambda = 1.552104574
patients living in a retirement home lambda = 0.584252292
Weekly social care costs for 98.1 Gamma, alpha = 15.36583528, 98.1
patients living in their own home lambda = 0.156634407
548 APPENDIX H
Weekly social care costs for 467 Gamma, alpha = 15.36583528, 467
patients living in a residential home lambda = 0.032903288
Weekly social care costs for 678 Gamma, alpha = 15.36583528, 678
patients living in a nursing home lambda = 0.022663474
Daily cost of hospital stay 241.68 Gamma, alpha = 15.36583528, 241.68
lambda = 0.063579259
The conventional way to interpret a cost-effectiveness analysis is to look at the option that is
optimal based on mean results from the probabilistic sensitivity analysis. These findings are
summarised in Table 88 below:
Table 88: Cost-effectiveness findings from probabilistic sensitivity analysis – first year following
investment in extra lists
Strategy Cost Incremental Effectiveness Incremental Incremental 95% CI
Cost ffectiveness C/E ratio
(ICER)
No hospital £46.4K 2.3212
investment for
early surgery
Hospital £47.4K £1.0K 2.3637 0.0425 £/QALY 22542 Cost saving
investment for -
early surgery dominanted
(<48 hours)
The PSA shows that there is a high uncertainty as to whether “investment for early surgery” is
cost-effective compared to “no investment for early surgery”. This uncertainty can be graphically
represented by plotting the results of the incremental analysis for all the 10,000 simulations into a
cost-effectiveness plane. Each point on the scatter plot represents the ICER of investment for
early surgery versus no investment for early surgery for each simulation. The dotted line
represents the £20,000/QALY threshold while the ellipse delimits the 95% confidence interval.
APPENDIX H 549
ICE Scatterplot of
Hospital investment for early surgery (<48 hours) vs. No hospital investment for early surgery
£60K
£50K
£40K
£30K
£20K
Incremental Cost
£10K
£0K
-£10K
-£20K
-£30K
-£40K
-£50K
-£60K
-1.10 -0.90 -0.70 -0.50 -0.30 -0.10 0.10 0.30 0.50 0.70 0.90
Incremental Effectiveness
We found that the strategy of “investment in extra operating lists” was cost-effective in 50% of
the simulations, both at a willingness to pay of £20,000 per QALY and of 30,000 per QALY.
20.5.9.2 Scenario analysis: second year

following implementation
We now compare the non-investment strategy versus the investment strategy, where for the
latter we use data referring to the second year following the introduction of the additional
operating lists. The findings of the deterministic and of the probabilistic cost-effectiveness
analysis are summarised in Table 89 and Table 90 below.
Table 89: Cost-effectiveness results - deterministic analysis – second year following investment
in extra lists
Cost effectiveness Cost-
effectiveness
ratio (ICER)
No hospital investment for early £46.4K 2.32
surgery
Hospital investment for early £47.3K £0.8K 2.413 0.093 £/QALY 9070
surgery (<48 hours)
(with probability of early surgery
from second year of
investment=84.49% and with total
number of patients operated in that
550 APPENDIX H
year = 441)
Table 90: Cost-effectiveness findings from probabilistic sensitivity analysis – first year following
investment in extra lists
Cost effectiveness Cost-
effectiveness
ratio (ICER)
No hospital investment for early £46.4K 2.321
surgery
Hospital investment for early £47.3K £0.8K 2.415 0.094 £/QALY 8933
surgery (<48 hours)
(with probability of early surgery
from second year of
investment=84.49% and with total
number of patients operated in that
year = 441)
The strategy of introducing extra theatre list is therefore cost-effective from the second year of
implementing the change aimed at reducing the waiting time to surgery for hip fracture patients.
20.5.10 Discussion
Our analysis showed that adding extra operating lists as a way of undertaking surgery within 48
hours from admission is slightly above the threshold of 20K/QALYs in the first year of
implementation, but becomes clearly cost-effective from the second year onwards.
However, our cost-effectiveness estimates are likely to be conservative in that we did not look at
the impact of early surgery on the presence of complications. This was because no information on
complications was available from Siegmeth (2005)308, and the other studies from the clinical
review that did report data on complications could not be used since they did not exclude
patients unfit for surgery from their population.
As resources and treatment effects data are based on information received from two specific
hospital settings (John Radcliffe hospital in Oxford and the Peterborough and Stamford Hospital
Foundation Trust), our findings may not be generalised to the whole NHS. For example, for some
hospitals the addition of extra operating lists may not be feasible if no spare theatre capacity is
available for this purpose.
In non-linear models, such as Markov models, there is often a difference between the deterministic and
probabilistic results and in such cases the probabilistic results should take precedence. The findings of the
PSA reported in section 20.5.9.1 show that there is a high uncertainty as to whether “investment for early
surgery” is cost-effective compared to “no investment for early surgery”. If we consider a 95% confidence
interval the base case results did not reach statistical significance (as reported in table 85). Moreover, we
found that the strategy of “investment in extra operating lists” was cost-effective in only 50% of the
simulations, both at a willingness to pay of £20,000 per QALY and of 30,000 per QALY.
A possible extension of the model could look at the possibility of introducing extra operating lists
during the weekend, which would be more expensive than weekdays, as personnel would have to
be paid up to a time and a third more in salary (BMA contract 2003). Patients admitted at
weekends or public holidays tend to do worse (Foss 2006)97). However, most large hospitals have
trauma lists at the weekend, with planned trauma lists built into job plans. The reason why extra
lists were introduced during weekdays in the model that we have developed is because it was
APPENDIX H 551
acknowledged that there are more competing patients for planned trauma lists in those days, for
example patients requiring specialist reconstructions such as pelvic fractures or complex joint
injuries.
552 APPENDIX H
20.6 Cost-effectiveness analysis of Hospital MDR vs Usual care
20.6.1 Introduction
The GDG identified as a high priority area for economic analysis the multidisciplinary management
in hospital for hip fracture patients.
In the economic plan, the clinical question (number 13) linked to this high priority area is the
following:
“What is the clinical effectiveness and cost-effectiveness of the following hospital-based

multidisciplinary rehabilitation programmes:
• Hip Fracture Programme (HFP),
• Geriatric Orthopaedic Rehabilitation Unit (GORU), and
• Mixed Assessment and Rehabilitation Unit (MARU)
versus each other and versus usual inpatient rehabilitation for hip fracture patients?”
The GDG felt that there were sufficient similarities between the GORU and MARU rehabilitation
programmes, and therefore decided to group the evidence for these interventions under the
same category of “GORU/MARU”. A detailed discussion of the main characteristics of each
rehabilitation programme is presented in Chapter 12 of this Guideline, especially in sections 12.1
and 12.2.
A review of the literature was conducted. The literature search and review methods can be found
in section 3. Despite some cost-effectiveness studies were identified, none represented a full cost-
utility analysis which addressed our clinical question. As a consequence, the GDG felt that an
original economic model of the listed interventions was essential in order to inform their
recommendations.
• When published data was not available we used expert opinion to populate the
model.
• We followed the methods of the NICE reference case. Therefore costs were
calculated from a NHS and personal social services perspective. Health gain was
measured in terms of quality-adjusted life-years (QALYs) gained. Both future costs
and QALYs were discounted at 3.5%.

APPENDIX H 553
female) hospitalised for surgery. The model spans over a life-time horizon.
20.6.3 Software
20.6.4 Structure of the model
20.6.4.1 Model cycles at time 0
We develop a Markov model with a cycle length of 3 months. Thus, all events are calculated on a
3 month basis at the end of which patients are in one of the possible health states. As the time
horizon in our model is lifetime, these cycles will keep repeating for the duration of the life
expectancy of the population in the studies.
The specific health states of our Markov model have been determined on the basis of the findings
of the clinical review. During cycle 0 the health states are determined by the types of
complications experienced while in hospital (and while undergoing their rehabilitation
programme). Using evidence from the clinical review, we assume that during cycle 0, patients can
occupy one of the following health states: “not recovered and with no complications”, “not
recovered and with pressure sores”, “not recovered and with moderate delirium”, “not recovered
and with severe delirium”, and “dead”.
This is a graphic representation of cycle 0 of the Markov model:

554 APPENDIX H
Patients who have had

Cycle 0
1st cycle surgery for hip fracture
(0-3 months)
Hospital stay
HFP
or GORU/MARU
Hospital
Intervention
MDR
or usual care
Dead
Not recovered
with no
complications
Not recovered
and with
Not recovered and with pressure sores
severe delirium
Not recovered and with
moderate delirium
Figure 158: Cycle 0 Markov model
The above diagram illustrates that throughout their hospital stay (and hence, while still
undergoing their rehabilitation programme) patients will be considered as “not recovered”. Some
of these “not recovered” patients will not develop any complications, but others will experience
delirium (moderate or severe), or pressure sores.
Evidence and treatment effects on complications – Cycle 0 of the Markov model

The clinical review found evidence of complications only from RCTs of HFP vs usual care. The
following complications were identified:
Table 91: Types of complications identified in the clinical review

Type of complication as reported in the Source
clinical review
Pressure sores Vidan (2005)344
Heart failure Vidan (2005)344
Pneumonia Vidan (2005)344
Confusion Vidan (2005)344
Chest infection, cardiac problem, bedsore Swanson (1998)325
Stroke, emboli Swanson (1998)325
Delirium Marcantonio (2001)203

APPENDIX H 555
Severe delirium Marcantonio (2001)203
The GDG decided to include the evidence on pressure sores from Vidan (2005)344 and on delirium
from Marcantonio (2001)203. This was because of the good quality of the evidence; the reliable
ascertainment of these complications, and their well recognised impact on costs of hospital stay.
The findings of Vidan (2005)344 on “confusion” were not considered in the economic model since
they were not statistically significant and because they did not distinguish between “moderate”
and “severe” confusion, so it was not possible to use these findings alongside those of
Marcantonio (2001)203 on delirium.
The evidence on complications from Swanson (1998)325 was not included in the economic model
since the paper only provided a composite figure for chest infections, cardiac problems and
bedsores and did not distinguish among the different types of complications. As a consequence, it
was not possible to determine the loss in health-related Quality of Life (QoL) due to each
complication and the associated costs.
The evidence on pneumonia (Vidan 2005)344 was also not included in the economic model,
because it showed no difference between the intervention and control group.
The GDG decided to exclude the remaining complications (heart failure, and stroke) due to the
weaker evidence of effectiveness in prevention and the unreliable ascertainment of the
conditions. In particular, it was pointed out that ‘heart failure’ is very difficult to define and
diagnose clinically, and that ‘stroke’ is a whole series of different conditions with hugely differing
origins and outcomes. It should also be noted that it is unlikely that we have introduced a bias in
our model because of the exclusion of these specific outcomes. In fact, despite the clinical review
reported that the relative risk for heart failure and stroke was large and in favour of usual care, it
was also true that they had wide confidence intervals, which meant that the difference was not
statistically significant. Moreover, the GDG agreed that the lower event numbers associated with
usual care was due to the fact that people had been less intensively monitored compared to the
intervention arms of the studies, so that some events may have been missed in the control arm.
As a consequence, the model only looked at the following complications: pressure sores (from
Vidan 2005)344, moderate delirium and severe delirium (Marcantonio 2001)203.
The clinical review did not find evidence of complications for GORU/MARU vs usual care. The GDG
decided to consider the sample complications from the HFP (pressure sores, moderate and severe
delirium) and assume that there was no difference between the intervention and usual care (and
hence to consider a RR equal to 1). This assumption was subject to a sensitivity analysis. Table 91
below reports the transition probabilities for cycle 0 of the Markov model.
Table 92: Transition probabilities - cycle 0 of the Markov model

Transition Probability Usual care HFP GORU
Probability moderate delirium* 22.0% 20.9% (RR 0.95) 22% (RR 1.00)
Probability severe delirium* 28.12% 11.25% (RR 0.4) 28.12% (RR 1.00)
556 APPENDIX H
Probability pressure sores** 16.46% 5.10 % (RR 0.31) 16.46% (RR 1.00)
*= source: Marcantonio (2001)203.
**= source: Vidan (2005)344
20.6.4.2 Cycles 1 – onwards
As for the health states for cycle 1 – onwards, we again used the findings of the clinical review and
assume that, after their hospital discharge (and therefore, after their hospital-based MDR or their
usual care has been completed), patients can transit between the following health states:
“recovered”, “not recovered”, and “dead”.
Vidan (2005)344, Stenvall (2007)320 and Shyu (2008)305 report findings regarding the effectiveness
of hospital MDR programmes versus usual care to help patients recover their pre-fracture
Activities of Daily Living (ADL) levels. The “recovered” health state in our model refers therefore
to the case in which patients have gone back to their pre-fracture ADL levels.
This is a graphic representation of cycles 1 to 3 of Markov model, following hospital discharge:

Cycles 1 to 3
surgery for hip fracture
(3-6 months;
6-9 months;
9-12 months) HFP
Intervention
ororusual
GORU/MARU
care
or usual care
Recovered
Dead
Not recovered
Figure 159: Cycles 1 to 3 of the Markov model
The above diagram illustrates that, up until 12 months, patients who are in the “recovered”
health state can stay in the same state in the following cycles, or can transit to the “dead” health
state.
However, patients in the “not recovered” health state can stay in the same state at the end of
each cycle, or transit to the “recovered” or “dead” states. This is because, from the clinical review,
we only have data regarding the transition of patients from the “not recovered” to the
APPENDIX H 557
“recovered” health state, and these data are only available up until 12 months follow up period.
No clinical data are available regarding the possible transition of the “recovered” patients to the
“not recovered health state”.
From 12 months onwards, we assume that patients will no longer transit from the “not
recovered” to the “recovered” health state, and that patients can only remain in the state they
are in or transit to the “dead” state. This is because no clinical data are available from the clinical
review after that point. Hence, the relevant transitions between health states after 12 months will
be:

Cycle 4 – onwards surgery for hip fracture
(after 12 months)
HFP
Intervention
GORU/MARU
ororusual care
or usual care
Recovered
Dead
Not recovered
Figure 160: Cycle 4 - onwards of the Markov model
That is, from cycle 4 onwards, patients who are in the “recovered” health state will stay in that
state or transit to the “dead” state. Similarly, patients in the “not recovered” health state will
remain in that state or transit to the “dead” state. The GDG noted that the assumption that
people remain in the same health state from 12 months onwards is clinically reasonable, as from
that time patients’ health state will no longer be influenced by their hip fracture. All possible
events after this time (e.g. death, falls, needs for care home etc) will take place at rates that are
consistent and in line with those of the general population and that therefore will no longer be a
consequence of the hip fracture nor of the specific rehabilitation programme received.
Whether they are “recovered” or “not recovered”, the place of residence at hospital discharge for
patients will also be affected by whether they received usual care, HFP or GORU/MARU as a form
of rehabilitation programme. This circumstance is represented in Figure Figure 161: Place of
residence at discharge below:
558 APPENDIX H
Residential place
at discharge Living own
home
“Recovered”
Dead
or
“Not recovered”
Living long
term care
Figure 161: Place of residence at discharge
No evidence is available from the clinical review regarding whether patients discharged to their
own home would then transit to the “living in long term care” setting in subsequent cycles of the
model, and vice versa. Hence, we make the assumption that patients will keep living in the same
place of residence they had when they were discharged from hospital, and that they can only
transit to the “dead” state in the following cycles.
Evidence and treatment effects on recovery of ADL levels and on place of residence at discharge
Table 93 reports the levels of the transition probabilities used in the model
Table 93: Transition probability of Not Recovery of ADL pre-fracture levels

Transition probability of Not Usual care HFP GORU/MARU
Recovery of ADL pre-fracture levels
At 3 months(1) 0.73 0.5767 (RR=0.79) 0.5694 (RR=0.78)
(1)
At 6 months 0.67 0.5293 (RR=0.79) 0.5226 (RR=0.78)
(2)
At 9 months 0.63 0.4977 (RR=0.79) 0.4914 (RR=0.78)
(3)
At 12 months 0.59 0.4661 (RR=0.79) 0.4602 (RR=0.78)
(1) 344
Data at 3 and 6 months from Vidan
(2)
Data at 9 months obtained with a linear extrapolation from the transition probabilities in
Vidan344
(3)
Data at 12 months pooled from Vidan, Shyu and Stenvall305,320,344
APPENDIX H 559
As for the place of residence following hospital discharge, we use the following treatment effects
in our model:
Usual care HFP GORU/MARU

Probability of returning to own home* 0.71 0.8094 (RR=1.14) 0.7881 (RR=1.11)
*source: NCGC meta-analysis of clinical trials
20.6.5 Evidence and treatment effects on mortality
In our model we distinguished two types of mortality: short-term mortality (within 12 months
from the start of the rehab programme) and long-term mortality (after 12 months).
SHORT-TERM MORTALITY
In order to take into account the difference in mortality due to the intervention, we used the data
from the RCTs included in our meta-analysis to estimate mortality. The data available from the
RCTs can be found in Table Table 93.
Table 94: Proportion of patients dead at different time points

6 months 12 months
Usual care1 16.73% 21.38%
2
HFP NA 17.32% (RR 0.81)
2
GORU/MARU 13.22% (RR 0.79) 20.31% (RR 0.95)
1 Data pooled from the usual care arms of RCTs in the clinical review
2 RR calculated compared to usual care
Data were available for usual care and GORU at 6 and 12 months from randomisation. Only 12
month data were available for the HFP intervention.
When more than one time points was available (i.e. for the usual care and GORU/MARU arms),
the probability of dying was calculated from the data reported in Table 4 as follows:
Prob_die_y to x = (% dead time x - % dead time y)/(1 - % dead time y)
Where:
Prob_die_y to x is the probability of dying from time y to the following time x
• “% dead time x” is the proportion of patients dead at time x

• “ % dead time y” is the proportion of patients dead at time y
To convert probabilities into a 3-month transition probability, which is the cycle length of the
model, we used the formula:
1 - exp((ln(1- Prob_die_y to x))/((x-y)/3)

560 APPENDIX H
Where x and y are the initial and final time points of the interval considered, exp(a)=exponential
of a; and ln(a)=natural log of a.
LONG-TERM MORTALITY
The mean age of the patients when entering the model was 81 as this was the mean age of
patients in the RCTs.
Life expectancy in people who were alive one year after a hip fracture was assumed to be the
same as the general population in England and Wales, as reported in a study (Parker1992, citing
Elmerson1988)268. The remaining life expectancy for the participants of the RCTs was obtained
from the Life Tables for the general population of England and Wales in the year 2005-2007 from
the Government Actuary Department
(http://www.gad.gov.uk/Documents/Demography/EOL/ILT%202005-07/wltewm0507.xls).
The value was adjusted for the ratio male/female corresponding to the patients characteristics in
the RCTs as follows:
Total LE = LEfemale * %female + LEmale * %male
20.6.6 Utilities data
20.6.6.1 Utilities for cycle 0 (0-3 months)
Utilities indicate the preference for health states on a scale from 0 (death) to 1 (perfect health).
Quality of life values are attached to all health states.
Stage 0 of the model refers to the first three months of the Markov model. They capture the time
that the patients spend in hospital, during which they undergo a surgical treatment of the
fracture, following which the rehabilitation process starts.
The utility weights for the health states in cycle 0 are summarised in Table 94.
Table 95: Utility weights for cycle 0

Table 5: Utility Base case value Source
weights for cycle 0Health
state
“Not recovered, no 0.314 ADL levels from Kennie
complications” (1988)176; EQ-5D scores from
Tidermark (2002)328
“Not recovered and with 0.19 Essex (2009)86
pressure sores”
“Not recovered and with 0.314 ADL levels from Kennie
moderate delirium” (1988)176; EQ-5D scores from
Tidermark (2002)328
“Not recovered and with 0.25 NICE clinical guideline on
severe delirium” Delirium224
APPENDIX H 561
We assume that the utility for the “not recovered, no complication” health state in the first three
months is the same as that of the “Not recovered” health state after the hospital discharge (i.e.
after the first cycle). The following paragraph explains how the utility for the “not recovered, no
complication” health state is obtained.
The NICE guideline on Delirium224 reports utility weights for patients with moderate and severe
delirium using the finding of Ekman (2007)76 on patients with dementia. Ekman (2007)76 estimates
that the mean utility score for mild, moderate and severe dementia correspond to 0.62, 0.40 and
0.25 respectively. As for pressure sores, Essex (2009)86 reports an EQ-5D score of 0.19 for patients
experiencing this complication. EQ-5D scores were obtained from a survey of a sample of 6
patients with pressure ulcers.
We proceeded by selecting the lowest EQ-5D score between the “not recovered with no
complication” health state and the EQ-5D linked with that particular complication (moderate
delirium, severe delirium or pressure sores). Thus, being the utility for “moderate delirium” 0.4,
and being this utility higher than the one of the “not recovered with no complication” health state
(0.4 vs 0.314), we selected the latter also for the “not recovered and with moderate delirium”
health state.
However, the utility score for patients with severe delirium identified in the literature was lower
than then the score for the “not recovered, no complications” health state (0.25 vs 0.314).
Similarly, the utility score for pressure sores identified in the literature was lower than the one of
the “not recovered, no complications” health state (0.19 vs 0.314). Hence, we used the EQ-5D
score for those specific complications (severe delirium, pressure sores) in our model.
20.6.6.2 Utilities for cycles 1 – onwards (3

months – onwards)
In order to assign an utility level to each of the health states for the model in cycles 1-onwards
(that is, “recovered” and “not recovered”), we proceeded by using the RCT included in our clinical
review by Kennie et al (1988)176 which reports the number of patients (in the treatment and
control group) classified according to their level of independence in activities of daily living before
admission (i.e. before the hip fracture) and at entry into study (i.e. before the rehabilitation
program has started). This information is summarised in Table 95 and Table 96 below.
Table 96: ADL levels before admission for treatment and control group (source: Kennie et al
1988)176
Independence in activities of Treatment group Control group
daily living before admission (n=54) (n=54)
(Katz index)
A 21 28
B 14 11
C 6 6
D 3 3
E 2 1
F 2 1
562 APPENDIX H
G 1 1
Not classified 5 3
Table 97: ADL levels at entry into study for treatment and control group (source: Kennie et al
1988)176
Independence in activities of Treatment group Control group
daily living at entry into (n=54) (n=54)
study (Katz index)
A 0 0
B 1 0
C 1 0
D 2 3
E 18 19
F 23 16
G 7 15
Not classified 2 1
Source: Kennie et al (1988)176
We use the data for the “independence in ADL before admission” to calculate the proportion of
independent and dependent patients that are in the “recovered” health state. Similarly, we use
the information on ADL for patients at entry into study to calculate the proportion of independent
and dependent patients that are in the “not recovered” health state.
As a consequence, we have:
• % of patients with A-B score in the “recovered” state:

(21+14) (from the treatment group) + (28+11) (from the control group)/100 = 74%
• % of patients with A-B score in the “not recovered” state:

1/100 = 1%
Hence, in the “recovered” health state, 74% of patients have an ADL score of A-B, and 26% of
patients in the same state have an ADL score of C-G. On the other hand, in the “not recovered”
health state, only 1% of patients have ADL score of A-B, the rest having an ADL score of C-F.
For each of these two states we calculated the composite utility, that is the utility for the
“independent” and for the “dependent” patients. Tidermark (2002)328 reports EQ-5D scores
associated with ADL scores of A-B and C-F for hip fracture patients at 4 months after the fracture.
These weights correspond to: 0.68 for ADLs of A-B, and to 0.31 for ADLs of C-G.
Using the proportion of patients who were reported as independent and as dependent before
admission for the “recovered” health state we have:
74% * 0.68 = 0.053
26% * 0.31= 0.081

APPENDIX H 563
Thus, the utility weight for “recovered” health state corresponds to 0.584
As for the “Not recovered” health state we have:
1% *0.68 = 0.0068
99%*0.31 = 0.307
Thus, the utility weight for “not recovered” health state is: 0.314. We summarise these findings in
Table 98:
Table 98: Utility weights for health states in cycles 1 -onwards

Health state EQ-5D Source
“Recovered” 0.584 ADL levels from Kennie
(1988)176; EQ-5D scores from
Tidermark (2002)328
“Not recovered” 0.314 ADL levels from Kennie
(1988)176; EQ-5D scores from
Tidermark (2002)328
20.6.7 Calculating QALYs gained
For each strategy (HFP, GORU/MARU and usual inpatient rehabilitation), the expected QALYs in
each cycle are calculated as follows:
Expected QALYs = Σ (Ui x Pi )
where
Ui = the utility score for health state i
Pi = the proportion of patients in health state i
and where health state i could be any of the health states reported in the Figures 147 and 148.
The proportion of patients in each health state depends on the effectiveness of the treatment,
and on the proportion of patients still alive, which falls as the number of cycles and therefore age
increases.
incremental QALYs gained associated with a treatment strategy are calculated as the difference
between the expected QALYs with that strategy and the expected QALYs with the comparator.
564 APPENDIX H
20.6.8 Cost data
20.6.8.1 Cost data: cycle 0 (hospital stay)
During hospital stay, the costs will depend on the rehabilitation programme, the length of hospital
stay and health state related costs. We analyse each category in turn.
Cost of the rehabilitation programme

The NICE “Guide to the methods of technology appraisal” points out that national data based on
healthcare resource groups (HRGs), such as the Payment by Results tariff, are a valuable source of
information for resource use and costs and should be considered for use whey they are
appropriate and available (“Guide to the methods of technology appraisal”, 2008, page 40).
However, data based on HRGs may not be appropriate in all circumstances, especially when the
definition of the HRG is broad or the mean cost probably does not reflect resource use in relation
to the interventions we are evaluating.
In our case, we would be using the HRG4 as the source to cost our rehab programmes. In the
document: “Casemix Service HRG4 - Guide to unbundling” it is pointed out that the HRG4 refers
to cases of Discrete Rehab services:
“[..] only discrete rehabilitation activity and costs should be reported using the rehabilitation HRG4
categories, for the reference costs collection.”
And the 2007 document on Collection Guidance on Reference Costs for 2006-07 specifies that:
“Rehabilitation HRGs are only generated where care is identified as taking place under a specialist
rehabilitation consultation or within a discrete rehabilitation ward or unit. [..] Where a patient is
not admitted specifically to a rehabilitation unit or where rehabilitation treatment is undertaken
without transfer to a specialist consultant, or without transfer to a rehabilitation unit, this should
not be reported as discrete rehabilitation”.
It would therefore seem that whilst this definition could apply to the GORU/MARU model (where
a patient is discharged from the orthopaedic unit and admitted to a separate geriatric
orthopaedic unit to receive the rehabilitation), it could not reflect the case of a HFP, where a
patient is not usually discharged to the care of a specialist rehabilitation consultant.
Thus, whilst we could use the HRG4 to cost a GORU and a MARU programme, we would not be
able to use it to cost a HFP.
As a consequence, the GDG decided to evaluate the cost of the different rehabilitation
programmes using the level of resources specified in the different RCTs included in the clinical
review. When necessary, such levels have been adjusted by expert opinion to reflect a pattern of
care closer to the UK health care setting (see below).
The resources used in the different RCTs have been reported as incremental resources used with
respect to the usual care arm of the study. Using information on unit costs for NHS personnel
provided by the PSSRU 2009, we were then able to estimate the incremental cost of both HFP and
GORU/MARU with respect to usual care.
Moreover, it is important to note that the level of resources used in the two hospital-based MDR
programmes are calculated in such a way to reflect the length of hospital stay of the patients in
our model. Thus, we use the length of stay for the HFP to calculate the incremental resources and
APPENDIX H 565
costs for that programme, as follows. Similarly, we use the length of stay for GORU/MARU to
calculate the incremental resources and costs for that rehab programme.
Tables 9 – 11 summarise the incremental resources used in the HFP and the GORU/MARU
programme, compared to usual care.
Table 99: Incremental resource use for GORU/MARU programme versus usual care
Staff resources Incremental resources Source Unit cost (source: Incremental
used, based on a LOS of PSSRU 2008/09), cost
32.88 days £ per hour
Orthogeriatrician Two consultant ward Kennie et al £108 £372.6
rounds (0.25/hour per (1988)176
patient each) and one
weekly conference
(0.25/hour = 0.75 hour
per week per patient
0.75*4.6 weeks = 3.45
hours per patients
Physiotherapist 8.5 hours per patient Naglie 2002222 £23 £195.5
Occupational 5 hr/patient GDG adjustment £23 £115

therapist from the 7.5
hr/pt reported in
Naglie222
Nurse Initial assessment Naglie 2002222 Nurse team £75.6
within 72 hours (0.5 leader: £27 £58.8
hour per patient) and Nurse day ward:
twice weekly £21
assessment afterwards
(0.25*2)/hour per
patient
0.5+0.5*4.6 weeks= 2.8
hours per patient
Social worker -0.4 hour per patient Naglie 2002222 £29 (from -£11.6
community data)
Dietician -0.4 hour per patient Naglie 2002222 £23/ -£9.2
Total incremental cost for GORU/MARU over usual care: £721 (with
generic
nurse, Band
5);
£738 (with
team leader
nurse, Band
6)
566 APPENDIX H
Table 100: Incremental resource use and incremental cost for HFP over usual care
Staff resources Incremental resources Source Incremental cost
used based on a LOS of (using PSSRU
25.5 days 2008/09 unit costs)
Orthogeriatrician Initial assessment 0.5 Cameron (1993)44; £108*6.625=£715.50
hour per patient, and Shyu (2008)305;
subsequently 0.25 hour Marcantonio203
per day:
0.50 + 0.25*24.5 =6.625
hour per patient
Physiotherapist 0.5 hour per patient per Cameron (1993)44 £23*12.75=£293.25
or nurse day:
0.50*25.5=12.75 hours
Total incremental cost of HFP over usual care: £1009
Hence, the incremental cost for HFP over usual care is £1009, while for the GORU/MARU
programme it is £721 (with generic nurse) or £738 (with team leader nurse).
Health state related costs in cycle 0

To calculate the health state costs during the hospital stay, we used the NHS reference cost for
excess bed days reported in table 28 below. The excess bed day cost is the cost per day for days
exceeding the trimpoint, a cut-off that determines patients with exceptionally long stay, and as
such usually estimates the cost of care without the cost of procedures (i.e. without the cost of the
surgery. These costs reflect the presence of complications experienced by hip fracture patients
during their entire hospital stay. Moreover, they distinguish between “major” and “intermediate”
complications, thus allowing users to take into account the different degrees of resource use.
Table 101: National Schedule of Reference Costs Year : '2008-09' - NHS Trusts and PCTs
combined Non-Elective Inpatient (Long Stay) Excess Bed Day HRG Data for hip procedures
Currency Currency Description Activity National
Code Average Unit
Cost
HA11A Major Hip Procedures Category 2 for Trauma 360 £243
with Major CC
HA11B Major Hip Procedures Category 2 for Trauma 620 £242
with Intermediate CC
HA11C Major Hip Procedures Category 2 for Trauma 162 £220
without CC
HA12B Major Hip Procedures Category 1 for Trauma 9,760 £237
with CC
APPENDIX H 567
HA12C Major Hip Procedures Category 1 for Trauma 1,230 £226

without CC
HA13A Intermediate Hip Procedures for Trauma with 14,891 £240
Major CC
HA13B Intermediate Hip Procedures for Trauma with 12,856 £249
Intermediate CC
HA13C Intermediate Hip Procedures for Trauma 2,972 £223
without CC
HA14A Minor Hip Procedures for Trauma with Major 5,195 £234
CC
HA14B Minor Hip Procedures for Trauma with 5,808 £245
Intermediate CC
The GDG decided to calculate a weighted average cost of the different categories of hip fractures
taking into account the level of activity associated with each procedure.
To cost the health state “not recovered with pressure sores” we use evidence from Bennett
(2004)17 regarding the cost of pressure ulcer treatment in the UK. The paper calculates the daily
cost of treating pressure ulcers looking at resources such as nurse time (dressing changes, patient
repositioning and risk assessment) dressings, antibiotics, diagnostic tests, and support surfaces.
These costs do not include inpatient costs, but assume that the patients are cared for in an
institutional setting (hospital or long-term care).
Pressure ulcers can have a different “grade”, ranging from 1 to 4 as their complexity increases.
However, the GDG emphasised that the published evidence on the incidence of the different
types of pressure sores in hip fracture patients reports many contradictory findings from which it
is difficult to draw definitive conclusions when it comes to costs. We followed the evidence in
Rademakers (2007)278 and assumed that 97% of the pressure ulcers were of grade 2, and 3% of
grade 3 or 4.
Bennett (2004)17 reports a daily cost for grade 2 pressure sores of £42, and of £50 for grade 3 and
4. These daily costs refer to patients who do not develop any further complications linked to the
pressure sores (such as critical colonisation, cellulites, or osteomyelitis), as no evidence on such
conditions was available from the RCTs included in our clinical review. Table 102 reports the total
daily cost for the “not recovered with pressure sores” health state.
Table 102: Total daily hospital cost for patients with pressure sores
Category of cost Level of cost
Daily inpatient hospital cost without £220.07
complications
Daily cost of grade 2 pressure sore 0.97*£45
Daily cost for grade 3 and 4 pressure sore 0.03*£50
Total daily cost for patients with pressure £265.22
sores
For the cost of the health state “not recovered with moderate delirium” we used the mean
weighted average cost for minor complications (£237), and for the cost of the health state “not
568 APPENDIX H
recovered with severe delirium”, we used the mean weighted average cost for major and
intermediate complications (£242.89). One limit with this approach is that all patients with
moderate delirium are assumed to have undergone a Major Hip Procedures Category 1 for
Trauma. Even if the difference between the two cost figures is quite low (£5.89) we test the
impact of this assumption on the base case findings in a sensitivity analysis.
It has to be emphasised that this approach to calculate the health state costs in cycle 0 is
necessary in that only figures regarding the total length of hospital stay are available from the
evidence included in our clinical review. Ideally, we would have needed information regarding the
additional length of hospital stay for the patients experiencing a particular complication, both for
the control and for the intervention groups, but this information was not available from the
clinical review. Moreover, even if Marcantonio (2001)203 reports the hospital days of delirium per
episode, it does not distinguish between the two types of delirium (moderate and severe) that
correspond to our health states in cycle 0 of the Markov model, and only gives an overall figure
for all types of delirium.
Table 103: Daily inpatient average cost for health states in cycle 0
Health state Average daily cost Source
Not recovered and with £220.07 Mean weighted average of excess bed days costs –
no complications NHS reference costs 2008-08 Major, Intermediate
and Minor Hip procedures with no complications
Not recovered and with £265.22 See Table 29
pressure sores
Not recovered and with £237 Mean weighted average of excess bed days costs –
moderate delirium NHS reference costs 2008-08. Major, Intermediate
and Minor Hip procedures with minor complications
Not recovered and with £242.89 Mean weighted average of excess bed days costs –
severe delirium NHS reference costs 2008-08. Major, Intermediate
and minor hip procedures with intermediate and
major complications
Evidence and treatment effects on length of hospital stay

The studies included in the clinical review comparing the GORU/MARU programme vs usual care
only reported the total length of hospital stay for patients in the intervention arm of the study.
Hence, no information was available to evaluate the number of days patients spent in the
orthopaedic ward and the number of days they spent in the orthogeriatric rehabilitation hospital
ward.
To calculate the length of stay at baseline (i.e. the usual care arm of the model), we pooled the
data for the usual care arm from all RCTs included in the clinical review. Table 104 reports the
relevant values for hospital length of stay used in the model:
APPENDIX H 569
Table 104: Mean length of hospital stay

Mean length of stay - usual care (days) 31.56
Mean difference length of stay - HFP (days) -6.06

Mean difference length of stay - GORU/MARU (days) 1.32
20.6.8.2 Cost data: cycle 1 - onwards
From cycle 1 – onwards, the costs for our model will depend on the place of discharge (whether
own home or residential or nursing home), which in turn will affect the level of health care
services and social care used, and on the probability of hospital readmissions.
Hospital readmissions
The RCTs on HFP versus usual care included in the clinical review did not report any information
over the reasons for hospital readmissions nor the associated length of stay.
Two RCTs on GORU/MARU versus usual care (Galvard 1995 and Stenvall 2007)107,320 reported data
on length of stay following readmission available from two RCTs on GORU/MARU. However, the
reasons for readmissions (whether orthopaedic-related or any other medical reason) were only
given in Galvard (1995)107.
Given the lack of data from the clinical review, the GDG decided to assume that readmissions
were composed by an equal proportion of patients are readmitted for surgery, medicine and
rehabilitation reasons. This assumption was also supported by unpublished data on readmissions
following hip fracture obtained from a GDG member and based on hospital records from
Peterborough and Stamford NHS Foundation Trust.
As for the length of stay following a hospital readmission, we followed the most recent clinical
paper (Stenvall 2007)320 and assumed a LOS for readmission for usual care is 11 days and in the
intervention (whether GORU/MARU or HFP) is 7 days.
The cost data for the hospital readmissions were obtained from Czoski-Murray (2007)63, which
reports the unit costs for inpatient stay (at 2002 prices) for surgery (£381), medicine (£282) and
rehabilitation (£188). These costs are based on Netten et al (2002)241. The mean unit cost for
inpatient stay for readmissions (at 2009 prices) was estimated at £367.00. This price has been
obtained using the annual percentage increases for prices of hospital and community health
services (HCHS) for 2002/03 – 2008/09 reported in the PSSRU 2009 report61.
Community care costs for the “recovered” and “not recovered” health states when discharged to
own home
To analyse the costs associated with the “recovered” and the “not recovered” health states we
need to take in to consideration whether patients are discharged to a long-term care setting or to
their own home.
The GDG decided that in determining the level of community (that is, health care and social care)
resources used after the hip fracture and after the rehabilitation programme it was important to
reflect the level of “dependency” and “independency” in activities of daily living of patients in each
of the health state.
570 APPENDIX H
The PSSRU 2009 identifies five different “packages” of community care provided in the home
setting of the patient (also known as “domiciliary care”), according to the different level of
dependency in the activities of daily living of the recipients. These packages of care are
summarised in Table 105 below.
Table 105: Weekly costs of community care packages – excluding accommodation and living
expenses. Source PSSRU 2009.
Community Description of the Weekly cost Average weekly Average weekly
care package level of functional (excluding cost of social care cost of health care
ability of the recipient accommodation, services services
of care living expenses
and
independently
provided home
care)
“Very low Mrs A. had problems £49 £41.3 £7.70 for a 11.7
cost” with three activities of (£18.10 of home minutes of GP
daily living: stairs, care (one hour of surgery visit (one
getting around weekly local every four weeks)
outside, and bathing. authority-
Her problems organised home
stemmed from a care)) and £23.20
previous stroke. of meals on
wheals)
“Low cost” Mrs B. had problems £87(1) £72 of home care £14.3 (of which
with three activities of (4 hours of local £6.60 of
daily living: stairs, authority- community nurse
getting around outside organised home (one visit per
and bathing. Her care) month) and £7.70
problems stemmed of one GP visit (one
from arthritic every four weeks))
conditions and
cardiovascular
disease.
“Median Mrs C. had problems £188 £181 of home care £7.70 for a 11.7
cost” with four activities of (10 hours of weekly minutes of GP
daily living: stairs, local authority- surgery visit (one
getting around organised home every four weeks)
outside, dressing and care)
bathing.
“High cost” Mr D. had problems £273 £216 £58
with seven activities of (of which £181 of £26 of community
daily living: stairs, home care (10 nurse (once a
getting around outside hours of weekly week); £24 for two
and inside the house, local authority- monthly OT visits;
APPENDIX H 571
using the toilet, organised home £7.70 for a 11.7

transferring between care) and £35 for a minutes of GP
chair and bed, day centre surgery visit (one
dressing and bathing. attended once a every four weeks)
His problems week)
stemmed from
arthritic conditions
and a previous stroke.
“Very high Mrs E suffered from £576 £542 of home care £34
cost” dementia and needed (30 hours of weekly £26 of community
help with nine local authority- nurse (once a
activities of daily organised home week);
living: stairs, getting care) £7.70 for a 11.7
around outside and minutes of GP
inside the house, using surgery visit (one
the toilet, transferring every four weeks)
between chair and
bed, dressing, bathing,
washing and feeding.
(1) Please note that the cost figure reported in the PSSRU 2009 for “low cost” is not correct
(£129) as the cost for the independently provided health care has not been subtracted (£42).
The correct figure should be £87.
We used the data from Kennie (1988)176 to determine the proportion of patients with level of
independence from A to G to attribute the community care costs to the “recovered” and “not
recovered” health state.
For both health states (“recovered” and “not recovered”), we assume that patients with ADL
score A or B do not incur in any domicilary care cost. However, we assume that the same type of
patients will each visit the GP once weekly.
The weekly health and social care costs are calculated by multiplying the weekly unit cost of the
different type of care (as obtained from the PSSRU 2009) times the proportion of patients with
the corresponding ADL score in the specific health state and times the level of resources used
(which depend on the level of dependency). The health and social care costs for the “recovered”
and “not recovered” health states are described in Table 106 and in Table 107 below.
Table 106: Health and social care costs for patients in the “recovered” health state discharged at
their own home
ADL % ADL in Unit health care Health cost Unit social Social care costs
recovered costs for care costs for recovered
state recovered state
state
A 0.454 7.7 3.4958 N/A N/A
B 0.231 7.7 1.7787 N/A N/A
C 0.112 7.7 0.8624 41.3 4.6256
D 0.056 7.7 0.4312 41.3 2.3128
572 APPENDIX H
E 0.028 14.3 0.4004 72 2.016

F 0.028 7.7 0.2156 181 5.068
G 0.018 58 1.044 216 3.888
NC 0.073 34 2.482 542 39.566
Total 10.7101 Total 57.4764
Annual health 556.925 Annual 2988.77
care cost social care
cost
Table 107: Health and social care costs for patients in the “not recovered” health state
discharged at their own home
ADL % ADL in Unit health care Health cost for Unit social Social care costs
not costs (£) recovered state care costs (£) for recovered
recovered (£) state (£)
state
A 0 0 0 N/A N/A
B 0.009 7.7 0.0693 N/A N/A
C 0.009 7.7 0.0693 41.3 0.3717
D 0.046 7.7 0.3542 41.3 1.8998
E 0.342 14.3 4.8906 72 24.624
F 0.362 7.7 2.7874 181 65.522
G 0.204 58 11.832 216 44.064
NC 0.028 34 0.952 542 15.176
20.9548 151.658
Annual health 1089.65 Annual social 7886.19
care cost care cost
Hence, the annual health and social care costs for the “recovered” and the “not recovered” health
state are:
Table 108: Annual health and social care costs for the “recovered” and the “not recovered”
health state
Annual health care costs £557 £2989
Annual social care costs £1090 £7886
Total community care costs £1647 £10875
While the health care costs will be fully funded by the NHS, the social care costs will only be
generally partially funded by the local councils71, 348, 72, 144. It was not possible to identify a
national average for the social care costs funded by local authorities in the published literature,
and as a consequence an assumption had to be made regarding the proportion of this care that
was publicly funded. In the base case analysis, we assume that 60% of social care costs are funded
APPENDIX H 573
by the local authorities, and are therefore includable in the model, and we then test this
assumption in a sensitivity analysis.
Community care costs for the “recovered” and “not recovered” health states when discharged to
long term care
The cost of long term care used in the model was estimated from the unit cost of stay in private
nursing homes, private residential care, voluntary residential care and local authority residential
care facility for older people. The care package costs per permanent residential week are
described in Table 109.
Table 109: Weekly long term care costs for patients not discharged to their own home.
(Source: PSSRU 2009).
Type of long term care Weekly health care costs Weekly fees
(minus living
expences)
Private nursing home £30.80 £678
£30.00 (GP weekly home visit)
£0.80 (community nursing)
Private residential care £26.3 £467
Voluntary residential care £28.7 £470
Local authority residential care £20.9 £902
£10.60 (GP weekly surgery visit)
These unit costs include the cost of external services such as community nursing, GP services as
well as personal living expenses. They also include capital costs for the local authority residential
care, and fees for the private and voluntary residential care. We subtracted £9.20, the cost of
personal living expenses per week, from each unit cost and estimated £717.05, the weighted
average of £708.80, £493.80, £489.80 and £913.80, to be the weekly unit cost of long term care.
By also subtracting the health care costs, we get: £557.64 as the weekly fees for long term care
(£28997 per year). The (weighted) health cost per week is £27 (£1404 per year).
The weighting is based on the distribution of residents, 65 years and older, in care homes in 1996.
It was reported that in nursing homes, local authority, private and voluntary residential homes the
number of residents were 5746, 5476, 2791 and 3664 respectively (Netten et al 1998)240. A
similar approach is also followed in the cost-effectiveness analysis conducted in the NICE Delirium
Guideline224.
It is important to note that, contrary to the community care packages for domiciliary care, we
could not distinguish the level of long-term residential care according to the level of
“dependency” in ADL of the patients in the different health state. Hence, the same figure for
community costs had to be used both for the “recovered” and “not recovered” health states if not
discharged at their own home.
574 APPENDIX H
As with the domiciliary care, the health care costs in table 18 will be fully funded by the NHS, but
the residential fees for long term care will only be generally partially funded by the local councils.
Moreover, only a very small proportion of patients (about 2%) qualifies for fully funded NHS care
(the so called “continuing care”)71, 348, 72, 144. It was not possible to identify a national average for
this figure in the published literature, and as a consequence an assumption had to be made
regarding the proportion of residential costs in long term care paid by local authorities. In the
base case analysis, we assume that 60% of residential fees costs are funded by the local
authorities, and then change this assumption in a sensitivity analysis.
20.6.9 Cost-effectiveness findings for base-case analysis
In the base case analysis, HFP is the dominant strategy (more effective, less costly) than both
GORU/MARU and usual care.
Table 110: Cost-effectiveness findings from the deterministic base case analysis
Strategy Cost Incremental Effectiveness Incremental Incremental cost-
(£000) Cost* (QALYs) Effectiveness* effectiveness
(£000) (QALYs)
HFP £34 3.75
GORU/MARU £36 £2 3.62 -0.13 (Dominated by
HFP)
Usual care £59 £26 2.73 -1.02 (Dominated by
HFP)
*Compared with HFP
Table 110 below shows the breakdown of the different cost categories for the three strategies of
the deterministic base case
Table 111: Cost breakdown for usual care, HFP and GORU/MARU
Resource item Usual Care HFP GORU
Rehab cost (initial costs)* - 1009 729
Complications* - -548 217
Readmission 969.5 762.2 535.3
Health care costs – living in 9178 4032 3738
own home
Social care costs – living in 14,000 5,000 5,000
own home
Health care costs – 2,615 1,801 1930
residential and nursing
home
Social care costs (fees) - 32,000 22,000 24,000
residential and nursing
home
Total cost 58762.50 33595.2 35203.3
* calculated incrementally vs usual care
APPENDIX H 575
20.6.9.1 Sensitivity analyses
In order to check how robust the findings in the deterministic base case analysis reported in
table 109 are, we ran a series of sensitivity analyses.
The results were not sensitive to changes in several parameters (length of hospitals stay, cost
of long-term care, proportion of long-term care borne by the NHS and PSS).
However, the results were sensitive to changes in the probability of returning home for both
HFP and GORU/MARU. In the base case analysis, the probability of returning home for the
HFP is 0.81 (RR of HFP vs usual care: 1.14), and for GORU/MARU it is 0.79 (RR of GORU/MARU
vs usual care: 1.11). The findings of a two-way sensitivity analysis on such probabilities are
reported in the graph below.
Net Monetary Be nefit (wtp=20000.) Sensitivity Analysis on

p_ReturnHome _GORU and p_Re turnHome _HFP
1.0
HFP
GORU/MARU
Usual care
p_ReturnHome_HFP
0.8
0.5
0.3
0.0
0.0 0.3 0.5 0.8 1.0
p_ReturnHome_GORU
A threshold sensitivity analysis shows that:
a) If the probability of returning home for HFP <0.77 (it is 0.81 in the base case scenario),
then GORU/MARU is the most cost-effective option at a willingness to pay threshold of
£20,000 per QALY.
576 APPENDIX H
b) If probability of returning home for GORU/MARU <0.83 (it is 0.79 in the base case
scenario), then HFP is the most cost-effective option at a willingness to pay threshold of
£20,000 per QALY.
A two-way sensitivity analysis on a) the proportion of social care costs borne by the NHS and
PSS for patients living in their own home and b) the proportion of social care costs borne by
the NHS and PSS for patients living in a residential or nursing accommodation found that HFP
is always the most cost-effective option.
A probabilistic sensitivity analysis was performed to assess the robustness of the model
results to plausible variations in the model parameters.
Probability distributions were assigned to each model parameter, where there was some
measure of parameter variability. We then re-calculated the main results 10,000 times, and
each time all the model parameters were set simultaneously, selecting from the respective
parameter distribution at random. Table 112 describes the type and properties of the
distributions used in the probabilistic sensitivity analysis.
Baseline risk Beta Bounded on 0 – 1 interval.
Derived from sample size,
number of patients
experiencing events
Cost Gamma Bounded at 0, positively
skewed. Derived from mean
and standard error
Utility Beta Bounded on 0 – 1 interval.
Derived from mean and
sample size
Risk ratio, length of stay Lognormal Bounded at 0. Derived from
log and standard error of log
Mean differences (e.g. in Normal Derived from mean and
length of stay, time of standard deviation
therapies, etc.)
Table 113 summarises the distribution, parameters and expected values for each variable of
the model.
Table 113: Probabilistic sensitivity analysis: formulas and expected value

Variable name Formula Expected Deterministic
value value
APPENDIX H 577
Cost per hospital bed day (patients with Gamma 237.13 237
moderate delirium) alpha = 15.366,
lambda = 0.0648;
Cost per hospital bed day (patients with Gamma 220.14 220.07
no complications) alpha = 15.366, lambda =
0.0698;
pressure sores) alpha = 15.366, lambda =
0.057984;
severe delirium) alpha = 15.366, lambda =
0.0633;
Annual health care costs – “not Gamma 2988.91 2989
recovered” patients living in their own alpha = 15.366, lambda =
home 0.005141;
Annual health care costs for “recovered” Gamma 558.76 557
patients living in their own home alpha = 15.366, lambda =
0.0275;
Annual social care costs for “not Gamma 7888.09 7886
recovered” patients living in their own alpha = 15.366, lambda =
home 0.001948;
Annual social care costs for “recovered” Gamma 1097.57 1090

patients living in their own home alpha = 15.366, lambda =
0.014;
Annual cost for fees in long term care – Gamma 28992.45 28997
“not recovered” patients alpha = 15.366, lambda =
0.00053;
Annual cost for fees in long term care – Gamma 28992.45 28997
“recovered” patients alpha = 15.366, lambda =
0.00053;
Annual health care costs for “not Gamma 1409.72 1404
recovered” patients in long term care alpha = 15.366, lambda =
0.0109;
Annual health care costs for “recovered” Gamma 1409.72 1404
patients in long term care alpha = 15.366, lambda =
0.0109;
Cost of hospital bed day for readmissions Gamma 324.18 324.01
alpha = 15.366, lambda =
0.0474;
Cost per hour of day ward nurse Gamma 21 21
0.7317;
Cost per hour of a dietician Gamma 23 23
0.668;
Cost per hour of a geriatrician Gamma 108 108
0.1423;
Cost per hour of an occupational Gamma 23 23
578 APPENDIX H
therapist alpha = 15.366, lambda =

0.668;
Cost per hour of a physiotherapist Gamma 23 23
0.668;
Cost per hour of a social worker Gamma 29 29
0.5298;
Cost per hour of a team lead nurse Gamma 27 27
0.5691;
Initial age None – from meta 81 81
analysis of RCTs
Length of stay (days) – usual care Log-Normal, 31.56 31.56
u (mean of logs) =
3.439942259
sigma (std dev of logs) =
0.154584841;
Length of stay (days) – mean difference – 1.32 1.32
GORU/MARU Normal, Mean = 1.32, Std
Dev = 0.03322; Expected
value: 1.32
Length of stay (days) – mean difference – Normal, Mean = -6.06, -6.06 -6.06
HFP Std Dev = 0.3593
Length of stay for hospital readmissions Triangular, Min = 4, 7 7
– GORU/MARU Likeliest = 7, Max = 10;
– HFP Likeliest = 7, Max = 10;
– usual care Likeliest = 11, Max = 15;
Proportion of patients with ADL scores C- Beta 0.99 0.99
G in the “not recovered” health state Integer parameters only,
n = 108, r = 107;
Proportion of patients with ADL scores C- Beta 0.31 0.31
G in the “recovered” health state Integer parameters only,
n = 108, r = 34;
Probability moderate delirium – usual Beta 0.22 0.22
care Integer parameters only,
n = 64 , r = 14;
Probability pressure sores –usual care Beta 0.1646 0.1646
Integer parameters only,
n = 164, r = 27;
Probability die at 12 months – usual care Beta 0.2138 0.2138
n = 870 , r = 186;
Probability die at 6 months – usual care Beta 0.1673 0.1673
n = 263, r = 44 ;
Probability die 6 to 12 months – usual Beta 0.0548 0.0558
care Integer parameters only,
APPENDIX H 579
n = 219, r = 12;
Probability of hospital readmission at 12 Beta 0.2578 0.26

months – usual care Integer parameters only,
n = 640 , r = 165;
Probability of not recovery of pre- Beta 0.59 0.59
fracture ADL levels at 12 months – usual Integer parameters only,
care n = 283 , r = 167;
Probability of not recovery of pre- Beta 0.728 0.73
fracture ADL levels at 3 months – usual Integer parameters only,
care n = 125, r = 91;
Probability of not recovery of pre- Beta, Integer parameters 0.672 0.67
fracture ADL levels at 6 months – usual only, n = 125, r = 84;
care
Proportion of social care costs funded by Triangular, Min = 0.3, 0.6 0.6
the NHS or local authorities – patients Likeliest = 0.6, Max = 0.9;
living in their own home
Proportion of long term fee costs funded Triangular, Min = 0.3, 0.6 0.6
by the NHS or local authorities – patients Likeliest = 0.6, Max = 0.9;
living in long term care
Probability severe delirium – usual care Beta 0.28125 0.28125

n = 64 , r = 18;
Proportion of men - HFR and None – from meta 0.76
GORU/MARU analysis of RCTs
Proportion of men - usual care None – from meta 0.79
analysis of RCTs
Relative risk of die – 12 months – Log-Normal, u (mean of 0.95 0.95
GORU/MARU logs) = -0.05969, sigma
(std dev of logs) =
0.129622261;
Relative risk of die – 12 months – HFP Log-Normal, u (mean of 0.81 0.81
logs) = -0.22022, sigma
(std dev of logs) =
0.140960518;
Relative risk of die – 6 months – Log-Normal, u (mean of 0.79 0.79
(std dev of logs) =
0.220399212;
Relative risk – moderate delirium - HFP Log-Normal, u (mean of 0.95 0.95
logs) = -0.10966, sigma
(std dev of logs) =
0.341655183;
Relative risk – not recovery – Log-Normal, u (mean of 0.78 0.78
(std dev of logs) =
0.107452415;
Relative risk – not recovery – HFP Log-Normal, u (mean of 0.79 0.78
580 APPENDIX H
logs) = -0.24094, sigma

(std dev of logs) =
0.102123395;
Relative risk pressure sores – HFP Log-Normal 0.31 0.31
u (mean of logs) = -
1.24407, sigma (std dev
of logs) = 0.381796535;
Relative risk of readmissions – Log-Normal 0.86 0.86
GORU/MARU u (mean of logs) = -
of logs) = 0.131073023;
Relative risk of readmissions - HFP Log-Normal 1.14 1.14
u (mean of logs) =
of logs) = 0.135536774;
Relative risk of returning to own home – Log-Normal 1.11 1.11
GORU/MARU u (mean of logs) =
of logs) = 0.0347056;
Relative risk of returning to own home – Log-Normal 1.14 1.14
HFP u (mean of logs) =
of logs) = 0.058435349;
Relative risk – severe delirium – HFP Log-Normal 0.4 0.4
u (mean of logs) = -
of logs) = 0.407720564;
Time input of dietician (incremental over Normal -0.4 -0.4
usual care) – GORU/MARU Mean = -0.4, Std Dev =
0.0332;
Time input of nurse - (incremental over Normal 2.8 2.8
usual care) – GORU/MARU Mean = 2.8, Std Dev =
0.358;
Time input of occupational therapist - Normal 5 5
(incremental over usual care) – Mean = 5, Std Dev = 0.64;
GORU/MARU
Time input of physiotherapist Normal 8.5 8.5

(incremental over usual care) – Mean = 8.5, Std Dev =
GORU/MARU 1.09;
Time input of social worker - Normal -0.4 -0.4
(incremental over usual care) – Mean = -0.4, Std Dev =
GORU/MARU 0.32;
Transition probability from “not Beta 0.272 0.27
recovered” to “recovered” health state – Integer parameters only,
0 to 3 months n = 125, r = 34;
recovered” to “recovered” health state – Integer parameters only, 7
3 to 6 months n = 91, r = 7;
APPENDIX H 581

recovered” to “recovered” health state – Integer parameters only, 1
6 to 12 months n = 499212, r = 60837
EQ-5D score for “Recovered” health state Beta 0.68 0.68
Real-numbered
parameters, alpha =
35.36, beta = 16.64;
EQ-5D score for “Not recovered with no Beta 0.31 0.31
complications” health state Real-numbered
parameters, alpha = 3.72,
beta = 8.28;
EQ-5D score for “Not recovered with Beta 0.19 0.19
pressure sores” health state Real-numbered
parameters, alpha =
0.952227, beta =
4.059492;
EQ-5D score for “Not recovered with Beta 0.25 0.25
severe delirium” health state Real-numbered
parameters, alpha = 293,
beta = 880;
The conventional way to identify the most cost-effective strategy is to look at the option that is
optimal based on the mean costs and mean QALYs averaged across all of the probabilistic
simulations. These findings are summarised in Table 114.
Table 114: Cost-effectiveness analysis from probabilistic analysis

Strategy Cost Incremental Effectiveness Incremental Incremental C/E
Cost effectiveness (ICER)
HFP £34K 3.74
GORU/MARU £36K £2K 3.61 -0.13 (Dominated)
Usual care £59K £25K 2.73 -1.01 (Dominated)
The probabilistic results are very similar to the deterministic ones indicating that HFP is dominant
(has lower cost and more QALYs) compared with the two alternatives.
These findings are described in Figures 162, 163 and 164. Each point on the second scatter plot
represents the incremental cost and QALYs gained for HFP vs GORU for one simulation. The
dotted line represents the £20,000/QALY threshold and the ellipse delimits the 95% confidence
space.
582 APPENDIX H
ICE Scatte rplot of

HFP vs. GORU/M ARU
£50K
£40K
£30K
£20K
Incremental Cost
£10K
£0K
-£10K
-£20K
-£30K
-£40K
-£50K
-£60K
-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0
Figure 162: Incremental cost-effectiveness scatter plot: HFP vs GORU/MARU
The scatter plot of HFP vs usual care shows the high certainty of HFP being cost-effective as all the
dots in the 95% confidence ellipse are below the £20,000/QALY threshold and more than 95% are
cost saving.
APPENDIX H 583
ICE Scatte rplot of

GORU/M ARU v s. Usual care
£11K
£6K
£1K
-£4K
Incremental Cost
-£9K
-£14K
-£19K
-£24K
-£29K
-£34K
-£39K
-£44K
-£49K
-£54K
-£59K
-0.80 -0.30 0.20 0.70 1.20 1.70 2.20
Figure 163: Incremental cost-effectiveness scatter plot: GORU/MARU vs. usual care
584 APPENDIX H
ICE Scatte rplot of

HFP vs. Usual care
£20K
£10K
£0K
Incremental Cost
-£10K
-£20K
-£30K
-£40K
-£50K
-£60K
-£70K
-0.50 0.00 0.50 1.00 1.50 2.00 2.50
Figure 164: Incremental cost-effectiveness scatter plot - HFP vs usual care
However, when we compared HFP with GORU the 95% CI showed a greater uncertainty as HFP
was dominant in the lower bound and GORU was dominant in the upper bound. The uncertainty
can be graphically represented by plotting the results of the incremental analysis for all the
10,000 simulations into a cost-effectiveness plane.
We also found that, at a willingness to pay equal to £20,000 per QALY, HFP was the optimal
strategy in 70% of the simulations; GORU/MARU was the most cost-effective intervention in 30%
of simulations, and usual care was never the optimal strategy. These findings are summarised in
table 42 below:
Table 42: Probability most cost-effective intervention at a willingness to pay of £20,000 and
£30,000 per QALY
Strategy Probability most cost-effective Probability most cost-effective
intervention at a WTP of £20,000 per intervention at a WTP of £30,000
QALY per QALY
HFP 0.70 0.80
GORU/MARU 0.30 0.20
Usual care 0 0
APPENDIX H 585
20.6.10 Discussion
The optimal strategy in a cost-effectiveness analysis is the one with the highest incremental net
benefit averaged across all the probabilistic simulations. This was HFP.
The model showed that usual care was clearly not the optimal strategy.
However, there was some uncertainty about which strategy was the most cost-effective between
HFP and GORU/MARU. In particular the results were sensitive to the proportion of patients
returning home after their rehabilitation: if the probability of returning home after undergoing a
GORU/MARU programme was 83% (instead of 79% in the base case) then GORU is the optimal
strategy.
Our analysis had to rely on several assumptions.
Firstly, no evidence was available which compared directly HFP vs GORU/MARU. As a

consequence, only an indirect comparison between the two hospital MDR programmes was
possible. This meant that findings had to be pooled in the usual care arm of the different RCTs
included in our clinical review, thus assuming that “usual care arms” in all such studies were
sufficiently similar. However, the GDG agreed that the population included in the RCTs on HFP
and the population included in the RCTs on GORU were sufficiently similar and that therefore our
findings were not affected by counfounding factors.
Secondly, no data were available regarding the presence and incidence of complications in the
GORU/MARU programme versus usual care. The assumption that in this case the relative risk for
that rehab programme was equal to 1 implies that we may have underestimated the efficacy of
GORU/MARU in reducing the presence of postoperative complications, and as a consequence,
that we may have overestimated its costs and decrement in quality of life compared to HFP.
However, when we changed the probabilities of complications for GORU/MARU in a one-way
sensitivity analysis, the findings of the cost-effectiveness analysis did not change, and HFP was still
the dominant strategy.
Finally, the finding of the meta-analysis of clinical trials regarding the length of stay showed a
longer length of stay for the GORU/MARU programme versus usual care (mean difference (days):
1.32). However, the inclusion of the study by Galvard (1995)107 in the meta-analysis may have
biased this finding. This is because Galvard (1995)107 reports a mean length of stay of 53.3 days for
the intervention (GORU) group and of 28 days for usual care. This finding, according to the
authors, was due to the fact that GORU was a new rehabilitation programme that had just been
implemented in their hospital, and the hospital staff was not yet experienced in the management
of the programme, which could have resulted in a longer length of stay for patients in the
intervention group. As a consequence, we may have overestimated the costs of hospital stay for
GORU/MARU. However, when we changed the length of hospital stay for the GORU/MARU
programme in a one-way sensitivity analysis, the findings of the cost-effectiveness analysis did not
change, and HFP was still the dominant strategy.
586 APPENDIX H
20.7 Cost-effectiveness analysis of Community MDR vs Usual care
20.7.1 Introduction
The GDG identified the multidisciplinary management in the community for hip fracture patients
as a high priority area for economic analysis.
The clinical question linked to this high priority area is the following:
What is the comparative effectiveness of community -based multidisciplinary rehabilitation

models versus usual care?
A review of the literature was conducted followed by economic modelling of the cost-
effectiveness of the listed interventions in England and Wales. The literature search and review
methods can be Chapter 3. Despite some cost-effectiveness studies were identified, none
represented a full cost-utility analysis which addressed our clinical question. As a consequence,
the GDG felt that an original economic model was essential in order to support their
recommendations.
• When published data was not available we used expert opinion to populate the
model.
• We followed the methods of the NICE reference case. Therefore costs were
calculated from the UK NHS and PSS perspective. Health gains were measured in
terms of quality-adjusted life-years (QALYs) gained.
female) hospitalised for surgery. The model spans over a life-time horizon.
20.7.3 Software
20.7.4 Economic evaluation type
We conduct a cost-utility analysis, where health outcomes are measured as Quality-Adjusted Life-
Years (QALYs). The cost effectiveness outcome of the model is measured as cost per QALY gained.
APPENDIX H 587
20.7.5 Time horizon and discount rates used
The model spans over a life-time horizon. All costs considered in the model were calculated at on
the basis of a four-months follow-up time and hence were not discounted. However, we used a
discount rate of 3.5% for the health gains, as these were calculated throughout the remaining life
of the cohort of patients.
20.7.6 Structure of the model
The structure of our model reflects the findings of the RCT by Crotty et al (2002)60. The paper
reports SF-36 scores for surviving patients, both in the community MDR and in the usual care arm
of the study, at a 4 months follow up.
We develop a decision tree with Markov states, where a hip fracture patients can either receive a
community based MDR programme or usual inpatient rehabilitation. Following this decision node,
a chance node determines whether patients survive or die following their specific rehabilitation
programme. The probability associated to this chance node is derived from Crotty et al (2002)60 at
a 4 months follow up. Subsequently, patients who are alive after the 4-months follow up period
transit in a Markov state, “alive after follow up”. Patients will then either stay in that state or
transit to the “dead” state in the following cycles.
The structure of the model is the following:
Alive after
follow up
Survive
Community M
MDR Die
C
Die
Hip fracture
Alive after
patients follow up
Survive M
Die
C
Usual care
Die
C = chance node
M = Markov node
Figure 165: Model structure - community MDR vs usual care
20.7.7 Utility data
Utility weights are calculated using SF-36 scores obtained from Crotty et al (2002)60. The paper
only reports total scores for the physical and mental components. Following personal
communications with the authors, we were able to access individual SF-36 scores, reported in
Table 115 below:
588 APPENDIX H
Table 115: SF-36 scores based on Crotty et al (2002)60.

SF-36 domain, Mean (SD) Conventional Care, n=29 Early Discharge, n=30
Physical functioning 28.8 (25.2) 41.2 (26.6)
Social functioning 62.1 (40.0) 72.5 (32.4)
Role-physical 61.2 (41.0) 53.3 (40.9)
Role-emotional 83.9 (31.6) 77.8 (38.5)
Mental health 77.9 (14.2) 80.1 (19.8)
Vitality 45.0 (21.9) 54.2 (24.3)
Bodily pain 61.4 (30.9) 65.1 (24.4)
General health 61.8 (30.1) 69.3 (24.1)
Source: primary data supplied by the authors of Crotty et al (2002)60
Using the Ara-Brazier method7, we mapped the individual SF-36 scores in EQ-5D utility weights.
We found that the EQ-5D weight for patients undergoing community MDR is 0.732, and for
patients undergoing usual inpatient rehabilitation is 0.643. As the effectiveness data refer to
findings at 4 months 60, we used these utility weights for cycle 0 only. For cycle 1-onwards we
assume that there is no difference in the utility score of the two groups of patients, and use the
EQ-5D score of the control group also for patients in the community MDR arm of the model.
20.7.8 Mortality
The mortality rates for the community MDR and usual care patients have been adjusted to take
into account the baseline characteristics of the two groups, which were very different in the two
arms of Crotty et al (2002)60, since 62% of patients were female in the COMMUNITY MDR versus
75% in the usual care group, and the median age for COMMUNITY MDR patients was 81.6 versus
83.5 years in the usual care arm.
First, we have calculated the age and gender-adjusted mortality rate (AMR) for the general UK
population as per characteristics in usual care arm and the same for community MDR arm. Then,
we have calculated the Standardised Mortality Rate (SMR) as = MR/AMR, both the usual care and
the community MDR arm. We have then assumed that the average age for the overall population
in the model was 80 years of age, and we have determined the probability of death using the
formula: SMR*pDeath[80].
We have found that that probability of death at 4 months for the patients in the usual care arm
corresponds to 0.07239, and for patients in the community MDR group is equal to 0.067. The
relative risk of the mortality rate for community MDR compared to usual care is 0.925.
20.7.9 Calculating QALYs gained
For each strategy (community MDR and usual inpatient rehabilitation), the expected QALYs in the
“survived” health state at each cycle are calculated as follows:
Expected QALYs = Σ (Usurvived x Psurvived )
where: Usurvived = the utility score for the patients who are still alive and Psurvived = the proportion of
alive patients
APPENDIX H 589
The proportion of patients in the “alive” health state depends on the effectiveness of the
treatment, and on the proportion of patients still alive, which falls as the number of cycles and
therefore age increases.
incremental QALYs gained associated with a treatment strategy are calculated as the difference
between the expected QALYs with that strategy and the expected QALYs with the comparator.
20.7.10 Cost analysis
20.7.10.1 Cost for the community MDR and

inpatient rehabilitation
programmes.
While in hospital, we assume that there is no difference in the level and type of resources used by
patients in the two groups, as no evidence of the contrary was found in the literature. Moreover,
as patients receive their inpatients rehabilitation services without being discharged to a different
ward, they will still be under the same HRG recorded at admission. Thus, the rehabilitation that
patients receive while in hospital is not a type of discrete rehabilitation service, that is, a service
that can be cost using its own HRG, since: “rehabilitation HRGs are only generated where care is
identified as taking place under a specialist rehabilitation consultation or within a discrete
rehabilitation ward or unit. [..] Where a patient is not admitted specifically to a rehabilitation unit
or where rehabilitation treatment is undertaken without transfer to a specialist consultant, or
without transfer to a rehabilitation unit, this should not be reported as discrete rehabilitation”
(Collection Guidance on Reference Costs for 2006-0770).
As a consequence, we use the reference cost for excess bed days reported in the National
Schedule of Reference Costs Year: '2008-09' - NHS Trusts and PCTs combined Non-Elective
Inpatient (Long Stay).
Crotty et al (2002)60 report evidence on the presence of complications experienced by hip fracture
patients in the two groups while in acute care. None of these complications were statistically
significant different between usual care and community MDR (the complications were:
pneumonia, pressure sores, confusion, wound infection and urinary tract infection). Moreover, no
additional information was provided in the paper as to whether those complications resulted in a
prolonged length of hospital stay for patients in the community MDR scheme. Thus, we used the
weighted average NHS reference cost for excess bed days for major, intermediate and minor hip
procedures with all types of complications, amounting to £241.68 per day.
As for the daily cost of the community MDR scheme, we use the NHS reference cost (2008-09)
reported for “Hospital at Home/ Early Discharge Schemes - Fractured Neck of Femur”, which
corresponds to £94 per day.
We conduct a sensitivity analysis on these values in section 32.1 of this chapter.
20.7.10.2 Length of stay
Crotty et al (2002)60 reports the following findings for the length of stay for the community MDR
and the usual inpatient rehabilitation:
Table 116: Length of stay in hospital and in own home

590 APPENDIX H
Length of stay community MDR (at home stay) 20.3 (mean, days)
Length of stay community MDR (at home stay) (in 7.8 (mean, days)
hospital stay)
Length of stay usual care (in hospital stay) 14.3 (mean, days)
20.7.10.3 Hospital readmissions and related

length of stay
Crotty et al (2002)60 gives information about the levels of readmissions during the four months
follow up of the study. The paper distinguishes between related readmissions and unrelated
readmissions, and gives the length of stay for both cases. However what these related and
unrelated admissions were was not clear in the paper. We consider surgery and the rehabilitation
admissions to be the “related” readmissions, and we consider the cost of a bed day in medicine
for the cost of not-related admissions.
These unit bed day costs are based on Czoski-Murray (2007)63, which reports the cost per day for
hospital stay in an orthopaedic, rehabilitation or general medicine ward at 2002 prices. We
assume that the “related readmissions” take place either for orthopaedic or for rehabilitation
reasons, and that the “unrelated readmission” are those in the generic medicine ward.
Taking into account of the inflation index, the cost per day of hospital stay for a related
readmission corresponds to £367.85 (assuming that half of these readmissions took place for
surgery and half for rehabilitation reasons) and to £364.61 for unrelated readmission.
Table 117: Evidence on readmissions (Crotty et al, 2002)60

Number of not related readmission for usual care 0.43
Mean difference for unrelated readmissions 0.38
Number of related readmission for usual care 0.27
Mean difference for related readmissions -0.05
Length of hospital stay for not related readmissions (usual care) 4.9
Mean difference for length of hospital stay for unrelated readmissions -0.3
Length of hospital stay for related readmissions (usual care) 3.6
Mean difference for length of hospital stay for related readmissions 0.1
20.7.10.4 Social services costs

APPENDIX H 591
For community services, Crotty et al (2002)60 intended any of the following: outpatient
rehabilitation; private therapy, district nursing, day care, respite care, employment rehabilitation
training, carer time off work and Meals on Wheels. As we do not have data regarding the exact
amount of resources for each of the above categories that were actually used by patients in the
two arms of the study, we assume that the weekly cost of social care is given by a weighted
average of the five categories of packages of care reported in the PSSRU 200961 and discussed in
section 18.2.2 of the hospital MDR model. We assume that an equal proportion of patients used
each type of social care package. However, in a sensitivity analysis we look at the case in which all
patients used a “very low cost” type of social care package and when all of them used a “very high
cost” package of care.
Only a proportion of the social care costs will generally be funded by local authorities71, 348, 72, 144.
It was not possible to identify a national average for the social care costs funded by local
authorities in the published literature, and as a consequence an assumption had to be made
regarding the proportion of this care that was publicly funded. In the base case analysis, we
assume that 60% of social care costs are borne by local authorities, and are therefore includable
in the model, and we then test this assumption in a sensitivity analysis.
As no further data were given regarding the use of social care services after the 4 months follow
up, we adopted a conservative approach and assumed that after that period there was no
difference in the use of social services that could be due to the different rehabilitation scheme
used.
20.7.10.5 Primary care costs
Crotty et al (2002)60 point out that: “[..] patients [in the community MDR scheme] tended to call
the GPs if problems arose and this invariably meant a visit to the home for the GP” (Crotty et al
2002, page 1160). On the other hand, no details were provided regarding whether all GP visits to
patients in the community MDR scheme took in fact place in the patients’ own home. Similarly,
no information was given regarding where GP visit took place for patients in the usual care arm.
As a consequence, we have assumed that the unit cost for a GP visit for patients in the usual care
scheme is the average between the cost of a GP visit at the patient’s own home (£117) and a GP
surgery visit (£76) as reported in the PSSRU 200961, and corresponds to £96.5.
As no further data were given regarding the use of primary care services after the 4 months follow
up, we adopted a conservative approach and assumed that after that period there was no
difference in the use of GP services that could be ascribed to the different rehabilitation scheme
used.
Table 118: Evidence on GP visits (Crotty et al, 2002)

Number of GP visits (usual care) 4.5
Number of GP visits (community MDR) 3.3
(mean difference: -1.2)
20.7.11 Cost effectiveness findings
The cost-effectiveness findings for the deterministic base case analysis is presented in Table 119:
Cost-effectiveness analysis - deterministic base case below:
592 APPENDIX H
Table 119: Cost-effectiveness analysis - deterministic base case

Strategy Cost Incremental Effectiveness Incremental Incremental cost-
Cost Effectiveness effectiveness ratio
Usual care £6469.1 3.0827 QALYs

Community £6903.2 £434.1 3.1283 QALYs 0.0456 QALYs 9521 £/QALYs
MDR
Hence, the community MDR scheme is a cost-effective treatment for the rehabilitation of hip
fracture patients in the deterministic case scenario. Table 120 reports a breakdown of costs for
the relevant resources used in the community MDR and in the usual care group.
Table 120: Cost breakdown for community MDR and usual care
Resource item Usual Care Community MDR
Rehab cost 3456 3793
Readmission 1124 1657
Domiciliary social care 1453 1133
GP visits 434 318
Total cost £6467 £6901
20.7.11.1 Sensitivity analysis
We now proceed by investigating how robust the findings of the deterministic analysis are by
conducting a series of sensitivity analysis.
To begin with we note that the model is not sensitive to changes in the level of social services paid
by the NHS (from 0 to 100%), as community MDR is still cost-effective.
Moreover, when the cost per week of social services is varied between the minimum (£41 per
week) and the maximum (£542) the option with the highest net benefit is still community MDR.
However, our findings are sensitive to the length of hospital stay (both for community MDR and
for usual care patients) and on the length of rehabilitation programme at home, as well as on the
daily cost of hospital stay following surgery and on the daily cost of the community MDR
programme. These findings are summarized in Table 121 below.
Table 121: Threshold sensitivity analysis

Variable Values Strategy with the highest net benefit
Length of stay in hospital for ≥ 9.78 (days) Usual care
community MDR patients ‹ 9.78 (days) Community MDR
Length of stay at home for ≥ 25.38 (days) Usual care
community MDR patients ‹ 25.38 (days) Community MDR
Length of stay in hospital for usual ≥ 12.32 (days) Community MDR
APPENDIX H 593
care patients ‹ 12.32 (days) Usual care

Daily cost for hospital stay > £168.18 Community MDR
Daily cost for community MDR < £ 117.53 Community MDR
rehab programme
Our cost-effectiveness findings are not sensitive to changes in the cost per day in hospital of the
related readmissions and to changes in the proportion of social care costs borne by local
authorities.
The following figure summarise the findings of a two-ways sensitivity analysis on the length of
stay in hospital and at home (the vertical axe reports the length of stay at home for community
MDR patients and the horizontal axe the length of stay in hospital for usual care).
Net Monetary Benefit (wtp=20000.) Sensitivity Analysis on

LOS_community_M DR_at_home and LOS_in_hospital_usual_care
100.0
LOS_in_hospital_usual_care
Community Rehab
Usual care
75.3
50.5
25.8
1.0
1.0 25.8 50.5 75.3 100.0
LOS_community_MDR_at_home
Figure 166: Two-way sensitivity analysis on length of stay at home and in hospital
A probabilistic sensitivity analysis was performed to assess the robustness of the model results to
plausible variations in the model parameters.
Probability distributions were assigned to each model parameter, where there was some measure
of parameter variability. We then re-calculated the main results 10000 times, and each time all
the model parameters were set simultaneously, selecting from the respective parameter
distribution at random.
594 APPENDIX H
Baseline risk Beta Bounded on 0 – 1 interval. Derived from

sample size, number of patients
experiencing events
Cost Gamma Bounded at 0, positively skewed. Derived
from mean and standard error
Utility Beta Bounded on 0 – 1 interval. Derived from
mean and sample size
Risk ratio, length of stay Lognormal Bounded at 0. Derived from log and
standard error of log
Mean differences (e.g. in Normal Derived from mean and standard deviation
length of stay, time of
therapies, etc.)
Table 123 summarises the expected values of the variables in our model from the different
distributions used in the PSA.
Table 123: Distribution and parameters - probabilistic sensitivity analysis

Name Baseline Expected value Distribution and parameters
value
Probability use of community 0.72 0.7187 Beta, Integer parameters only, n =
services for usual care 32, r = 23
Weekly social care unit cost for 216 216 Gamma, alpha = 15.36583528,
“high need” patients (£) lambda = 0.071138126
“low need” patients lambda = 0.213414379
“median need” patients (£) lambda = 0.085365752
“very high need” patients (£) lambda = 0.02835025
“very low need” patients (£) lambda = 0.37477647
NHS reference costs for 94 94 Gamma, alpha = 15.36583528,
community MDR (£; daily) lambda = 0.163466333
APPENDIX H 595
NHS reference cost for usual care 240 240 Gamma, alpha = 15.36583528,
lambda = 0.064024314
Mean difference in GP visits for -1.2 -1.2 Normal, Mean = -1.2, Std Dev =
community MDR 0.0957
Number of GP visits for usual 4.5 4.5 Normal, Mean = 4.5, Std Dev = 0.646
care
Length of stay (days) at own 20.3 20.3 Log-Normal, u (mean of logs) =
home for community MDR 3.006198781, sigma (std dev of logs)
programme = 0.094043657
Length of stay (days) in hospital 7.8 7.8 Log-Normal, u (mean of logs) =
for community MDR patients 2.028128367, sigma (std dev of logs)
= 0.228014764
Length of stay (days) unrelated 4.9 4.9 Log-Normal, u (mean of logs) =
readmissions 1.264935055, sigma (std dev of logs)
= 0.80535725
Length of stay (days) related 3.6 3.6 Log-Normal, u (mean of logs) =
readmissions 1.061775585, sigma (std dev of logs)
= 0.662054772
Length of stay (days) in hospital 14.3 14.3 Log-Normal, u (mean of logs) =
– usual care 2.650611207, sigma (std dev of logs)
= 0.138912419
Probability mortality – 0.067 0.067 Beta, Real-numbered parameters,
community MDR alpha = 2.278, beta = 31.722
Probability mortality usual care 0.0724 0.0724 Beta, Real-numbered parameters,
alpha = 2.31648, beta = 29.68352
Proportion of patients with “very 0.2 0.2 Dirichlet; Alpha list (proportion of
low”/”low”/ patients with very low social care
”median”/”high”/”very high” costs; proportion of patients with
social care costs low social care costs; proportion of
patients with median social care
costs; proportion of patients with
high social care costs; proportion of
patients with very high social care
costs)
Proportion of social care costs 0.6 0.6 Triangular, Min = 0.30, Likeliest =
funded by the NHS 0.60, Max = 0.90;
EQ-5D score (community MDR) 0.732 0.732 Beta, Real-numbered parameters,
596 APPENDIX H
alpha = 24.888, beta = 9.112
EQ-5D score (usual care) 0.643 0.643 Beta, Real-numbered parameters,

alpha = 20.576, beta = 11.424
Number of not related 0.43 0.43 Normal, Mean = 0.43, Std Dev =
readmission for usual care 0.0617
Number of related readmission 0.27 0.27 Normal, Mean = 0.27, Std Dev =
for usual care 0.387
Mean difference for length of 0.1 0.1 Normal, Mean = 0.1, Std Dev =
hospital stay for related 0.0145
readmissions
Mean difference for related -0.05 -0.05 Normal, Mean = -0.05, Std Dev = 0.04
readmissions
Mean difference for length of -0.3 -0.3 Normal, Mean = -0.3, Std Dev =
hospital stay for unrelated 0.03442
readmissions
Mean difference for unrelated 0.38 0.38 Normal, Mean = 0.38, Std Dev =
readmissions 0.545;
Relative Risk use of community 0.78 0.78 Log-Normal, u (mean of logs) = -
service 0.265778098, sigma (std dev of logs)
= 0.186100721
Unit cost for a GP visit 76 76 Gamma, alpha = 15.36583528,
lambda = 0.202182043
Unit cost for related 352 352 Gamma, alpha = 15.36583528,
readmissions lambda = 0.043652941
Unit cost for unrelated 249 249 Gamma, alpha = 15.36583528,
readmissions lambda = 0.061710182
The cost-effectiveness findings of the PSA are summarized in Table 124 below:
Table 124: cost-effectiveness finding from probabilistic sensitivity analysis

Strategy Cost Incr Cost Eff Incr Eff Incremental cost- 95% CI on ICERs
effectiveness ratio
Usual care £6466.6 3.0827 QALYs
Community £6901.2 £434.6 3.1283 0.0456 9533 £/QALYs Cost saving -
Rehab QALYs QALYs dominated
APPENDIX H 597
The PSA shows that there is a high uncertainty as to whether community MDR is cost-effective
compared to usual care. This uncertainty can be graphically represented by plotting the results of
the incremental analysis for all the 10,000 simulations into a cost-effectiveness plane. Each point
on the scatter plot represents the ICER of community MDR versus usual care for each simulation.
The dotted line represents the £20,000/QALY threshold while the ellipse delimits the 95%
confidence interval.
Figure 167: Incremental cost-effectiveness scatter plot - Community MDR vs usual care
From the simulations conducted for the PSA, we found that at a willingness to pay equal to
£20,000 per QALY, community MDR was the optimal strategy in 50% of the simulations. At a
willingness to pay of £30,000 per QALY, community MDR was the optimal strategy in 60% of the
simulations.
Table 125: Probability most cost-effective intervention at a willingness to pay of £20,000 and
30,000 per QALY
Strategy Probability most cost- Probability most cost-effective
effective intervention at a intervention at a WTP of £30,000
WTP of £20,000 per QALY per QALY
Community 0.50 0.60

MDR
598 APPENDIX H
Usual care 0.50 0.40
20.7.12 Discussion
The model shows that community MDR is cost-effective in the rehabilitation of patients after a
hip fracture. However, this finding is rather sensitive to variations in the length of stay, both in
hospital and at home. Moreover, a PSA has shown that there is high uncertainty over the cost-
effectiveness of community MDR compared to usual care.
The model has several limitations, such as the fact that it is based on the clinical evidence derived
from only one RCT)60 based in Australia. Moreover, the evidence on treatment effects60 was
available only up to 4 months follow up. No information was available regarding the impact of
community MDR after that time point.
20.8 Cost analysis of cemented vs. uncemented implants (newer designs of arthroplasty)
In order to conduct a cost analysis for the cemented and uncemented implants, we need to consider the
following cost components: implants cost; accessories costs; length of stay; re-operation and theatre costs.
a) Cost of implants
The National Joint Registry (NJRv7) was accessed on February 11th 2011, in order to find out the five most
commonly used types of both cemented (stems with no head) and uncemented implants in the UK.
Furthermore, the NHS Supply catalogue 2011 was searched in order to obtain the most recent price for
each of these items. All this information is reported in Table 126 below:
Table 126: Price of new design cemented and uncemented stems most commonly used in the UK
Cemented implants – Price per item (£) Uncemented implants Price per item (£)
stems with no head
(from most common to
(from most common to less common)
less common)
Exeter 410.53 Corail 893.47
C-Stem 347.37 Taperloc Could not retrieve price

on NHS supply catalogue
CPT 393.68 789.47
JRI Furlong
Stanmore Could not retrieve price Accolade 684.21

on NHS supply catalogue
Charnely Could not retrieve price 789.47
on NHS supply catalogue SL-Plus
Average price for most 383.86 Average price for most 789.15
common cemented common uncemented
implants implants
APPENDIX H 599
b) Accessories costs for the cemented group of patients
The price for accessorises used when new design cemented stems are implanted are presented in Table 127
below.
Table 127: Prices for accessorises used with new design cemented stems
Resources Unit price Source of unit price
Pulse Lavage £27.86 NHS Supply Chain catalogue

2011
Biogel glove £1.07 NHS hospital*
Cement £45 Data supplied by GDG
member
Cement mixing kit £35 Data supplied by GDG
member
Cement Restrictor £35.88 Smith & Nephews**
(size 16/18/20mm)
Femoral canal brush Smith & Nephews**
£64.48
(NW 12.5/19MM)
MIXOR femoral pressurizer Smith & Nephews**
£14.70
(large/medium/small)
Sterilzed tray NHS hospital record***

£25
Total cost for accessories £248.99
*Peterborough and Stamford Hospital NHS Trust
**Price list available online at: http://browse.uk-

plc.net/Companies/SMITH__NEPHEW/products/CEMENT_ACCESSORIES.htm
(accessed 21st February 2011)
***Data from John Radcliffe NHS Hosptial Trust.
It is important to note that the cost of the sterilized tray could vary from £25 to £50 (source: John
Radcliffe NHS Hosptial Trust) depending on the number of instruments contained in the tray. It is
also relevant to point out that our cost calculation for the accessorises used with the new design
of cemented stems is very similar to the cost reported in Unnanuntana et al (2009)335, which
calculates the cost of accessorises used for a third-generation cement technique. They considered
two 40-g bags of bone cement without antibiotics, a vacuum mixing cartridge, cement pressurizer,
canal plug and distal cement centralizer, canal brush and cement scrapers. The average total cost
for the two 40-g batches of bone cement and all accessories used to achieve a third-generation
cement technique was estimated to be $386 (range, $351-$407) (January 2008 prices), which
correspond to £252 (range, £229 - £266) (converted using 2008 purchasing power parity).
The GDG noted that the accessorises costs determined in table 126 represent a through end of
the spectrum; at the “lower end” of the spectrum, the only accessories costs to consider would be
600 APPENDIX H
those for the cement, cement mixing kit, restrictor and sterilized tray, for an overall cost of
£140.88.
c) Cost of length of stay in hospital
To calculate the health state costs during the hospital stay, we use the NHS reference cost for excess bed
days reported in Table 128 below. The excess bed day cost is the cost per day for days exceeding the
trimpoint, a cut-off that determines patients with exceptionally long stay, and as such usually estimates the
cost of care without the cost of procedures (i.e. without the cost of the surgery. These costs reflect the
presence of complications experienced by hip fracture patients during their entire hospital stay. Moreover,
they distinguish between “major” and “intermediate” complications, thus allowing users to take into
account the different degrees of resource use.
Table 128: National Schedule of Reference Costs Year : '2008-09' - NHS Trusts and PCTs combined Non-
Elective Inpatient (Long Stay) Excess Bed Day HRG Data for hip procedures
Currency Currency Description Activity National Average
Code Unit Cost
HA11A Major Hip Procedures Category 2 for Trauma with 360 £243
Major CC
HA11B Major Hip Procedures Category 2 for Trauma with 620 £242
Intermediate CC
HA11C Major Hip Procedures Category 2 for Trauma 162 £220
without CC
HA12B Major Hip Procedures Category 1 for Trauma with CC 9,760 £237
HA12C Major Hip Procedures Category 1 for Trauma 1,230 £226
without CC
HA13A Intermediate Hip Procedures for Trauma with Major 14,891 £240
CC
HA13B Intermediate Hip Procedures for Trauma with 12,856 £249
Intermediate CC
HA13C Intermediate Hip Procedures for Trauma without CC 2,972 £223
HA14A Minor Hip Procedures for Trauma with Major CC 5,195 £234
HA14B Minor Hip Procedures for Trauma with Intermediate 5,808 £245
CC
Mean weighted average of excess bed days costs – NHS reference costs 2008-08 £240
Major, Intermediate and Minor Hip procedures with all types of complications
Using the evidence reported in Figved (2009)94, the mean LOS in hospital for patients in the cemented
group was 7.8 days and in the uncemented group 8.4 (p<0.52). This implies that the LOS costs for the
cemented group correspond to £1872 and for the uncemented group to £2016.
APPENDIX H 601
d) Re-operation costs
The cost of the re-operations in the two groups of patients is calculated by using the weighted average of
the NHS reference cost for non-elective inpatient short stay data for NHS Trusts and PCTs combined. The
different HRGs and unit costs associated with each type of surgical procedure and possible presence of
complications are summarised in Table 129 below.
Table 129: National Schedule of Reference Costs Year : '2008-09' - NHS Trusts and PCTs combined Non-
Elective Inpatient (Short Stay) HRG Data
Currency Code Currency Description Activity National

Average
Unit Cost
HB11A Major Hip Procedures for non Trauma Category 2 with 4 £1,793
Major CC
HB11B Major Hip Procedures for non Trauma Category 2 with 5 £2,001
CC
HB11C Major Hip Procedures for non Trauma Category 2 3 £1,765
without CC
HB12A Major Hip Procedures for non Trauma Category 1 with 16 £1,811
Major CC
HB12B Major Hip Procedures for non Trauma Category 1 with 89 £2,097
CC
HB12C Major Hip Procedures for non Trauma Category 1 88 £1,611
without CC
HB13Z Intermediate Hip Procedures for non Trauma Category 51 £2,771
2
HB14B Intermediate Hip Procedures for non Trauma Category 746 £1,671
1 with CC
HB14C Intermediate Hip Procedures for non Trauma Category 529 £1,785
1 without CC
HB15D Minor Hip Procedures for non Trauma Category 2 19 213 £912
years and over with CC
HB15E Minor Hip Procedures for non Trauma Category 2 19 124 £1,028
years and over without CC
HB15F Minor Hip Procedures for non Trauma Category 2 18 8 £779
years and under with CC
HB15G Minor Hip Procedures for non Trauma Category 2 18 73 £1,304
years and under without CC
HB16B Minor Hip Procedures for non Trauma Category 1 with 141 £2,057
CC
HB16C Minor Hip Procedures for non Trauma Category 1 111 £833
without CC
Mean weighted average of short-stay non-elective NHS reference costs 2009-10 £1,598.39
Major, Intermediate and Minor Hip procedures with all types of complications
Figved (2009)94 reports a re-operation rate of 6.3% for the cemented group and of 7.4% for the uncemented
group. The re-operation costs therefore correspond to £100.70 in the cemented group and to £118.28 for
the uncemented group.
e) Theatre time costs
Figved (2009)94 reports the duration of the operation for the cemented group which was 12.4 minutes
longer than for the uncemented group. Using a cost per minute for the theatre use of £20.50 (from
602 APPENDIX H
Peterborough and Stamford NHS Trust accountant data), the higher theatre costs for the cemented group
correspond to: £254.2
Summary of costs components
Patients who Patients who

received cemented received
implants uncemented
implants
Cost categories:
a) Implants £383.86 £789.15
b) Accessories costs for cemented implants £248.99 -
c) LOS £1872 £2016
d) Re-operations £100.70 £118.28
e) Incremental theatre costs for cemented group £254.2 -
Total costs £2859.75 £2923.43
It follows that the overall incremental cost of the newer design of uncemented implants over the cemented
ones £63.68. When the lower estimate for accessories costs is used (£140.88) the total costs for the
cemented group corresponds to £2751.64, and the incremental cost of the uncemented implants to
£171.79. These cost does not include the additional pain relief required by patients in the uncemented
group. However, the unit costs for analgesics is relatively low, as showed in Appendix H section 20.1.
APPENDIX I 603
21 Appendix I: High Priority Research

Recommendations
21.1 Imaging options in occult hip fracture

Research question: In patients with a continuing suspicion of a hip fracture but whose
radiographs are normal, what is the clinical and cost effectiveness of computed
tomography (CT) compared to magnetic resonance imaging (MRI), in confirming or
excluding the fracture?
Why this is important:

that advances in technology have made the reliability of CT comparable with that of MRI. If
modern CT can be shown to have similar reliability and accuracy to MRI, then this has
considerable implications because of its widespread availability out of hours and lower cost.
It is therefore a high priority to confirm or refute this assumption by direct randomised
comparison. The study design would need to retain MRI as the ’gold standard’ for cases of
uncertainty and to standardise the criteria, expertise and procedures for radiological
assessment. Numbers required would depend on the degree of sensitivity/specificity (the
key outcome criteria) set as target requirement for comparability, but need not necessarily
be very large.
Criteria for selecting high-priority research recommendations:
PICO question In patients with a continuing suspicion of a hip

fracture but whose radiographs are normal,
Each research recommendation should be what is the clinical and cost effectiveness of
formulated as an answerable question or a set computed tomography compared to magnetic
of closely related questions. This should use resonance imaging, in confirming or excluding
the PICO framework (patient, intervention, the fracture?
comparison and outcome)
Patient: patients with a continuing suspicion of
a hip fracture but whose radiographs are
normal
Intervention: Modern Computed Tomography
techniques e.g. 64-slice scanners with three
dimensional capabilities and spiral
multidetector CT (MDCT)
Comparison: Magnetic resonance imaging
604 APPENDIX I
Outcomes: Diagnostic accuracy including

sensitivity and specificity
Importance to patients or the population. The altered guidance would ensure the
What would be the impact of any new or availability of accurate diagnosis out of hours
altered guidance on the population (for and thus promote the benefits of prompt,
example, acceptability to patients, quality of accurate surgery to all patients in this group –
life, morbidity or disease prevalence, severity prompt pain relief, lower mortality, enhanced
of disease or mortality)? return to independent living, fewer
complications and shorter hospital stay.
Relevance to NICE guidance Demonstration of comparable sensitivity and
specificity with MRI would enable CT
How would the answer to this question change techniques to be recommended as
future NICE guidance (that is, generate new investigation of first choice in these
knowledge and/or evidence)? circumstances.
Relevance to the NHS Avoiding delay to surgery in hip fracture is
cost-effective. The altered guidance would
What would be the impact on the NHS and support this objective. CT is in addition
(where relevant) the public sector of any new available at lower NHS cost than MRI.
or altered guidance (for example, financial
advantage, effect on staff, impact on strategic
planning or service delivery)?
National priorities The question has a direct bearing on the

Department of Health Best Practice Tariff
Is the question relevant to a national priority initiative to achieve time-to-surgery not
area (such as a national service framework or exceeding 36 hours.
white paper)? The relevant document should
be specified.
Current evidence base There have been no studies comparing the

sensitivity and specificity of modern multi-
What is the current evidence base? What are detector CT techniques with the current gold
the problems with the current evidence base? standard (MRI) in the diagnosis of hip fracture.
(that is, why is further research required?) See Section 5.5.1 of the Full Guideline.
Reference should be made to the section of
the full guideline that describes the current
evidence base, including details of trials and
systematic reviews. The date on which the
final literature search was undertaken should
be specified.
Equality No specific equality issues.
Does the research recommendation address

equality issues? For example, does it focus on
groups that need special consideration, or
focus on an intervention that is not available
for use by people with certain disabilities?
Study design The research design of choice would be a two-

stage design comprising (1) an initial small-
It should also specify the most appropriate scale prospective randomised trial to test an
APPENDIX I 605
study design to address the proposed agreed minimum percentage variability

question(s). Primary research or secondary between methods followed (subject to
research (for example, systematic reviews) can outcome) by (2) a prospective cohort study
be recommended. using CT alone.
Feasibility It should be possible to undertake both

elements in a realistic timescale and at
Can the proposed research be carried out in a reasonable cost. This would not be the case if
realistic timescale and at an acceptable cost? a full-scale Phase 3 trial (as distinct from a
As part of cost-effectiveness analysis, formal prospective cohort) were considered essential.
value-of-information methods may also
sometimes be used to estimate the value for It would be ethically necessary to retain the
money of additional research. Are there any availability of MRI as opt-out gold standard
ethical or technical issues? throughout both studies.
Other comments The ideal study would compare both CT and

MRI in the same patients. This is, however,
Any other important issues should be impractical. The proposed research design has
mentioned, such as potential funders or some limitations, but does have the potential
outcomes of previous attempts to address this to provide useful evidence. The alternative of
issue or methodological problems. However, awaiting an “evolutionary” approach to
this is not a research protocol. progress in this area is less acceptable.
Importance The research is of high priority, since its

findings have the potential to alter future
How important is the question to the overall guidance on the diagnosis of occult hip
guideline? The research recommendation fracture.
should be categorised into one of the
following categories of importance:
• High: the research is essential to inform

future updates of key recommendations in the
guideline
• Medium: the research is relevant to the
recommendations in the guideline, but the
research recommendations are not key to
future updates
• Low: the research is of interest and will fill
existing evidence gaps.
606 APPENDIX I
21.2 Anaesthesia
Research question: What is the clinical and cost effectiveness of regional versus general
anaesthesia on postoperative morbidity in patients with hip fracture?
patients are sedated before regional anaesthesia is administered, and this is not taken into
best addressed by a randomised controlled trial. This would ideally be a multi-centre trial
including 3000 participants in each arm. This is achievable given that there are about
70,000 to 75,000 hip fractures a year in the UK. The study should have three arms that look
at spinal anaesthesia versus spinal anaesthesia plus sedation versus general anaesthesia;
this would separate those with regional anaesthesia from those with regional anaesthesia
plus sedation. The study would also need to control for surgery, especially type of fracture,
prosthesis and grade of surgeon. A qualitative research component would also be helpful to
study on patient preference for type of anaesthesia.
This needs to be multicentre and could be conducted in one year in the U.K.Sample size,
may need to have 3,000 in each limb which is achievable if one considers that there are 80,
000 hip fractures a year in the UK.
PICO question What is the clinical and cost effectiveness of

Each research recommendation should be regional versus general anaesthesia on
formulated as an answerable question or a set postoperative morbidity in patients with hip
of closely related questions. This should use the fracture?
PICO framework (patient, intervention, Patient: patients undergoing surgical repair for
comparison and outcome) hip fractures
Intervention: regional anaesthesia
Comparison: general anaesthesia
Outcome: postoperative morbidity
Importance to patients or the population. Improved survival following hip fracture.
What would be the impact of any new or Improved analgesia following surgery. Reduced
altered guidance on the population? (for complications such as acute delirium, nausea
example, acceptability to patients, quality of and vomiting.
life, morbidity or disease prevalence, severity
of disease or mortality).
Relevance to NICE guidance The study may give the evidence to give better
How would the answer to this question change guidance to anaesthetists. There have been no
future NICE guidance (that is, generate new studies comparing modern anaesthesia
knowledge and/or evidence)? techniques in this group of patients. The
current evidence is old and unreliable. The hip
fracture population is now older and has more
comorbidities than the population in which the
historical studies were conducted.
APPENDIX I 607
The studies are also important to help patients

and their carers make informed decisions about
the form of anaesthesia most appropriate for
them.
Importance : High
Relevance to the NHS There may be a reduction in length of stay in

What would be the impact on the NHS and patients receiving spinal anaesthesia, without
(where relevant) the public sector of any new sedation. Postoperative recovery should be
or altered guidance (for example, financial quicker.
National priorities SIGN recommend spinal but without any

Is the question relevant to a national priority evidence base. The evidence for benefit is weak
area (such as a national service framework or and was conducted over 30 years ago.
be specified.
Current evidence base No trial evidence was identified

What is the current evidence base? What are
the problems with the current evidence base?
(that is, why is further research required?)
Reference should be made to the section of the
full guideline that describes the current
systematic reviews. The date on which the final
literature search was undertaken should be
specified.
Equality This recommendation does not exclude any

Does the research recommendation address patient group. However, special consideration
equality issues? For example, does it focus on should be given to very frail older people with a
groups that need special consideration, or high prevalence of cognitive impairment.
Study design The study design for the proposed research

would be best addressed by an RCT. This would
It should also specify the most appropriate ideally have three arms (3000 participants
study design to address the proposed each) which looks at spinal versus spinal plus
question(s). Primary research or secondary sedation versus general anaesthsia, this would
research (for example, systematic reviews) can separate those with regional anaesthesia from
be recommended. those with regional anaesthesia plus sedation.
The study would also need to control for
surgery, especially type of fracture, prosthesis
and grade of surgeon.
A qualitative research component would also

be helpful to study on patient preference for
608 APPENDIX I
Feasibility Although the number of participants suggested

is relatively high, it is worth considering that
Can the proposed research be carried out in a there are over 80,000 patients admitted with
realistic timescale and at an acceptable cost? hip fractures each year. This should be feasible
As part of cost-effectiveness analysis, formal by conducting a multi-centre RCT.
sometimes be used to estimate the value for
money of additional research. Are there any
ethical or technical issues?
Other comments Potential funders include : The National

Any other important issues should be Institute for Health Research (NIHR), ASTRA
mentioned, such as potential funders or foundation.
outcomes of previous attempts to address this
issue or methodological problems. However,
this is not a research protocol.
Importance High. The research is essential to inform future

How important is the question to the overall updates of key recommendations in the
guideline? The research recommendation guideline.
should be categorised into one of the following
categories of importance:

guideline
future updates
APPENDIX I 609
21.3 Displaced intracapsular hip fractures

Research question:
What is the clinical and cost effectiveness of large head total hip replacement versus
hemiarthroplasty on functional status, reoperations and quality of life in patients with
displaced intracapsular hip fracture?

intervention.
PICO question Question: What is the clinical and cost

effectiveness of large head total hip
Each research recommendation should be replacement versus hemiarthroplasty on
formulated as an answerable question or a set functional status, reoperations and quality of
of closely related questions. This should use the life in patients with displaced intracapsular hip
PICO framework (patient, intervention, fracture?
comparison and outcome)
Patients: Patients sustaining displaced
intracapsular hip fractures
Intervention: Arthroplasty
Comparison: Either hemiarthroplasty (half a hip

replacement) or total hip replacement with a
large head
Outcome: Timely functional status, cost

effectiveness,re-operations and quality of life
at one year
610 APPENDIX I
Importance to patients or the population. Presently there are over 30,000 who sustain a
What would be the impact of any new or displaced intracapsular hip fracture per year in
altered guidance on the population? (for the United Kingdom. Whilst there is evidence
example, acceptability to patients, quality of that total hip replacement with a small femoral
life, morbidity or disease prevalence, severity head gives some advantages in specific groups
of disease or mortality). (3 small RCTs) the concern has been the risk of
dislocations. The technology has advanced and
it is now possible to perform large head
(>36mm) total hip replacement which
significantly reduces the risk of dislocation and
may improve function. The drawback is the
increased cost (between £1000 - £2000 or >10-
20% of the tariff)
Relevance to NICE guidance Presently the NICE recommendations

How would the answer to this question change recommend replacement arthroplasty and is
future NICE guidance (that is, generate new only specific about a defined group of
knowledge and/or evidence)? cognitively unimpaired, previously mobile and
with no significant comorbidities. There is
currently widespread practice in arthroplasty
and there is an increase use of more expensive
prosthesis. Surgeons are beginning to adopt
large head technology without evidence of
effectiveness, cost benefit or consideration of
complication rates.
This recommendation is considered high by the

NICE Hip Fracture Development Group as the
results of this study would advise NICE on
future recommendations for the large and
vulnerable group of patients
Relevance to the NHS The NHS would be in a better position to focus

resources on those in most need. Better
What would be the impact on the NHS and function of the large head total hip
(where relevant) the public sector of any new replacement may reduce care costs in both the
or altered guidance (for example, financial acute setting and rehabilitation.
National priorities Improving the care of those suffering fragility

fractures is a NHS priority. Hip fractures are the
Is the question relevant to a national priority largest cost of this group and account for two
area (such as a national service framework or thirds of all hospital days due to fractures and
white paper)? The relevant document should 87% of the costs (£385million 2007).
be specified.
Current evidence base One cohort study has been presented on large
head total hip replacement and three previous
What is the current evidence base? What are RCTs on small head total hip replacements have
APPENDIX I 611
the problems with the current evidence base? been published

Reference should be made to the section of the
full guideline that describes the current
systematic reviews. The date on which the final
literature search was undertaken should be
specified.
Equality Yes, very frail older people with a high

prevalence of cognitive impairment.
Study design Design: A randomised controlled trial of

displaced intracapsular fractures in previously
It should also specify the most appropriate mobile patients between hemiarthroplasty and
study design to address the proposed large head total hip replacement.
question(s). Primary research or secondary
research (for example, systematic reviews) can Outcome: Does large head arthroplasty
be recommended. improve recovery of mobility one year after
surgical management of displaced intracapsular
hip fracture
Feasibility The research would be ethically and technically

feasible.
Can the proposed research be carried out in a
realistic timescale and at an acceptable cost? The research costs would need to be
As part of cost-effectiveness analysis, formal considered in the context that participants
value-of-information methods may also would still need treatment if outside a trial
sometimes be used to estimate the value for which would set the research costs into proper
money of additional research. Are there any context and perspective.
Other comments The National Institute for Health Research

Any other important issues should be (NIHR) would be an appropriate funding source.
mentioned, such as potential funders or Industry support would off lay excess implant
outcomes of previous attempts to address this costs

How important is the question to the overall updates of key recommendations in the
guideline? The research recommendation guideline.
should be categorised into one of the following
categories of importance:
guideline
612 APPENDIX I

future updates
APPENDIX I 613
21.4 Intensive rehabilitation therapies after hip fracture

Research question:
What is the clinical and cost effectiveness of additional intensive physiotherapy and/or
occupational therapy (for example progressive resistance training) after hip fracture?
The rapid restoration of physical and self care functions is critical to recovery from hip
fracture, particularly where the goal is to return to the patient to preoperative levels of
around, 400 to 500 patients, and the primary outcome should be physical function and
PICO question Question: What is the clinical and cost

effectiveness of additional intensive
Each research recommendation should be physiotherapy and/or occupational therapy
formulated as an answerable question or a set (for example progressive, resistance training)
of closely related questions. This should use after hip fracture?
the PICO framework (patient, intervention,
comparison and outcome) Patients: All patients who have a fracture,
studies should consider all forms of surgical
treatment. Separate studies maybe needed for
those with severe cognitive impairment and
those without (depending on specifics of the
intervention)
Intervention: Progressive therapy protocols
Comparison: Usual care therapy
Outcome: Restoration of mobility, health

related quality of life, falls, residence, ADL
/IADL abilities, linked geriatric syndromes,
resource use.
Importance to patients or the population. Patients and their families value mobility very
What would be the impact of any new or highly. The ability to walk even short distances,
altered guidance on the population? (for can mean the difference between being able
614 APPENDIX I
example, acceptability to patients, quality of to live at home, or not. The step between
life, morbidity or disease prevalence, severity being able to walk outside and inside is greater
of disease or mortality). still. The same can be said for key skills like
dressing and bathing. The impact of improved
mobility, strength, balance and function would
have a substantial impact on the patient and
their family, as well as the requirement for
long term residential or at home care.
Relevance to NICE guidance It would enable NICE to come to a decision on

How would the answer to this question change whether to recommend more intensive
future NICE guidance (that is, generate new physiotherapy and/or occupational therapy, by
knowledge and/or evidence)? generating new evidence on clinical and cost
effectiveness. This is very important to the
guideline – at the moment we have made
several statements about volume (frequency
of therapy) but not of content. The guideline
would be strengthened considerably with this
additional information.
The level would be high.
Relevance to the NHS There would possibly be an increase in the

amount of therapy time that is needed, and
What would be the impact on the NHS and this would incur impacts on strategic planning
(where relevant) the public sector of any new and service delivery.
or altered guidance (for example, financial
National priorities Yes the national service framework for older

people.
Is the question relevant to a national priority
area (such as a national service framework or
be specified.
Current evidence base Very limited trial evidence, and no trial

evidence for some interventions.
the problems with the current evidence base?
Reference should be made to the section of
the full guideline that describes the current
systematic reviews. The date on which the
final literature search was undertaken should
be specified.
Equality Yes these are vulnerable older adults who

need special consideration, particularly if they
Does the research recommendation address have cognitive impairment or frailty. These
APPENDIX I 615
equality issues? For example, does it focus on types of services are not currently provided to
groups that need special consideration, or many hip fracture patients, and certainly not
focus on an intervention that is not available those with cognitive impairments
Study design Design: A randomised controlled trial of

intensive therapy (to be specified) versus usual
It should also specify the most appropriate care therapy
study design to address the proposed
question(s). Primary research or secondary Outcome: Mobility, function, health related
research (for example, systematic reviews) can quality of life, resource use, and costs (health
be recommended. and social care)
Feasibility The research would be ethically and

technically feasible.
Can the proposed research be carried out in a
realistic timescale and at an acceptable cost? The outcome and research question is
As part of cost-effectiveness analysis, formal sufficiently important to merit a large scale
value-of-information methods may also randomised controlled trial
Other comments The National Institute for Health Research

Any other important issues should be (NIHR) HTA would be an appropriate funding
mentioned, such as potential funders or source.
outcomes of previous attempts to address this

updates of key recommendations in the
How important is the question to the overall guideline.
guideline? The research recommendation

guideline
future updates
616 APPENDIX I
21.5 Early Supported Discharge in Care Home patients

Research question:
What is the clinical and cost effectiveness of early supported discharge on mortality, quality
of life and functional status in patients with hip fracture who are admitted from a care
home?
Given the patient frailty and comorbidities, rehabilitation may have no effect on clinical
outcomes for this group. However, the fact that they already live in a home where they are
supported by trained care staff, clearly provides an opportunity for a systematic approach
to rehabilitation. Early multidisciplinary rehabilitation based in care homes or nursing
homes would take advantage of the day-to-day care arrangements already in place and
provide additional NHS support to deliver naturalistic rehabilitation, where problems are
tackled in the patient’s residential.
care required.
The research would follow a two-stage design: (1) an initial feasibility study to refine the
collaboration between the Hip Fracture Programme team, care-home staff and other
community-based professionals, and (2) a cluster randomized controlled comparison (with
two or more intervention units and matched control units) set against agreed outcome
criteria. The latter should include those specified above, together with measures of the
impact on care-home staff activity and cost, as well as qualitative data from patients on
relevant quality-of-life variables.
APPENDIX I 617
PICO question Patients: Elderly hip fracture patients

admitted from a care/nursing home
Each research recommendation should be Intervention: Structured multidisciplinary
formulated as an answerable question or a rehabilitation
set of closely related questions. This should
use the PICO framework (patient, Comparison: Standard care
intervention, comparison and outcome).
Outcome: Reduction in hospital LOS, short
and long-term functional improvement,
reduction in readmission to hospital,
reduction in upgrade from care to nursing
home dependency.
Importance to patients or the population. Reduced dependency
What would be the impact of any new or
altered guidance on the population? (for
example, acceptability to patients, quality of
life, morbidity or disease prevalence, severity
of disease or mortality).
Relevance to NICE guidance The answer to this question is key to

How would the answer to this question guidance on early supported discharge in hip
change future NICE guidance (that is, fracture patients admitted from care home,
generate new knowledge and/or evidence)? who represent a significant proportion of
patients
With this information available NICE would

be in a position to recommend early
supported discharge in this group of patients.
Importance : High
Relevance to the NHS Reduction in hospital LOS will allow greater

efficiency with respect to usage of trauma
What would be the impact on the NHS and beds.
(where relevant) the public sector of any new
or altered guidance (for example, financial Reduction in re-admissions, upgraded
advantage, effect on staff, impact on dependency to nursing homes represent
strategic planning or service delivery)? significant cost savings
National priorities A number of national guidelines now

recommend the need for research in care
Is the question relevant to a national priority home patients following hip fracture (SIGN,
area (such as a national service framework or Orthopaedic Blue Book, NIHR HTA review,
white paper)? The relevant document should Cochrane review).
be specified.
Current evidence base No trial evidence was identified.

the problems with the current evidence
base? (that is, why is further research
618 APPENDIX I
required?) Reference should be made to the

section of the full guideline that describes the
current evidence base, including details of
trials and systematic reviews. The date on
which the final literature search was
undertaken should be specified.
Equality Yes, very frail older people with a high

prevalence of cognitive impairment.
Study design This will comprise: a systematic literature

review, focusing on rehabilitation in care
It should also specify the most appropriate homes; a qualitative interview study with
study design to address the proposed care home residents, their families, care
question(s). Primary research or secondary home staff, allied health professionals and
research (for example, systematic reviews) inpatient orthopaedic staff regarding
can be recommended. discharge planning and rehabilitation for
these patients; and an evaluation of a pilot
early supported multidisciplinary
rehabilitation service compared to usual care.
Feasibility The research would be ethically and

technically feasible, at an acceptable level of
Can the proposed research be carried out in a cost.
realistic timescale and at an acceptable cost?
As part of cost-effectiveness analysis, formal
Other comments Potential funders include :The National

Any other important issues should be Institute for Health Research (NIHR), BUPA,
mentioned, such as potential funders or Alzheimer’s Society
outcomes of previous attempts to address
this issue or methodological problems.
However, this is not a research protocol.
Importance High. The research is essential to inform

future updates of key recommendations in
How important is the question to the overall the guideline.
guideline? The research recommendation

future updates of key recommendations in
APPENDIX I 619
the guideline
future updates
22 Appendix J: Excluded studies
Chapter Study ID Reasons for exclusion
Diagnosis Lubovsky et al (2005)196 The trial was excluded because of the very
small sample size. Only 13 patients
included and only 6 patients received CT
and MRI. The results were reported in a
way that did not allow calculations of
sensitivity and specificity.
Timing of surgery Davis et al (1988)68 No baseline characteristics, no adjustment
for comorbidity.
Timing of surgery Franzo et al (2005)101 No clear explanation of adjustment and no
baseline characteristics for each group.
Timing of surgery Gdalevich et al (2004)108 No baseline characteristics, no adjustment
for comorbidity.
Timing of surgery Hoenig et al (1997)146 Not only surgical delay investigated, unable
to extract raw data.
Timing of surgery Kenzora et al (1984)177 No baseline characteristics, no adjustment
for comorbidity.
Timing of surgery Mackenzie wt al (2006)198 Letter/short correspondence.
Timing of surgery McGuire et al (2004)209 The aim of the study is on day of the week
of admission.
Timing of surgery Moran et al (2005)215 No baseline characteristics, no adjustment
for comorbidity.
Timing of surgery Novack et al (2007)243 Adjusted hazard ratios given.
Timing of surgery Rae et al (2007)280 Baseline characteristics not given for each
group.
Timing of surgery Rogers et al (1995)290 No baseline characteristics, no adjustment
for comorbidity.
620 APPENDIX J
Timing of surgery Sebestyen et al (2008)302 No adjustment for comorbidity.

Timing of surgery Shabat et al (2003)304 Inadequate methodology.
Timing of surgery Sircar et al (2007)312 No baseline characteristics, no adjustment
for comorbidity.
Timing of surgery Sund & Liski (2005)322 Adjusted odds ratios for provider
characteristics.
Analgesia Gorodetskyi et al Not a study of nerve blocks.
(2007)122
Analgesia Mannion et al (2005)202 No ’control’ group without the nerve
block.
Analgesia Marhofer et al (1998)205 No ’control’ group without the nerve
block.
Analgesia Mutty et al (2007)221 No proximal femoral fractures included.
Analgesia Piangatelli et al (2004)272 No ’control’ group without the nerve block.
Analgesia Schiferer et al (2007)301 Inclusion of participants with other
conditions. The trialists were unable to
provide separate results for only the hip
fracture participants.
Analgesia Turker et al (2003)332 No ‘control’ group without the nerve block
Analgesia Van Leeuwen et al No ’control’ group without the nerve block.
(2000)342
Anaesthesia Alonso Chico et al (2003)6 Not a trial of different types of anaesthesia
but a comparison of different drugs within
one form of anaesthesia.
Anaesthesia Barna (1981)13 No randomisation of patients.
Anaesthesia Ben-David et al (2000)16 Not a trial of different types of anaesthesia
Anaesthesia Coleman et al (1988)56 The study was excluded as it involved a
change in the types of drugs used only, not
a change in the method of anaesthesia.
Anaesthesia Critchley et al (1995)57 Not a trial of different types of anaesthesia
Anaesthesia Darling et al (1994)64 The study was excluded as it was not felt
relevant to this review as no clinical
outcomes were reported.
Anaesthesia Dyson et al (1988)75 Lack of outcome data for the anaesthesia
comparison.
Anaesthesia El-Zahaar et al (1995)79,334 This trial was excluded because separate
results for patients having surgery for a hip
fracture were not presented.
Anaesthesia Favarel-Garrigues et al The trial was excluded as it was not
(1996)92 considered a comparison of different forms
of anaesthesia, only of a modification of
anaesthetic technique.
Anaesthesia Hemmingsen & Nielsen Not a trial of different types of anaesthesia
(1991)143 but a comparison of different drugs within
Anaesthesia Marhofer et al (1999)204 Not a comparison of anaesthetic methods.
Anaesthesia Matot et al (2003)207 Compared techniques outside the scope of
APPENDIX J 621
this review.
208
Anaesthesia Maurette et al (1993) The trial was excluded as it was a trial of
different drugs with the same anaesthetic
technique, not a comparison of different
types of anaesthesia.
Anaesthesia Naja et al (2000)223 No randomisation of patients.
Anaesthesia Nishikawa et al (2002)242 Not a comparison of different types of
anaesthesia.
Anaesthesia Owen & Hutton (1982)252 Not a comparison of anaesthetic
techniques.
Anaesthesia Sinclair et al (1997)311 Not a comparison of different types of
anaesthesia.
Anaesthesia Sutcliffe & Parker No randomisation of patients.
(1994)323
Anaesthesia Tonczar & Hammerle The study was excluded as it involved a
(1981)329 neuroleptic anaesthesia and the only
outcome measures were plasma
catecholamines, cortisol, blood pressure
and changes in heart rate.
Anaesthesia Ungemach (1987)333 The trial was excluded as it was a
comparison of different drugs within one
type of anaesthesia (general anaesthesia)
and not a comparison of different
anaesthetic techniques.
Surgeon Seniority Claque et al (2002) 52 Retrospective study, unclear if adjusted for
confounders. Not stated how patients
were allocated to surgeons.
Surgeon Seniority Englesbe et al (2009) 82 Compares outcomes at time when new
trainees start compared to other times of
the year. Not about surgeon seniority.
Surgeon Seniority Evans et al (1979) 87 No results or data for surgeon seniority
analysis.
Surgeon Seniority Faraj & Drakau (2007) 90 No adjustment for confounders and no
indication of how patients were allocated
to surgeons.
Surgeon Seniority Fung et al (2007) 104 No outcome of interest.
Surgeon Seniority Giannoudis et al (1998) No outcome of interest.
111
Surgeon Seniority Grimley et al (1980) 126 Compares hospitals outcomes rather than
surgeon seniority. Unclear if retrospective
or prospective. No indication of how
patients were allocated to surgeons.
Surgeon Seniority Harper & Walsh (1985) 135 Unclear if retrospective or prospective, no
adjustment for confounders.
Surgeon Seniority Holmberg et al (1987) 149 Unclear if retrospective or prospective, no
adjustment for confounders.
Surgeon Seniority Holt et al (1994) 151 No adjustment for confounders.
Surgeon Seniority Levi & Gebuhr (2000) 192 Unclear if retrospective or prospective, no
adjustment for confounders, no outcomes
measured by surgeon seniority only
reports in words there was no difference
622 APPENDIX J
between registrars and consultants.

181
Surgeon Seniority Kukla et al (2001) Unclear if retrospective or prospective.
Examines years of experience but
inexperienced surgeons were supervised.
Results presented as a continuous variable.
Surgeon Seniority Parker et al (1994) 271 Not surgeon seniority, investigates the use
of a special "Hip Fracture Team".
Surgeon Seniority Sarvilinna et al (2002) 299 Retrospective study, no adjustment for
confounders.
Surgeon Seniority Sehat et al (2006) 303 Not about surgeon seniority.
Surgeon Seniority Weinrauch (2006) 350 Not stated how patients were assigned to
surgeons. Not stated the total number of
surgeons involved nor how many involved
in each category. Does not adjust for any
confounders.
Internal fixation vs Bhandari et al (2003) 21 Systematic review, used Cochrane review
arthroplasty instead.
Internal fixation vs Bjorgul et al (2006) 25 Non-randomised study.
arthroplasty
Internal fixation vs Bray et al (1988) 35 Excluded from Cochrane review due to
arthroplasty inadequate randomisation. Patients were
allocated according to day of week and
surgeon preference. In addition to the low
numbers recruited five were lost to follow-
up.
Internal fixation vs El-Abed et al (2005) 78 Excluded from Cochrane review as non-
arthroplasty randomised study, type of procedure used
was by the preference of the attending
surgeon on the day of admission.
Internal fixation vs Gjertsen et al (2010) 118 Non-randomised study.
arthroplasty
Internal fixation vs Haentjens et al (2005) 130 Non-randomised study.
arthroplasty
Internal fixation vs Heetveld et al (2009) 142 Non-randomised study.
arthroplasty
Internal fixation vs Hunter (1974) 156 Excluded from Cochrane review as non-
arthroplasty randomised study.
Internal fixation vs Hunter (1969) 155 Excluded from Cochrane review as non-
Internal fixation vs Neander (2000) 238 Excluded from Cochrane review due to
arthroplasty inadequate randomisation procedure. The
first 20 patients were randomised with
closed envelopes but the last 80 were
allocated according to the day of week
they were admitted (Monday to Thursday
total hip replacement, Friday to Sunday
reduction and fixation).
Internal fixation vs Parker (1992) 261 Excluded from Cochrane review as non-
Internal fixation vs Riley (1978) 285 Excluded from Cochrane review as study
arthroplasty provided no adequate data.
APPENDIX J 623
Internal fixation vs Rodriguez et al (1987) 289 Excluded from Cochrane review as non-
Internal fixation vs Rogmark & Johnell (2006) Systematic review, used Cochrane review
292
Internal fixation vs Sikorski & Barrington This comparison excluded from Cochrane
arthroplasty (1981) 309 review due to poor methodological quality.
Internal fixation vs Stewart (1984) 321 Excluded from Cochrane review as non-
Internal fixation vs Wang et al (2009) 347 Systematic review, used Cochrane review
Hemiarthroplasty vs Goh et al (2009) 120 Systematic review, used Cochrane review
total hip instead.
replacement
Hemiarthroplasty vs Haentjens et al (2005) 130 Non-randomised study.
total hip
replacement
Hemiarthroplasty vs Heetveld et al (2009) 142 Non-randomised study.
total hip
replacement
Hemiarthroplasty vs Kavcic et al (2006) 172 Methodology not reported. Only mentions
total hip patients were randomly selected. No
replacement indication of allocation concealment,
method of randomisation, blinding, or
inclusion/exclusion criteria.
Cement Ahn et al (2008) 2 Systematic review that includes
randomised and non-randomised studies.
Used Cochrane review.
Cement Bajammal et al (2008) 10 Systematic review of cement use in
appendicular fractures, not just hip
fractures. Used Cochrane review.
Cement Christie et al (1994) 50 Excluded from Cochrane review as
biometric study with no clinical outcome
measures. No methods given for RCT, no
outcomes from our protocol.
Cement Clark et al (2001) 53 Excluded from Cochrane review as non-
randomised study.
Cement Dorr et al (1986) 73 Cemented vs uncemented
hemiarthroplasty not a randomised
comparison
Cement Faraj & Branfoot (1999) 89 Excluded from Cochrane review as non-
randomised study, use of cement was at
operating surgeon's preference.
Cement Field & Rushton (2005) 93 Excluded from Cochrane review because of
a limited number of cases using what is at
present an experimental new cup.
Cement Georgescu et al (2004) 109 Excluded from Cochrane review because of
a lack of reported results within the
conference abstract
Cement Gierer et al (2002) 112 Excluded from Cochrane review as non-
randomised study, use of cement was at
operating surgeon's preference.
624 APPENDIX J
Cement Graf et al (2000) 123 Excluded from Cochrane review as non-

randomised study.
Cement Johnson et al (2001) 164 Excluded from Cochrane review as non-
randomised study.
Cement Karpmann et al (1992) 170 Excluded from Cochrane review as there
was inadequate reporting of the trial.
Attempts were made to contact the
trialists for further information, without
success.
Cement Khan et al (2002) 178 Systematic review, excluded as used
Cochrane review instead.
Cement Lachiewicz et al (2008) 184 Elective hip replacement patients, not hip
fracture patients.
Cement Leidinger et al (2002) 190 Excluded from Cochrane review as
variations of cementing technique are not
part of the protocol
Cement Pitto et al (2000) 273 Excluded from Cochrane review as small
numbers and only outcome measure is
transoesophagel echocardiography shown
embolism. No methods given for RCT, no
outcomes from our protocol.
Cement Sadr & Arden (1977) 295 Excluded from Cochrane review as unclear
whether randomised, the use of Proplast
coated prosthesis is no longer prevalent,
small study of 40 patients with limited
reporting of outcomes for the 25 assessed
patients at follow up.
Cement Vochteloo et al (2009) 345 Protocol for a randomised study, study not
completed.
Surgical approach to Barden et al (2001) 12 Excluded from Cochrane review as not a
hemiarthroplasty comparison of different surgical
approaches.
Surgical approach to Cashman & Cashman Elective hip replacement patients.
hemiarthroplasty (2008) 47
Surgical approach to Chan & Hoskin (1975) 49 No adjustment for confounders.
hemiarthroplasty
Surgical approach to Enocson et al (2009) 83 About total hip replacement.
hemiarthroplasty
Surgical approach to Enocson et al (2010) 84 Possible double counting of the included
hemiarthroplasty study ENOCSON 2008. No adjustment for
confounders.
Surgical approach to Keene et al (1993) 175 Not about surgical approach.
hemiarthroplasty
Surgical approach to Lafosse et al (2007) 187 About a minimumally invasive approach.
hemiarthroplasty
hemiarthroplasty
hemiarthroplasty
Surgical approach to Unwin & Thomas (1994) No adjustment for confounders.
336
hemiarthroplasty
APPENDIX J 625
Surgical approach to Wang et al (2010) 346 About a minimumally invasive approach.

hemiarthroplasty
Surgical approach to Yang et al (2010) 356 About a minimumally invasive approach.
hemiarthroplasty
Surgical approach to Widman & Isacson Excluded from Cochrane review as not a
hemiarthroplasty (2001)353 comparison of different surgical
approaches.
Screws/nails Baumgaertner et al No relevant outcomes.
(1998)15
Screws/nails Benum et al (1994)18 Abstract only.
Screws/nails Butt et al (1995)40 Does not meet our inclusion criteria:
includes trochanteric and subtrochanteric
combined.
Screws/nails Davis et al (1988)67 Does not meet our inclusion criteria:
includes trochanteric and associated
subtrochanteric combined.
Screws/nails Dujardin et al (2001)74 Experimental nail not used commercially.
Screws/nails Kuwabara et al (1998)183 Unable to obtain paper.
Screws/nails Lee et al (2007)188 Does not meet our inclusion criteria: all
high energy trauma (subtrochanteric
fractures).
Screws/nails Mehdi et al (2000)212 Abstract only.
Screws/nails Michos et al (2001)213 Abstract only.
Screws/nails Mott et al (1993)217 Abstract only.
Screws/nails Pahlpatz & Langius Does not meet our inclusion criteria:
(1993)253 Includes trochanteric and subtrochanteric
fractures combined.
Screws/nails Rahme & Harris (2007)281 Does not meet our inclusion criteria: all
high energy trauma (subtrochanteric).
Surgical procedures Giancola et al (2008)110 No cost figures were reported.
(economic evidence)
Surgical procedures Gill & Ursic (2007)114 Inadequate methodological design and
(economic evidence) limited applicability to the UK NHS.
Surgical procedures Kim et al (2005)179 Proximal femoral nail compared to long-
(economic evidence) stem cementless calcar-replacement
prosthesis (not an included intervention).
Surgical procedures Marinelli et al (2008)206 Inadequate methodology.
(economic evidence)
Surgical procedures Rogmark et al (2003)291 The study does not distinguish patients on
(economic evidence) the basis of whether they received
hemiarthroplasty or total hip replacement.
Mobilisation Binder et al (2004)24 The comparison is not versus usual care.
Mobilisation Galea et al (2008)106 The comparison is not versus usual care,
both have a targeted plan.
Mobilisation Graham (1968)124 The intervention is weight bearing at 2
weeks or 12 weeks. Not relevant to our
review question.
Mobilisation Mangione et al (2005)201 The comparison is not versus usual care.
Mobilisation Resnick et al (2007)283 Does not answer our review question:
augmented mobilisation vs. usual care.
626 APPENDIX J
Mobilisation Tsauo et al (2005)330 Does not answer our review question:

community mobilisation vs. usual care.
Mobilisation Yu-yahiro et al (2009)359 Does not answer our review question:
community mobilisation vs usual care.
MDR Fordham et al (1986)96 Discussion paper with a cost benefit
analysis
MDR Giusti et al (2006)117 Does not meet our inclusion criteria for
MDR team: medicine; nursing;
physiotherapy; occupational therapy; and
social care. Additional components may
include: nutrition; pharmacy; and clinical
psychology.
MDR Gonzalez-Montalvo et al Mixed intervention, acute orthogeriatric
(2010)121 unit model, plus early surgery.
MDR Ho et al (2009)145 Letter to editor.
MDR Holt et al (2010)152 Does not meet our inclusion criteria: no
outcomes reported that were prioritised in
our protocol. Survival analysis rather than
mortality.
MDR Iliffe et al (2010)159 Protocol only, not full results.
MDR Kuisma (2002)180 Does not meet our inclusion criteria for
psychology.
MDR O’Cathain (1994)245 Observational study.
MDR Olsson et al (2007)248 Does not meet our inclusion criteria for
psychology.
MDR Pryor & Williams Observational study.
(1989)275
MDR Richards et al (1998)284 Mixed population, only 31% hip fracture
patients.
MDR Ryan et al (2006)293 Does not answer our review question.
Intervention is intensity of multidisciplinary
rehab (intensive: 6 or more face-to-face
sessions per week from MDR team vs. less
intensive: 3 or less face-to-face sessions
per week).
MDR Shyu et al (2010)307 Reports 2 year follow up. 1 year data
already included, which is the longest time
point stated in our protocol.
MDR Uy et al (2008)338 Very low number of patients. N = 11
Hospital MDR Cameron et al (2000)41 The studies included in the HTA were
(economic evidence) grouped in a different way to that
considered for our clinical review, and
therefore its cost analysis was not
APPENDIX J 627
applicable for our review question.

55
Community MDR Coast et al (1998) Mixed population with only 31% hip
(economic evidence) fracture patients.
Community MDR Van Balen et al (2002)340 Patients in the early supported discharge
(economic evidence) scheme were only discharged to a nursing
home with rehabilitation facilities and not
to their own home.
Patient views Boutin-Lester & Gibson Only 1 / 5 of the patients had HF. This
(2002) 32 patient also had osteoporosis.
Patient views Closs & Briggs (2002) 54 Words used by patients to describe pain,
not hip fracture patients only.
Patient views Franchignoni (2002) 100 Only 5/55 patients had hip fracture.
Patient views Gjertsen et al (2008) 119 Not qualitative research into patient views.
Patient views Hallstrom et al (2000) 131 7/9 patients had cervical fractures.
Patient views Harrison (2006) 138 Very brief summary of MSc thesis, unable
to obtain a copy of thesis.
Patient views Hedman et al (2008) 141 Compares level of care received between
cognitively impaired and cognitively intact
hip fracture patients in two Swedish
hospitals.
Patient views Huang & Acton (2009) 154 Patient views about the period after
discharge from rehabilitation in Taiwan.
Patient views Lin & Lu (2005) 194 Caregivers views after discharge from
hospital not patient views.
Patient views Lin (2006) 193 Not a patient view study.
Patient views Magasi et al (2009) 199 About choice of a rehabilitation facility in
the US, not applicable to UK.
Patient views Resnick et al (2005) 282 Patient views on a specific exercise
programme adopted at a centre in the
USA.
Patient views Robinson (1999) 287 Patient views about adapting to life after
rehabilitation.
Patient views Smith et al (1997) 315 Review of article on report about patient
views on discharge information. Unable to
obtain a copy of full report with qualitative
research.
Patient views Webster (1976) 349 Not qualitative research of patient views.
Patient education Allegrante et al (2007) 5 Not patient education intervention alone.
Patient education Bhandari & Tornetta About which way of communicating risk
(2004) 22 ratios to patients.
Patient education Elinge et al (2003) 80 Group learning programme started 3
months after fracture.
Patient education Gill & Ursic (1994) 115 Education for nurses not patients.
Patient education Jackson (2010) 160 Education intervention for healthcare
professionals not patients.
Patient education Tappen et al (2003) 326 Effect of video intervention of recovery
from hip surgery. Unclear how patients
were allocated to interventions.
Patient education Yoon et al (2008) 357 Non-randomised study.
628 BIBLIOGRAPHY
Bibliography
Bibliography
1. Adams HA, Wolf C, Michaelis G, Hempelmann G. Postoperative course and endocrine

stress reaction of geriatric patients with para-articular hip fractures. Prospective
randomized study comparing spinal anesthesia and halothane intubation narcosis.
Anästhesie, Intensivtherapie, Notfallmedizin 1990, 25(4):263-70. (Guideline Ref ID:
ADAMS1990)
2. Ahn J, Man LX, Park S, Sodl JF, Esterhai JL. Systematic review of cemented and
uncemented hemiarthroplasty outcomes for femoral neck fractures. Clinical
Orthopaedics and Related Research 2008, 466(10):2513-8. (Guideline Ref ID: AHN2008)
3. Ahrengart L, Tornkvist H, Fornander P, Thorngren KG, Pasanen L, Wahlstrom P et al. A

randomized study of the compression hip screw and Gamma nail in 426 fractures.
Clinical Orthopaedics and Related Research 2002,(401):209-22. (Guideline Ref ID:
AHRENGART2002)
4. Al-Ani AN, Samuelsson B, Tidermark J, Norling A, Ekstrom W, Cederholm T et al. Early

operation on patients with a hip fracture improved the ability to return to independent
living. A prospective study of 850 patients. Journal of Bone & Joint Surgery - American
Volume 2008, 90A(7):1436-42. (Guideline Ref ID: ALANI2008)
5. Allegrante JP, Peterson MG, Cornell CN, Mackenzie CR, Robbins L, Horton R et al.
Methodological challenges of multiple-component intervention: lessons learned from a
randomized controlled trial of functional recovery after hip fracture. Hss Journal 2007,
3(1):63-70. (Guideline Ref ID: ALLEGRANTE2007)
6. Alonso Chico A, Cruz Pardos P, Alvarez Grau J, Pachoco Jimenez A, Arregui Martinez de
Lejarza M, anchez Garcia ML et al. Comparison of the hemodynamic response in
subarachnoid anesthesia with bupivacaine versus bupivacaine with fentanyl in
traumatology surgery in elderly patients. Revista Espanola de Anestesiologia y
Reanimacion 2003, 50(1):17-22. (Guideline Ref ID: ALONSOCHICO2003)
7. Ara R, Brazier J. Deriving an algorithm to convert the eight mean SF-36 dimension scores
into a mean EQ-5D preference-based score from published studies (where patient level
data are not available). Value in Health 2008, 11(7):1131-43. (Guideline Ref ID:
ARA2008)
BIBLIOGRAPHY 629
8. Archibald G. Patients' experiences of hip fracture. Journal of Advanced Nursing 2003,

44(4):385-92. (Guideline Ref ID: ARCHIBALD2003)
9. Aune AK, Ekeland A, Odegaard B, Grogaard B, Alho A. Gamma nail vs compression screw
for trochanteric femoral fractures. 15 reoperations in a prospective, randomized study
of 378 patients. Acta Orthopaedica Scandinavica 1994, 65(2):127-30. (Guideline Ref ID:
AUNE1994)
10. Bajammal SS, Zlowodzki M, Lelwica A, Tornetta P, III, Einhorn TA, Buckley R et al. The
use of calcium phosphate bone cement in fracture treatment. A meta-analysis of
randomized trials. Journal of Bone & Joint Surgery - American Volume 2008, 90(6):1186-
96. (Guideline Ref ID: BAJAMMAL2008)
11. Baker RP, Squires B, Gargan MF, Bannister GC. Total hip arthroplasty and
hemiarthroplasty in mobile, independent patients with a displaced intracapsular
fracture of the femoral neck. A randomized, controlled trial. Journal of Bone & Joint
Surgery - American Volume 2006, 88(12):2583-9. (Guideline Ref ID: BAKER2006)
12. Barden B, Kroger K, Loer F. [Intraoperative Doppler ultrasound of the femoral vein for
maintaining venous patency in hip joint prosthesis implantation]. Unfallchirurg 2001,
104(2):138-42. (Guideline Ref ID: BARDEN2001)
13. Barna B. Comparison of spinal and general anesthesia in the surgical treatment of hip
fractures. Orvosi Hetilap 1981, 122(19):1135-8. (Guideline Ref ID: BARNA1981)
14. Barton TM, Gleeson R, Topliss C, Greenwood R, Harries WJ, Chesser TJ. A comparison of
the long gamma nail with the sliding hip screw for the treatment of AO/OTA 31-A2
fractures of the proximal part of the femur: a prospective randomized trial. Journal of
Bone & Joint Surgery - American Volume 2010, 92(4):792-8. (Guideline Ref ID:
BARTON2010)
15. Baumgaertner MR, Curtin SL, Lindskog DM. Intramedullary versus extramedullary
fixation for the treatment of intertrochanteric hip fractures. Clinical Orthopaedics and
Related Research 1998,(348):87-94. (Guideline Ref ID: BAUMGAERTNER1998)
16. Ben-David B, Frankel R, Arzumonov T, Marchevsky Y, Volpin G. Minidose bupivacaine-

fentanyl spinal anesthesia for surgical repair of hip fracture in the aged. Anesthesiology
2000, 92(1):6-10. (Guideline Ref ID: BENDAVID2000)
17. Bennett G, Dealey C, Posnett J. The cost of pressure ulcers in the UK. Age and Ageing
2004, 33(3):230-5. (Guideline Ref ID: BENNETT2004)
18. Benum P, Grontvedt T, Braten M, Rossvoll I, Walloe A, Ekeland A et al. Gamma nailing
versus CHS in intertrochanteric and subtrochanteric femoral fractures: a prospective
randomised multicentre study. Acta Orthopaedica Scandinavica Supplementum 1994,
260:33-4. (Guideline Ref ID: BENUM1994)
19. Bergeron E, Lavoie A, Moore L, Bamvita JM, Ratte S, Gravel C et al. Is the delay to
surgery for isolated hip fracture predictive of outcome in efficient systems? Journal of
Trauma-Injury Infection & Critical Care 2006, 60(4):753-7. (Guideline Ref ID:
BERGERON2006)
630 BIBLIOGRAPHY
20. Berggren D, Gustafson Y, Eriksson B, Bucht G, Hansson LI, Reiz S et al. Postoperative
confusion after anesthesia in elderly patients with femoral neck fractures. Anesthesia
and Analgesia 1987, 66(6):497-504. (Guideline Ref ID: BERGGREN1987)
21. Bhandari M, Devereaux PJ, Swiontkowski MF, Tornetta P, III, Obremskey W, Koval KJ et
al. Internal fixation compared with arthroplasty for displaced fractures of the femoral
neck. A meta-analysis. Journal of Bone & Joint Surgery - American Volume 2003, 85-
A(9):1673-81. (Guideline Ref ID: BHANDARI2003)
22. Bhandari M, Tornetta P. Communicating the risks of surgery to patients. European

Journal of Trauma 2004, 30(3):177-81. (Guideline Ref ID: BHANDARI2004)
23. Bigler D, Adelhoj B, Petring OU, Pederson NO, Busch P, Kalhke P. Mental function and
morbidity after acute hip surgery during spinal and general anaesthesia. Anaesthesia
1985, 40(7):672-6. (Guideline Ref ID: BIGLER1985)
24. Binder EF, Brown M, Sinacore DR, Steger-May K, Yarasheski KE, Schechtman KB. Effects
of extended outpatient rehabilitation after hip fracture: a randomized controlled trial.
JAMA 2004, 292(7):837-46. (Guideline Ref ID: BINDER2004)
25. Bjorgul K, Reikeras O. Hemiarthroplasty in worst cases is better than internal fixation in
best cases of displaced femoral neck fractures: a prospective study of 683 patients
treated with hemiarthroplasty or internal fixation. Acta Orthopaedica 2006, 77(3):368-
74. (Guideline Ref ID: BJORGUL2006)
26. Blomfeldt R, Tornkvist H, Eriksson K, Soderqvist A, Ponzer S, Tidermark J. A randomised

controlled trial comparing bipolar hemiarthroplasty with total hip replacement for
displaced intracapsular fractures of the femoral neck in elderly patients. Journal of Bone
& Joint Surgery - British Volume 2007, 89(2):160-5. (Guideline Ref ID: BLOMFELDT2007)
27. Blomfeldt R, Tornkvist H, Ponzer S, Soderqvist A, Tidermark J. Internal fixation versus

hemiarthroplasty for displaced fractures of the femoral neck in elderly patients with
severe cognitive impairment. Journal of Bone & Joint Surgery - British Volume 2005,
87(4):523-9. (Guideline Ref ID: BLOMFELDT2005)
28. Borkan JM, Quirk M. Expectations and outcomes after hip fracture among the elderly.
International Journal of Aging & Human Development 1992, 34(4):1992-350. (Guideline
Ref ID: BORKAN1992)
29. Borkan JM, Quirk M, Sullivan M. Finding meaning after the fall: injury narratives from
elderly hip fracture patients. Social Science & Medicine 1991, 33(8):947-57. (Guideline
Ref ID: BORKAN1991)
30. Bottle A, Aylin P. Mortality associated with delay in operation after hip fracture:
observational study. British Medical Journal 2006, 332(7547):947-51. (Guideline Ref ID:
BOTTLE2006)
31. Bottle A, Jarman B, Aylin P, Taylor R. Some way to go for consistent implementation of
guidance on hip fracture. British Medical Journal 2004, 328(7448):1097. (Guideline Ref
ID: BOTTLE2004)
BIBLIOGRAPHY 631
32. Boutin-Lester P, Gibson RW. Patients' perceptions of home health occupational therapy.
Australian Occupational Therapy Journal 2002, 49(3):146-54. (Guideline Ref ID:
BOUTINLESTER2002)
33. Bowman AM. Sleep satisfaction, perceived pain and acute confusion in elderly clients
undergoing orthopaedic procedures. Journal of Advanced Nursing 1997, 26(3):550-64.
(Guideline Ref ID: BOWMAN1997)
34. Branfoot T, Faraj AA, Porter P. Cemented versus uncemented Thompson's prosthesis: a
randomised prospective functional outcome study. Injury 2000, 31:280-1. (Guideline Ref
ID: BRANFOOT2000)
35. Bray TJ, Smith-Hoefer E, Hooper A, Timmerman L. The displaced femoral neck fracture.
Internal fixation versus bipolar endoprosthesis. Results of a prospective, randomized
comparison. Clin Orthop Relat Res 1988,(230):127-40. (Guideline Ref ID: BRAY1988)
36. Brichant JF, Blom-Peters L, Buffels R, Lamy M. Central neural blockage failed to decrease
deep venous thrombosis in pateints undergoing hip surgery and receiving low molecualr
weight heparin. British Journal of Anaesthesia 1995, 74(Suppl 1):75. (Guideline Ref ID:
BRICHANT1995)
37. Bridle SH, Patel AD, Bircher M, Calvert PT. Fixation of intertrochanteric fractures of the
femur. A randomised prospective comparison of the gamma nail and the dynamic hip
screw. Journal of Bone & Joint Surgery - British Volume 1991, 73(2):330-4. (Guideline Ref
ID: BRIDLE1991)
38. British Orthopaedic Association. (2007) Advisory book on consultant trauma and
orthopaedic services. London: British Orthopaedic Association. (Guideline Ref ID:
BOA2007A)
39. British Orthopaedic Association. (2007) The care of patients with fragility fractures.
(Guideline Ref ID: BOA2007)
40. Butt MS, Krikler SJ, Nafie S, Ali MS. Comparison of dynamic hip screw and gamma nail: a
prospective, randomized, controlled trial. Injury 1995, 26(9):615-8. (Guideline Ref ID:
BUTT1995)
41. Cameron I, Crotty M, Currie C, Finnegan T, Gillespie L, Gillespie W et al. Geriatric

rehabilitation following fractures in older people: A systematic review. Health
Technology Assessment 2000, 4(2):i-102. (Guideline Ref ID: CAMERON2000C)
42. Cameron I, Lyle D, Quine S. Accelerated rehabilitation after proximal femoral fracture: a
randomised controlled trial. Proceedings of the Annual Conference of the Australian
Association of Gerontology 1991, 26:87-90. (Guideline Ref ID: CAMERON1991)
43. Cameron ID. Coordinated multidisciplinary rehabilitation after hip fracture. Disability
and Rehabilitation 2005, 27(18-19):1081-90. (Guideline Ref ID: CAMERON2005)
44. Cameron ID, Lyle DM, Quine S. Accelerated rehabilitation after proximal femoral
fracture: a randomized controlled trial. Disability and Rehabilitation 1993, 15(1):29-34.
(Guideline Ref ID: CAMERON1993)
632 BIBLIOGRAPHY
45. Cameron ID, Lyle DM, Quine S. Cost effectiveness of accelerated rehabilitation after
proximal femoral fracture. Journal of Clinical Epidemiology 1994, 47(11):1307-13.
(Guideline Ref ID: CAMERON1994A)
46. Casati A, Aldegheri G, Vinciguerra E, Marsan A, Fraschini G, Torri G. Randomized

comparison between sevoflurane anaesthesia and unilateral spinal anaesthesia in
elderly patients undergoing orthopaedic surgery. European Journal of Anaesthesiology
2003, 20(8):640-6. (Guideline Ref ID: CASATI2003)
47. Cashman JP, Cashman WF. Comparison of complications in transtrochanteric and

anterolateral approaches in primary total hip arthroplasty. Orthopedics 2008,
31(11):1085. (Guideline Ref ID: CASHMAN2008)
48. Chakladar A, White SM. Cost estimates of spinal versus general anaesthesia for
fractured neck of femur surgery. Anaesthesia 2010, 65(8):810-4. (Guideline Ref ID:
CHAKLADAR2010)
49. Chan RN, Hoskinson J. Thompson prosthesis for fractured neck of femur. A comparison
of surgical approaches. Journal of Bone & Joint Surgery - British Volume 1975, 57(4):437-
43. (Guideline Ref ID: CHAN1975)
50. Christie J, Burnett R, Potts HR, Pell AC. Echocardiography of transatrial embolism during
cemented and uncemented hemiarthroplasty of the hip. J Bone Joint Surg Br 1994,
76(3):409-12. (Guideline Ref ID: CHRISTIE1994)
51. Chudinov A, Berkenstadt H, Salai M, Cahana A, Perel A. Continuous psoas compartment

block for anesthesia and perioperative analgesia in patients with hip fractures. Regional
Anesthesia and Pain Medicine 1999, 24(6):563-8. (Guideline Ref ID: CHUDINOV1999)
52. Clague JE, Craddock E, Andrew G, Horan MA, Pendleton N. Predictors of outcome
following hip fracture. Admission time predicts length of stay and in-hospital mortality.
Injury 2002, 33(1):1-6. (Guideline Ref ID: CLAGUE2002)
53. Clark DI, Ahmed AB, Baxendale BR, Moran CG. Cardiac output during hemiarthroplasty
of the hip. A prospective, controlled trial of cemented and uncemented prostheses. J
Bone Joint Surg Br 2001, 83(3):414-8. (Guideline Ref ID: CLARK2001)
54. Closs SJ, Briggs M. Patients' verbal descriptions of pain and discomfort following
orthopaedic surgery. International Journal of Nursing Studies 2002, 39(5):563-72.
(Guideline Ref ID: CLOSS2002)
55. Coast J, Richards SH, Peters TJ, Gunnell DJ, Darlow MA, Pounsford J. Hospital at home or
acute hospital care? A cost minimisation analysis. British Medical Journal 1998,
316(7147):1802-6. (Guideline Ref ID: COAST1998)
56. Coleman SA, Boyce WJ, Cosh PH, McKenzie PJ. Outcome after general anaesthesia for
repair of fractured neck of femur. A randomized trial of spontaneous v. controlled
ventilation. British Journal of Anaesthesia 1988, 60(1):43-7. (Guideline Ref ID:
COLEMAN1988)
57. Critchley LA, Stuart JC, Conway F, Short TG. Hypotension during subarachnoid
anaesthesia: haemodynamic effects of ephedrine. British Journal of Anaesthesia 1995,
74(4):373-8. (Guideline Ref ID: CRITCHLEY1995)
BIBLIOGRAPHY 633
58. Crotty M, Kittel A, Hayball N. Home rehabilitation for older adults with fractured hips:
how many will take part? Journal of Quality in Clinical Practice 2000, 20(2-3):65-8.
(Guideline Ref ID: CROTTY2000A)
59. Crotty M, Whitehead C, Miller M, Gray S. Patient and caregiver outcomes 12 months
after home-based therapy for hip fracture: a randomized controlled trial. Archives of
Physical Medicine and Rehabilitation 2003, 84(8):1237-9. (Guideline Ref ID:
CROTTY2003)
60. Crotty M, Whitehead CH, Gray S, Finucane PM. Early discharge and home rehabilitation
after hip fracture achieves functional improvements: a randomized controlled trial.
Clinical Rehabilitation 2002, 16(4):406-13. (Guideline Ref ID: CROTTY2002)
61. Curtis L. (2009) Unit costs of health & social care 2009. Canterbury, Kent: Personal
Social Services Research Unit. (Guideline Ref ID: CURTIS2009)
62. Cuvillon P, Ripart J, Debureaux S, Boisson C, Veyrat E, Mahamat A et al. Analgesia after
hip fracture repair in elderly patients: the effect of a continuous femoral nerve block: a
prospective and randomised study. Annales Françaises d'Anesthésie et de Réanimation
2007, 26(1):2-9. (Guideline Ref ID: CUVILLON2007)
63. Czoski-Murray C, De-Nigris E, Brazier JE, Walters S. (2007) A prospective controlled

study of the costs and health related quality of life following hip fracture. School of
Health and Related Research, University of Sheffield. (Guideline Ref ID:
CZOSKIMURRAY2007)
64. Darling JR, Murray JM, Hainsworth AM, Trinick TR. The effect of isoflurane or spinal
anesthesia on indocyanine green disappearance rate in the elderly. Anesthesia &
Analgesia 1994, 78(4):706-9. (Guideline Ref ID: DARLING1994)
65. Davis FM, Laurenson VG. Spinal anaesthesia or general anaesthesia for emergency hip
surgery in elderly patients. Anaesthesia & Intensive Care 1981, 9(4):352-8. (Guideline
Ref ID: DAVIS1981)
66. Davis FM, Woolner DF, Frampton C, Wilkinson A, Grant A, Harrison RT et al. Prospective,
multi-centre trial of mortality following general or spinal anaesthesia for hip fracture
surgery in the elderly. British Journal of Anaesthesia 1987, 59(9):1080-8. (Guideline Ref
ID: DAVIS1987)
67. Davis TR, Sher JL, Checketts RG, Porter BB. Intertrochanteric fractures of the femur: a
prospective study comparing the use of the Kuntscher-Y nail and a sliding hip screw.
Injury 1988, 19(6):421-6. (Guideline Ref ID: DAVIS1988)
68. Davis TR, Sher JL, Porter BB, Checketts RG. The timing of surgery for intertrochanteric
femoral fractures. Injury 1988, 19(4):244-6. (Guideline Ref ID: DAVIS1988A)
69. Davison JN, Calder SJ, Anderson GH, Ward G, Jagger C, Harper WM et al. Treatment for
displaced intracapsular fracture of the proximal femur. A prospective, randomised trial
in patients aged 65 to 79 years. Journal of Bone & Joint Surgery - British Volume 2001,
83(2):206-12. (Guideline Ref ID: DAVISON2001)
70. Department of Health. (2007) Reference Costs 2006-07 Collection Guidance. London:
(Guideline Ref ID: DH2007)
634 BIBLIOGRAPHY
71. Department of Health. (2009) The national framework for NHS continuing healthcare
and NHS-funded nursing care. (Guideline Ref ID: DH2009)
72. Department of Health. (2010) Prioritising need in the context of Putting People First: a
whole system approach to eligibility for social care - guidance on eligibility criteria for
adult social care, England 2010. (Guideline Ref ID: DH2010A)
73. Dorr LD, Glousman R, Hoy AL, Vanis R, Chandler R. Treatment of femoral neck fractures
with total hip replacement versus cemented and noncemented hemiarthroplasty.
Journal of Arthroplasty 1986, 1(1):21-8. (Guideline Ref ID: DORR1986)
74. Dujardin FH, Benez C, Polle G, Alain J, Biga N, Thomine JM. Prospective randomized
comparison between a dynamic hip screw and a mini-invasive static nail in fractures of
the trochanteric area: preliminary results. Journal of Orthopaedic Trauma 2001,
15(6):401-6. (Guideline Ref ID: DUJARDIN2001)
75. Dyson A, Henderson AM, Chamley D, Campbell ID. An assessment of postoperative

oxygen therapy in patients with fractured neck of femur. Anaesthesia & Intensive Care
1988, 16(4):405-10. (Guideline Ref ID: DYSON1988)
76. Ekman M, Berg J, Wimo A, Jonsson L, McBurney C. Health utilities in mild cognitive
impairment and dementia: a population study in Sweden. International Journal of
Geriatric Psychiatry 2007, 22(7):649-55. (Guideline Ref ID: EKMAN2007)
77. Ekstrom W, Karlsson-Thur C, Larsson S, Ragnarsson B, Alberts KA. Functional outcome in

treatment of unstable trochanteric and subtrochanteric fractures with the proximal
femoral nail and the Medoff sliding plate. Journal of Orthopaedic Trauma 2007,
21(1):18-25. (Guideline Ref ID: EKSTROM2007)
78. El-Abed K, McGuinness A, Brunner J, Dallovedova P, O'Connor P, Kennedy JG.

Comparison of outcomes following uncemented hemiarthroplasty and dynamic hip
screw in the treatment of displaced subcapital hip fractures in patients aged greater
than 70 years. Acta Orthopaedica Belgica 2005, 71(1):48-54. (Guideline Ref ID:
ELABED2005)
79. El-Zahaar MS, Al-Kawally HM, Said AS. A double-blind randomized study of the effects of
torniquet use and type of anesthetic techniques on the incidence of deep vein
thrombosis (DVT) in orthopedic surgery. Journal of Neurological & Orthopaedic
Medicine & Surgery 1995, 16(2):70-4. (Guideline Ref ID: ELZAHAAR1995)
80. Elinge E, Lofgren B, Gagerman E, Nyberg L. A group learning programme for old people
with hip fracture: A randomized study. Scandinavian Journal of Occupational Therapy
2003, 10(1):27-33. (Guideline Ref ID: ELINGE2003)
81. Emery RJ, Broughton NS, Desai K, Bulstrode CJ, Thomas TL. Bipolar hemiarthroplasty for
subcapital fracture of the femoral neck. A prospective randomised trial of cemented
Thompson and uncemented Moore stems. Journal of Bone & Joint Surgery - British
Volume 1991, 73(2):322-4. (Guideline Ref ID: EMERY1991)
82. Englesbe MJ, Fan Z, Baser O, Birkmeyer JD. Mortality in medicare patients undergoing
surgery in July in teaching hospitals. Annals of Surgery 2009, 249(6):871-6. (Guideline
Ref ID: ENGLESBE2009)
BIBLIOGRAPHY 635
83. Enocson A, Hedbeck CJ, Tidermark J, Pettersson H, Ponzer S, Lapidus LJ. Dislocation of
total hip replacement in patients with fractures of the femoral neck. Acta Orthopaedica
2009, 80(2):184-9. (Guideline Ref ID: ENOCSON2009)
84. Enocson A, Lapidus G, Tornkvist H, Tidermark J, Lapidus LJ. Direction of hip arthroplasty
dislocation in patients with femoral neck fractures. International Orthopaedics 2010,
34(5):641-7. (Guideline Ref ID: ENOCSON2010)
85. Enocson A, Tidermark J, Tornkvist H, Lapidus LJ. Dislocation of hemiarthroplasty after

femoral neck fracture: better outcome after the anterolateral approach in a prospective
cohort study on 739 consecutive hips. Acta Orthopaedica 2008, 79(2):211-7. (Guideline
Ref ID: ENOCSON2008)
86. Essex HN, Clark M, Sims J, Warriner A, Cullum N. Health-related quality of life in hospital
inpatients with pressure ulceration: assessment using generic health-related quality of
life measures. Wound Repair & Regeneration 2009, 17(6):797-805. (Guideline Ref ID:
ESSEX2009)
87. Evans JG, Prudham D, Wandless I. A prospective study of fractured proximal femur:
factors predisposing to survival. Age and Ageing 1979, 8(4):246-50. (Guideline Ref ID:
EVANS1979)
88. Evans PD, Wilson C, Lyons K. Comparison of MRI with bone scanning for suspected hip
fracture in elderly patients. Journal of Bone & Joint Surgery - British Volume 1994,
76(1):158-9. (Guideline Ref ID: EVANS1994)
89. Faraj AA, Branfoot T. Cemented versus uncemented Thompson's prostheses: A

functional outcome study. Injury 1999, 30(10):671-5. (Guideline Ref ID: FARAJ1999)
90. Faraj AA, Drakau NN. Press-fit hemiarthroplasty for elderly with femoral neck fracture:
High complication rates in operations performed by younger surgeons. European
Journal of Orthopaedic Surgery and Traumatology 2007, 17(3):267-71. (Guideline Ref ID:
FARAJ2007)
91. Farnworth MG, Kenny P, Shiell A. The costs and effects of early discharge in the
management of fractured hip. Age and Ageing 1994, 23(3):190-4. (Guideline Ref ID:
FARNWORTH1994)
92. Favarel-Garrigues JF, Sztark F, Petitjean ME, Thicoipe M, Lassie P, Dabadie P.

Hemodynamic effects of spinal anesthesia in the elderly: single dose versus titration
through a catheter. Anesthesia & Analgesia 1996, 82(2):312-6. (Guideline Ref ID:
FAVARELGARRIGUES1996)
93. Field RE, Rushton N. Five-year clinical, radiological and post-mortem results of the
Cambridge Cup in patients with displaced fractures of the neck of the femur. Journal of
Bone and Joint Surgery - Series B 2005, 87(10):1344-51. (Guideline Ref ID: FIELD2005)
94. Figved W, Opland V, Frihagen F, Jervidalo T, Madsen JE, Nordsletten L. Cemented versus
uncemented hemiarthroplasty for displaced femoral neck fractures. Clin Orthop Relat
Res 2009, 467(9):2426-35. (Guideline Ref ID: FIGVED2009)
636 BIBLIOGRAPHY
95. Fletcher AK, Rigby AS, Heyes FL. Three-in-one femoral nerve block as analgesia for
fractured neck of femur in the emergency department: a randomized, controlled trial.
Annals of Emergency Medicine 2003, 41(2):227-33. (Guideline Ref ID: FLETCHER2003)
96. Fordham R, Thompson R, Holmes J, Hodkinson C. (1986) A cost-benefit study of

geriatric-orthopaedic management of patients with fractured neck of femur. Discussion
paper 14. York: University of York: Centre for Health Economics. (Guideline Ref ID:
FORDHAM1986)
97. Foss NB, Kehlet H. Short-term mortality in hip fracture patients admitted during
weekends and holidays. British Journal of Anaesthesia 2006, 96(4):450-4. (Guideline Ref
ID: FOSS2006)
98. Foss NB, Kristensen BB, Bundgaard M, Bak M, Heiring C, Virkelyst C et al. Fascia iliaca
compartment blockade for acute pain control in hip fracture patients: a randomized,
placebo-controlled trial. Anesthesiology 2007, 106(4):773-8. (Guideline Ref ID:
FOSS2007)
99. Foss NB, Kristensen MT, Kristensen BB, Jensen PS, Kehlet H. Effect of postoperative
epidural analgesia on rehabilitation and pain after hip fracture surgery: a randomized,
double-blind, placebo-controlled trial. Anesthesiology 2005, 102(6):1197-204.
(Guideline Ref ID: FOSS2005A)
100. Franchignoni F, Ottonello M, Benevolo E, Tesio L. Satisfaction with hospital

rehabilitation: Is it related to life satisfaction, functional status, age or education?
Journal of Rehabilitation Medicine 2002, 34(3):105-8. (Guideline Ref ID:
FRANCHIGNONI2002)
101. Franzo A, Francescutti C, Simon G. Risk factors correlated with post-operative mortality
for hip fracture surgery in the elderly: a population-based approach. European Journal
of Epidemiology 2005, 20(12):985-91. (Guideline Ref ID: FRANZO2005)
102. Frihagen F, Nordsletten L, Madsen JE. Hemiarthroplasty or internal fixation for

intracapsular displaced femoral neck fractures: randomised controlled trial. British
Medical Journal 2007, 335(7632):1251-4. (Guideline Ref ID: FRIHAGEN2007)
103. Frihagen F, Nordsletten L, Tariq R, Madsen JE. MRI diagnosis of occult hip fractures.
Acta Orthopaedica 2005, 76(4):524-30. (Guideline Ref ID: FRIHAGEN2005)
104. Fung W, Jonsson A, Buhren V, Bhandari M. Classifying intertrochanteric fractures of the

proximal femur: does experience matter? Medical Principles & Practice 2007, 16(3):198-
202. (Guideline Ref ID: FUNG2007)
105. Furstenberg AL. Expectations about outcome following hip fracture among older
people. Social Work in Health Care 1986, 11(4):33-47. (Guideline Ref ID:
FURSTENBERG1986)
106. Galea MP, Levinger P, Lythgo N, Cimoli C, Weller R, Tully E et al. A targeted home- and
center-based exercise program for people after total hip replacement: a randomized
clinical trial. Archives of Physical Medicine and Rehabilitation 2008, 89(8):1442-7.
(Guideline Ref ID: GALEA2008)
BIBLIOGRAPHY 637
107. Galvard H, Samuelsson SM. Orthopedic or geriatric rehabilitation of hip fracture

patients: a prospective, randomized, clinically controlled study in Malmo, Sweden.
Aging-Clinical & Experimental Research 1995, 7(1):11-6. (Guideline Ref ID:
GALVARD1995)
108. Gdalevich M, Cohen D, Yosef D, Tauber C. Morbidity and mortality after hip fracture:
the impact of operative delay. Archives of Orthopaedic and Trauma Surgery 2004,
124(5):334-40. (Guideline Ref ID: GDALEVICH2004A)
109. Georgescu N, Iancu C, Trandabat C, Sirbu P, Alexa O. (2004) A prospective

randomised study comparing two types of hemiarthroplasty for displaced fractures of the
femoral neck in elderly patients Journal of Bone & Joint Surgery - British Volume
86-B (Suppl III): 338-Abstract (Guideline Reference ID: GEORGESCU2004)
110. Giancola R, Antonini G, Delle Rose G, Crippa C. Percutaneous compression plating

versus gamma nail for the treatment of pertrochanteric hip fractures. Strategies in
Trauma & Limb Reconstruction 2008, 3(1):9-14. (Guideline Ref ID: GIANCOLA2008)
111. Giannoudis PV, McGuigan J, Shaw DL. Ionising radiation during internal fixation of
extracapsular neck of femur fractures. Injury 1998, 29(6):469-72. (Guideline Ref ID:
GIANNOUDIS1998)
112. Gierer P, Landes J, Grubwinkler M, Gradl G, Lob G, Andress HJ. [The femoral neck
fracture in the elderly patient - cemented or cementless hip arthroplasty?]. Zentralbl
Chir 2002, 127(6):514-8. (Guideline Ref ID: GIERER2002)
113. Gilchrist WJ, Newman RJ, Hamblen DL, Williams BO. Prospective randomised study of an
orthopaedic geriatric inpatient service. British Medical Journal 1988, 297(6656):1116-8.
(Guideline Ref ID: GILCHRIST1988)
114. Gill JB, Jensen L, Chin PC, Rafiei P, Reddy K, Schutt RC, Jr. Intertrochanteric hip fractures
treated with the trochanteric fixation nail and sliding hip screw. Journal of Surgical
Orthopaedic Advances 2007, 16(2):62-6. (Guideline Ref ID: GILL2007)
115. Gill KP, Ursic P. The impact of continuing education on patient outcomes in the elderly
hip fracture population. Journal of Continuing Education in Nursing 1994, 25(4):181-5.
(Guideline Ref ID: GILL1994)
116. Gille J, Gille M, Gahr R, Wiedemann B. Acute pain management in proximal femoral
fractures: femoral nerve block (catheter technique) vs. systemic pain therapy using a
clinic internal organisation model. Anaesthesist 2006, 55(4):414-22. (Guideline Ref ID:
GILLE2006)
117. Giusti A, Barone A, Oliveri M, Pizzonia M, Razzano M, Palummeri E et al. An analysis of

the feasibility of home rehabilitation among elderly people with proximal femoral
fractures. Archives of Physical Medicine and Rehabilitation 2006, 87(6):826-31.
(Guideline Ref ID: GIUSTI2006A)
118. Gjertsen JE, Vinje T, Engesaeter LB, Lie SA, Havelin LI, Furnes O et al. Internal screw
fixation compared with bipolar hemiarthroplasty for treatment of displaced femoral
neck fractures in elderly patients. Journal of Bone & Joint Surgery - American Volume
2010, 92(3):619-28. (Guideline Ref ID: GJERTSEN2010)
638 BIBLIOGRAPHY
119. Gjertsen JE, Vinje T, Lie SA, Engesaeter LB, Havelin LI, Furnes O et al. Patient
satisfaction, pain, and quality of life 4 months after displaced femoral neck fractures: a
comparison of 663 fractures treated with internal fixation and 906 with bipolar
hemiarthroplasty reported to the Norwegian Hip Fracture Register. Acta Orthopaedica
2008, 79(5):594-601. (Guideline Ref ID: GJERTSEN2008)
120. Goh SK, Samuel M, Su DH, Chan ES, Yeo SJ. Meta-analysis comparing total hip
arthroplasty with hemiarthroplasty in the treatment of displaced neck of femur
fracture. Journal of Arthroplasty 2009, 24(3):400-6. (Guideline Ref ID: GOH2009)
121. Gonzalez-Montalvo JI, Alarcon T, Mauleon JL, Gil-Garay E, Gotor P, Martin-Vega A. The
orthogeriatric unit for acute patients: a new model of care that improves efficiency in
the management of patients with hip fracture. Hip International 2010, 20(2):229-35.
(Guideline Ref ID: GONZALEZMONTALVO2010)
122. Gorodetskyi IG, Gorodnichenko AI, Tursin PS, Reshetnyak VK, Uskov ON. Non-invasive
interactive neurostimulation in the post-operative recovery of patients with a
trochanteric fracture of the femur. A randomised, controlled trial. Journal of Bone &
Joint Surgery - British Volume 2007, 89(11):1488-94. (Guideline Ref ID:
GORODETSKYI2007)
123. Graf C, Scherer MA, Gumppenberg S. Cemented vs.cement-free endoprothetics in the

traumatology of older people. Hefte zur der Unfallchirurg 2000, 282:343. (Guideline Ref
ID: GRAF2000)
124. Graham J. Early or delayed weight-bearing after internal fixation of transcervical

fracture of the femur. A clinical trial. Journal of Bone & Joint Surgery - British Volume
1968, 50(3):562-9. (Guideline Ref ID: GRAHAM1968)
125. Grimes JP, Gregory PM, Noveck H, Butler MS, Carson JL. The effects of time-to-surgery
on mortality and morbidity in patients following hip fracture. American Journal of
Medicine 2002, 112(9):702-9. (Guideline Ref ID: GRIMES2002A)
126. Grimley EJ, Wandless I, Prudham D. A prospective study of fractured proximal femur;
hospital differences. Public Health 1980, 94(3):149-54. (Guideline Ref ID: GRIMLEY1980)
127. Guyer P, Landolt M, Keller H, Eberle C. The Gamma Nail in per- and intertrochanteric
femoral fractures--alternative or supplement to the dynamic hip screw? A prospective
randomized study of 100 patients with per- and intertrochanteric femoral fractures in
the surgical clinic of the City Hospital of Triemli, Zurich, September 1. Aktuelle
Traumatologie 1991, 21(6):242-9. (Guideline Ref ID: GUYER1991)
128. Guyer P, Landolt M, Keller H, Eberle C. The Gamma nail in per- and intertrochanteric
femoral fractures - alternative or complimentary to the DHS? A prospective randomised
study. In: Marti RK, Dunki Jacobs PB, eds. Proximal femoral fractures. Operative
technique and complications. Volume 2, 1993. pp 481-98. London: Medical Press
Limited. (Guideline Reference ID: Ref ID: GUYER1993A)
129. Haddad FS, Williams RL. Femoral nerve block in extracapsular femoral neck fractures.
Journal of Bone & Joint Surgery - British Volume 1995, 77(6):922-3. (Guideline Ref ID:
HADDAD1995)
BIBLIOGRAPHY 639
130. Haentjens P, Autier P, Barette M, Boonen S, Belgian Hip Fracture Study Group.
Predictors of functional outcome following intracapsular hip fracture in elderly women.
A one-year prospective cohort study. Injury 2005, 36(7):842-50. (Guideline Ref ID:
HAENTJENS2005)
131. Hallstrom I, Elander G, Rooke L. Pain and nutrition as experienced by patients with hip
fracture. Journal of Clinical Nursing 2000, 9(4):639-46. (Guideline Ref ID:
HALLSTROM2000)
132. Handoll HHG, Cameron ID, Mak JCS, Finnegan TP. Multidisciplinary rehabilitation for
older people with hip fractures. Cochrane Database of Systematic Reviews 2009, Issue
4:CD007125. (Guideline Ref ID: HANDOLL2009)
133. Handoll HHG, Sherrington C. Mobilisation strategies after hip fracture surgery in adults.
Cochrane Database of Systematic Reviews 2007, Issue 1:CD001704. (Guideline Ref ID:
HANDOLL2007)
134. Hardy DC, Descamps PY, Krallis P, Fabeck L, Smets P, Bertens CL et al. Use of an
intramedullary hip-screw compared with a compression hip-screw with a plate for
intertrochanteric femoral fractures. A prospective, randomized study of one hundred
patients. Journal of Bone & Joint Surgery - American Volume 1998, 80(5):618-30.
(Guideline Ref ID: HARDY1998)
135. Harper MC, Walsh T. Ender nailing for peritrochanteric fractures of the femur. An
analysis of indications, factors related to mechanical failure, and postoperative results.
Journal of Bone & Joint Surgery - American Volume 1985, 67(1):79-88. (Guideline Ref ID:
HARPER1985)
136. Harper WM. Treatment of intracapsular proximal femoral fractures. A prospective

randomised trial comparing cemented and uncemented Thompson hemiarthroplasty in
the treatment of displaced intracapsular proximal femoral fractures. 1994. University
of Leicester. (Guideline Reference ID: HARPER1994)¶
137. Harrington P, Nihal A, Singhania AK, Howell FR. Intramedullary hip screw versus sliding
hip screw for unstable intertrochanteric femoral fractures in the elderly. Injury 2002,
33(1):23-8. (Guideline Ref ID: HARRINGTON2002)
138. Harrison SA. Focus on research... Patients' perspectives of written information supplied
in an acute trauma setting following fractured neck of femur. British Journal of
Occupational Therapy 2006, 69(8):385. (Guideline Ref ID: HARRISON2006)
139. Hauer K, Rost B, Rutschle K, Opitz H, Specht N, Bartsch P et al. Exercise training for
rehabilitation and secondary prevention of falls in geriatric patients with a history of
injurious falls. Journal of the American Geriatrics Society 2001, 49(1):10-20. (Guideline
Ref ID: HAUER2001)
140. Hauer K, Specht N, Schuler M, Bartsch P, Oster P. Intensive physical training in geriatric
patients after severe falls and hip surgery. Age and Ageing 2002, 31(1):49-57. (Guideline
Ref ID: HAUER2002)
141. Hedman AR, Heikkila K, Grafstrom M, Stromberg L. Hip fractures and cognitive state:
patient outcomes and proxies' perceptions of the rehabilitation period. International
Journal of Older People Nursing 2008, 3(3):178-86. (Guideline Ref ID: HEDMAN2008)
640 BIBLIOGRAPHY
142. Heetveld MJ, Rogmark C, Frihagen F, Keating J. Internal fixation versus arthroplasty for
displaced femoral neck fractures: what is the evidence? Journal of Orthopaedic Trauma
2009, 23(6):395-402. (Guideline Ref ID: HEETVELD2009)
143. Hemmingsen C, Nielsen JE. Intravenous ketamine for prevention of severe hypotension
during spinal anaesthesia. Acta Anaesthesiologica Scandinavica 1991, 35(8):755-7.
(Guideline Ref ID: HEMMINGSEN1991)
144. HM Government. (2009) Shaping the future of care together. (Guideline Ref ID:
HMG2009)
145. Ho WWS, Kwan Dai DL, Liu KW, Chow KM, Lau E, Woo J et al. To investigate the effect
and cost-effectiveness of implementing an orthogeriatric intervention for elderly
patients with acute hip fracture: the experience in Hong Kong. Journal of the American
Geriatrics Society 2009, 57(11):2153-4. (Guideline Ref ID: HO2009)
146. Hoenig H, Rubenstein LV, Sloane R, Horner R, Kahn K. What is the role of timing in the
surgical and rehabilitative care of community-dwelling older persons with acute hip
fracture? Archives of Internal Medicine 1997, 157(5):513-20. (Guideline Ref ID:
HOENIG1997A)
147. Hoffman CW, Lynskey TG. Intertrochanteric fractures of the femur: a randomized
prospective comparison of the Gamma nail and the Ambi hip screw. Australian and New
Zealand Journal of Surgery 1996, 66(3):151-5. (Guideline Ref ID: HOFFMAN1996)
148. Hollingworth W, Todd C, Parker M, Roberts JA, Williams R. Cost analysis of early
discharge after hip fracture. British Medical Journal 1993, 307(6909):903-6. (Guideline
Ref ID: HOLLINGWORTH1993)
149. Holmberg S, Kalen R, Thorngren KG. Treatment and outcome of femoral neck fractures.
An analysis of 2418 patients admitted from their own homes. Clinical Orthopaedics and
Related Research 1987,(218):42-52. (Guideline Ref ID: HOLMBERG1987)
150. Holmes J, House A. Psychiatric illness predicts poor outcome after surgery for hip
fracture: a prospective cohort study. Psychological Medicine 2000, 30(4):921-9.
(Guideline Ref ID: HOLMES2000)
151. Holt EM, Evans RA, Hindley CJ, Metcalfe JW. 1000 femoral neck fractures: the effect of
pre-injury mobility and surgical experience on outcome. Injury 1994, 25(2):91-5.
(Guideline Ref ID: HOLT1994)
152. Holt G, Smith R, Duncan K, McKeown DW. Does delay to theatre for medical reasons
affect the peri-operative mortality in patients with a fracture of the hip? Journal of Bone
& Joint Surgery - British Volume 2010, 92(6):835-41. (Guideline Ref ID: HOLT2010)
153. Hood G, Edbrooke DL, Gerrish SP. Postoperative analgesia after triple nerve block for
fractured neck of femur. Anaesthesia 1991, 46(2):138-40. (Guideline Ref ID: HOOD1991)
154. Huang TT, Acton GJ. Ways to maintain independence among Taiwanese elderly adults
with hip fractures: a qualitative study. Geriatric Nursing 2009, 30(1):28-35. (Guideline
Ref ID: HUANG2009)
BIBLIOGRAPHY 641
155. Hunter GA. A comparison of the use of internal fixation and prosthetic replacement for
fresh fractures of the neck of the femur. Br J Surg 1969, 56(3):229-32. (Guideline Ref ID:
HUNTER1969)
156. Hunter GA. A further comparison of the use of internal fixation and prosthetic
replacement for fresh fractures of the neck of the femur. Br J Surg 1974, 61(5):382-4.
(Guideline Ref ID: HUNTER1974)
157. Huusko TM, Karppi P, Avikainen V, Kautiainen H, Sulkava R. Randomised, clinically

controlled trial of intensive geriatric rehabilitation in patients with hip fracture:
subgroup analysis of patients with dementia. British Medical Journal 2000,
321(7269):1107-11. (Guideline Ref ID: HUUSKO2000)
158. Huusko TM, Karppi P, Avikainen V, Kautiainen H, Sulkava R. Intensive geriatric

rehabilitation of hip fracture patients: a randomized, controlled trial. Acta Orthopaedica
Scandinavica 2002, 73(4):425-31. (Guideline Ref ID: HUUSKO2002A)
159. Iliffe S, Kendrick D, Morris R, Skelton D, Gage H, Dinan S et al. Multi-centre cluster
randomised trial comparing a community group exercise programme with home based
exercise with usual care for people aged 65 and over in primary care: protocol of the
ProAct 65+ trial. Trials [Electronic Resource] 2010, 11(1):6. (Guideline Ref ID: ILIFFE2010)
160. Jackson SE. The efficacy of an educational intervention on documentation of pain

management for the elderly patient with a hip fracture in the emergency department.
JEN: Journal of Emergency Nursing 2010, 36(1):10-5. (Guideline Ref ID: JACKSON2010)
161. Jensen J, Rasmussen TB, Christensen S, Holm-Moller S, Lauritzen J. Internal fixation or

prosthetic replacement in fresh femoral neck fractures. Acta Orthopaedica Scandinavica
1984, 55(6):712. (Guideline Ref ID: JENSEN1984)
162. Johansson T. Displaced femoral neck fractures: a prospective randomized study of

clinical outcome, nutrition and costs. 2002. Linkoping University. (Guideline Reference
ID: JOHANSSON2002)¶
163. Johansson T, Bachrach-Lindstrom M, Aspenberg P, Jonsson D, Wahlstrom O. The total

costs of a displaced femoral neck fracture: comparison of internal fixation and total hip
replacement. A randomised study of 146 hips. International Orthopaedics 2006, 30(1):1-
6. (Guideline Ref ID: JOHANSSON2006)
164. Johnson MM, Thompson B, Hartford JM. Complications of cemented versus

noncemented hemiarthroplasty for femoral neck fractures [abstract]. Mid-America
Orthopaedic Association 19th annual meeting; 2001 April 18-22; Amelia Island (FL).
http://www.jbjs.org/abstracts/maoa01/absÿ001.htm [accessed 24-11-2003]. (Guideline
Ref ID: JOHNSON2001)
165. Jones SF, White A. Analgesia following femoral neck surgery. Lateral cutaneous nerve
block as an alternative to narcotics in the elderly. Anaesthesia 1985, 40(7):682-5.
(Guideline Ref ID: JONES1985)
166. Jonsson B, Sernbo I, Carlsson A, Fredin H, Johnell O. Social function after cervical hip
fracture. A comparison of hook-pins and total hip replacement in 47 patients. Acta
Orthopaedica Scandinavica 1996, 67(5):431-4. (Guideline Ref ID: JONSSON1996)
642 BIBLIOGRAPHY
167. Juelsgaard P, Sand NP, Felsby S, Dalsgaard J, Jakobsen KB, Brink O et al. Perioperative
myocardial ischaemia in patients undergoing surgery for fractured hip randomized to
incremental spinal, single-dose spinal or general anaesthesia. European Journal of
Anaesthesiology 1998, 15(6):656-63. (Guideline Ref ID: JUELSGAARD1998)
168. Kamel HK, Iqbal MA, Mogallapu R, Maas D, Hoffmann RG. Time to ambulation after hip
fracture surgery: relation to hospitalization outcomes. Journals of Gerontology Series A-
Biological Sciences & Medical Sciences 2003, 58(11):1042-5. (Guideline Ref ID:
KAMEL2003)
169. Kamitani K, Higuchi A, Asahi T, Yoshida H. Postoperative delirium after general

anesthesia vs. spinal anesthesia in geriatric patients. Masui 2003, 52(9):972-5.
(Guideline Ref ID: KAMITANI2003)
170. Karpman RR, Lee TK, Moore BM. Austin-Moore versus bipolar hemi-arthroplasty for
displaced femoral neck fractures: a randomized prospective study. Orthopaedic
Transactions 1992, 16:749-50. (Guideline Ref ID: KARPMAN1992)
171. Karumo I. Recovery and rehabilitation of elderly subjects with femoral neck fractures.
Annales Chirurgiae et Gynaecologiae 1977, 66(3):170-6. (Guideline Ref ID:
KARUMO1977A)
172. Kavcic G, Hudoklin P, Mikek M, Hussein M. Hemiarthroplasty versus total arthroplasty

for treatment of femoral neck fractures. European Journal of Trauma 2006, 32(Suppl
1):24. (Guideline Ref ID: KAVCIC2006)
173. Keating JF, Grant A, Masson M, Scott NW, Forbes JF. Displaced intracapsular hip
fractures in fit, older people: a randomised comparison of reduction and fixation,
bipolar hemiarthroplasty and total hip arthroplasty. Health Technology Assessment
2005, 9(41):iii-x, 1. (Guideline Ref ID: KEATING2005)
174. Keating JF, Grant A, Masson M, Scott NW, Forbes JF. Randomized comparison of
reduction and fixation, bipolar hemiarthroplasty, and total hip arthroplasty. Treatment
of displaced intracapsular hip fractures in healthy older patients. Journal of Bone & Joint
Surgery - American Volume 2006, 88(2):249-60. (Guideline Ref ID: KEATING2006)
175. Keene GS, Parker MJ, Pryor GA. Mortality and morbidity after hip fractures. British
Medical Journal 1993, 307(6914):1248-50. (Guideline Ref ID: KEENE1993)
176. Kennie DC, Reid J, Richardson IR, Kiamari AA, Kelt C. Effectiveness of geriatric
rehabilitative care after fractures of the proximal femur in elderly women: a
randomised clinical trial. British Medical Journal 1988, 297(6656):1083-6. (Guideline Ref
ID: KENNIE1988)
177. Kenzora JE, McCarthy RE, Lowell JD, Sledge CB. Hip fracture mortality. Relation to age,
treatment, preoperative illness, time of surgery, and complications. Clinical
Orthopaedics and Related Research 1984,(186):45-56. (Guideline Ref ID:
KENZORA1984A)
178. Khan RJK, MacDowell A, Crossman P, Keene GS. Cemented or uncemented

hemiarthroplasty for displaced intracapsular fractures of the hip - A systematic review.
Injury 2002, 33(1):13-7. (Guideline Ref ID: KHAN2002B)
BIBLIOGRAPHY 643
179. Kim SY, Kim YG, Hwang JK. Cementless calcar-replacement hemiarthroplasty compared
with intramedullary fixation of unstable intertrochanteric fractures. A prospective,
randomized study. Journal of Bone & Joint Surgery - American Volume 2005,
87(10):2186-92. (Guideline Ref ID: KIM2005)
180. Kuisma R. A randomized, controlled comparison of home versus institutional

rehabilitation of patients with hip fracture. Clinical Rehabilitation 2002, 16(5):553-61.
(Guideline Ref ID: KUISMA2002)
181. Kukla C, Heinz T, Gaebler C, Heinze G, Vecsei V. The standard Gamma nail: a critical
analysis of 1,000 cases. Journal of Trauma-Injury Infection & Critical Care 2001,
51(1):77-83. (Guideline Ref ID: KUKLA2001)
182. Kullenberg B, Ysberg B, Heilman M, Resch S. Femoral nerve block as pain relief in hip
fracture. A good alternative in perioperative treatment proved by a prospective study.
Lakartidningen 2004, 101(24):2104-7. (Guideline Ref ID: KULLENBERG2004)
183. Kuwabara H, Wada T, Minagi Y, Iwasaki T, Tsuji H. Compression hip screw and gamma
nail for intertrochanteric fractures - randomized prospective study. Hokkaido Journal of
Orthopaedics & Traumatology 1998, 40(2):29-33. (Guideline Ref ID: KUWABARA1998)
184. Lachiewicz PF, Kelley SS, Soileau ES. Survival of polished compared with precoated
roughened cemented femoral components. A prospective, randomized study. Journal of
Bone & Joint Surgery - American Volume 2008, 90(7):1457-63. (Guideline Ref ID:
LACHIEWICZ2008)
185. Laffosse JM, Accadbled F, Molinier F, Chiron P, Hocine B, Puget J. Anterolateral mini-
invasive versus posterior mini-invasive approach for primary total hip replacement.
Comparison of exposure and implant positioning. Archives of Orthopaedic and Trauma
Surgery 2008, 128(4):363-9. (Guideline Ref ID: LAFFOSSE2008)
186. Laffosse JM, Chiron P, Molinier F, Bensafi H, Puget J. Prospective and comparative study
of the anterolateral mini-invasive approach versus minimally invasive posterior
approach for primary total hip replacement. Early results. International Orthopaedics
2007, 31(5):597-603. (Guideline Ref ID: LAFFOSSE2007)
187. Laffosse JM, Chiron P, Tricoire JL, Giordano G, Molinier F, Puget J. Prospective and
comparative study of minimally invasive posterior approach versus standard posterior
approach in total hip replacement. Revue de Chirurgie Orthopedique et Reparatrice de l
Appareil Moteur 2007, 93(3):228-37. (Guideline Ref ID: LAFFOSSE2007A)
188. Lee PC, Hsieh PH, Yu SW, Shiao CW, Kao HK, Wu CC. Biologic plating versus
intramedullary nailing for comminuted subtrochanteric fractures in young adults: a
prospective, randomized study of 66 cases. Journal of Trauma-Injury Infection & Critical
Care 2007, 63(6):1283-91. (Guideline Ref ID: LEE2007)
189. Lefaivre KA, Macadam SA, Davidson DJ, Gandhi R, Chan H, Broekhuyse HM. Length of
stay, mortality, morbidity and delay to surgery in hip fractures. Journal of Bone & Joint
Surgery - British Volume 2009, 91(7):922-7. (Guideline Ref ID: LEFAIVRE2009)
190. Leidinger W, Hoffmann G, Meierhofer JN, Wolfel R. Reduction of severe cardiac

complications during surgical repair of femoral neck fractures with cemented hip
644 BIBLIOGRAPHY
arthroplasty. [German]. Unfallchirurg 2002, 105(8):675-9. (Guideline Ref ID:

LEIDINGER2002)
191. Leung KS, So WS, Shen WY, Hui PW. Gamma nails and dynamic hip screws for
peritrochanteric fractures. A randomised prospective study in elderly patients. Journal
of Bone & Joint Surgery - British Volume 1992, 74(3):345-51. (Guideline Ref ID:
LEUNG1992)
192. Levi N, Gebuhr P. Early failure and mortality following intramedullary fixation of
peritrochanteric fractures. International Journal of Risk and Safety in Medicine 2000,
13(2-3):99-101. (Guideline Ref ID: LEVI2000)
193. Lin PC, Hung SH, Liao MH, Sheen SY, Jong SY. Care needs and level of care difficulty
related to hip fractures in geriatric populations during the post-discharge transition
period. Journal of Nursing Research 2006, 14(4):251-60. (Guideline Ref ID: LIN2006)
194. Lin PC, Lu CM. Hip fracture: family caregivers' burden and related factors for older
people in Taiwan. Journal of Clinical Nursing 2005, 14(6):719-26. (Guideline Ref ID:
LIN2005)
195. Little NJ, Verma V, Fernando C, Elliott DS, Khaleel A. A prospective trial comparing the
Holland nail with the dynamic hip screw in the treatment of intertrochanteric fractures
of the hip. Journal of Bone & Joint Surgery - British Volume 2008, 90(8):1073-8.
(Guideline Ref ID: LITTLE2008)
196. Lubovsky O, Liebergall M, Mattan Y, Weil Y, Mosheiff R. Early diagnosis of occult hip
fractures: MRI versus CT scan. Injury 2005, 36(6):788-92. (Guideline Ref ID:
LUBOVSKY2005)
197. Macaulay W, Nellans KW, Garvin KL, Iorio R, Healy WL, Rosenwasser MP et al.
Prospective randomized clinical trial comparing hemiarthroplasty to total hip
arthroplasty in the treatment of displaced femoral neck fractures: winner of the Dorr
Award. Journal of Arthroplasty 2008, 23(6 Suppl 1):2-8. (Guideline Ref ID:
MACAULAY2008)
198. Mackenzie DG, Wild S, Muir R. Mortality associated with delay in operation after hip
fracture: Scottish data provide additional information... British Medical Journal 2006,
332(7549):1093. (Guideline Ref ID: MACKENZIE2006B)
199. Magasi S, Durkin E, Wolf MS, Deutsch A. Rehabilitation consumers' use and
understanding of quality information: a health literacy perspective. Archives of Physical
Medicine and Rehabilitation 2009, 90(2):206-12. (Guideline Ref ID: MAGASI2009)
200. Majumdar SR, Beaupre LA, Johnston DWC, Dick DA, Cinats JG, Jiang HX. Lack of
association between mortality and timing of surgical fixation in elderly patients with hip
fracture: results of a retrospective population-based cohort study. Medical Care 2006,
44(6):552-9. (Guideline Ref ID: MAJUMDAR2006)
201. Mangione KK, Craik RL, Tomlinson SS, Palombaro KM. Can elderly patients who have
had a hip fracture perform moderate- to high-intensity exercise at home? Physical
Therapy 2005, 85(8):727-39. (Guideline Ref ID: MANGIONE2005)
BIBLIOGRAPHY 645
202. Mannion S, Hayes I, Loughnane F, Murphy DB, Shorten GD. Intravenous but not
perineural clonidine prolongs postoperative analgesia after psoas compartment block
with 0.5% levobupivacaine for hip fracture surgery. Anesthesia and Analgesia 2005,
100(3):873-8. (Guideline Ref ID: MANNION2005)
203. Marcantonio ER, Flacker JM, Wright RJ, Resnick NM. Reducing delirium after hip
fracture: a randomized trial. Journal of the American Geriatrics Society 2001, 49(5):516-
22. (Guideline Ref ID: MARCANTONIO2001)
204. Marhofer P, Faryniak B, ller C, Koinig H, Kapral S, Mayer N. Cardiovascular effects of 6%

hetastarch and lactated Ringer's solution during spinal anesthesia. Regional Anesthesia
& Pain Medicine 1999, 24(5):399-404. (Guideline Ref ID: MARHOFER1999)
205. Marhofer P, Schrogendorfer K, Wallner T, Koinig H, Mayer N, Kapral S. Ultrasonographic

guidance reduces the amount of local anesthetic for 3-in-1 blocks. Regional Anesthesia
and Pain Medicine 1998, 23(6):584-8. (Guideline Ref ID: MARHOFER1998)
206. Marinelli M, Soccetti A, Panfoli N, de Palma L. Cost-effectiveness of cemented versus

cementless total hip arthroplasty. A Markov decision analysis based on implant cost.
Journal of Orthopaedics and Traumatology 2008, 9(1):23-8. (Guideline Ref ID:
MARINELLI2008)
207. Matot I, Oppenheim-Eden A, Ratrot R, Baranova J, Davidson E, Eylon S et al.

Preoperative cardiac events in elderly patients with hip fracture randomized to epidural
or conventional analgesia. Anesthesiology 2003, 98(1):156-63. (Guideline Ref ID:
MATOT2003)
208. Maurette P, Bonada G, Djiane V, Erny P. A comparison between lidocaine alone and
lidocaine with meperidine for continuous spinal anesthesia. Regional Anesthesia 1993,
18(5):290-5. (Guideline Ref ID: MAURETTE1993)
209. McGuire KJ, Bernstein J, Polsky D, Silber JH. The 2004 Marshall Urist Award: delays until
surgery after hip fracture increases mortality. Clinical Orthopaedics and Related
Research 2004,(428):294-301. (Guideline Ref ID: MCGUIRE2004)
210. McKenzie PJ, Wishart HY, Smith G. Long-term outcome after repair of fractured neck of
femur. Comparison of subarachnoid and general anaesthesia. British Journal of
Anaesthesia 1984, 56(6):581-5. (Guideline Ref ID: MCKENZIE1984)
211. McLaren AD, Stockwell MC, Reid VT. Anaesthetic techniques for surgical correction of
fractured neck of femur. A comparative study of spinal and general anaesthesia in the
elderly. Anaesthesia 1978, 33(1):10-4. (Guideline Ref ID: MCLAREN1978)
212. Mehdi SA, Kinninmonth AWG, MacLeod C, McKenzie E, James PJ. Extracapsular hip
fracture fixation: aa prospective randomised comparison of the intramedullary hip
screw with the sliding hip screw. Injury 2000, 31:287. (Guideline Ref ID: MEHDI2000)
213. Michos I, Brakoulakis E, Pastroudis A, Loutriotis A, Adamopoulos G. The Gamma nail

system compared to sliding nail and plate for peritrochanteric fractures. Journal of Bone
& Joint Surgery - British Volume 2001, 83(Suppl 2):193. (Guideline Ref ID: MICHOS2001)
214. Miedel R, Ponzer S, Tornkvist H, Soderqvist A, Tidermark J. The standard Gamma nail or
the Medoff sliding plate for unstable trochanteric and subtrochanteric fractures. A
646 BIBLIOGRAPHY
randomised, controlled trial. Journal of Bone & Joint Surgery - British Volume 2005,
87(1):68-75. (Guideline Ref ID: MIEDEL2005)
215. Moran CG, Wenn RT, Sikand M, Taylor AM. Early mortality after hip fracture: is delay
before surgery important? Journal of Bone & Joint Surgery - American Volume 2005,
87A(3):483-9. (Guideline Ref ID: MORAN2005)
216. Moseley AM, Sherrington C, Lord SR, Barraclough E, St George RJ, Cameron ID. Mobility
training after hip fracture: a randomised controlled trial. Age and Ageing 2009,
38(1):74-80. (Guideline Ref ID: MOSELEY2009)
217. Mott MP, Kronik JL, Fitzgerald RHJr, Morawa LG, Georgiadis GM, Salot WH. Gamma nail
versus the sliding hip screw: a prospective randomized comparison. Orthopaedic
Transactions 1993, 17:1049. (Guideline Ref ID: MOTT1993)
218. Mouzopoulos G, Stamatakos M, Arabatzi H, Vasiliadis G, Batanis G, Tsembeli A et al. The

four-year functional result after a displaced subcapital hip fracture treated with three
different surgical options. International Orthopaedics 2008, 32(3):367-73. (Guideline Ref
ID: MOUZOPOULOS2008)
219. Muller ME, Allgower N, Schneider R, Willenegger H. The comprehensive classification of

fractures of long bones. In: Allgower M, ed. Manual of internal fixation, 1991. Berlin:
Springer-Verlag. (Guideline Reference ID: Ref ID: MULLER1991)
220. Murgue D, Ehret B, Massacrier-Imbert S, Durand O, Gibaud F, Maakel A et al. Equimolar

nitrous oxide/oxygen combined with femoral nerve block for emergency analgesia of
femoral neck fractures. Journal Europeen Des Urgences (Jeur) 2006, 19(1):9-14.
(Guideline Ref ID: MURGUE2006)
221. Mutty CE, Jensen EJ, Manka MA, Jr., Anders MJ, Bone LB. Femoral nerve block for
diaphyseal and distal femoral fractures in the emergency department. Journal of Bone &
Joint Surgery - American Volume 2007, 89(12):2599-603. (Guideline Ref ID: MUTTY2007)
222. Naglie G, Tansey C, Kirkland JL, Ogilvie-Harris DJ, Detsky AS, Etchells E et al.
Interdisciplinary inpatient care for elderly people with hip fracture: A randomized
controlled trial. Canadian Medical Association Journal 2002, 167(1):25-32. (Guideline
Ref ID: NAGLIE2002)
223. Naja Z, el Hassan MJ, Khatib H, Ziade MF, Lonnqvist PA. Combined sciatic-paravertebral
nerve block vs. general anaesthesia for fractured hip of the elderly. Middle East Journal
of Anesthesiology 2000, 15(5):559-68. (Guideline Ref ID: NAJA2000)
224. National Clinical Guideline Centre. (2010) Delirium: diagnosis, prevention and
management. (Guideline Ref ID: NCGC2010)
225. National Clinical Guideline Centre. (2010) Reducing the risk of venous
thromboembolism (deep vein thrombosis and pulmonary embolism) in patients
admitted to hospital. London: (Guideline Ref ID: NCGC2010A)
226. National Institute for Clinical Excellence. Guidance on the selection of prostheses for
primary total hip replacement http://www.nice.org.uk/ta2 [accessed 6-10-2008].
(Guideline Ref ID: TA022000)
BIBLIOGRAPHY 647
227. National Institute for Health and Clinical Excellence. (2004) The assessment and
prevention of falls in older people. London: Royal College of Nursing. (Guideline Ref ID:
NICE2004)
228. National Institute for Health and Clinical Excellence. (2005) The management of
pressure ulcers in primary and secondary care. London: Royal College of Nursing.
(Guideline Ref ID: NICE2005)
229. National Institute for Health and Clinical Excellence. (2006) Nutrition support in adults:
oral nutrition support, enteral tube feeding and parenteral nutrition. London: National
Institute for Health and Clinical Excellence. (Guideline Ref ID: NICE2006A)
230. National Institute for Health and Clinical Excellence. (2006) Supporting people with
dementia and their carers in health and social care. (Guideline Ref ID: NICE2006)
231. National Institute for Health and Clinical Excellence. (2008) Prevention and treatment of
surgical site infection. London: National Institute for Health and Clinical Excellence.
(Guideline Ref ID: NICE2008B)
232. National Institute for Health and Clinical Excellence. Social value judgements: principles
for the development of NICE guidance
http://www.nice.org.uk/media/C18/30/SVJ2PUBLICATION2008.pdf [accessed 13-10-
2009]. (Guideline Ref ID: NICE2008A)
233. National Institute for Health and Clinical Excellence. The guidelines manual 2009
http://www.nice.org.uk [accessed 13-1-2009]. (Guideline Ref ID: NICE2009)
234. National Institute for Health and Clinical Excellence. (2010) Alendronate, etidronate,
risedronate, raloxifene and strontium ranelate for the primary prevention of
osteoporotic fragility fractures in postmenopausal women. London: National Institute
for Health and Clinical Excellence. (Guideline Ref ID: NICE2010)
235. National Institute for Health and Clinical Excellence. (2010) Alendronate, etidronate,
risedronate, raloxifene, strontium ranelate and teriparatide for the secondary
prevention of osteoporotic fragility fractures in postmenopausal women. London:
National Institute for Health and Clinical Excellence. (Guideline Ref ID: NICE2010A)
236. National Institute for Health and Clinical Excellence. (2010) Denosumab for the
prevention of osteoporotic fractures in postmenopausal women. London: National
Institute for Health and Clinical Excellence. (Guideline Ref ID: NICE2010B)
237. National Institute for Health and Clinical Excellence. (2010) Reducing the risk of venous
thromboembolism (deep vein thrombosis and pulmonary embolism) in patients
admitted to hospital. London: National Institute for Health and Clinical Excellence.
(Guideline Ref ID: NICE2010C)
238. Neander G. (2000) Reduction and fixation versus total hip arthroplasty in the treatment
of displaced femoral neck fractures. Results after four years of a prospective
randomised study in 100 patients. Displaced femoral neck fractures. Studies on
osteosynthesis and total hip arthroplasty. Stockholm: Division of Orthopaedics,
Karolinska Institutet, Danderyds Hospital. (Guideline Ref ID: NEANDER2000)
648 BIBLIOGRAPHY
239. Neander G, Adolphson P, von Sivers K, Dahlborn M, Dalen N. Bone and muscle mass
after femoral neck fracture. A controlled quantitative computed tomography study of
osteosynthesis versus primary total hip arthroplasty. Archives of Orthopaedic and
Trauma Surgery 1997, 116(8):470-4. (Guideline Ref ID: NEANDER1997)
240. Netten A, Bebbington A, Darton R, Forder J, Miles K. (1998) Survey of Care Homes for
Elderly People. Final Report. Canterbury, Kent: Personal Social Services Research Unit.
(Guideline Ref ID: NETTEN1998)
241. Netten A, Curtis L. (2002) Unit costs of health and social care 2001. Personal and Social
Services Research Unit (PSSRU). (Guideline Ref ID: NETTEN2002)
242. Nishikawa K, Yamakage M, Omote K, Namiki A. Prophylactic IM small-dose

phenylephrine blunts spinal anesthesia-induced hypotensive response during surgical
repair of hip fracture in the elderly. Anesthesia & Analgesia 2002, 95(3):751-6.
(Guideline Ref ID: NISHIKAWA2002)
243. Novack V, Jotkowitz A, Etzion O, Porath A. Does delay in surgery after hip fracture lead
to worse outcomes? A multicenter survey. International Journal for Quality in Health
Care 2007, 19(3):170-6. (Guideline Ref ID: NOVACK2007)
244. O'Brien PJ, Meek RN, Blachut PA, Broekhuyse HM, Sabharwal S. Fixation of
intertrochanteric hip fractures: gamma nail versus dynamic hip screw. A randomized,
prospective study. Canadian Journal of Surgery 1995, 38(6):516-20. (Guideline Ref ID:
OBRIEN1995)
245. O'Cathain A. Evaluation of a Hospital at Home scheme for the early discharge of
patients with fractured neck of femur. Journal of Public Health Medicine 1994,
16(2):205-10. (Guideline Ref ID: OCATHAIN1994)
246. Office for Economic Cooperation and Development (OECD). OECD PPPs and exchange
rates http://stats.oecd.org/Index.aspx?DatasetCode=SNA_TABLE4 [accessed 11-1-
2009]. (Guideline Ref ID: OECD2009)
247. Oldmeadow LB, Edwards ER, Kimmel LA, Kipen E, Robertson VJ, Bailey MJ. No rest for
the wounded: Early ambulation after hip surgery accelerates recovery. ANZ Journal of
Surgery 2006, 76(7):607-11. (Guideline Ref ID: OLDMEADOW2006)
248. Olsson LE, Karlsson J, Ekman I. Effects of nursing interventions within an integrated care
pathway for patients with hip fracture. Journal of Advanced Nursing 2007, 58(2):116-25.
(Guideline Ref ID: OLSSON2007A)
249. Olsson LE, Nystrom AE, Karlsson J, Ekman I. Admitted with a hip fracture: patient
perceptions of rehabilitation. Journal of Clinical Nursing 2007, 16(5):853-9. (Guideline
Ref ID: OLSSON2007)
250. Orosz GM, Magaziner J, Hannan EL, Morrison RS, Koval K, Gilbert M et al. Association of
timing of surgery for hip fracture and patient outcomes. JAMA 2004, 291(14):1738-43.
(Guideline Ref ID: OROSZ2004)
251. Ovesen O, Andersen M, Poulsen T, Nymark T, Overgaard S, Rock ND. The trochanteric
gamma nail versus the dynamic hip screw: a prospective randomised study. One-year
BIBLIOGRAPHY 649
follow-up of 146 intertrochanteric fractures. Hip International 2006, 16(4):293-8.

(Guideline Ref ID: OVESEN2006)
252. Owen H, Hutton P. Doxapram and the fractured femur. Anaesthesia 1982, 37(3):301-4.
(Guideline Ref ID: OWEN1982)
253. Pahlpatz PVM, Langius FB. Comparing the Gamma nail and the Dynamic Hip Screw in
the treatment of pertrochanteric fractures. Preliminary results of a prospective
randomised study. In: Marti RK, Dunki Jacobs PB, eds. Proximal femoral fractures.
Operative techniques and complications. Volume 2., 1993. pp 475-80. London: Medical
Press Limited. (Guideline Reference ID: Ref ID: PAHLPATZ1993)
254. Pajarinen J, Lindahl J, Michelsson O, Savolainen V, Hirvensalo E. Pertrochanteric femoral

fractures treated with a dynamic hip screw or a proximal femoral nail. A randomised
study comparing post-operative rehabilitation. Journal of Bone & Joint Surgery - British
Volume 2005, 87(1):76-81. (Guideline Ref ID: PAJARINEN2005)
255. Pajarinen J, Lindahl J, Savolainen V, Michelsson O, Hirvensalo E. Femoral shaft

medialisation and neck-shaft angle in unstable pertrochanteric femoral fractures.
International Orthopaedics 2004, 28(6):347-53. (Guideline Ref ID: PAJARINEN2004)
256. Palm H, Jacobsen S, Krasheninnikoff M, Foss NB, Kehlet H, Gebuhr P et al. Influence of
surgeon's experience and supervision on re-operation rate after hip fracture surgery.
Injury 2007, 38(7):775-9. (Guideline Ref ID: PALM2007)
257. Palm H, Krasheninnikoff M, Jacobsen S. Surgical treatment of proximal femoral fracture.

Ugeskrift for Laeger 2006, 168(35):2891-6. (Guideline Ref ID: PALM2006)
258. Park SR, Kang JS, Kim HS, Lee WH, Kim YH. Treatment of intertrochanteric fracture with
the Gamma AP locking nail or by a compression hip screw--a randomised prospective
trial. International Orthopaedics 1998, 22(3):157-60. (Guideline Ref ID: PARK1998)
259. Parker M, Johansen A. Hip fracture. British Medical Journal 2006, 333(7557):27-30.
(Guideline Ref ID: PARKER2006)
260. Parker M, Pryor G, Gurusamy K. Cemented versus uncemented hemiarthroplasty for

intracapsular hip fractures: a randomised controlled trial in 400 patients. Journal of
Bone & Joint Surgery - British Volume 2010, 92B(1):116-22. (Guideline Ref ID:
PARKER2010)
261. Parker MJ. Internal fixation or arthroplasty for displaced subcapital fractures in the
elderly? Injury 1992, 23(8):521-4. (Guideline Ref ID: PARKER1992B)
262. Parker MJ, Griffiths R, Appadu B. Nerve blocks (subcostal, lateral cutaneous, femoral,
triple, psoas) for hip fractures. Cochrane Database of Systematic Reviews 2002, Issue
1:CD001159. (Guideline Ref ID: PARKER2002A)
263. Parker MJ, Gurusamy K. Arthroplasties (with and without bone cement) for proximal
femoral fractures in adults. Cochrane Database of Systematic Reviews 2006, Issue
650 BIBLIOGRAPHY
264. Parker MJ, Gurusamy K. Internal fixation versus arthroplasty for intracapsular proximal
femoral fractures in adults. Cochrane Database of Systematic Reviews 2006, Issue
4:CD001708. (Guideline Ref ID: PARKER2006C)
265. Parker MJ, Gurusamy KS, Azegami S. Arthroplasties (with and without bone cement) for
proximal femoral fractures in adults. Cochrane Database of Systematic Reviews 2010,
266. Parker MJ, Handoll HHG, Griffiths R. Anaesthesia for hip fracture surgery in adults.
PARKER2004B)
267. Parker MJ, Khan RJ, Crawford J, Pryor GA. Hemiarthroplasty versus internal fixation for
displaced intracapsular hip fractures in the elderly. A randomised trial of 455 patients.
Journal of Bone & Joint Surgery - British Volume 2002, 84(8):1150-5. (Guideline Ref ID:
PARKER2002)
268. Parker MJ, Myles JW, Anand JK, Drewett R. Cost-benefit analysis of hip fracture
treatment. Journal of Bone & Joint Surgery - British Volume 1992, 74(2):261-4.
(Guideline Ref ID: PARKER1992)
269. Parker MJ, Pervez H. Surgical approaches for inserting hemiarthroplasty of the hip.
PARKER2002B)
270. Parker MJ, Pryor GA, Myles JW. Early discharge after hip fracture. Prospective 3-year
study of 645 patients. Acta Orthopaedica Scandinavica 1991, 62(6):563-6. (Guideline Ref
ID: PARKER1991)
271. Parker MJ, Pryor GA, Myles JW. The value of a special surgical team in preventing
complications in the treatment of hip fractures. International Orthopaedics 1994,
18(3):184-8. (Guideline Ref ID: PARKER1994)
272. Piangatelli C, De AC, Pecora L, Recanatini F, Testasecca D. Levobupivacaine versus

ropivacaine in psoas compartment block and sciatic nerve block in orthopedic surgery
of the lower extremity. Minerva Anestesiologica 2004, 70(12):801-7. (Guideline Ref ID:
PIANGATELLI2004)
273. Pitto RP, Blunk J, Kossler M. Transesophageal echocardiography and clinical features of
fat embolism during cemented total hip arthroplasty. A randomized study in patients
with a femoral neck fracture. Archives of Orthopaedic and Trauma Surgery 2000, 120(1-
2):53-8. (Guideline Ref ID: PITTO2000)
274. Pownall E. Using a patient narrative to influence orthopaedic nursing care in fractured
hips. Journal of Orthopaedic Nursing 2004, 8(3):151-9. (Guideline Ref ID:
POWNALL2004)
275. Pryor GA, Williams DR. Rehabilitation after hip fractures. Home and hospital
management compared. Journal of Bone & Joint Surgery - British Volume 1989,
71(3):471-4. (Guideline Ref ID: PRYOR1989)
276. Puolakka TJ, Laine HJ, Tarvainen T, Aho H. Thompson hemiarthroplasty is superior to
Ullevaal screws in treating displaced femoral neck fractures in patients over 75 years. A
BIBLIOGRAPHY 651
prospective randomized study with two-year follow-up. Annales Chirurgiae et

Gynaecologiae 2001, 90(3):225-8. (Guideline Ref ID: PUOLAKKA2001)
277. Racle JP, Benkhadra A, Poy JY, Gleizal B, Gaudray A. Comparative study of general and
spinal anesthesia in elderly women in hip surgery. Annales Françaises d'Anesthésie et de
Réanimation 1986, 5(1):24-30. (Guideline Ref ID: RACLE1986)
278. Rademakers L, Vainas T, van Zutphen S, Brink P, van Helden S. Pressure ulcers and
prolonged hospital stay in hip fracture patients affected by time-to-surgery. European
Journal of Trauma and Emergency Surgery 2007, 33(3):238-44. (Guideline Ref ID:
RADEMAKERS2007)
279. Radford PJ, Needoff M, Webb JK. A prospective randomised comparison of the dynamic
hip screw and the gamma locking nail. Journal of Bone & Joint Surgery - British Volume
1993, 75(5):789-93. (Guideline Ref ID: RADFORD1993)
280. Rae HC, Harris IA, McEvoy L, Todorova T. Delay to surgery and mortality after hip
fracture. ANZ Journal of Surgery 2007, 77(10):889-91. (Guideline Ref ID: RAE2007A)
281. Rahme DM, Harris IA. Intramedullary nailing versus fixed angle blade plating for
subtrochanteric femoral fractures: a prospective randomised controlled trial. Journal of
Orthopaedic Surgery 2007, 15(3):278-81. (Guideline Ref ID: RAHME2007)
282. Resnick B, Orwig D, Wehren L, Zimmerman S, Simpson M, Magaziner J. The Exercise Plus
Program for older women post hip fracture: participant perspectives. Gerontologist
2005, 45(4):539-44. (Guideline Ref ID: RESNICK2005)
283. Resnick B, Orwig D, Yu-Yahiro J, Hawkes W, Shardell M, Hebel JR et al. Testing the
effectiveness of the exercise plus program in older women post-hip fracture. Annals of
Behavioral Medicine 2007, 34(1):67-76. (Guideline Ref ID: RESNICK2007A)
284. Richards SH, Coast J, Gunnell DJ, Peters TJ, Pounsford J, Darlow MA. Randomised
controlled trial comparing effectiveness and acceptability of an early discharge, hospital
at home scheme with acute hospital care. British Medical Journal 1998,
316(7147):1796-801. (Guideline Ref ID: RICHARDS1998)
285. Riley TBH. Knobs or screws? - a prospective trial of prosthetic replacement against
internal fixation of subcapital fractures [Abstract]. Journal of Bone & Joint Surgery -
British Volume 1978, 60(1):136. (Guideline Ref ID: RILEY1978)
286. Rizzo PF, Gould ES, Lyden JP, Asnis SE. Diagnosis of occult fractures about the hip.
Magnetic resonance imaging compared with bone-scanning. Journal of Bone & Joint
Surgery - American Volume 1993, 75(3):395-401. (Guideline Ref ID: RIZZO1993)
287. Robinson SB. Transitions in the lives of elderly women who have sustained hip fractures.
Journal of Advanced Nursing 1999, 30(6):1341-8. (Guideline Ref ID: ROBINSON1999)
288. Roden M, Schon M, Fredin H. Treatment of displaced femoral neck fractures: a

randomized minimum 5-year follow-up study of screws and bipolar hemiprostheses in
100 patients. Acta Orthopaedica Scandinavica 2003, 74(1):42-4. (Guideline Ref ID:
RODEN2003)
652 BIBLIOGRAPHY
289. Rodriguez J, Herrara A, Canales V, Serrano S. Epidemiologic factors, morbidity and

mortality after femoral neck fractures in the elderly. A comparative study: internal
fixation vs. hemiarthroplasty. Acta Orthopaedica Belgica 1987, 53(4):472-9. (Guideline
Ref ID: RODRIGUEZ1987)
290. Rogers FB, Shackford SR, Keller MS. Early fixation reduces morbidity and mortality in
elderly patients with hip fractures from low-impact falls. Journal of Trauma-Injury
Infection & Critical Care 1995, 39(2):261-5. (Guideline Ref ID: ROGERS1995)
291. Rogmark C, Carlsson A, Johnell O, Sembo I. Costs of internal fixation and arthroplasty for
displaced femoral neck fractures: a randomized study of 68 patients. Acta Orthopaedica
Scandinavica 2003, 74(3):293-8. (Guideline Ref ID: ROGMARK2003)
292. Rogmark C, Johnell O. Primary arthroplasty is better than internal fixation of displaced
femoral neck fractures: a meta-analysis of 14 randomized studies with 2,289 patients.
Acta Orthopaedica 2006, 77(3):359-67. (Guideline Ref ID: ROGMARK2006)
293. Ryan T, Enderby P, Rigby AS. A randomized controlled trial to evaluate intensity of
community-based rehabilitation provision following stroke or hip fracture in old age.
Clinical Rehabilitation 2006, 20(2):123-31. (Guideline Ref ID: RYAN2006)
294. Sadowski C, Lubbeke A, Saudan M, Riand N, Stern R, Hoffmeyer P. Treatment of reverse

oblique and transverse intertrochanteric fractures with use of an intramedullary nail or
a 95 degrees screw-plate: a prospective, randomized study. Journal of Bone & Joint
Surgery - American Volume 2002, 84-A(3):372-81. (Guideline Ref ID: SADOWSKI2002)
295. Sadr B, Arden GP. A comparison of the stability of proplast-coated and cemented
Thompson prostheses in the treatment of subcapital femoral fractures. Injury 1977,
8(3):234-7. (Guideline Ref ID: SADR1977)
296. Safran DG, Graham JD, Osberg JS. Social supports as a determinant of community-based
care utilization among rehabilitation patients. Health Services Research 1994, 28(6):729-
50. (Guideline Ref ID: SAFRAN1994)
297. Safran O, Goldman V, Applbaum Y, Milgrom C, Bloom R, Peyser A et al. Posttraumatic

painful hip: sonography as a screening test for occult hip fractures. Journal of
Ultrasound in Medicine 2009, 28(11):1447-52. (Guideline Ref ID: SAFRAN2009)
298. Santini S, Rebeccato A, Bolgan I, Turi G. Hip fractures in elderly patients treated with
bipolar hemiarthroplasty: comparison between cemented and cementless implants.
Journal of Orthopaedics and Traumatology 2005, 6(2):80-7. (Guideline Ref ID:
SANTINI2005)
299. Sarvilinna R, Pajamaki J, Sovelius R, Puolakka T, Huhtala H, Jarvinen M. Bipolar

hemiartroplasty in the treatment of hip fracture: are special centres necessary for the
optimal outcome? Scandinavian Journal of Surgery: SJS 2002, 91(2):182-5. (Guideline
Ref ID: SARVILINNA2002)
300. Saudan M, Lubbeke A, Sadowski C, Riand N, Stern R, Hoffmeyer P. Pertrochanteric

fractures: is there an advantage to an intramedullary nail?: a randomized, prospective
study of 206 patients comparing the dynamic hip screw and proximal femoral nail.
Journal of Orthopaedic Trauma 2002, 16(6):386-93. (Guideline Ref ID: SAUDAN2002)
BIBLIOGRAPHY 653
301. Schiferer A, Gore C, Gorove L, Lang T, Steinlechner B, Zimpfer M et al. A randomized

controlled trial of femoral nerve blockade administered preclinically for pain relief in
femoral trauma. Anesthesia and Analgesia 2007, 105(6):1852-4. (Guideline Ref ID:
SCHIFERER2007)
302. Sebestyen A, Boncz I, Sandor J, Nyarady J. Effect of surgical delay on early mortality in
patients with femoral neck fracture. International Orthopaedics 2008, 32(3):375-9.
(Guideline Ref ID: SEBESTYEN2008B)
303. Sehat K, Baker RP, Pattison G, Price R, Harries WJ, Chesser TJ. The use of the long
gamma nail in proximal femoral fractures. Injury 2005, 36(11):1350-4. (Guideline Ref ID:
SEHAT2005)
304. Shabat S, Heller E, Mann G, Gepstein R, Fredman B, Nyska M. Economic consequences

of operative delay for hip fractures in a non-profit institution. Orthopedics 2003,
26(12):1197-9. (Guideline Ref ID: SHABAT2003)
305. Shyu YI, Liang J, Wu CC, Su JY, Cheng HS, Chou SW et al. Interdisciplinary intervention
for hip fracture in older Taiwanese: benefits last for 1 year. Journals of Gerontology
Series A-Biological Sciences & Medical Sciences 2008, 63(1):92-7. (Guideline Ref ID:
SHYU2008A)
306. Shyu YI, Liang J, Wu CC, Su JY, Cheng HS, Chou SW et al. A pilot investigation of the
short-term effects of an interdisciplinary intervention program on elderly patients with
hip fracture in Taiwan. Journal of the American Geriatrics Society 2005, 53(5):811-8.
(Guideline Ref ID: SHYU2005)
307. Shyu Y-I, Liang J, Wu C-C, Su J-Y, Cheng H-S, Chou S-W et al. Two-year effects of
interdisciplinary intervention for hip fracture in older Taiwanese. Journal of the
American Geriatrics Society 2010, 58(6):1081-9. (Guideline Ref ID: SHYU2010)
308. Siegmeth AW, Gurusamy K, Parker MJ. Delay to surgery prolongs hospital stay in
patients with fractures of the proximal femur. Journal of Bone & Joint Surgery - British
Volume 2005, 87(8):1123-6. (Guideline Ref ID: SIEGMETH2005A)
309. Sikorski JM, Barrington R. Internal fixation versus hemiarthroplasty for the displaced
subcapital fracture of the femur. A prospective randomised study. Journal of Bone &
Joint Surgery - British Volume 1981, 63-B(3):357-61. (Guideline Ref ID: SIKORSKI1981)
310. Sikorski JM, Davis NJ, Senior J. The rapid transit system for patients with fractures of
proximal femur. British Medical Journal Clinical Research Ed 1985, 290(6466):439-43.
(Guideline Ref ID: SIKORSKI1985)
311. Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and

length of hospital stay after repair of proximal femoral fracture: randomised controlled
trial. BMJ 1997, 315(7113):909-12. (Guideline Ref ID: SINCLAIR1997)
312. Sircar P, Godkar D, Mahgerefteh S, Chambers K, Niranjan S, Cucco R. Morbidity and

mortality among patients with hip fractures surgically repaired within and after 48
hours. American Journal of Therapeutics 2007, 14(6):508-13. (Guideline Ref ID:
SIRCAR2007)
654 BIBLIOGRAPHY
313. Skinner P, Riley D, Ellery J, Beaumont A, Coumine R, Shafighian B. Displaced subcapital

fractures of the femur: a prospective randomized comparison of internal fixation,
hemiarthroplasty and total hip replacement. Injury 1989, 20(5):291-3. (Guideline Ref ID:
SKINNER1989)
314. Slauenwhite CA, Simpson P. Patient and family perspectives on early discharge and care
of the older adult undergoing fractured hip rehabilitation. Orthopaedic Nursing 1998,
17(1):30-6. (Guideline Ref ID: SLAUENWHITE1998)
315. Smith M, Rousseau N, Lecouturier J, Gregson B, Bond J, Rodgers H. Are older people
satisfied with discharge information? Nursing Times 1997, 93(43):52-3. (Guideline Ref
ID: SMITH1997)
316. Sonne-Holm S, Walter S, Jensen JS. Moore hemi-arthroplasty with and without bone
cement in femoral neck fractures. A clinical controlled trial. Acta Orthopaedica
Scandinavica 1982, 53(6):953-6. (Guideline Ref ID: SONNEHOLM1982)
317. Soreide O, Molster A, Raugstad TS. Internal fixation versus primary prosthetic
replacement in acute femoral neck fractures: a prospective, randomized clinical study.
British Journal of Surgery 1979, 66(1):56-60. (Guideline Ref ID: SOREIDE1979)
318. Spansberg NL, Anker-Moller E, Dahl JB, Schultz P, Christensen EF. The value of
continuous blockade of the lumbar plexus as an adjunct to acetylsalicyclic acid for pain
relief after surgery for femoral neck fractures. European Journal of Anaesthesiology
1996, 13(4):410-2. (Guideline Ref ID: SPANSBERG1996)
319. Stenvall M, Olofsson B, Lundstrom M, Englund U, Borssen B, Svensson O et al. A

multidisciplinary, multifactorial intervention program reduces postoperative falls and
injuries after femoral neck fracture. Osteoporosis International 2007, 18(2):167-75.
(Guideline Ref ID: STENVALL2007A)
320. Stenvall M, Olofsson B, Nyberg L, Lundstrom M, Gustafson Y. Improved performance in

activities of daily living and mobility after a multidisciplinary postoperative
rehabilitation in older people with femoral neck fracture: a randomized controlled trial
with 1-year follow-up. Journal of Rehabilitation Medicine 2007, 39(3):232-8. (Guideline
Ref ID: STENVALL2007)
321. Stewart HD. Pugh's nail fixation versus Thompson's prosthesis for displaced subcapital
fractures of the femur. Injury 1984, 15(4):227-31. (Guideline Ref ID: STEWART1984)
322. Sund R, Liski A. Quality effects of operative delay on mortality in hip fracture treatment.
Quality and Safety in Health Care 2005, 14(5):371-7. (Guideline Ref ID: SUND2005)
323. Sutcliffe AJ, Parker M. Mortality after spinal and general anaesthesia for surgical fixation
of hip fractures. Anaesthesia 1994, 49(3):237-40. (Guideline Ref ID: SUTCLIFFE1994)
324. Svenningsen S, Benum P, Nesse O, Furset OI. Femoral neck fractures in the elderly--a
comparison of 3 treatment methods. Nordisk Medicin 1985, 100(10):256-9. (Guideline
Ref ID: SVENNINGSEN1985)
325. Swanson CE, Day GA, Yelland CE, Broome JR, Massey L, Richardson HR et al. The
management of elderly patients with femoral fractures. A randomised controlled trial of
BIBLIOGRAPHY 655
early intervention versus standard care. Medical Journal of Australia 1998, 169(10):515-
8. (Guideline Ref ID: SWANSON1998)
326. Tappen RM, Whitehead D, Folden SL, Hall R. Effect of a video intervention on functional
recovery following hip replacement and hip fracture repair. Rehabilitation Nursing
2003, 28(5):148-53. (Guideline Ref ID: TAPPEN2003)
327. Tidermark J, Ponzer S, Svensson O, Soderqvist A, Tornkvist H. Internal fixation compared

with total hip replacement for displaced femoral neck fractures in the elderly. A
randomised, controlled trial. Journal of Bone & Joint Surgery - British Volume 2003,
85(3):380-8. (Guideline Ref ID: TIDERMARK2003B)
328. Tidermark J, Zethraeus N, Svensson O, Tornkvist H, Ponzer S. Femoral neck fractures in

the elderly: functional outcome and quality of life according to EuroQol. Quality of Life
Research 2002, 11(5):473-81. (Guideline Ref ID: TIDERMARK2002)
329. Tonczar L, Hammerle AF. The impairment of stress parameters by hip joint close
operations and the influence of anaesthesia. Preliminary results of a prospective study.
Unfallchirurgie 1981, 7(3):138-41. (Guideline Ref ID: TONCZAR1981)
330. Tsauo JY, Leu WS, Chen YT, Yang RS. Effects on function and quality of life of
postoperative home-based physical therapy for patients with hip fracture. Archives of
Physical Medicine and Rehabilitation 2005, 86(10):1953-7. (Guideline Ref ID:
TSAUO2005)
331. Tuncer S, Sert OA, Yosunkaya A, Mutlu M, Celik J, Okesli S. Patient-controlled femoral
nerve analgesia versus patient-controlled intravenous analgesia for postoperative
analgesia after trochanteric fracture repair. Acute Pain 2003, 4(3-4):105-8. (Guideline
Ref ID: TUNCER2003)
332. Türker G, Uçkunkaya N, Yavasçaoglu B, Yilmazlar A, Ozçelik S. Comparison of the

catheter-technique psoas compartment block and the epidural block for analgesia in
partial hip replacement surgery. Acta Anaesthesiologica Scandinavica 2003, 47(1):30-6.
(Guideline Ref ID: TÜRKER2003)
333. Ungemach J. Inhalation anesthesia or "balanced anesthesia"? A comparative

perioperative study in geriatric patients. Anaesthesist 1987, 36(6):288-91. (Guideline Ref
ID: UNGEMACH1987)
334. Ungemach JW, Andres FJ, Eggert E, Schoder K. The role of anaesthesia in geriatric
patients with hip fractures: a prospective study. European Journal of Anaesthesiology
1993, 10(5):380. (Guideline Ref ID: UNGEMACH1993)
335. Unnanuntana A, Dimitroulias A, Bolognesi MP, Hwang KL, Goodman SB, Marcus RE.
Cementless femoral prostheses cost more to implant than cemented femoral
prostheses. Clin Orthop Relat Res 2009, 467(6):1546-51. (Guideline Ref ID:
UNNANUNTANA2009)
336. Unwin AJ, Thomas M. Dislocation after hemiarthroplasty of the hip: a comparison of the
dislocation rate after posterior and lateral approaches to the hip. Annals of the Royal
College of Surgeons of England 1994, 76(5):327-9. (Guideline Ref ID: UNWIN1994)
656 BIBLIOGRAPHY
337. Utrilla AL, Reig JS, Munoz FM, Tufanisco CB. Trochanteric gamma nail and compression
hip screw for trochanteric fractures: a randomized, prospective, comparative study in
210 elderly patients with a new design of the gamma nail. Journal of Orthopaedic
Trauma 2005, 19(4):229-33. (Guideline Ref ID: UTRILLA2005)
338. Uy C, Kurrle SE, Cameron ID. Inpatient multidisciplinary rehabilitation after hip fracture
for residents of nursing homes: a randomised trial. Australasian Journal on Ageing 2008,
27(1):43-4. (Guideline Ref ID: UY2008)
339. Valentin N, Lomholt B, Jensen JS, Hejgaard N, Kreiner S. Spinal or general anaesthesia
for surgery of the fractured hip? A prospective study of mortality in 578 patients. British
Journal of Anaesthesia 1986, 58(3):284-91. (Guideline Ref ID: VALENTIN1986)
340. van Balen R, Steyerberg EW, Cools HJ, Polder JJ, Habbema JD. Early discharge of hip
fracture patients from hospital: transfer of costs from hospital to nursing home. Acta
Orthopaedica Scandinavica 2002, 73(5):491-5. (Guideline Ref ID: VAN2002)
341. van Dortmont LM, Douw CM, van Breukelen AM, Laurens DR, Mulder PG, Wereldsma JC
et al. Cannulated screws versus hemiarthroplasty for displaced intracapsular femoral
neck fractures in demented patients. Annales Chirurgiae et Gynaecologiae 2000,
89(2):132-7. (Guideline Ref ID: VANDORTMONT2000)
342. Van Leeuwen FL, Bronselaer K, Gilles M, Sabbe MB, Delooz HH. The 'three in one' block
as locoregional analgesia in an emergency department. European Journal of Emergency
Medicine 2000, 7(1):35-8. (Guideline Ref ID: VANLEEUWEN2000)
343. van Vugt AB, Oosterwijk WM, Goris RJ. Osteosynthesis versus endoprosthesis in the
treatment of unstable intracapsular hip fractures in the elderly. A randomised clinical
trial. Archives of Orthopaedic and Trauma Surgery 1993, 113(1):39-45. (Guideline Ref ID:
VANVUGT1993)
344. Vidan M, Serra JA, Moreno C, Riquelme G, Ortiz J. Efficacy of a comprehensive geriatric
intervention in older patients hospitalized for hip fracture: a randomized, controlled
trial. Journal of the American Geriatrics Society 2005, 53(9):1476-82. (Guideline Ref ID:
VIDAN2005)
345. Vochteloo AJ, Niesten D, Riedijk R, Rijnberg WJ, Bolder SB, Koeter S et al. Cemented
versus non-cemented hemiarthroplasty of the hip as a treatment for a displaced
femoral neck fracture: design of a randomised controlled trial. BMC Musculoskeletal
Disorders 2009, 10:56. (Guideline Ref ID: VOCHTELOO2009)
346. Wang G, Gu GS, Li D, Sun DH, Zhang W, Wang TJ. Comparative study of anterolateral
approach versus posterior approach for total hip replacement in the treatment of
femoral neck fractures in elderly patients. Chinese Journal of Traumatology 2010,
13(4):234-9. (Guideline Ref ID: WANG2010)
347. Wang J, Jiang B, Marshall RJ, Zhang P. Arthroplasty or internal fixation for displaced
femoral neck fractures: which is the optimal alternative for elderly patients? A meta-
analysis. International Orthopaedics 2009, 33(5):1179-87. (Guideline Ref ID:
WANG2009)
348. Wanless D. (2006) Securing good care for older people: taking a long-term view. London:
King's Fund. (Guideline Ref ID: WANLESS2006)
BIBLIOGRAPHY 657
349. Webster R. Fractured femur: a patient's viewpoint. New Zealand Nursing Journal 1976,
69(12):6-8. (Guideline Ref ID: WEBSTER1976)
350. Weinrauch P. Intra-operative error during Austin Moore hemiarthroplasty. Journal of

Orthopaedic Surgery 2006, 14(3):249-52. (Guideline Ref ID: WEINRAUCH2006)
351. Weller I, Wai EK, Jaglal S, Kreder HJ. The effect of hospital type and surgical delay on
mortality after surgery for hip fracture. Journal of Bone & Joint Surgery - British Volume
2005, 87B(3):361-6. (Guideline Ref ID: WELLER2005)
352. White IW, Chappell WA. Anaesthesia for surgical correction of fractured femoral neck. A
comparison of three techniques. Anaesthesia 1980, 35(11):1107-10. (Guideline Ref ID:
WHITE1980)
353. Widman J, Isacson J. Lateral position reduces blood loss in hip replacement surgery: a
prospective randomized study of 74 patients. International Orthopaedics 2001,
25(4):226-7. (Guideline Ref ID: WIDMAN2001)
354. Williams MA, Oberst MT, Bjorklund BC. Posthospital convalescence in older women
with hip fracture. Orthopaedic Nursing 1994, 13(4):55-64. (Guideline Ref ID:
WILLIAMS1994)
355. Wykes C, Pryor J, Jeeawody B. The concerns of older women during inpatient
rehabilitation after fractured neck of femur. International Journal of Therapy &
Rehabilitation 2009, 16(5):261-70. (Guideline Ref ID: WYKES2009)
356. Yang C, Zhu Q, Han Y, Zhu J, Wang H, Cong R et al. Minimally-invasive total hip
arthroplasty will improve early postoperative outcomes: A prospective, randomized,
controlled trial. Irish Journal of Medical Science 2010, 179(2):285-90. (Guideline Ref ID:
YANG2010)
357. Yoon RS, Macaulay W, Torres G, Nellans KW, Siris ES, Bigliani LU et al. Assessment of
inpatient fragility fracture education and outpatient follow-up at an urban tertiary care
institution. Endocrine Practice 2008, 14(1):58-68. (Guideline Ref ID: YOON2008)
358. Young Y, Resnick B. Don't worry, be positive: improving functional recovery 1 year after
hip fracture. Rehabilitation Nursing 2009, 34(3):110-7. (Guideline Ref ID: YOUNG2009A)
359. Yu-Yahiro JA, Resnick B, Orwig D, Hicks G, Magaziner J. Design and implementation of a
home-based exercise program post-hip fracture: the Baltimore hip studies experience.
PM&R 2009, 1(4):308-18. (Guideline Ref ID: YUYAHIRO2009)
360. Ziden L, Frandin K, Kreuter M. Home rehabilitation after hip fracture. A randomized
controlled study on balance confidence, physical function and everyday activities.
Clinical Rehabilitation 2008, 22(12):1019-33. (Guideline Ref ID: ZIDEN2008)
361. Ziden L, Kreuter M, Frandin K. Long-term effects of home rehabilitation after hip
fracture - 1-year follow-up of functioning, balance confidence, and health-related
quality of life in elderly people. Disability and Rehabilitation 2010, 32(1):18-32.
(Guideline Ref ID: ZIDEN2010A)
658 BIBLIOGRAPHY
362. Ziden L, Scherman MH, Wenestam CG. The break remains--Elderly people's experiences
of a hip fracture 1 year after discharge. Disability and Rehabilitation 2010, 32(2):103-13.
(Guideline Ref ID: ZIDEN2010)
363. Ziden L, Wenestam CG, Hansson-Scherman M. A life-breaking event: early experiences

of the consequences of a hip fracture for elderly people. Clinical Rehabilitation 2008,
22(9):801-11. (Guideline Ref ID: ZIDEN2008A)
364. Zou J, Xu Y, Yang H. A comparison of proximal femoral nail antirotation and dynamic hip
screw devices in trochanteric fractures. Journal of International Medical Research 2009,
37(4):1057-64. (Guideline Ref ID: ZOU2009)

Full Guideline PDF 183081997

Uploaded by

Copyright:

Available Formats

Full Guideline PDF 183081997

Uploaded by

Document Information

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

Full Guideline PDF 183081997

Uploaded by

Copyright:

Available Formats

The management of hip fracture in adults

These changes can be seen in the short version of the

Produced by the National Clinical Guideline Centre

First published 2011

© National Clinical Guideline Centre 2011

2.2 REMIT .................................................................................................................................................. 11

10.4 USE OF CEMENT IN ARTHROPLASTY ...................................................................................................... 112

16 APPENDIX D: LITERATURE SEARCH STRATEGIES .......................................................................................... 262

Guideline development group members

Professor Cameron Swift (Chair) Emeritus Professor of Health Care of the

Mr Tim Chesser Consultant Trauma and Orthopaedic Surgeon

Mr Anthony Field Patient Member

Dr Richard Griffiths Consultant Anaesthetist

Mr Robert Handley Consultant Trauma and Orthopaedic Surgeon

Mrs Karen Hertz Advanced Nurse Practitioner

Dr Sally Hope General Practitioner

Dr Antony Johansen Consultant Orthogeriatrician

Professor Sarah (Sallie) Lamb Professor of Rehabilitation, Director of

Professor Opinder Sahota Consultant Physician

Mrs Tessa Somerville Patient Member

Mrs Heather Towndrow Clinical Manager, Day Rehabilitation and Falls

Mr Martin Wiese Consultant in Emergency Medicine

NCGC staff members of the guideline

Dr Saoussen Ftouh Senior Research Fellow / Project Manager

Ms Joanna Ashe Information Scientist

Miss Elisabetta Fenu Senior Health Economist

Dr Jennifer Hill Operations Director

Dr Antonia Morga Health Economist

Dr Sarah Riley Research Fellow

Mr Carlos Sharpin Senior Information Scientist / Research Fellow

Professor Judy Adams Consultant and Honorary Professor of

Mrs Pamela Holmes Practice Development Manager (SCIE)

Mr Martyn Parker Consultant Orthopaedic Surgeon

Dr Luigi Siciliani Senior Lecturer in Health Economics

Acronyms and abbreviations

ADL Activities of Daily Living

IQR Interquartile range

SCIE Social Care Institute for Excellence

Figure 1: Types of hip fracture (Parker M & Johansen A, 2006)259,270

2 Development of the guideline

2.1 What is a NICE clinical guideline?

NICE clinical guidelines can:

• provide recommendations for the treatment and care of people by health

• be used to develop standards to assess the clinical practice of individual health

• be used in the education and training of health professionals

• help patients to make informed decisions

• improve communication between patient and health professional

We produce our guidelines using the following steps:

• Guideline topic is referred to NICE from the Department of Health

• The scope is prepared by the National Clinical Guideline Centre (NCGC)

• The NCGC establishes a guideline development group

• There is a consultation on the draft guideline.

• The final guideline is produced.

The NCGC and NICE produce a number of versions of this guideline:

• the NICE guideline lists the recommendations

• the quick reference guide (QRG) presents recommendations in a suitable format

The remit for this guideline is:

To prepare a clinical guideline on the management of fractured neck of femur.

2.3 Who developed this guideline?

2.4 What this guideline covers