Upload
doandien
View
247
Download
1
Embed Size (px)
Citation preview
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257
Understanding place of death for patients with non malignant conditions: a systematic literature review
Murtagh FEM,1 Bausewein C,1 Petkova H,1 Sleeman KE,1 Dodd RH,1 Gysels M,1 Johnston B,2 Murray S,3 Banerjee S,4 Shipman C,1 Hansford P,5 Wakefield D,1 Gomes B,1 and Higginson IJ 1
1 King’s College London, Cicely Saunders Institute, Dept of Palliative Care, Policy &
Rehabilitation 2 School of Nursing and Midwifery, College of Medicine, Dentistry and Nursing, University
of Dundee 3 Primary Palliative Care Research Group, Centre for Population Health Sciences,
University of Edinburgh 4 King’s College London, Institute of Psychiatry 5 St Christopher’s Hospice, Sydenham, London
Published September 2012
This project is funded by the Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 2
Address for correspondence:
Dr FEM Murtagh
Department of Palliative Care, Policy and Rehabilitation
King's College London
Cicely Saunders Institute
Bessemer Road, London SE5 9PJ
Email: [email protected]
This report should be referenced as follows:
Murtagh FEM, Bausewein C, Petkova H, Sleeman KE, Dodd RH, Gysels M, Johnston B,
Murray S, Banerjee S, Shipman C, Hansford P, Wakefield D, Gomes B, and Higginson IJ.
Understanding place of death for patients with non malignant conditions: a systematic
literature review. Final report. NIHR Service Delivery and Organisation programme; 2012
Relationship statement:
This document is an output from a research project that was funded by the NIHR Service
Delivery and Organisation (SDO) programme based at the National Institute for Health
Research Evaluations, Trials and Studies Coordinating Centre (NETSCC) at the University
of Southampton. The management of the project and subsequent editorial review of the
final report was undertaken by the NIHR Service Delivery and Organisation (SDO)
programme. From January 2012, the NIHR SDO programme merged with the NIHR
Health Services Research (NIHR HSR) programme to establish the new NIHR Health
Services and Delivery Research (NIHR HS&DR) programme. Should you have any queries
please contact [email protected].
Copyright information:
This report may be freely reproduced for the purposes of private research and study and
extracts (or indeed, the full report) may be included in professional journals provided
that suitable acknowledgement is made and the reproduction is not associated with any
form of advertising. Applications for commercial reproduction should be addressed to:
NETSCC, HS&DR.
National Institute for Health Research
Evaluation, Trials and Studies Coordinating Centre
University of Southampton
Alpha House, Enterprise Road
Southampton SO16 7NS
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 3
Disclaimer:
This report presents independent research funded by the National Institute for Health
Research (NIHR). The views expressed are those of the authors and not necessarily
those of the NHS, the NIHR or the Department of Health. If there are verbatim
quotations included in this publication the views and opinions expressed by the
interviewees are those of the interviewees and not necessarily those of the NHS, the
NIHR or the Department of Health.
Criteria for inclusion
Reports are published if (1) they have resulted from work for the SDO programme
including those submitted post the merge to the HS&DR programme, and (2) they are of
a sufficiently high scientific quality as assessed by the reviewers and editors. The
research in this report was commissioned by the SDO programme as project number
09/1005/01. The contractual start date was in January 2010. The final report began
editorial review in April 2011 and was accepted for publication in September 2012. The
authors have been wholly responsible for all data collection, analysis and interpretation,
and for writing up their work. The SDO editorial team have tried to ensure the accuracy
of the authors’ report and would like to thank the reviewers for their constructive
comments on the final report documentation. However, they do not accept liability for
damages or losses arising from material published in this report.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 4
Contents
Contents ................................................................................................... 4
List of tables .............................................................................................. 9
List of figures ........................................................................................... 10
Glossary of terms/abbreviations ................................................................. 16
Acknowledgements ................................................................................... 18
Executive Summary .................................................................................. 19
Background ............................................................................................. 19
Aim ........................................................................................................ 19
Methods .................................................................................................. 19
Results.................................................................................................... 20
Conclusions ............................................................................................. 22
The Report .............................................................................................. 24
1 Background ........................................................................................ 24
1.1 Major cost implications .................................................................. 24
1.2 Increasing public priority ................................................................ 24
1.3 Mismatch between preference and reality ......................................... 25
1.4 The evidence in cancer .................................................................. 25
1.5 The evidence in non malignant conditions ........................................ 26
1.6 Why non malignant conditions are different ...................................... 27
1.7 Why this systematic literature review is needed ................................ 27
1.8 Aims and objectives ...................................................................... 28
2 Theoretical modelling .......................................................................... 29
2.1 A theoretical model in cancer .......................................................... 29
2.2 Developing a theoretical model for non malignant conditions .............. 30
3 Review questions ................................................................................ 33
4 Methods ............................................................................................ 33
4.1 Identifying the evidence ................................................................. 33
4.1.1 Sources of evidence ................................................................. 33
4.1.2 Electronic databases ................................................................ 34
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 5
4.1.3 Search terms .......................................................................... 34
4.1.4 Outcomes ............................................................................... 36
4.1.5 Refinement of search strategy ................................................... 37
4.1.6 Reference list and cited reference searching ................................ 37
4.1.7 Hand searching ....................................................................... 37
4.2 Inclusion of evidence in the review .................................................. 38
4.2.1 Study outcomes ...................................................................... 38
4.2.2 Study population ..................................................................... 38
4.2.3 Study design ........................................................................... 39
4.2.4 Country of origin and non English papers .................................... 39
4.2.5 Other non malignant conditions ................................................. 39
4.2.6 Cross checks for inclusion of evidence ........................................ 39
4.2.7 Categorization of included studies .............................................. 40
4.2.8 Exclusions .............................................................................. 40
4.3 Study quality assessment ............................................................... 40
4.3.1 Quantitative evidence............................................................... 40
4.3.2 Qualitative evidence ................................................................. 43
4.4 Grading the strength of evidence .................................................... 44
4.5 Data extraction ............................................................................. 45
4.6 Synthesis of evidence .................................................................... 45
4.6.1 Presentation of quantitative evidence ......................................... 45
4.6.2 Questions addressed by the evidence ......................................... 46
5 Results: overview ............................................................................... 48
5.1 Evidence identified ........................................................................ 48
5.2 Inclusion/exclusion checking ........................................................... 48
5.3 Presentation of results ................................................................... 48
6 Results: preferences for place of care .................................................... 50
6.1 Evidence identified ........................................................................ 50
6.2 Study characteristics and quality of the evidence .............................. 50
6.3 Preferences for place of care and factors influencing this .................... 55
7 Results: preferences for place of death .................................................. 62
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 6
7.1 Evidence identified ........................................................................ 62
7.2 Study characteristics and quality of the evidence .............................. 62
7.3 Preferences for place of death and factors associated with this ........... 74
8 Results: evidence on place of death ...................................................... 81
8.1 Identified evidence ........................................................................ 81
8.2 Quality of evidence ........................................................................ 81
8.3 Heterogeneity of evidence .............................................................. 87
8.4 Categorization of variables associated with place of death .................. 88
8.5 Disease-specific analyses ............................................................... 88
9 Results: factors associated with home death .......................................... 90
9.1 Factors operating in the general non-cancer population ...................... 90
9.2 Personal and demographic factors ................................................... 93
9.3 Disease-related factors .................................................................. 97
9.4 Environmental factors - social support ............................................101
9.5 Environmental factors – health and social care input.........................102
9.6 Environmental factors – macrosocial factors ....................................102
9.7 Symptoms, function, illness burden, and trajectory ..........................103
10 Results: factors associated with hospital death ...................................106
10.1 Factors operating in the general non-cancer population ..................106
10.2 Personal and demographic factors ...............................................109
10.3 Disease-related factors ..............................................................114
10.4 Environmental factors – social support .........................................117
10.5 Environmental factors – health and social care input ......................119
10.6 Environmental factors – macrosocial factors .................................121
10.7 Symptoms, function, illness burden, and trajectory .......................125
10.8 Factors specific to the COPD population ........................................127
11 Results: factors associated with hospice death ....................................134
11.1 Factors operating in the general non cancer population ..................134
11.2 Factors specific to those already referred to palliative care .............134
12 Results: factors associated with nursing home death ...........................139
12.1 Factors operating in the general non-cancer population ..................139
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 7
12.2 Personal and demographic factors ...............................................141
12.3 Disease-related factors ..............................................................144
12.4 Symptoms, function, illness burden, and trajectory .......................146
12.5 Environmental factors – social support .........................................148
12.6 Environmental factors – health and social care input ......................148
12.7 Environmental factors – macrosocial factors .................................149
12.8 Factors specific to dementia populations .......................................150
12.9 Factors specific to nursing home populations ................................153
13 Results: qualitative evidence ............................................................161
13.1 Chronic heart failure ..................................................................161
13.2 Chronic obstructive pulmonary disease ........................................163
13.3 Long term neurological conditions ...............................................166
13.4 End stage kidney disease ...........................................................168
13.5 Older people, including stroke and dementia .................................169
13.5.1 Outcomes: place of care and death, preferences, and transitions
173
14 Results: transitions in care towards end of life ....................................177
14.1 Transition into hospital ...............................................................190
14.2 Transition into nursing home ......................................................191
14.3 Transition back home .................................................................196
14.4 Transition to palliative care .........................................................197
15 Discussion ......................................................................................200
15.1 What is the prevalence of a home death preference? .....................200
15.2 How does this preference vary by diagnosis? ................................200
15.3 What determines preferences for place of care and place of death? .201
15.4 What factors are associated with actual place of death? .................203
15.5 What are the key transitions in care at the end of life and what
influences these transitions ....................................................................209
15.6 The work of the Management Fellow within this review ..................210
15.6.1 Communication, planning ahead and enabling smooth transitions
210
15.6.2 Disease specific issues ..........................................................211
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 8
15.7 Strengths of this review .............................................................211
15.8 Limitations ...............................................................................213
15.8.1 Identifying the evidence .......................................................213
15.8.2 Heterogeneity of evidence .....................................................213
15.8.3 Quality and extent of the evidence .........................................215
16 Conclusions ....................................................................................216
16.1.1 Practice and policy implications..............................................218
16.1.2 Research implications ...........................................................219
References .............................................................................................220
Appendix 1: Consultation participants ........................................................236
Appendix 2: Final search strategy ..............................................................238
Appendix 3: All included studies ................................................................240
Appendix 4: Excluded papers ....................................................................262
Appendix 5: Quality score for qualitative studies .........................................278
Appendix 6: Data extraction for qualitative studies ......................................282
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 9
List of tables
Table 1. Sources of data: electronic databases and journals ...................... 34
Table 2. Included papers which report preferred place of care using patient
data ............................................................................................. 51
Table 3. Included papers which report preferred place of care as reported by
proxies ............................................................................................. 53
Table 4. Included papers which report preferred place of care reported by
chart reviews and secondary analysis ......................................................... 54
Table 5. Conventions used in the ‘bubble’ plots ........................................ 55
Table 6. Evidence on factors associated with preference for home treatment
as reported by patients ............................................................................. 58
Table 7. Evidence on factors associated with preference for hospice enrolment
as reported by patients ............................................................................. 60
Table 8. Included papers which report preferences for place of death in
advanced non malignant conditions reported by patients ............................... 64
Table 9. Included papers which report preferences for place of death as
reported by families .................................................................................. 69
Table 10. Included papers which report preferences for place of death as
reported by professionals .......................................................................... 71
Table 11. Included papers which report preferences for place of death as
reported from records ............................................................................... 73
Table 12. Conventions used in the ‘bubble’ plots ........................................ 74
Table 13. Evidence on factors associated with GPs awareness of patients’
preferred place of death ............................................................................ 80
Table 14. Included papers reporting multivariate analyses on factors affecting
place of death .......................................................................................... 82
Table 15. Evidence on factors associated with home death in the general
advanced non-cancer population ................................................................ 91
Table 16. Conventions used in the graphs ................................................. 93
Table 17. Evidence on factors associated with hospital death in the general non
cancer population ....................................................................................107
Table 18. Evidence on factors associated with hospital death in the COPD
population ............................................................................................128
Table 19. Evidence on factors associated with nursing home death in the
general non-cancer population ..................................................................140
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 10
Table 20. Evidence on factors associated with nursing home death in the
dementia population ................................................................................150
Table 21. Evidence on factors associated with nursing home death among
nursing home residents ............................................................................153
Table 22. The definition of ‘transition’, as adopted for this review ...............177
Table 23. Included papers on transitions towards end of life in non malignant
conditions ............................................................................................178
Table 24. Participants in the consultation to develop the theoretical model ..236
Table 25. Terms used for the final search strategy (these were refined and
adapted to each electronic database) .........................................................238
Table 26. The qualitative studies’ methodological quality scores (according to
the method developed by Hawker et al. 2002) ............................................278
Table 27. Data Extraction Table CHF .......................................................282
Table 28. Data Extraction Table COPD .....................................................287
Table 29. Data Extraction Table Long Term Neurological Conditions ............293
Table 30. Data Extraction Table End Stage Kidney Disease ........................296
Table 31. Data Extraction Table Older People, including stroke and dementia ....
............................................................................................298
List of figures
Figure 1. Model of the factors affecting place of care and place of death (15) 29
Figure 2. Proposed theoretical model of the factors affecting place of care and
death in non malignant conditions .............................................................. 31
Figure 3. Terms used to derive the search strategy (using PICOS) .............. 35
Figure 4. Quality scale used for quantitative evidence (adapted from Edwards
(68, 69)) ............................................................................................. 42
Figure 5. Quality scale used for qualitative evidence, from Hawker (70) ....... 43
Figure 6. Algorithm to assess strength of evidence (15) ............................. 45
Figure 7. PRISMA flowchart for identification of evidence ............................ 49
Figure 8. Preferred place of care as reported by patients themselves ........... 55
Figure 9. Preferred place of death as reported by renal patients .................. 75
Figure 10. Preferred place of death as reported by patients with CHF ......... 76
Figure 11. Preferred place of death as reported by patients (mixed diagnoses)
.......................................................................................... 77
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 11
Figure 12. Preferred place of death as reported by families (mixed diagnoses) .
.......................................................................................... 78
Figure 13. Preferred place of death as reported by professionals (mixed
diagnoses) .......................................................................................... 79
Figure 14. Association of age with home death ........................................ 94
Figure 15. Association of gender with home death .................................... 95
Figure 16. Association of ethnicity with home death ................................. 95
Figure 17. Association of marital status with home death .......................... 96
Figure 18. Association of further education with home death ..................... 97
Figure 19. Association of CHF with home death ........................................ 98
Figure 20. Association of COPD with home death ..................................... 98
Figure 21. Association of dementia with home death ................................ 99
Figure 22. Association of renal disease with home death ..........................100
Figure 23. Association of stroke with home death ....................................100
Figure 24. Association of co-morbidity with home death ...........................101
Figure 25. Association of no informal carer/living alone with home death ...101
Figure 26. Association of hospital bed availability with home death ...........102
Figure 27. Association of income with home death ..................................102
Figure 28. Association of activity of daily living (ADL) impairment with home
death .........................................................................................103
Figure 29. Association of cognitive impairment with home death ...............104
Figure 30. Association of ‘expected death’ with home death .....................104
Figure 31. Association of trajectory of illness with home death ..................105
Figure 32. Association of age with hospital death ....................................109
Figure 33. Association of gender with hospital death................................109
Figure 34. Association of ethnicity with hospital death .............................110
Figure 35. Association of marital status with hospital death ......................111
Figure 36. Association of education with hospital death ............................112
Figure 37. Association of preference for place of death with hospital death .113
Figure 38. Association of personal income with hospital death ..................113
Figure 39. Association of CHF (including heart disease) with hospital death 114
Figure 40. Association of COPD with hospital death .................................115
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 12
Figure 41. Association of dementia with hospital death ............................115
Figure 42. Association of renal disease with hospital death .......................116
Figure 43. Association of stroke (cerebrovascular disease) with hospital death .
.........................................................................................116
Figure 44. Association of comorbidity with hospital death .........................117
Figure 45. Association of living situation with hospital death .....................117
Figure 46. Association of place of residence with hospital death ................118
Figure 47. Association of carer situation with hospital death .....................118
Figure 48. Association of hospital bed availability with hospital death ........119
Figure 49. Association of health care provision with hospital death ............120
Figure 50. Association of palliative care provision with hospital death ........121
Figure 51. Association of population age with hospital death .....................121
Figure 52. Association of population density with hospital death ................122
Figure 53. Association of population ethnicity with hospital death ..............122
Figure 54. Association of population education with hospital death ............123
Figure 55. Association of population religious membership with hospital death
123
Figure 56. Association of socioeconomic factors with hospital death ...........124
Figure 57. Association of cognitive impairment with hospital death ............125
Figure 58. Association of functional status with hospital death ..................125
Figure 59. Association of prognosis predictor with hospital death ..............126
Figure 60. Association of illness trajectory with hospital death ..................126
Figure 61. Association of geographical region with hospital death ..............127
Figure 62. Association of age with hospital death ....................................129
Figure 63. Association of albumin level with hospital death .......................129
Figure 64. Association of sepsis with hospital death .................................130
Figure 65. Association of comorbidity with hospital death .........................130
Figure 66. Association of intubation with hospital death ...........................131
Figure 67. Association of post intubation PH with hospital death................131
Figure 68. Association of prognosis with hospital death ............................132
Figure 69. Association of trajectory with hospital death ............................133
Figure 70. Association of ventilation with hospital death ...........................133
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 13
Figure 71. Association of age with inpatient hospice death .......................135
Figure 72. Association of gender with inpatient hospice death ...................135
Figure 73. Association of ethnicity with inpatient hospice death ................136
Figure 74. Association of diagnosis with inpatient hospice death ...............136
Figure 75. Association of symptoms (pain) with inpatient hospice death.....137
Figure 76. Association of living situation with inpatient hospice death ........137
Figure 77. Association of age with nursing home death ............................141
Figure 78. Association of gender with nursing home death .......................142
Figure 79. Association of ethnicity with nursing home death .....................142
Figure 80. Association of marital status with nursing home death ..............143
Figure 81. Association of education with nursing home death ...................143
Figure 82. Association of dementia with nursing home death ....................144
Figure 83. Association of stroke with nursing home death ........................144
Figure 84. Association of CHF with nursing home death ...........................145
Figure 85. Association of COPD with nursing home death .........................145
Figure 86. Association of ESKD with nursing home death .........................146
Figure 87. Association of co-morbidity with nursing home death ...............146
Figure 88. Association of functional limitations with nursing home death ....147
Figure 89. Association of trajectory with nursing home death ...................147
Figure 90. Association of living situation with nursing home death .............148
Figure 91. Association of PC provision with nursing home death ................148
Figure 92. Association of socioeconomic status with nursing home death ...149
Figure 93. Association of age with nursing home death in those who die with
dementia ≥ 65 years ...............................................................................151
Figure 94. Association of gender with nursing home death in those who die
with dementia ≥ 65 years ........................................................................151
Figure 95. Association of country of residence with nursing home death in
those who die with dementia ≥ 65 years ....................................................152
Figure 96. Association of bed availability with nursing home death in those
who die with dementia ≥ 65 years .............................................................152
Figure 97. Association of age with nursing home death in the nursing home
population .........................................................................................154
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 14
Figure 98. Association of gender with nursing home death in the nursing home
population .........................................................................................154
Figure 99. Association of marital status with nursing home death in the
nursing home population ..........................................................................155
Figure 100. Association of individual care preferences with nursing home death
in the nursing home population .................................................................155
Figure 101. Association of acceptance by home of resident’s choice with nursing
home death in the nursing home population ...............................................156
Figure 102. Association of diagnosis with nursing home death in the nursing
home population .....................................................................................156
Figure 103. Association of cognitive status with nursing home death in the
nursing home population ..........................................................................157
Figure 104. Association of functional status with nursing home death in the
nursing home population ..........................................................................157
Figure 105. Association of hospitalisation with nursing home death in the
nursing home population ..........................................................................158
Figure 106. Association of length of nursing home stay with nursing home
death in the nursing home population ........................................................158
Figure 107. Association of healthcare provision with nursing home death in the
nursing home population ..........................................................................159
Figure 108. Association of number of nursing home beds with nursing home
death in the nursing home population ........................................................159
Figure 109. Association of nursing home transfer policy with nursing home
death in the nursing home population ........................................................160
Figure 110. Association between age and transition into hospital ................190
Figure 111. Association between functional status and transition into hospital ...
.........................................................................................191
Figure 112. Association between age and transition into nursing home ........192
Figure 113. Association between dementia or cognitive impairment and
transition into nursing home .....................................................................193
Figure 114. Association between pre-admission living situation and transition
into nursing home ...................................................................................194
Figure 115. Association between length of hospital stay and transition from
hospital into nursing home .......................................................................195
Figure 116. Association between pre-admission social support and transition
from hospital into nursing home ................................................................196
Figure 117. Association between age and transition back to home ..............197
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 15
Figure 118. Association between age and transition into palliative care
programmes .........................................................................................198
Figure 119. Association between ethnicity and transition into palliative care
programmes .........................................................................................199
Figure 120. Model of the factors affecting place of death in non malignant
conditions (high strength evidence only) ....................................................204
Figure 121. Model of the factors affecting place of death in non malignant
conditions (high & moderate strength evidence) ..........................................205
Figure 122. Factors with high strength evidence of association with home death
in non malignant conditions ......................................................................206
Figure 123. Factors with high strength evidence of association with nursing
home death in non malignant conditions ....................................................207
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 16
Glossary of terms/abbreviations
Abbreviations: chronic heart failure - CHF
chronic obstructive pulmonary disease - COPD
dementia – not abbreviated
end-stage kidney disease - ESKD
motor neurone disease – MND
multiple sclerosis – MS
Parkinson’s disease – PD
multiple systems atrophy – MSA
progressive supranuclear palsy – PSP
stroke – not abbreviated
long term neurological condition - LTNC
Advanced disease: when treatments and other interventions are predominantly
palliative and/or focused on improving quality of life rather than
curing the condition.
Advanced disease is defined as Class III & IV in the New York Heart
Association classification1 for CHF, Stage III & IV disease for COPD2,
stage 5 disease GFR <15 mL/min/1.73 m2 or needing renal
replacement therapy (for ESKD) (1). For dementia, stroke, or
progressive degenerative neurological conditions, the literature does
not provide a consistent definition of disease stage.
Non malignant condition:
condition other than cancer, in particular for the purposes of this
review chronic heart failure (CHF), chronic obstructive pulmonary
disease (COPD), dementia, end-stage kidney disease (ESKD),
stroke, and long-term neurological conditions (as defined for the
National Service Framework for Long Term Conditions (2): motor
neurone disease (MND), multiple sclerosis (MS), Parkinson’s disease
(PD), multiple systems atrophy (MSA), and progressive
supranuclear palsy (PSP)
1 NYHA. The Stages of Heart Failure - NYHA classification. 2002. Available at: http://www.abouthf.org/questions_stages.htm
Accessed on 14th Nov 2010. 2 Global Initiative for Chronic Obstructive Lung Disease (GOLD). Global Strategy for the Diagnosis, Management, and
Prevention of COPD. 2007. Available at: http://www.goldcopd.com/GuidelinesResources. Accessed on 14th Nov 2010.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 17
Transition in care:
a change in setting or place of care, a change in focus of care, or a
distinct change/movement in the patient’s emotional or
psychological journey. (This definition was agreed through expert
consultation).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 18
Acknowledgements
We want to acknowledge the work of all the people involved in this
systematic literature review, and their hard work in pulling together very
disparate evidence. We also want to thank Mrs Penny Hansford, the
Management Fellow on this project.
We would like to thank the funders of this research, NIHR Service Delivery
and Organisation Programme, for making this research possible.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 19
Executive Summary
Background
Healthcare towards the end of life places major resource burdens on the
NHS, with between 10-20% of all healthcare expenditure spent on care in
the last year of life. As well as cost, the quality of end of life care is a
growing priority for the public and for patients and their families, and this
includes being cared for and dying in the place of their choice.
Most cancer patients (50-70%) prefer a home death, but this is not the
reality they experience; in the UK, 59% of all deaths occur in hospitals, a
further 17% in care homes, and only 18% at home, with marked regional
variations.
For those with conditions other than cancer, the proportions dying at home
differ markedly according to condition, with only 12% of deaths from
respiratory or neurological causes occurring at home, and almost all
dementia deaths occurring in care homes (55%) or hospital (39%).
High quality end of life care needs to occur in the preferred setting, and be
supported by appropriate resources for that setting. The factors and
preferences which influence place of death in cancer have been described,
but much less is understood about these factors and preferences in non
malignant conditions. This systematic literature review synthesizes the
available evidence for the first time.
Aim
For those with advanced non malignant conditions, to identify, critically
appraise, and synthesize the published evidence on:
Preferences regarding place of care and place of death
Factors influencing place of death
Key transitions towards end of life
Methods
Using standard systematic review methods, we identified, reviewed, and
synthesized the national and international evidence on these areas. Both
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 20
qualitative and quantitative evidence was identified, extracted, quality
assessed, and synthesized. The strength of quantitative evidence was
graded high, medium and low quality, and a narrative synthesis of
qualitative evidence was produced to complement the quantitative findings.
Results
A conceptual model of the factors influencing place of care and death in
advanced non malignant conditions has been developed, and in the context
of this model, factors where there is moderate/strong quality and strength
of evidence can be described:
Personal and demographic factors:
Hospital deaths are more likely among minority ethnic groups
Those more likely to live alone (single or widowed) are less likely to
die at home, and more likely to die in care homes, while those who
are married are more likely to experience home death
Disease-related factors:
Higher levels of co-morbidity reduce the chance of home death, and
increase the chance of hospital death.
Illnesses where there is a longer trajectory of functional impairment
(even if severe) are associated with increased home death (possibly
because of the time available for planning and preparation)
Older patients, especially those with dementia, are most likely to die
in care homes
Disease-related factors:
Age: no effect on home death older age▼ hospital death age > 75 ▲NH death
Primary disease: Heart disease (not CHF) ▲
home death CHF no effect on home death COPD ▲ home death dementia ▲ home death dementia ▲ NH death
Associated co-morbidities: co-morbidity ▼home death
Environmental factors
Health and social care input: Service provision: increased hospital bed availability ▼ home
death increased hospital bed availability ▲ hospital
death palliative care provision ▼ hospital death
Social support: Living alone ▼ home
death
Personal and demographic factors: Gender: no effect home or hospital death female ▲NH death
Ethnicity: black ▼ home death non white ▲ hospital death non white ▼ NH death
Education: no effect nursing home death further education ▲ home death further education ▼ hospital death
Marital status: single/widowed ▼hospital death single/widowed ▲ NH death married ▲ home death
partner no effect hospital death divorced ▼hospital death
Socioeconomic status: higher household income ▲home death
Place of care and place of death
Specific symptoms. illness burden and trajectories: Degree of impaired mobility/disability severe functional impairment ▲
home death functional status - no effect on hospital death
Macro-social factors
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 21
Environmental factors:
Increased availability of hospital beds is consistently associated with
reduced likelihood of home death and greater likelihood of hospital
death, although the effect is small
Greater palliative care provision (across conditions) reduces the
chance of hospital death
Although environmental factors, including health and social care input, are
perhaps most amenable to influence, there is relative little evidence as yet
in this area.
The combined quantitative and qualitative evidence also reveals that:
Just under half of patients with advanced non malignant conditions
report a preference for home death (this is notably lower than
among cancer patients).
Among older people, across conditions, preferences for place of care
and death are complex, highly dependent on circumstances, and
may change over time.
Across all conditions, considerations of carer/family burden (as well
as personal considerations) are a major influence on the preferences
of those with advanced disease, and this resonates with evidence on
the factors which precipitate transition into hospital or nursing home
care, and likelihood of death in those places.
For those with chronic heart failure and chronic obstructive
pulmonary disease:
o the lack of a clearly predictable pattern of illness has a marked
impact on awareness of deterioration and subsequent perspectives
about place of care and death
o preferences may not always be consistent with a desire for ‘open’
awareness of death
Those who prefer hospital care may relate hospital to a sense of
safety and/or the perception of better chance of survival.
Those with long term neurological conditions may prefer to plan
ahead, with often profound concern about burden on their family
carers.
Dementia is associated with greater odds of nursing home death,
while co-morbidity is associated with reduced odds of home death.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 22
Other factors associated with increased likelihood of home death in
non malignant conditions include being married, having further
education, and having higher household income.
Living alone, the absence of an informal carer, and cognitive
impairment or dementia all reduce the likelihood of home death,
increase the likelihood of transition into nursing home care, and
increase likelihood of nursing home death.
Conclusions
The evidence on the factors influencing place of death in non malignant
conditions is complex and inter-related. However, clear implications for
practice, policy and research emerge from this synthesis of the evidence.
Practice and policy implications:
1. A preference for home death (while still common among non cancer
patients) is less prevalent than for cancer patients, so attention
should be given to achieving preferred place of care and death in non
malignant conditions, wherever that place is.
2. For those with non-cancer conditions, the presence or absence of a
family or informal carer is a key component in achieving home death.
Effective and sustained carer support, especially in context of longer
illnesses trajectories, is likely to increase home death rates.
3. Minority ethnic groups, and those with lower socio-economic status
achieve lower rates of home death; best practice initiatives need to
target this imbalance, while still taking preferred place into account.
4. For older people, advance care planning is especially important, as
their preferences are complex, highly dependent on circumstances,
and may well evolve over time. The advance care planning they
require therefore needs skilled professionals who can re-visit
preferences and planning sensitively and frequently in response to
change.
5. In addition, older people are more likely to die in care homes,
especially if there is cognitive impairment or dementia. Resources
need to be targeted both at improving care home deaths and
ensuring the opportunity for home death among older people
(especially those who live alone) is adequately supported, when
desired and feasible.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 23
6. For those with end-stage heart or respiratory disease, the conception
of an ‘open’ awareness of approaching death and forward planning of
health care to accommodate decline (derived largely from models of
care in cancer), is not always appropriate. Again, skilled advance care
planning is important, to work across the range of awareness, and to
respond to rapidly changing circumstances.
7. In those conditions with longer trajectories, for instance some long
term neurological conditions, advance planning may be welcomed
and home death can be supported even when there is marked
functional impairment.
8. The increase in the ageing population, with correspondingly higher
levels of co-morbidity, will likely mitigate against home deaths and in
favour of hospital deaths, unless innovative approaches can be
developed to support those with complex co-morbidities in the
community.
9. Increased resources for the delivery of palliative care early in the
illness trajectory across non malignant conditions will likely reduce
hospital deaths.
Research implications:
There are major gaps in the current evidence to inform practice and policy,
and research is particularly needed:
10.To provide insights into how preferences for place of care and death
among those with advanced non malignant conditions evolve over time and with advancing illness, and what factors shape these
preferences.
11.To understand how duration and trajectory of illness affect transitions in place of care, and place of death.
12.To determine how health and social care provision in non cancer conditions can influence place of care and death.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 24
The Report
1 Background
The numbers of UK deaths are predicted to rise by 17% between 2012 and
2030 (3). Planning and providing the health and social services to address
this rise is an urgent priority, and will require either substantial increase in
provision for institutional end of life care, or considerable expansion of
community services to support home deaths, or both. Understanding the
factors which influence home deaths in non malignant conditions will play a
significant part in informing this planning process.
1.1 Major cost implications
Overall, health care towards the end of life places major resource burdens
on the NHS. Between 10-20% of healthcare expenditure is spent on care in
the last year of life (4-6); in the UK, this is currently between eight and 16
billion pounds per annum (£8,000 – 16,000 million) (7). Much of this cost
relates to hospital rather than home care (6, 8). End of life care at home is,
in general, less costly than end of life care in hospital (9), and a recent
National Audit Office report highlights the considerable cost savings which
could be achieved in acute healthcare by reducing hospital admissions and
increasing home and community care in the last year of life (8).
1.2 Increasing public priority
However cost, while important, is only one consideration. The quality of life,
care, and death of patients at the end of life is a growing public priority.
Across Europe and the UK, the public have themselves identified quality of
life as a top priority towards end of life (10). End of life care is now at the
forefront of NHS health policy for England, as highlighted by the End of Life
Care Strategy (11). Attention is focused on improving end of life care for all,
regardless of diagnosis (11), with the aim of delivering high quality care
(12) which is responsive to patient choice (13). In the context of increasing
public awareness and demand for choice, to be cared for and to die at home
(if this is the preferred option) has become a key NHS objective. The End of
Life Care Strategy emphasises choice in place of death, and encourages
commissioners to provide the services to support an increased number of
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 25
home deaths (11). The Marie Curie ‘Delivering Choice’ programme3 has also
recognised this priority, and seeks to enable local service development to
support home deaths. Delivering high quality care in the setting of choice
represents best use of the considerable resources expended on care
towards the end of life.
1.3 Mismatch between preference and reality
A key component of high quality care is the extent to which care is adapted
to the preferences, expectations and values of patients and their families
(14). Understanding preferences for place of care and death, along with
factors influencing place of death, and transitions in place of care is
essential to both improve care and ensure it is of the highest quality.
Reviews of the evidence on preferences, factors, and influences on place of
death for cancer patients have informed the provision and development of
generalist and specialist end of life services (15-17). However, less
attention has been given to preferences, factors, and influences on place of
death for the greater numbers of people dying with non malignant
conditions. It is timely therefore to review this evidence in those with non
malignant disease, especially given the future changes in mortality in an
ageing population, with increasing numbers dying at older ages and with
chronic non malignant conditions (18).
1.4 The evidence in cancer
Among cancer patients and their families, most (50-70%) prefer to die at
home (15, 16), although there is variation across studies, and according to
whether the views of patients, or relatives are sought (19). Despite the
overall high preference for home death (at least among cancer patients),
many do not achieve this preference. Hospital death remains the most
common occurrence in developed countries (18). In the UK in 2005, 59% of
all deaths occurred in NHS hospitals, but only 18% occurred at home, rising
to 35% if care homes are included (20).
3 www.mariecurie.org.uk/en-gb/healthcare-professionals/innovation/Delivering-Choice-Programme/
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 26
1.5 The evidence in non malignant conditions
In the UK, most people die from causes other than cancer, with only 27% of
deaths from cancer (20). While 23% of cancer deaths are at home, the
situation is very different in non malignant conditions. Only 12% of deaths
from respiratory causes and neurological causes occur at home, for
instance, while in dementia most deaths occur in care home settings (55%)
or hospital (39%) (21). The proportion of home deaths is also falling over
time, from 31% in 1974 to 18% by 2005, and this fall is most pronounced
among those with non malignant disease (3, 20). What proportion of those
with non malignant disease prefer home death, and what influences these
preferences, has not been systematically appraised. Factors associated with
home death have been reviewed in cancer, but not in non malignant
conditions. In the latter, symptomatology and dependency may be a more
important influence on place of death than, for example, age or disease
type (22). There may be clear distinctions between preferences for place of
care and place of death (23). Some studies also report somewhat
unexpected detail: for instance, Seymour and colleagues report high
preference for hospital care and death among Chinese people (24), and
Tang et al report considerable discordance between patient and family
carers preference for place of death (25). This evidence needs to be
understood in order to determine how end of life care can best be delivered.
Understanding the current evidence on preferences for place of care and
death, and factors which influence both the preference and the reality of
place of death will contribute to developing best models of care, both
through the synthesis of evidence and through the identification of gaps in
it. Several authors have identified the need for different models of end-of-
life care, in end-stage heart failure (26-28) and advanced chronic
obstructive pulmonary disease (29-31). Evidence reviews on end-of-life care
in dementia raise the challenges of prognostication (and hence achieving
appropriate emphasis on and timing in provision of end of life care), of
achieving good symptom control, and of high quality communication (32),
and again emphasize the need for different models of care (33-35). Some of
the evidence is focused on populations in one particular setting (36), or on
older people (37, 38), rather than on specific disease groups. Overall, the
predominant focus has been on drawing together evidence on symptoms
and other needs in the different non malignant conditions, and considering
how models might differ from that established for cancer patients, rather
than on understanding the current evidence on preferences and factors
influencing place of care and death, and transitions in care.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 27
1.6 Why non malignant conditions are different
There are several reasons why considerations of place of care and death for
those with non malignant conditions may be very different for those with
cancer. First, those with non malignant conditions are often older, especially
for diseases such as dementia or organ failure (20, 22). Second, the
trajectory of illness differs for those with non malignant disease, with, for
instance, a more fluctuant course in organ failure (39), and poorer function
and protracted decline in dementia (32). This may influence treatment and
other preferences (40), affect advance planning (41), and hinder
appropriate transitions in care (42). Third, prognostication is more difficult
for those with non malignant disease, which makes prediction and
anticipation of end of life more challenging (43-45). And fourth,
communication with patient and family about this more uncertain prognosis
may be particularly difficult (32, 46, 47). Models of end of life care in non
malignant disease, where they exist, are largely derived from the cancer
model, with limited evaluation of how suitable they might be in advanced
non malignant disease (48), and generalist end of life care for those with
non malignant conditions remains diffusely conceptualised, with a limited
evidence base and little testing or evaluation of interventions (49, 50).
1.7 Why this systematic literature review is needed
In the face of these considerations, there is some evidence on the
preferences for place of care and death, influences on place of death, and
transitions in care for those with non malignant disease. Older patients (51)
and those dying of certain non malignant diseases (52) are less likely to die
at home. They do experience similar (or even greater) levels of symptoms
and distress as those with cancer (27, 47, 53-56). But despite these
considerable needs, they access fewer services towards the end of life than
cancer patients (57), and their preferences for place of death are often not
met (52). Factors that influence place of death, include illness-related
factors such as illness trajectory (51), and service-related factors, such as
availability of hospital beds (52). But to date this evidence has not been
systematically reviewed, and in particular, comparisons between diseases
and across ages to inform care as patterns of disease and mortality change
have not been made.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 28
1.8 Aims and objectives
In this project, we therefore aim to identify, critically appraise, and
synthesize the evidence on preferences regarding place of care and place of
death, factors influencing place of death, and key transitions towards end of
life, for patients with advanced non malignant conditions.
Specifically, we:
1. Identify the extent to which people with non malignant disease actually
want to die at home, describing the proportion with a home preference,
the quantity, quality, and strength of the evidence, and identifying gaps
in the evidence;
2. Identify what determines preferences for place of care and place of
death among those with non malignant conditions and their families, and
how preferences change according to different influences and over time;
3. Determine which factors are associated with place of care and death
(and hence reasons for variation in place of death) for patients with non
malignant disease, with a focus on the influence of health and social care
resources at home;
4. Identify key transitions in care at the end of life for patients with non
malignant disease, and perceived barriers/facilitators to these
transitions.
In order to concentrate the review, we focused specifically on six non
malignant diseases: chronic heart failure (CHF), chronic obstructive
pulmonary disease (COPD), dementia, end-stage kidney disease (ESKD),
long-term neurological conditions (as defined for the National Service
Framework for Long Term Conditions (2): motor neurone disease (MND),
multiple sclerosis (MS), Parkinson’s disease (PD), multiple systems atrophy
(MSA), and progressive supranuclear palsy (PSP)), and stroke.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 29
2 Theoretical modelling
2.1 A theoretical model in cancer
An explanatory model for the factors influencing the place of care and death
for patients with advanced cancer has already been developed by Gomes
and Higginson (15) (see Figure 1).
Figure 1. Model of the factors affecting place of care and place of death (15)
This incorporates patient, family, and professional perspectives, but is
essentially patient-centred, placing the patient in the heart of the model. It
adopts a bio-psychosocial framework, appropriate for the integrative,
comprehensive and multidimensional study of the individual, as well as a
useful conceptual framework for provision of health services. It is also
appropriate to palliative care, as it comprises the holistic principles of care
based on the whole person. It can be described as a multi-theoretical model
that comprises contributions from an ecological perspective (58) and from
crisis theory applied to coping with physical illness (59, 60). Place of care
and death are analysed taking into account the physical environment and its
relationship to people at individual, interpersonal, organizational and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 30
community levels. Crisis theory enables understanding of serious illness as
a life crisis or disruption which people have to cope with in order to (re)-
establish a balance. Adaptive coping with illness has already been
conceptualised as depending on the characteristics of the patient, aspects of
the illness, and specific features of the physical and social environment (59,
60).
2.2 Developing a theoretical model for non malignant conditions
Bearing these perspectives in mind, the model developed by Gomes and
Higginson has been used to explain place of care and place of death for
cancer patients, based on dynamic interactions between three groups of
factors. We have extended this model, using both theoretical and clinical
perspectives on the potential factors which influence place of care and death
for those with non malignant conditions. This extended model incorporates
contributions from the chronic illness literature (61-64), including
experiential and societal perspectives pertinent to non malignant disease.
In addition, and because of the limited theoretical context for non malignant
end of life care, we conducted a detailed consultation exercise to
incorporate those factors considered most relevant for non malignant
conditions. In this way, the extended model (see Figure 2 overleaf) was
derived from 1) theory, and 2) expert consensus through extensive
consultation with the Project Advisory Group and invited experts in the
different disease areas (see APPENDIX 1 for details of those who
participated in the consultation).
We believe it is important to define and conceptualise this area prior to
reviewing the evidence in order to enable a more objective assessment of
the review findings, and especially to inform our subsequent consideration
of gaps in the evidence.
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 29
Figure 2. Proposed theoretical model of the factors affecting place of care and death in non malignant conditions
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 32
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 33
3 Review questions
Our review questions are:
Among those with non malignant disease and their families:
1. What is the prevalence of a home death preference?
2. How does this preference (and the evidence relating to it) vary
by age, type of disease, stage of disease, and co-morbidity?
3. What determines, shapes, or changes preferences for place of
care and place of death?
4. What factors are associated with actual place of death?
5. How do health and social care resources influence actual place of
death?
6. What are the key transitions in care at the end of life?
7. What influences these transitions?
For each question, we report our findings, with details of the quantity,
quality, and strength of the evidence. We also identify the gaps in the
evidence.
4 Methods
4.1 Identifying the evidence
4.1.1 Sources of evidence
A variety of data sources and procedures as itemised below were used to
identify relevant evidence:
six electronic databases (Medline, Embase, Psychinfo, Cinahl, British
Nursing Index, and ASSIA)
reference lists of relevant primary studies identified in the electronic
search
cited reference searching of all included studies
Figure 3 provides further details of the electronic databases and journals
searched.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 34
Table 1. Sources of data: electronic databases and journals
Electronic Databases
MEDLINE (OVID Medline ® In-Process & Other Non-Indexed
Citations and Ovid Medline ® from 1950 to 30 March 2010)
EMBASE (OVID From 1980 to 6 April 2010)
PSYCINFO (OVID, from 1806 to 30 March 2010)
CINAHL (From 1980 to 7 April 2010)
British Nursing Index and Archive (BNI) (OVID from 1985 to 7 April
2010)
Applied Social Sciences Index and Abstracts (ASSIA) from 1987 to
13 April 2010
4.1.2 Electronic databases
Six electronic databases were selected to ensure a diverse and
comprehensive search, including databases with different traditions and
backgrounds: medical (MEDLINE); biomedical and pharmacological
(EMBASE); psychological (PSYCINFO); nursing and allied health literature
(CINAHL); nursing and midwifery (BNI); and social (ASSIA) literature. The
search strategy was developed and refined in all six databases according to
the terms used in each database, due to differences in subject headings
and/or Boolean characters in each database.
4.1.3 Search terms
The search strategy was developed by deriving terms from the PICOS
(Population, Intervention, Comparator, Outcome, Study Design) approach
from the NIHR Centre for Reviews and Dissemination in York (65) as
adapted to the aims and objectives of this review. Figure 3below provides a
graphical representation of these terms.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 35
Figure 3. Terms used to derive the search strategy (using PICOS)
Extensive scoping searches were then undertaken to develop and refine the
search. We used ‘sentinel’ papers (key papers identified as representing the
area of interest for the review) to test the sensitivity and specificity of our
search strategies. Given the very large number of studies identified in the
preliminary literature scoping, it was decided to use a search strategy with
mid range precision, appropriate to the aims and time constraints of the
review, and to include reference list and cited reference searching for all
included studies. Reference list and cited reference searching proved a
more precise way to capture evidence in what was a very wide ranging
search.
This systematic review focuses on actual and preferred place of death for
patients with non-malignant diseases, as well as preferences and
transitions in relation to this. Therefore the search strategy aims to use
terms within the following areas (terms in bold type correspond to Figure 3:
(1) Non-malignant diseases (six main examples: Chronic Obstructive
Pulmonary Disease (COPD); Chronic Heart Failure (CHF); Dementia;
End-stage Chronic Kidney Disease (ESKD); Long Term Neurological
Conditions (LTNC) and Stroke)
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 36
(2) Advanced disease
(3) Outcomes (place of death; preferences for place of care or
death; transitions in care, including place of care)
(4) Factors and variables influencing the outcomes
(5) Study design (no limitations on design included in the search
strategy)
4.1.4 Outcomes
For the purpose of this systematic review, we considered the outcomes as
the dependent variables. Factors that influenced the outcomes were
considered to be the independent variables. Some variables operated both
as dependent and independent variables – for example, the review was
interested in preferences about place of death as an outcome, but
preference for place of death may also operate as a factor influencing
actual place of death.
A transition is defined as ‘a change in setting or place of care, a change in
focus of care, or a distinct change/movement in the patient’s emotional or
psychological journey’ (see GLOSSARY). This definition was agreed after
expert consultation. This review was interested mostly in transitions of
setting, care, or focus of care i.e. change in focus from active to palliative
care, recognizing that most of the published literature concerns changes in
the setting of care. A note needs to be made of some country differences in
palliative care provision, which can influence the definition of “transitions”
and, in consequence, the search. For instance, in a Canadian definition of
transition (66), transition is defined as: 1) a change in location of where
the patient was cared for; or 2) a change in which service (specialist
groupings, primary care) provided care. We have adopted a wider
definition:
“a change in setting or place of care, a change in focus of care, or a
distinct change / movement in the patient’s emotional or
psychological journey”.
because of feedback during our consultation with experts and clinicians on
the theoretical model that the transition of a patient in terms of their
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 37
emotional acceptance or denial of their illness can be one of the key
determining factors in the outcomes of interest (i.e. place of care or death).
We do not include “preferences” in the search terms for the outcome
“preferences for place of care or death” because our scoping searches
indicated that this did not improve the sensitivity of our searches, while
making them less specific.
4.1.5 Refinement of search strategy
The disease terms were discussed and refined extensively with experts in
the six non malignant conditions, and the final search strategy is available
in APPENDIX 2. The MeSH headings for each illness were presented to
clinical experts in the six non-malignant disease areas, discussed and
refined following discussion. For example, following one expert
recommendation (Prof. Turner-Stokes), we introduced Huntington’s disease
in the search terms for long term neurological conditions. The severe
progression of this illness is typically predictable, which is why it was
considered likely that studies of this condition may be informative about
patient preferences. We also added the term “advance care planning”
following expert consultation.
4.1.6 Reference list and cited reference searching
In addition to the above database searches, the reference lists of all
included papers were searched. Cited reference searching was undertaken
for all included studies, using Scopus and Web of Science.
The reference lists of studies which fulfil the inclusion criteria apart from
containing original data (i.e. relevant systematic reviews, discussion
papers, editorials and commentaries, but not books, letters, guidelines,
authors’ response, case reports or case series, were also searched.
The decision was taken not to hand-search either palliative or disease-
specific journals, given the inclusion of all relevant journals into the
electronic databases, the diversity of conditions and the concentration of
disease-specific evidence into more recent years (and hence greater
likelihood of electronic indexing of this evidence).
4.1.7 Hand searching
We originally intended to include hand-searching of key journals not
indexed in the electronic databases, but initial hand searches did not add
additional evidence, particularly once reference list and cited reference
searching was included, so hand searching was omitted, and our resources
focused on detailed reference list and cited reference searching.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 38
4.2 Inclusion of evidence in the review
This systematic literature review aimed to capture all studies which
reported original data on place of death, preferences for place of
care/death, and transitions for patients with non-malignant conditions,
including transitions in place of care. Of particular interest were studies
which reported factors associated with any of these outcomes, and where
the factors were analysed using multivariate methods.
Studies were included in the systematic review on the basis of the following
criteria:
1. Outcomes
2. Population
3. Study design
These criteria are modified from the PICOS (Population, Intervention,
Comparator, Outcome, Study design) approach laid out in the CRD’s
revised ‘Guidance for Undertaking Reviews in Healthcare’, published by the
NHS Centre for Reviews and Dissemination in Jan 2009 (67).
4.2.1 Study outcomes
Studies were included if they relate to any of these outcomes: Place of death
Preferences for place of care or death
Transitions in care4
4.2.2 Study population
Studies were included if the study population included or related to
participants:
Over 18 years
With non-malignant disease (≥ 50% of participants with non cancer
conditions)
With advanced disease 5
4 Full definition is in GLOSSARY on page 15
5 Full definition is in GLOSSARY on page 15
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 39
If limited information about participants’ diagnosis (or other selection
criteria) was given, the authors of the paper were approached in order to
clarify this. If no further detail was forthcoming, a decision was made as to
whether the study population was likely to include ≥ 50% non cancer
conditions on the basis of setting/context. For instance, studies conducted
within a whole population or representative sample of a whole population
were judged likely to contain ≥ 50% non cancer conditions, as were studies
in nursing home populations; these studies were included. For studies
conducted within specific palliative care services, a judgement was made on
the basis of information about service and setting provided within each
paper as to whether the study population was likely to include ≥ 50% non
cancer conditions or not.
4.2.3 Study design
All studies which reported original data were included. We included studies
regardless of study type or publication status, and (as already mentioned)
approached the authors where necessary for the full study details.
Systematic and non systematic reviews, discussion papers, commentaries
or editorials which related to the population and outcomes of interest, but
which did not contain original data were excluded, but retained for
reference list and cited reference searching.
4.2.4 Country of origin and non English papers
Studies containing original data which met the population, outcome and
study design criteria above were included regardless of the country in which
the research took place. We included both English and non English
language papers, and made every effort to incorporate the non-English
papers into the review through translation by members of the wider team,
and our collaborators (through which we have access to a wide range of
language skills); when this was not possible, details are provided.
4.2.5 Other non malignant conditions
Although the search concentrates on six conditions as exemplars of non-
malignant conditions, we included studies with focus on other non-
malignant illness. However, our search was not targeted to these other
conditions, and conclusions in relation to these conditions are inevitably
constrained.
4.2.6 Cross checks for inclusion of evidence
Three reviewers were involved in the selection process of studies for
inclusion. One reviewer applied the selection criteria to the title and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 40
abstract of each identified study, and categorised into: (a) ‘definitely
included’, (b) ‘definitely excluded’, (c) ‘uncertain’, and (d) ‘excluded but of
interest’ (this category relates to any systematic or non-systematic
reviews, discussion papers, commentaries or editorials, which did not
contain original data). All papers in the categories (a) ‘definitely included’
and (c) ‘uncertain’ were then reviewed by the second and third reviewers
(independent of each other) to assess separately for inclusion/exclusion. In
cases of disagreement, inclusion/exclusion was then discussed between the
three reviewers to reach consensus. In order to assess the reliability of the
inclusion/exclusion decision process, a random sample (10%) of (b)
‘definitely excluded’ were also reviewed for inclusion/exclusion
independently, and the agreement reported.
4.2.7 Categorization of included studies
Included studies were categorized by three criteria: (1) population or
setting, (2) diagnosis, and (3) the outcomes of interest, namely place of
care and place of death, preferences, or transitions.
4.2.8 Exclusions
We excluded studies with no assessment of an outcome for interest for this
review, or with a non-eligible population (e.g. under 18 years of
age/paediatric, > 50% cancer diagnosis, or early stage of disease). We also
excluded studies without original data, such as reviews, discussion papers,
papers reporting duplicate data, editorials, commentaries, case histories,
historical overviews.
4.3 Study quality assessment
For assessing the quality of individual studies, we sought quality
assessment scales which met specific quality criteria relevant for this
review:
4.3.1 Quantitative evidence
Focusing on observational studies (we anticipated few, if any, interventional
studies since observational designs are the most frequently used designs
used to test real world associations), the criteria which we considered
important included: study design, uniformity or heterogeneity in disease
stage, sample and group description, representativeness of sample, explicit
inclusion/exclusion criteria, acceptable response rate, adjustment for
confounders, and reliability and objectivity in outcome assessment. In
reviewing factors, methods to control for confounding, and use of multi-
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 41
variable analysis, if appropriately conducted, was rated more highly than
unadjusted and uni-variate designs. Longitudinal design, especially with
regard to preferences, was more highly rated. Using these criteria we
therefore assessed whether there were appropriate pre-existing quality
scales.
One scale met almost all of our criteria; that developed by Edwards (68)
and subsequently adapted by Higginson (69) (see Figure 4). We made
minor modifications, predominantly around clarifying the language to help
ensure consistency of quality scoring. This scale does not address
uniformity or heterogeneity in disease stage; however, we found that few
studies reported stage of disease, so this was not likely to be a
discriminatory criterion. We also decided to consider multi-variate analysis
within the grading of evidence rather than the quality assessment.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 42
Figure 4. Quality scale used for quantitative evidence (adapted from
Edwards (68, 69))
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 43
4.3.2 Qualitative evidence
For qualitative studies, the quality criteria we sought included: credibility of
findings (transparency, coherence, resonance with other knowledge, and
corroboration / triangulation), rationale and appropriateness of design,
degree of coverage (details of coverage, maximising inclusion, reasons for
non-participation, discussion of access), transparency of reporting,
rationale/rigor in analysis, and context/interpretation. We identified the
scale proposed by Hawker (70) as most appropriate for our review, and
adopted it with only one small modification; the original scale assesses
ethics and bias together (see item 6, in Figure 5), which we divided into two
separate items and assessed individually. Following the original scoring
system, each area was rated on a 4-point scale from 1 (very poor) to 4
(good), where the total minimum score equalled 10, and the total
maximum score, 40.
Figure 5. Quality scale used for qualitative evidence, from Hawker (70)
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 44
4.4 Grading the strength of evidence
For the quantitative evidence, we graded studies as high, medium, or low
quality in the same way as a similar review of evidence on factors affecting
place of death in cancer (see Figure 6).
This grading follows a previous system used on risk factors (71), and the
patient-centred approach of the SORT taxonomy (72). According to this
grading system, studies are high quality if they perform a multi-variate
analysis and have an individual quality score ≥ 70%, or medium quality if
they perform a multi-variate analysis and have a individual quality score <
70%. A minimum of three high quality studies where ≥ 70% of studies
report similar findings constitutes high strength evidence, whereas at least
three medium quality studies with >50% agreement constitutes medium
strength evidence. Given the extent of confounding in unadjusted and uni-
variate analyses, we focused on multi-variate studies as far as possible,
unless there was a complete absence of multi-variate analyses when we
report what these lower levels of evidence might suggest.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 45
Figure 6. Algorithm to assess strength of evidence (15)
With the qualitative evidence, we undertook a narrative synthesis, with
integration of quantitative and qualitative findings at the level of the
discussion. The heterogeneity of qualitative findings made it too challenging
to integrate evidence at the level of the results.
4.5 Data extraction
For each included quantitative study, data was extracted by individual team
members onto an electronic data extraction form. All data extraction was
re-checked by a second member of the team. For reasons of time,
qualitative studies were only extracted once, although the final synthesis
was reviewed by a second researcher (CS) who was not involved in the
original qualitative extractions, and who reflected on consistency and
inclusion of findings.
4.6 Synthesis of evidence
4.6.1 Presentation of quantitative evidence
For each outcome, we report those factors which are shown to have no
association with the outcome of interest, as well as those factors which do
have an association with the outcome of interest. This is because it is also
important to understand which factors have been studied and found not to
be relevant. However, only statistically significant factors with association
with the outcome of interest are taken forward into the final models.
We have excluded factors which are relevant to a specific local, regional, or
national context – for example, payment source of Medicare or Medicaid in
the US in relation to place of death – although we have included
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 46
geographical and ecological factors in order to gain insight into the relative
importance of these, both independently and in relation to other factors.
4.6.2 Questions addressed by the evidence
Findings are summarized descriptively, and conclusions drawn as to the
quantity, quality, and strength of the evidence, for each review question in
turn. We report:
1. What is the prevalence of a home death preference? Data on prevalence
of preference is grouped and mean prevalence weighted by study size and
range is reported.
2. How does this preference (and the evidence relating to it) vary by age,
and diagnosis? Evidence is broken down by age and type of disease where
possible. It was not possible to further analyse according to stage of
disease (early or late stage), or co-morbidity because of limitations in the
evidence. As far as possible from the available evidence, we make
comparisons across the six non malignant diseases.
3. What determines, shapes, or changes preferences for place of care and
place of death? This is addressed predominantly by the qualitative
evidence. Findings on preferences are integrated in order to summarize
the available evidence and also provide new insights from the process of
integration itself, while retaining the character and essence of the
observations from participants in the studies or made by the authors of
the studies.
4. What factors are associated with actual place of death? We group factors
according to the theoretical model (Figure 2) and analyse for the direction
of effect (for and against home death, and for and against hospital death)
and consistency of findings (number of studies reporting the effect out of
total studies included). We also compare direction and consistency of
factors between diseases. We extract the odds ratios (where available)
reported in included studies for those factors supported by sufficiently
strong evidence. We also conduct a sensitivity analysis to determine if a
different grading threshold alters findings.
5. How do health and social care resources influence actual place of death?
We intended to analyse included studies which relate to health and social
care resources, again for direction of effect (home or hospital) and
consistency (number of findings reporting the same effect out of the total
number of studies), but also report the relative impact of different health
and social care resources, using odds ratios where sufficiently strong
evidence exists. However, there was insufficient detail within studies to
make this feasible.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 47
6. What are the key transitions in care at the end of life? We describe the
key transitions identified, and the evidence which supports this, as well as
the qualitative evidence on transitions in care.
In a final step, we draw conclusions about the extent of the evidence and
gaps in it.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 48
5 Results: overview
5.1 Evidence identified
We identified 10,350 articles from the six electronic searches, excluding
duplicates. After screening the title and abstract, we excluded 10,112
records, leaving 238 eligible articles which met our inclusion criteria. We
then conducted reference list and citation searching of these records, and
identified a further 600 studies. Following de-duplication, this generated
559 records, of which 11 were unobtainable; either not available in full text
format (n=5), or were in a language for which we could not obtain a
translation (n=6). Therefore 548 full-text articles were assessed in detail
for eligibility, of which we excluded 258, based on a decision which followed
the protocol for exclusion. A record of each excluded study and the reason
for exclusion is available in APPENDIX 4. Finally, we included 290 studies,
of which 234 used quantitative methods, 55 used qualitative methods, and
1 was mixed (both qualitative and quantitative) methods. Figure 7 overleaf
demonstrates the flow of evidence through the review. A complete list of all
included studies is available in APPENDIX 3.
5.2 Inclusion/exclusion checking
Of the 290 included papers, 11 (3.8%) required consensus discussion to
reach agreement on inclusion/exclusion. Seven of these papers had
originally been identified as ‘uncertain’ for inclusion.
Only two papers were identified as incorrectly excluded in the review of a
10% random sample of excluded papers by a second reviewer.
5.3 Presentation of results
We go on to present the review findings according to each review question.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 49
Figure 7. PRISMA flowchart for identification of evidence
13,736 records identified through database searching
10,350 records after duplicates removed
10,112 records excluded 10,350 records screened
238 records included
234 quantitative studies
600 additional records identified (after title and
abstract screening) through reference and
citation searching of the 238 included studies
Eligibility
Screening
Included
Identification
548 full-text articles assessed for
eligibility
559 records included after
duplicates removed
258 full-text articles excluded, with reasons
11 studies unobtainable (5 studies unobtainable
in full text; 6 studies unable to translate:
Japanese n=4, Icelandic n=1, Korean n=1)
55 qualitative studies
290 included (QUAN and QUAL)
1 mixed method (quantitative and qualitative) study,
included in both the quantitative and qualitative analyses
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 50
6 Results: preferences for place of care
6.1 Evidence identified
Just six of the included papers described preferences for place of care
among people with advanced non malignant conditions. Two of these
papers provided information on preferences for both place of care and place
of death (73, 74). Because of the established differences between patient
and proxy perspectives regarding preferences, and the marked limitations
in using records to capture preferences, we present the evidence according
to the data sources used: either i) data derived directly from patients
themselves, ii) data derived from proxy sources (families or professionals),
and iii) data derived from medical records. As there may be differences in
perceptions and preferences on place of care versus place of death, we
report preferences for place of care here, and address preferences for place
of death subsequently.
6.2 Study characteristics and quality of the evidence
Studies describing preferences for place of care reported i) by patients with
non malignant conditions themselves included 449 patients, ii) by
professional and family caregivers documented information on 327
residents, and iii) by review of medical records provided information on
24,880 patients. (This latter evidence came from medical records of 232
patients, plus secondary analysis from a minimum data set from a further
24,648 patients). Five of the six studies came from the USA and one from
the UK.
i) Data from patients themselves on preferred place of care Three
papers reported patients’ preferences for place of care (73-75) (see Table
2). As data and results from the two studies published by Fried are
identical, we treat the two papers as one study. Among these included
studies, the study design varied; one study collected cross-sectional data
(75, 76) while the other study was a prospective cohort study conducted
over 24 months (74). The study populations and settings also differed; one
study included community dwelling persons with COPD, CHF and
pneumonia two months after hospitalisation (75, 76) while the remaining
study described preferences of patients suffering from COPD, CHF and
cancer (74).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 51
Table 2. Included papers which report preferred place of care using patient data
Paper Country Design Data
collection
Study
setting
Participants Target
age
group
Mean age,
yrs (SD)
(range or
proportions
if mean and
SD not
provided)
Disease
group
N Preferences for
place of care
Quality
score
Factors
Fried 1999
USA cross-sectional
quantitative and qualitative telephone interviews
2 months after hospitalization
community dwelling persons with CHF, COPD, or pneumonia
≥ 65 y 75.6 (6.8)
COPD 26% CHF 47% Pneumonia 37%
246 Home 46% Hospital 54%
50%
6 factors; preference was not associated with age, sex, ethnicity, education, diagnosis, or functional status.
Fried 2000
USA cross-sectional
telephone interviews
2 months after hospitalization
community dwelling persons with CHF, COPD, or pneumonia
≥ 65 y 75.6 (6.8)
COPD 26% CHF 47% Pneumonia 37%
246 Home 46% Hospital 54%
63%
Bi-variate analysis: 6 factors for home treatment: race, education, finances, living with spouse, religiosity, ADL status (all relative risk) White 1.58 (0.80-3.14) College or more 1.42 (1.09-1.86) money left over at end of month 1.20 (0.91-1.58) lives with spouse 1.29 (0.98-1.68) deeply religious 1.32 (1.00-1.74) ≥ 2 ADL dependencies 1.60 (1.18-2.15) also factors on perceptions about treatment at home and in the hospital
Casarett 2006
USA
prosp cohort study (secondary data analysis)
multiple interviews over a period of up to 24 months
ambulatory primary care and specialty clinics, general medicine inpatient units
Ambulatory and inpatient care
≥ 60y 73 (range 60-93)
Cancer 32% COPD 30% CHF 38%
203 Hospice enrolment 30%
75% 20 factors influencing hospice enrolment, bi-variate analysis
TOTAL PARTICIPANTS 449 (Fried 1999 and Fried 2000 relate to the same study)
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 52
Both studies on patient preferences which collected data directly from
patients interviewed the patients directly, Fried over the telephone (75, 76)
and Casarett in consecutive personal interviews (74). In the study by Fried
et al, patients were asked about their last hospital admission and a
hypothetical scenario where they could get similar investigations and
treatment, and their preferences in relation to this (75, 76); 46%
preferred home care, and 54% preferred hospital care. In the second
study by Casarett and colleagues, patients were interviewed to determine
whether patient preferences were a barrier to hospice enrolment (74);
30% enrolled in hospice. Both studies by Fried et al were scored as
moderate quality and the paper by Casarett as high quality (74).
ii) Data about patients preference for place of care as reported by
proxies Just one study reported information from after-death interviews
with family and staff caregivers of nursing home residents on decisions
made about specific life-sustaining treatments (LSTs) among residents in
long-term care (LTC) settings (77) (see Table 3). Detailed information on
patients’ diagnoses was not reported in this study. This paper was of low
quality, however it reported that 32% preferred hospitalisation and
45% preferred not to be hospitalised.
iii) Data about patients preference for place of care reported using
medical records Three studies presented data from chart reviews (78, 79)
or a medical database (80) (see Table 4). One by Anderson reports
advance care planning documents from nursing home residents with end
stage renal disease on peritoneal dialysis (78); 13% preferred nursing
home care (not to be hospitalised). A second study by Triplett analysed
advance directives of dementia patients (79); 84% preferred community
discharge from the nursing home. In the third study, by Arling et al,
reports on the preferences of nursing home residents with mixed diagnoses
were analysed from a minimum data set (80); just 2.5% preferred
hospice care, although this relates only to those with advance directives,
and what was specified in those directives. The study by Anderson (78) was
rated of low quality, the other two of moderate quality.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 53
Table 3. Included papers which report preferred place of care as reported by proxies
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (or proportions if mean not provided)
Disease group
N
Preferences for place of care
Quality score
Factors
Biola 2010
England cross-sectional
after-death interviews
27 nursing homes and 85 residential care/assisted living settings
both family and staff caregivers
not specified
decedents 86.6 (9.9)
not reported
654 (caregivers) 327 LTC residents
Hospitalisation 31.7% No hospitalisation 45.0% Decision depending on situation 23.2%
44%
17 factors (resident, family caregiver, facility) regarding hospitalisation assessed, none associated with decision about hospitalisation; no specific information given on type of analyses
TOTAL PARTICIPANTS 327
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 54
Table 4. Included papers which report preferred place of care reported by chart reviews and secondary analysis
Abbreviations: ESRD = end-stage renal disease, CAPD = continuous ambulatory peritoneal dialysis
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (or proportions if mean not provided)
Disease group
N
Preferences for place of care
Quality score
Factors
Anderson 2006
USA
retro-spective cohort study
chart review
nursing home ESRD on CAPD
not specified
62.7 (12.8) on admission to NH
ESRD patients on CAPD 100%
109 NH 13% (not to be hospitalised)
31%
4 factors for declining hospitalisation (multivariate analysis): poorer ADL scores (4.3, SD 3.7 v 8.8, SD 4.3) older age (69.0, SD 13.8 v 61.6, SD 12.5) more likely to have dementia (6/14 v 26/94)) less likely to have amputations (0/14 v 26/94)
Arling 2010
USA
retro-spective cohort study
secondary data from minimum data set
nursing home nursing home residents
not specified
77.3 (12.7)
Cancer 14% Dementia 16% Depression 24% Diabetes 26% End stage disease 3%
24,648
preference for community discharge (meaning back to home or supported living) 84%
63%
Triplett 2008
USA chart review
chart review
nursing home end-stage dementia
not specified
81.5 (7.1)
Dementia 100%
123 Hospice 2.5% 56%
TOTAL PARTICIPANTS:
24,880
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 52
6.3 Preferences for place of care and factors influencing this
The study published in two papers reporting preferences for place of care
from the patient perspective (see Figure 8) indicated that 46% chose home
as the preferred place of care and 54% preferred hospital (75, 76).
Table 5. Conventions used in the ‘bubble’ plots
Figure 8. Preferred place of care as reported by patients themselves
X axis represents each option for place of care
Y axis represents the proportion with preference for that place of care
The size of the ‘bubble’ represent the size of the study from which this
evidence is drawn
The colour of the bubble is consistent across plots for place of care; for
instance light blue represents a home preference, and red a hospital
preference
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 56
This study also identified six factors associated with preference for home
treatment (see Table 6). White race, higher education, better financial
situation, living with spouse, deep religiosity, and higher activities of daily
living (ADL) dependencies increased patients’ preference for home
treatment. However, this was only confirmed in uni-variate and not multi-
variate analysis, and constitutes weak evidence.
Table 6 groups these factors according to the proposed theoretical model
for factors influencing place of care and death in non malignant conditions
(see ‘Type of factor in Table 6, and also the theoretical model in Figure 2 on
page 28).
In the second study, 30% of patients were enrolled in hospice care over a
time period of 24 months (74). 20 factors associated with preference for
hospice enrolment were assessed. Concern about being kept alive by
machines, low-burden treatment that would increase chance of survival,
impairment of ADL and IADL, and number of moderate/severe symptoms
were all significantly associated with hospice enrolment (see Table 7).
In the study that used proxy data to report preferences, family and staff
caregivers reported that about a third of decedents made a decision that
they want to be hospitalised if indicated and 45% that they didn’t want to
be hospitalised (77). About 23% wanted to decide depending on the
situation.
In studies relying on medical records to report preferences, 13% of patients
with end stage renal disease and peritoneal dialysis in a nursing home
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 57
chose not to be hospitalised (78) and 2.5% of nursing home residents with
dementia (who had made advance directives) chose hospice as their
preference for care (79).
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 54
Table 6. Evidence on factors associated with preference for home treatment as reported by patients
Type of factor Factor Strength of evidence * Consistency Direction
Personal and demographic factors
Ethnicity: White Financial situation: money left over at end of month Education: any College education or more Religiosity: deeply religious
Low Low Low Low
(1/1) (1/1) (1/1) (1/1)
Preference for home treatment ▲ Preference for home treatment ▲ Preference for home treatment ▲ Preference for home treatment ▲
Environmental factors – social support
Living situation: lives with spouse
Low
(1/1)
Preference for home treatment ▲
Symptoms/ illness burden/ trajectory
ADL impairment: ≥ 2 ADL dependencies
Low
(1/1)
Preference for home treatment ▲
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 59
* using Figure 6: Algorithm to assess strength of evidence (15)
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 55
Table 7. Evidence on factors associated with preference for hospice enrolment as
reported by patients
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 61
Note: in the US context that ‘hospice enrolment’ means enrolment onto a program which provides the
kind of care provided under the Medicaid/Medicare hospice benefit. See page 126 for further
discussion of the differences between countries in this respect.
Type of factor Factor Strength of evidence
Consistency Direction
Personal and demographic factors
Age Gender: male Ethnicity: white Education: years
Low Low Low Low
(1/1) (1/1) (1/1) (1/1)
No significant effect No significant effect No significant effect No significant effect
Disease-related factors
Co-morbidity Concerned about being kept alive by machines Informed by physician that likely to die of illness Advance directive Prefers to avoid dying in a hospital Preferences for life-sustaining treatment Low-burden treatment that would increase chance of survival High-burden treatment that would increase chance of survival
Low Low Low Low Low Low Low Low
(1/1) (1/1) (1/1) (1/1) (1/1) (1/1) (1/1) (1/1)
No significant effect ▲ hospice enrolment No significant effect No significant effect No significant effect No significant effect ▲ hospice enrolment No significant effect
Environmental factors – social support
Informal caregivers present in the home
Low
(1/1)
No significant effect
Environmental factors – health and social care input
Care from a VA Medical Center Knowledge of hospice as an option for care
Low Low
(1/1) (1/1)
No significant effect No significant effect
Environmental factors – macrosocial factors
Financial situation at end of month
low
(1/1)
No significant effect
Symptoms/ illness burden/ trajectory
ADL impairment IADL impairment Global quality of life Number of moderate/severe symptoms
Low Low Low Low
(1/1) (1/1) (1/1) (1/1)
▲ hospice enrolment ▲ hospice enrolment No significant effect ▲ hospice enrolment
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 62
7 Results: preferences for place of death
7.1 Evidence identified
Seventeen included studies reported preferences for place of death, and
factors influencing these preferences. In total, this represents the
preferences of 9,495 participants, across eight countries. Studies which use
data derived direct from patients themselves included 1,639 patients, while
reports from families represent a further 1,350 participants and reports
from professionals another 5,929 participants. Two additional studies
represented a further 577 patients using medical records and databases to
identify reported preferences. Again, because of the established differences
between patient and proxy perspectives, and the marked limitations in
using records to capture preferences, we present the evidence on
preferences for place of death according to the data sources used: either i)
data derived directly from patients themselves, ii) data derived from proxy
sources (families or professionals), and iii) data derived from medical
records.
7.2 Study characteristics and quality of the evidence
i) Data from patients themselves on preferred place of death
The seven studies describing preferences for place of death in non-cancer
conditions reported by patients come from the following countries: USA (3
studies), Canada (2 studies), Spain and Saudi Arabia (1 study each). Full
details are presented in Table 8 overleaf. Most studies had a cross-sectional
design, although two studies differed: five studies reported cross-sectional
data (73, 81-84), one was a prospective cohort study (74), and one study
tested an intervention in a randomised controlled trial (85). The populations
and settings also varied. In five studies, participants had a mix of different
diagnoses, mainly COPD, CHF and cancer (<50%) (73, 74, 84-86).
Overall, the seven studies report prevalence of preference for home
death between 36-77%, which represents the preferences of 1639
participants. Mean weighted prevalence6 of home death preference
from these studies is 42.2%.
One study could not be included in this weighted calculation, since it did not
report actual preference. Home death preference varies markedly across the
studies, although the different studies included varying disease populations;
6 Weighted by size of study.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 63
two studies reported preferences of patients with end-stage renal disease;
in one of these studies patients had a range of ESKD, including dialysis and
pre-dialysis (82), and in the second study, patients were all treated with
haemodialysis (81). A further study interviewed only patients with
congestive heart failure (83) and the remaining study did not report the
diagnoses of participants (87). Comparisons across conditions will be made
later. Data was collected in hospitals, outpatient clinics, home care and
nursing home settings. Of the seven studies collecting data directly from
patients, five interviewed patients (73, 74, 83-85), and two used
questionnaires (81, 82). Quality scores varied from 50% and 77%. Three
studies were scored low with 50%, one study with 56% and 63% each and
two studies high; 75% and 77%, respectively.
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 58
Table 8. Included papers which report preferences for place of death in advanced non malignant conditions reported by
patients
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 65
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
Preferences for place of death
Quality score
Comments
Al-Jahdali 2009
Saudi Arabia
cross-sectional
questionnaire survey
Two tertiary hospitals
Haemodialysis patients
all ages
51.1 (15.5)
ESKD 100% 100 Home 73* - 77%** Hospital 70%
56%
*Wishes if medical condition deteriorates to impending death) ** Wishes if medical condition deteriorates to impending death & if hospitalisation and medical intervention would not improve condition
Casarett 2006
USA
Pro-spective cohort study
multiple interviews over a period of up to 24 months
ambulatory primary care and specialty clinics, general medicine inpatient units
Cancer, COPD, CHF
≥ 60 73 (range 60-93)
Cancer 32% COPD 30% CHF 38%
203 Home 71.4% 75%
Casarett 2005
USA RCT
Telephone interviews, medical records
nursing home
Nursing home residents and their surrogate decision makers
all ages
84 (range 54-102)
Cancer 4% Dementia 62% CHFC 22% Chronic lung disease 17%
205
Resident died in preferred place of death: all 47%, control group 33% (2 pts), intervention group 54% (6 pts)
77%
Intervention: structured interview identified residents whose goals for care, treatment preferences, and palliative care needs made them appropriate for hospice care. These residents’ physicians were notified and asked to authorize a hospice informational visit.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 66
Davison 2010
Canada cross-sectional
self-administered survey in clinic or at home
dialysis, transplantation, or predialysis clinics in a Canadian, university-based renal program
stage 4 and stage 5 CKD
≥ 18 68.2 (14.4)
stage 4 and stage 5 CKD 100%
584
Home 36.1% Hospital 27.4% Hospice 28.8 %
63%
Formiga 2004
Spain cross-sectional
interviews hospital
elderly patients hospitalized for heart failure
> 64 79 (8.7) decompensated heart failure 100%
80 Home 50% Hospital 40%
50%
Fried 1999
USA cross-sectional
telephone interviews
2 months
after hospitalization
community dwelling
persons with CHF, COPD, or pneumonia
≥ 65 75.6 (6.8)
COPD 26%
CHF 47% Pneumonia 37%
246
Home 43.1% Hospital 48.0% Do not know 8.9%
50%
Stajduhar 2008
Canada cross-sectional
face-to-face interview with patients and family care givers
tertiary hospital
seriously ill > 18
median 72 (range 54–95)
Cancer 40.5% CHF 31.2% COPD 22.5% Cirrhosis 5.8%
138
Home 53.6% Hospital 27.5% Other 18.8%
50%
TOTAL PARTICIPANTS:
1639
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 60
ii) Data about preferences for place of death as reported by proxies
Four studies on preferences reported by families were identified, and come
from USA (3 studies) and Spain (1 study) (see Table 9 for details). Two of
these report cross-sectional data (88, 89) and two are retrospective cohort
studies (90, 91). All studies were post-bereavement studies; none included
family carers of living patients. Two studies included relatives of mixed
populations (89, 90), one study the relatives of renal patients (88) and one
did not report the diagnoses of the deceased (91). In three studies (89-91),
family caregivers were interviewed, of which one study also used data from
death certificates (89). One study used postal questionnaires to receive
relatives’ views on patients’ preferences (88). The quality scores for these
studies were variable; one study scored very low with 25%, while other
studies attained moderate scores of 50%, 56% and 63%. Families
reported prevalence of preference for home death which varied
from 25-97%, which represents the reported preferences of 1264
participants. Mean weighted prevalence7 of home death preference,
as reported by families, was 65.5%.
The four studies on preferences for place of death in non-cancer conditions
reported by professionals come from Belgium (2 studies), the Netherlands
and Japan (1 study each). Details are presented in Table 10. Three studies
are retrospective cohort studies (92-94) and one is a cross-sectional study
(95). They present data from mixed patient populations (92-94) and again,
one study did not report any diagnoses (95). Three studies report data from
the Sentinel Networks of General Practitioners in Belgium (93, 94) and the
Netherlands (92). One study is distinct in reporting data collected from
nursing homes about the last five discharged residents who died (95). The
design of studies varied; three were mortality-follow back studies using
structured standardized questionnaires about registered non-sudden and
expected deaths in GP practices (92-94) and one collected data from
nursing homes (95). Study quality ranked between 56% and 69% with two
studies scoring 63%.
Professionals report a prevalence of patient preferences for home
death which varies from 30-69%, which represents the preferences of
5929 participants. Mean weighted prevalence8 of home death
preference from these studies is 47.8%.
7 Weighted by size of study.
8 Weighted by size of study.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 68
iii) Data about preferences for place of death as derived from
medical records
Just two studies used records to obtain information about preferences for
place of death. Both were retrospective cohort studies based in the US, and
one retrieved data from medical records (86) while the other used a medical
database (87). They report a prevalence of home death preference for
68% and 81.6% respectively; from the records of 577 patients.
Weighted mean prevalence of home death preference is 79.3%.
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 62
Table 9. Included papers which report preferences for place of death as reported by families
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
Preferences for place of death
Quality score
Comments
Cohen 2005
USA cross-sectional
postal question-aires
nephrology clinics
Families of patients who died after receiving dialysis*
all ages
not mentioned End stage renal disease
86
35.3% died in preferred place of death, 37.6% did not die in preferred place of death
25% *spouse 49.4%, adult child 17.6%, sibling 9.4%
Curtis 2002
USA
retro-spective cohort study
interviews for question-aire validation
community-based research and community engagement organization whose mission is to improve quality at life’s end
family members of patients who died
all ages
74.9 (15.52)
Cancer ≈26.6% COPD ≈6% Heart disease ≈28.6% Dementia ≈7% cerebrovascular disease 7.5% kidney disease 4% pneumonia 2.8% neurologic 2.8% chronic liver disease 3% other 27%
205 Home ≈25.3% 50%
validation study of the QODD tool; diagnoses given for 252 decedents but demographics for 205 as further information on them
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 70
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
Preferences for place of death
Quality score
Comments
Ramon 2006
Spain
retro-spective cohort study
interviews 6 areas in Catalonia/Spain
caregivers ≥ 65 81.4 (8) deceased
not reported 584 Home 97.3% other2.7%
63%
Tolle 2000
USA cross-sectional
death certificates and telephone interview
2-5 months after decedents death
family respondents
> 18
Family respondents 60 y Decedents 77 y
Cancer 24% Heart disease 23% Cerebrovascular 11% Pneumonia/influenza 6% COPD, Alzheimer etc. 36%
475
Home 43.7% Hospital 14% NH 19.7%
56%
TOTAL PARTICIPANTS:
1,350
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 71
Table 10. Included papers which report preferences for place of death as reported by professionals
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
Preferences for place of death
Quality score
Comments
Abarshi 2009
NL
retro-spective cohort study
mortality follow-back study with structured, standardized questionnaires for each death case
Dutch Sentinel Network of GPs
GPs asked to fill in short form on the care the deceased received in the last 3 months of life
>1 year
> 80 years: 53%, 41-80 years 44%
43% cancer 29% cardiovascular
637
Home 69% Hospital 2% Hospice 10% Care home 19%
56%
Factors associated with GPs awareness of place of death; logistic regression
Meeussen 2009
Belgium
retro-spective cohort study
mortality follow-back study with structured, standardized questionnaires for each death case
Belgian Sentinel Network of GPs
GPs; deceased patients
≥ 1 y
1-64 y: 12.5%; 65-79 y: 34.6; ≥ 85y: 53%
Cancer 45.8% ‘Non-cancer’ 54.2%
798
Home 57.6% Hospital 4.7% NH 30.9% Pall Care Unit 6.6%
63%
study examines factors associated with GPs awareness of patients preferred place of death; information on preferred place of death from 363 patients; only paper with multivariate analysis
Shinoda-Tagawa 2005
Japan cross-sectional
questionnaire
2,000/4,886 randomly selected NH; licensed nurses collected data for the last five recently discharged residents who had died
nurses reporting about NH residents
not specified
not reported
not reported 4175 NH 44.1%
63%
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 72
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
Preferences for place of death
Quality score
Comments
Van den Block 2007
Belgium
retro-spective cohort study
mortality follow-back study with structured, standardized questionnaires for each death case
Belgian Sentinel Network of GPs, all registered non-sudden or expected deaths in GP practice
GPs; deceased patients
≥ 1 y
1-64y 13%, 65-74 y 23%, 75-84y 34%, +85y 29%
Cancer 39% Cardiovascular disease 26% Not specified 30%
319 Home 30.1% Hospital 5.3% NH 11.6%
69%
TOTAL PARTICIPANTS:
5,929
National Institute for Health Research
Service Delivery and Organisation Programme
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 66
Table 11. Included papers which report preferences for place of death as reported from records
Paper
Country
Design
Data collection
Study setting or context
Participants
Age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
Preferences for place of death
Quality score
Comments
Groth-Juncker 1983
USA
retro-spective cohort study
medical records
comparison of home and institutional deaths
elderly home-bound patients
all ages
not reported Cancer 41% ‘Non-cancer’ 57%
98 Home 68% 50%
Pritchard 1998
USA
retro-spective cohort study
administrative database
Pts enrolled in observational phase of SUPPORT, who survived initial hospitalization & subsequently died during 6 month follow up period
patients enrolled in observational phase of SUPPORT
≥ 65 y
not reported not reported 479 Home 81.6% Hospital 19.4%
50%
TOTAL PARTICIPANTS
577
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 67
7.3 Preferences for place of death and factors associated with this
One of the main areas of interest for this review was comparison between
different non malignant conditions. For this reason, we present findings for
preference for place of death by disease group. We go on to describe factors
associated with preference for place of death.
Table 12. Conventions used in the ‘bubble’ plots
There was heterogeneity with regard to options offered when studying
preferences for place of death. In studies reporting preferences from
patients, options were mainly home, and sometimes also hospital. One
Canadian study also named hospice as a preference option (82) and one
study named ‘other’ as a category without further specification (84). The
only intervention study did not state the specific place but showed that
patients in the intervention group died more often within the preferred place
of death (85). However, this difference was not significant probably because
the numbers who actually died in the preferred place were small.
Preferences in patients with renal disease: (see Figure 9). In Al-
Jahdali’s study, patients on haemodialysis were asked in different scenarios
about their preferences on care at the end of life (81). In the case of
deterioration to impending death, 73% preferred to die at home. This
number increased to 84% if patients were receiving medication at home
and to 87% if they also were followed up at home.
Asked if hospitalisation and medical intervention would not improve their
condition, 77% would choose home as preferred place of death. However,
70% answered in the same question that they would prefer to be
transferred to hospital. In the second study with patients with a range of
end stage renal disease, home was the first choice in 36.1%, and hospital
X axis represents each option for place of care
Y axis represents the proportion with preference for that place of care
The size of the ‘bubble’ represent the size of the study from which this
evidence is drawn.
The colour of the bubble is consistent across plots for place of care; for
instance light blue represents a home preference, and red a hospital
preference
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 75
and hospice were similarly distributed around 27-28% (82). Using this
evidence, the weighted mean prevalence of home death preference
among ESKD patients is 41.8%.
Figure 9. Preferred place of death as reported by renal patients
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 69
Figure 10. Preferred place of death as reported by patients with CHF
Preferences in patients with chronic heart failure: (see Figure 10)
Only one study included patients with heart failure. In this study, 50%
chose home and 40% chose hospital as the preferred place of death
(83).
Preferences in patients with mixed non malignant conditions: (see
Figure 11). Five studies reported views of patients from mixed patients
groups (73, 74, 84-86) and one study did not report any specific disease
but stated that data were from patients from the SUPPORT study (87).
Home was the preferred place of death in about 70% of patients in a
prospective study with cancer, COPD and CHF patients (74) and also in a
retrospective study in elderly patients with cancer and non-cancer
diagnoses (86). It should be noted that both studies did not give any
alternative locations for preferred place of death. In one further study with
mixed patient groups, home was the top preferred place of death in 43%
(73). Patients were interviewed two months after discharge from hospital.
In a study with patients interviewed during their hospital stay 53% chose
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 77
home as the preferred place of death (84). This study also named a
category ‘place of death does not matter’ as third option, with almost 20%
of patients stating this. The fifth study reporting preferences from patients
with mixed diagnoses did not state were patients preferred to die but only
whether they died in their preferred location (85). Finally, in the study
presenting data from the SUPPORT study without specifying diseases, about
80% of patients chose home as preferred place of death and the remaining
20% chose hospital (87).
Figure 11. Preferred place of death as reported by patients (mixed
diagnoses)
Preferences for place of death as reported by families The only study
reporting views of families of patients who died from renal disease only
described that about a third of patients died in the preferred place of death
and about 38% did not die in preferred place of death (88). However, place
of death was not specified therefore we could not include it in the synthesis
of findings. Families of deceased patients with a variety of non-malignant
diagnoses reported that patients wanted to die at home in 25 – 44% (89,
90). In a retrospective cohort study from Spain, caregivers reported that
97% of patients wanted to die at home (91). This study did not give any
details of patients’ diagnoses, but deceased patients were elderly with an
average age of 81 years.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 78
Figure 12. Preferred place of death as reported by families (mixed
diagnoses)
Preferences for place of death as reported by professionals (see
Figure 13) The data coming from general practitioners in Belgium and
Netherlands indicated home as preferred place of death in between 30 -
69% of patients, hospital as preferred place in 2 – 5 % of patients, and
nursing/care home as preferences for place of death in 12 – 31% patients.
In the Japanese paper only nursing home was named as preferred place of
death, in 44% of patients.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 79
Figure 13. Preferred place of death as reported by professionals
(mixed diagnoses)
Factors influencing preferences for place of death The main interest of
this review is factors influencing place of death and also preferences for
place of death. None of the identified studies described above reported any
factors that influenced patients’ preferences. The nearest to this question
were data from two studies which evaluated factors that were associated
with general practitioners awareness for patients’ preferences for place of
death (92, 93). Both papers reported data from general practitioner (GP)
sentinel networks in Belgium and The Netherlands. These sentinel networks
are pre-existing nationwide health surveillance instruments (92). In both
countries, participating GPs were asked to fill in a short registration form
within one week of reporting a patient’s death on the care the deceased
received in the last three months of life (92, 93). This included questions
about patients’ preferences for place of death. Neither paper stated the
location of patients` preferences, such as home or hospital but focussed on
factors affecting knowledge of the preferences (whatever it was). The two
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 80
papers reported ten different factors associated with GPs awareness of
patients’ preferences (see Table 13). There was low strength evidence that
all ten factors were associated with GPs awareness of patients’ preferences.
The factors are grouped in relation to patients’ characteristics (financial
status), to goal of care/treatment domain, place of care, and service
provision/ use (informal care, GP contact, specialist palliative care support),
following the theoretical model (see Figure 2).
Table 13. Evidence on factors associated with GPs awareness of
patients’ preferred place of death
Type of factor
Factor
Strength of evidence
Consistency
Direction
Disease-related factors
Overall treatment goal (reference curative) life prolonging Palliative Main care domain (reference physical) Psychosocial Spiritual
Low Low
(1/1) (1/1)
▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death
Environmental factors – social support
Location of care other than home place of care 7 days before death (reference hospital) home care home hospice some/often informal care over last 3 months
Low Low Low
(1/1) (1/1) (1/1)
▼ GP awareness for preferred place of death ▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death
Environmental factors – health and social care input
Not hospitalised in last 3 months of life >7 GP contacts over last 3 months of life No use of specialized palliative care services
Specialist palliative care initiative delivered over last 3 months of life
Low Low Low Low
(1/1) (1/1) (1/1) (1/1)
▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death ▼ GP awareness for preferred place of death ▲ GP awareness for preferred place of death
Environmental factors – macrosocial factors
Financial status Average Above average
Low
(1/1)
▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 81
8 Results: evidence on place of death
8.1 Identified evidence
Of the 287 included papers, 110 papers reported quantitative evidence on
place of death. Of these 110, there were 35 papers (representing 34
studies) which undertook multivariate analysis on factors affecting place of
death (52, 91, 95-127) and we have focused our review on these papers
which undertake this more rigorous multi-variate analysis. Table 14
provides full details of these studies.
These studies of place of death in non malignant conditions come from the
following countries: USA (17 studies) , Japan (6 studies), Canada (3
studies), Australia (2 studies), Belgium (2 studies), the Netherlands (2
studies), England (1 study), Scotland (1 study), Wales (1 study), Singapore
(1 study), China (1 study), and Spain (1 study). One study undertook a
cross-country comparison between five countries (33).
Together, these studies reviewed over 6 million participants (specifically,
6,296,516 people, excluding the two ecological studies).
8.2 Quality of evidence
Since we analysed only studies which conducted multi-variate analyses, all
studies were either medium or high quality - 27 studies were high quality,
and 8 were of medium quality (see Figure 4 for details of how the quality
scores were derived). No study scored less than 50%, apart from one of the
ecological studies, for whom the generic quality score was not well suited.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 75
Table 14. Included papers reporting multivariate analyses on factors affecting place of death
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (or proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Afessa 2002
USA prospective cohort study
hospital admissions and outcome data
acute hospital COPD patients with ICU admission
all ages 63.1 (8.9) COPD 180 4 logistic regression
79% high
Ai-Ping 2005
Singapore retrospective cohort study
medical records
acute hospital COPD patients with ICU admission
all ages 70.0 (8.3) COPD 57 6 logistic regression
79% high
Babazono 1998
Japan analysis of routinely collected data
death registration data
regional whole regional population
70+ not given all diagnoses: cancer 18.7% cardiovascular disease 26.8% cerebrovascular disease 21.8% respiratory disease 17.1%
455 7 logistic regression
71% high
Bell 2009
USA prospective cohort study
longitudinal participant data
regional Japanese American men
all ages 84.4 all diagnoses: cancer 24.9% coronary heart disease 13.4% stroke 11.1% dementia 8.1% respiratory 4.0% other 38.6%
1,352 7 logistic regression
64% med
Clifford 1991 Australia analysis of routinely collected data
death registration data
regional whole regional population
all ages <65 26.3% 65-74 21.0% 75-84 32.3% 85+ 20.4%
all diagnoses: cancer 25.3% circulatory 45.1% respiratory 8.9% other 20.7%
7,697 5 logistic regression
64% medium
Cohen 2006
Belgium analysis of routinely collected data
death registrations and linked health care databases
regional whole population 1+ <65 17.0% 65-80 34.8% >80 48.1%
all diagnoses: cancer 26.9% heart failure 5.2% respiratory diseases 11.5% cerebrovascular diseases 9.0% nervous system diseases 2.3%
55,759 8 logistic regression
79% high
Enguidanos 2005
USA secondary analyses of routinely collected provider data
linked Medi-Cal and Medicare databases
Medi-Cal and Medicare
Medicare population
65+ 65-74 18.7% 75-84 35.4% ≥ 85 45.9%
all diagnoses: cancer 13.6% non cancer 86.4%
38,519 15 logistic regression
71% high
Fried 1999
USA prospective cohort study
longitudinal participant data
long term home care program
patients in long term home care program
65+ 80 (8.3) all diagnoses: proportions not given
620 7 logistic regression
71% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 83
Gruneir 2007 USA analysis of routinely collected data
death registrations and linked health care databases
national population dataset
whole population, except pregnancy-related, homicide/suicide, or accidental deaths
15+ <65 20.9% 65-74 21.2% 75-84 30.8% 85-94 22.8% 95+ 4.2%
all diagnoses: cancer 25.1% heart disease 30.5% cerebrovascular accident 7.4% COPD 5.0% CHF 3.6% Alzheimer's disease 1.0%
1,402,167 17 multi-level model using generalised estimating equations
79% high
Hansen 2002 USA retrospective case control study
medical records
national population dataset
whole population all ages ≥ 65 75%, <25 3.2%
all diagnoses: cancer 23% heart disease 31% stroke 7% COPD 5%
2,317,586 11 logistic regression
71% high
Haydar 2004 USA analysis of routine data
death registration data
home-based primary care service
CHF/dementia deaths in home based primary care service
65+ 87.5 (8.5) CHF and/or dementia: CHF 16.9% dementia 45.9% both 19.8%
172 6 logistic regression
57% med
Houttekier 2010
Belgium England Nether-lands Scotland Wales
analysis of routinely collected data
death registration data
national population datasets
all deaths from dementia
65+ 65-74 5.9% 75-84 34.4% ≥ 85 59.7%
dementia 30,281 4 logistic regression
79% high
Johnson 2005 USA secondary analyses of routinely collected provider data
national hospice provider data
national provider of hospice services (inpatient and community-based)
patients receiving inpatient and community-based hospice care
65+ overall age not provided
all diagnoses: cancer 35.5%
115,854 12 logistic regression
71% high
Klinkenberg 2005
Nether-lands
retrospective cohort study
proxy data from family members interviewed after the death
regional longitudinal study of older people
regional longitudinal study on ageing
55+ < 80 37.8% ≥ 80 62.2%
all diagnoses: cancer 24.0% non cancer 74.4%
270 10 logistic regression
71% high
Kwak 2008
USA retrospective cohort study
Medicaid claims data and death registration data
Medicare and Medicaid nursing home provision
nursing home residents
65+ 85.9 (8.1) all diagnoses: cancer 5.4% dementia 14.9% heart disease 36.7% other 43.1%
30,765 9 logistic regression
79% high
Levy 2004
USA retrospective cohort study
Medicare claims data
US nursing homes
nursing home residents
65+ approx 81 all diagnoses: proportions not given
51,187 18 logistic regression
71% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 84
Liu 2007
China prospective cohort study
hospital records
acute hospital COPD patients requiring mechanical ventilation
all ages 65.7 (11.6) COPD 138 5 logistic regression
64% med
McGregor 2007
Canada retrospective cohort study
administrative data plus nursing home records
publically funded nursing homes
nursing home residents
65+ 65-74 7.7% 75-84 32.9% ≥ 85 59.4%
all diagnoses: proportions not given
14,413 6 logistic reg
n
64% med
Menec 2009
Canada retrospective cohort study
administrative data plus nursing home and hospital databases
long term care facilities
all long term care residents from participating facilities
all ages ≤ 74 9.2% 75-84 29.5% 85-94 48.2% 95+ 13.1%
all diagnoses: frailty trajectory 41.5% organ failure trajectory 39.7% terminal illness trajectory 10.1% sudden death or other trajectory 8.6%
2,379 10 logistic reg
n
86% high
Miller 2003
USA retrospective cohort study
hospice program providers databases
hospice programs (in- patient and community-based) across seven states
patients receiving inpatient and community-based hospice care
all ages <65 17% 65-84 52% ≥ 85 43%
all diagnoses: cancer 47% dementia 11% debility 8% other 34%
28,797 9 logistic reg
n
79% high
Mitchell 2005
USA analysis of routinely collected data
death registration data
national population dataset
whole national population
65+ overall age not provided
all diagnoses excluding deaths from trauma: cancer 22.4% dementia 5.1% other 72.5%
1,740,374 3 multi-variate linear reg
n
64% med
Motiwala 2006
Canada retrospective cohort study
administrative health insurance database
region whole regional population
65+ 66-74 26.6% 75-84 40.6% ≥ 85 32.8%
all diagnoses: cancer 34% dementia 27.7%
58,689 6 logistic reg
n
71% high
Muramatsu 2008
USA prospective cohort study
longitudinal participant data and proxy data from family members interviewed after the death
national longitudinal study of non institutionalised older people
representative sample of older people
70+ overall age not provided
all diagnoses: proportions not given
3,362 16 logistic reg
n
71% high
Pritchard 1998
USA prospective cohort study and comparative analysis using routine databases
combination of longitudinal participant data and Medicare claims database
acute hospital patients acutely hospitalised who subsequently died within 6 months
65+ overall age not provided
all diagnoses: proportions not given
479 14 logistic reg
n
57% med
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 85
Ramon 2006
Spain retrospective cohort study
proxy data from family members interviewed after the death
acute hospital hospital deaths 65+ 81.4 (8) all diagnoses: proportions not given
584 8 logistic reg
n
71% high
Rosenwax 2008
Australia retrospective cohort study
death registrations and linked health care databases
regional population dataset
deaths from Alzheimer's disease compared with other causes
75+ 75-84 35.2% >85 64.8%
all diagnoses: Alzheimer’s disease 3.7%
15903 7 logistic reg
n
71% high
Sauvaget 1996
Japan ecological study
national government databases
national population datasets
whole national population
70+ overall age not provided
all diagnoses: proportions not given
ecological study
18 multiple linear reg
n
36% med
Shega 2008
USA retrospective cohort study
proxy data from family members interviewed after the death
outpatient geriatric clinics
all dementia deaths
all ages 85 dementia 135 2 logistic reg
n
79% high
Shinoda-Tagawa 2005
Japan cross-sectional national survey
individual and nursing home characteristics collected by survey
nationally representative sample of nursing homes
nursing home or hospital
all ages overall age not provided
all diagnoses: proportions not given
4,175 15 logistic reg
n
71% high
Silveira 2006
USA retrospective cohort study
death and birth registration data
national population dataset
whole national population
all ages 75 (14) all diagnoses: cancer 27% circulatory disease 35% cerebrovascular disease 9% COPD 6% renal disease 1%
351,110 10 logistic reg
n
79% high
Takezako 2007
Japan retrospective case-control study
nursing home records
single nursing home
nursing home all ages 86.6 (7.6) all diagnoses: dementia 65% cerebrovascular disease 41% heart disease 23% cancer 4% (multiple diagnoses for some participants)
86 3 logistic reg
n
71% high
Wachterman 2006 *
USA retrospective cohort study
death registration data plus proxy data from family members interviewed after the death
nationally representative sample of all deaths
sample of whole population
15+ overall age not provided
all diagnoses: cancer 26.9% chronic cardiovascular 25.7%
12,771** 4 logistic reg
n
79% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 86
* these papers report different components of the same study, hence **participants only counted once
Weitzen 2003 *
USA retrospective cohort study
death registration data plus proxy data from family members interviewed after the death
nationally representative sample of all deaths
sample of whole population with deaths from chronic illness
15+ 73.9 all chronic disease-related diagnoses: cancer 25.8% CHF 32.1% stroke 7.6% COPD 5.0%
(10,122)** 14 logistic reg
n
71% high
Yang 2006
Japan ecological study, including trends over time
death registration data
national population dataset
whole national population
all ages overall age not provided
all diagnoses: proportions not given
ecological study
3 logistic reg
n
71% high
Yasamura 2000
Japan retrospective cohort study
proxy data from family members interviewed after the death
region regional population
40+ 74.9 (11.2) all diagnoses: cancer 33.8% stroke 15.0% cardiovascular disease 12.2% senility 19.7%
213 5 logistic reg
n
71% high
TOTAL PARTICIPANTS: 6,286,394
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 80
8.3 Heterogeneity of evidence
Heterogeneity of studies Studies were heterogeneous in three main
areas:
Study design
Study population and setting
Methods of data collection
Heterogeneity in study design Three study designs accounted for most
of the evidence identified:
analysis of death registration data, with or without linkage to other routine
health databases
secondary analyses of routinely-collected data from provider organisations
post-death interviews with bereaved family members
In addition, there were five longitudinal studies (96, 99, 102, 111, 118),
one of which also incorporated data from Medicare alongside prospective
data (118). Shinoda-Tagawa et al conducted a cross-sectional national
survey (95) and two studies conducted a case control study (104, 123). The
remaining two studies were ecological (120, 126).
Heterogeneity in study population and setting: There was a marked
range in size of studies. Seven studies used complete national (103, 104,
106, 115, 120, 122, 126) or cross-national (106) datasets. In contrast, one
or two studies were conducted in a single hospital (97) or single nursing
home (123). There was a range of settings, from all deaths in a country or
region, to specific hospice, hospital or nursing home settings (see Table 14).
Heterogeneity in methods of data collection: Eight studies undertook
secondary analyses of routinely collected data from Medicare or other
provider organisations (97, 101, 107, 110, 112-114, 116). Seven studies
used proxy data from post-death interviews with family members (91, 108,
117, 121, 124, 125, 127). Five studies undertook analysis of death
registration data (98, 100, 105, 106, 115), and a further five studies
undertook analysis of death registrations alongside other linked datasets
(52, 103, 109, 119, 122).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 88
8.4 Categorization of variables associated with place of
death
This was one of the main challenges in synthesizing studies. Within the
included studies there was marked heterogeneity in approach, both to the
outcome of interest and to the potential factors influencing it. For instance,
place of death is sometimes categorised into a binary variable (e.g. home
deaths v deaths elsewhere, hospital deaths v deaths elsewhere) and
sometimes into several variables (e.g. home, hospital, hospice in patient,
nursing home). Since the key locations presented in the evidence are home
and hospital, we have synthesized evidence into:
i) factors associated with home deaths (v death elsewhere or death in any
other place)
ii) factors associated with hospital deaths (v death elsewhere or death in any
other place)
We have grouped papers which treat place of death as a binary variable
(e.g. home v all other places) together with papers which categorise place
of death into several variables (e.g. home v nursing home, home v
hospice). Because of this approach, it is not possible to assume that an
increase in a factor associated with greater likelihood of home death
necessarily means that a reduction in that same factor is associated with
greater likelihood of hospital deaths. However, if a factors operates in both
categories (is associated with likelihood of home death and likelihood of
hospital death), this adds to the validity of the evidence.
8.5 Disease-specific analyses
One of the main areas of interest for this review was comparison of factors
between different non malignant conditions. For this reason, we present
findings by disease group. However, there was a paucity of evidence
relating to specific disease groups (see Table 14); only three studies
reported factors associated with place of death specifically in COPD (96, 97,
111), one study reported findings specifically in CHF and dementia
combined (105), and three further studies reported findings specifically in
dementia (106, 119, 121). No included studies reported multi-variate
analysis of factors affecting place of death in stroke, ESKD, or LTNCs.
A further five studies reported, not on disease-specific populations, but on
populations in one setting – nursing homes (95, 109, 110, 112, 113, 123). All
other studies related, not to a specific disease group, but to a wider or more
general population of those with advanced non cancer diagnoses (these included
most or all non cancer diagnoses).
We have therefore reported, in turn, findings from:
this general non-cancer population
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 89
nursing home residents only
COPD only
CHF and dementia only (grouped because there is only one study which
reports multivariate analysis of factors associated with place of death in
CHF, and this study relates to both CHF and dementia)
Home and hospital death are addressed separately, although there was no
evidence on home death in COPD. The evidence on place of death for nursing
home residents relates to two places only: nursing home or hospital.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 90
9 Results: factors associated with home death
9.1 Factors operating in the general non-cancer population
In the general population with advanced non cancer conditions, we found
evidence for 15 factors associated with home death (see Table 15).
Interpretation of this evidence was complex. There was high strength and
consistent evidence about five factors associated with home death: age,
diagnosis, co-morbidity, functional impairment, and hospital bed availability
(see Table 15).
In summary:
There was strong evidence that, taken over the whole age range, age
has no association with place of death, although this is accompanied
by low strength evidence that being over 75 is associated with increased
likelihood of home death.
There was strong and consistent evidence that heart disease (but not
CHF), COPD, and dementia were associated with increased odds of
home death. However, this does depend on the comparator group; there
is low strength evidence, for instance, that COPD is associated with
reduced home death as compared with cancer.
There is strong evidence that co-morbidity is associated with reduced
home death.
There is also strong evidence that hospital bed availability is
associated with reduced likelihood of home death, although with
small effect.
Other factors require more complex interpretation, and will be addressed in turn
below.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 84
Table 15. Evidence on factors associated with home death in the general advanced non-cancer population
Type of factor Factor Strength of evidence
Consistency Direction
Personal and demographic factors
Age (each added year, all ages) Age 75-84 Age >85 Gender Ethnicity: Black White Latino Marital status Education
High Low Low Moderate Moderate Low Low Moderate Moderate
100% (5/5) 50% (1/2) 50% (1/2) 100% (2/2) 55% (5/9) 66% (2/3)
- (1/1) - (1/1)
66% (4/6) 66% (2/3)
No significant effect No significant effect 74-85 compared to <65 - ▲home death >85 - ▲ home death No significant effect Black - ▼home death White - ▲home death Latino - ▲home death Married - ▲home death Further education - ▲home death
Disease-related factors
Diagnosis**: CHF*** heart/circulaty disease*** COPD**** Dementia ESKD***** Stroke****** Co-morbidity
Moderate High High Low High Low Moderate Low Low High
100% (2/2) 75% (3/4) 100% (3/3)
- (1/1) -
75% (3/4)
- (1/1)
66% (2/3) 50% (1/2) 50% (1/2) 100% (3/3*)
CHF compared to other – no significant effect Heart/circulatory disease compared to other – ▲ home death COPD compared to other - ▲ home death COPD compared to cancer - ▼home death Dementia compared to other/no dementia - ▲ home death ESKD compared to cancer - ▼home death Stroke compared to other - no significant effect Stroke compared to cancer - ▼home death Stroke compared to cancer - ▲home death Co-morbidity - ▼home death
Environmental factors – social support
Living alone No informal carer
Moderate Low
66% (2/3)
- (1/1)
Living alone - ▼home death No significant effect
Environmental factors – health and social care input
Hospital bed availability
High
100% (4/4)
Increased hospital bed availability - ▼home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 92
Environmental factors – macrosocial factors
Household or area income
Moderate
100% (2/2)
Increased income - ▲home death
Symptoms/ illness burden/ trajectory
ADL impairment Cognitive impairment Expected death Trajectory of illness
High Low Low Low
100% (3/3) 50% (1/2) 50% (1/2)
- (1/1) Not comparable since different aspects of trajectory assessed (2/2)
No significant effect (severe ADL impairment may be associated with▲home death – see text) Severe impairment - ▲home death Some impairment – no effect Expected death - ▲home death Illness < 1 day - ▲home death Illness 1-28 days - ▼home death Decline over 5 months - ▲home death
*One study (Motiwala, 2006) relates to home death with home care
** Diagnosis only considered as a factor when reference group was either ‘other diseases’ or ‘cancer’
** Some studies define CHF specifically and others only report heart/circulatory disease (including cardiovascular disease)
****One study reported ‘chronic lung disease’ rather than COPD specifically
*****Includes genitourinary disease as well as renal disease
******Two studies define cerebrovascular disease, rather than stroke alone
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 86
9.2 Personal and demographic factors
A more detailed breakdown of the evidence on factors affecting place of
death elucidates this complex evidence more fully. There was evidence in
relation to the personal and demographic factors of age, gender, ethnicity,
marital status and education. Table 16 describes the conventions adopted
for the subsequent graphs for each factor.
Table 16. Conventions used in the graphs
X axis represents the value of the odds ratio; increased odds to right
Y axis represents each study
Where possible, a standard size of X axis is used (from 0.0 – 5.0) – odds
ratios above 5.0 are documented on the graph
The colour of the marker for each odds ratio represents study quality (red =
high quality, yellow = medium quality)
95% confidence intervals are plotted where available
Where feasible, odds ratios have been recalculated in line with the most
common approach adopted by the included studies reporting that factor, in
order to increase comparability between studies; for instance, most studies
report the odds ratio of the factor ‘female’ with reference to male. But a few
report the odds ratios for ‘male’ reference female - these have been inverted
to aid comparison.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 94
Figure 14. Association of age with home death
Seven studies report the relationship between age and home death (see
Figure 14), and all are high quality (Table 14). Four of these studies
reported findings which reach significance at the 95% level (27, 45, 48, 55)
(52, 117) (119, 126). Five of the seven studies treat age as a continuous
variable, but two (125, 126) categorise into either three or four groups to
understand the association of older age, beyond 65 years. There is a lack of
consistency in the findings; in a Belgium study (52) older age is associated
with reduced likelihood of home death, although the effect is small (OR
0.99). US (117) and Australian (119) studies show increased likelihood of
home death with age, although again the effect is small (OR 1.03 and 1.04,
respectively). Two further US studies (101, 128) show no significant
association between age and home death. Considering age as a continuous
variable, and across all groups, there is high strength evidence therefore,
that age is not associated with home death in the general non-cancer
population.
However, the two remaining studies focus specifically on older age groups
(125, 126). These studies have categorised age to compare those under 65,
with those aged 65-74 years, those aged 75-84 years, and those above 85
years (or some combination of these categories). These two papers are
consistent in suggesting that those > 85 are more likely to die at home (OR
1.39 and 3.53, respectively), although the strength of this evidence is
weak.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 95
Figure 15. Association of gender with home death
Nine studies report the relationship between gender and home death (see
Figure 15), and again all are high quality (Table 15). Five of these studies
reported findings which reach significance at the 95% level (52, 102, 119,
126, 128), but there is a lack of consistency in the findings. In some
studies, for instance in Belgium (52), the US (128), and Japan (126) being
female is associated with reduced likelihood of home death (OR 0.93, 0.89,
and 0.93). However US (124) and Australian (119) studies show increased
likelihood of home death in women, although again the effect is small (OR
1.4 and 1.23, respectively). Four further US studies (101, 117, 124, 125)
shows no significant association between gender and home death. There is
moderate strength evidence therefore, that gender is not associated with
home death in the general non-cancer population.
Figure 16. Association of ethnicity with home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 96
Evidence on ethnicity comes exclusively from the US, and should be
interpreted in this context. Four studies report the relationship between
ethnicity and home death (101, 117, 125, 128). All are high quality and all
report results significant at the 95% level. There is moderate strength
evidence that black ethnicity is associated with reduced likelihood of home
death, and low strength evidence that white or Latino ethnicity is associated
with increased likelihood of home death. Although findings in relation to a
number of other ethnic groups are reported in the studies, interpretation of
these additional findings in relation to other ethnic groups is difficult,
because of variation in the reference groups used.
Figure 17. Association of marital status with home death
Six studies report the relationship between marital status and home death
(Figure 17), and all are high quality except one (100) of medium quality
(Table 14). All report significant results, although there is some
inconsistency in the findings which may reflect variations in the reference
groups used, as well as national variations. Two US studies (124, 125)
indicate that being widowed or single rather than married increases
likelihood of home death; but this is in relation to home rather than hospital
deaths and excludes all nursing home deaths, which may well influence the
reference group. Conversely, four other studies (100, 117, 119, 128)
indicate that being married increases the likelihood of home death (OR
between 1.05 – 1.62). There is moderate evidence therefore that being
married (rather than single or widowed) is associated with increased
likelihood of home death.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 97
Figure 18. Association of further education with home death
Three studies report the relationship between home death and education
beyond secondary or high school (52, 117, 125). All are high quality
studies, and two (which categorise education into higher or no higher
education, rather than each additional year of education) report significant
associations between having higher education and dying at home (52, 125).
This constitutes moderate evidence that higher education is associated with
home death.
9.3 Disease-related factors
We have specifically assessed place of death in relation to six conditions
(see below). However, not all studies categorised diagnosis or cause of
death in the same way as adopted in this review. We have therefore
mapped the categories as closely as possible to the six main conditions
considered in this review:
Chronic Heart Failure (CHF)
Chronic Obstructive Pulmonary Disease (COPD)
Dementia
Stroke
End-stage Chronic Kidney Disease (ESKD)
Long-term Neurological Conditions (LTNC)
We specify any variation in the definitions both on the graphs and in the
text, and will reflect on the impact of this approach in the discussion.
Studies used a wide variety of reference groups – to achieve some level of
comparability, we report diagnostic categories which were related to cancer
and other conditions in general, rather than specific alternative diagnoses.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 98
Figure 19. Association of CHF with home death
For CHF, the evidence is contradictory. Six high quality studies reported
findings on this (101, 102, 117, 125, 126, 128), although only two had a
specific category for CHF. Of these, both indicate no association between
CHF and home death. The remaining four studies indicate that heart or
circulatory disease in general is associated with home death, and this may
represent in the inclusion of ischaemic heart disease (and consequent
sudden death at home) in this category. All studies are from US, with the
single exception of one from Japan (126).
Figure 20. Association of COPD with home death
The evidence on COPD is more consistent. Four high quality studies from
the US report on this (101, 102, 125, 128), and three specifically classify
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 99
COPD. They are consistent in reporting that COPD is associated with home
death, with the exception of one study which compares COPD with cancer,
rather than other conditions (125). There is strong evidence therefore that
people with COPD are more likely to die at home in comparison to other
diseases, but there is weak evidence that those with COPD are less likely to
die at home when they are specifically compared to those who die from
cancer.
Figure 21. Association of dementia with home death
Four studies report the association of dementia with home death (101, 102,
116, 119) – all are high quality studies, and three are consistent in
demonstrating the association of dementia with home death. The fourth
approaches significance. The odds ratios across all three studies which are
statistically significant at the 95% level indicate that home death is between
1.51 – 2.28 times more likely among those with dementia, when compared
with other conditions.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 100
Figure 22. Association of renal disease with home death
Only one study reports the association between deaths from renal disease
and home death (128), and this includes genitourinary disease along with
renal disease; this provides weak evidence that those with renal disease are
less likely to have a home death.
Figure 23. Association of stroke with home death
Five studies report on the association between stroke and home death (101,
102, 125, 126, 128), although there is variation in how stroke is classified –
sometimes including the wider category of cerebrovascular disease. Taking
studies which compare stroke to other conditions, the evidence is
inconsistent with two studies demonstrating no significant effect and one
demonstrating reduced likelihood of home death. Two further studies,
comparing stroke to cancer, are inconsistent.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 101
Figure 24. Association of co-morbidity with home death
Three studies review the association of co-morbidity with home death (116,
117, 119). Together, these provide consistent and high evidence that
increased co-morbidity is associated with lower odds of home death.
9.4 Environmental factors - social support
Figure 25. Association of no informal carer/living alone with home death
Four studies consider social factors such as availability of informal carers
and living alone (52, 102, 117, 125). These provide moderate evidence that
living alone or without an informal carer is associated with reduced chance
of home death. There is some inconsistency in home these are defined
within studies, and the reference groups used, which may account for the
low level of consistency.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 102
9.5 Environmental factors – health and social care input
Figure 26. Association of hospital bed availability with home death
When considering environmental factors such as health and social care
input, we found strong evidence that increased hospital bed availability was
associated with reduced likelihood of home death; this was identified in four
studies (52, 116, 117, 128). This was very consistent and identified in three
different countries - USA, Canada and Belgium – with markedly different
health systems.
9.6 Environmental factors – macrosocial factors
Figure 27. Association of income with home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 103
Few studies considered macrosocial factors, such as national or regional
policies, health or social systems and strategies. However, two studies did
consider the relationship between income and home death (117, 128); both
found significant evidence that higher income was associated with home
death, though with small increase in odds ratios.
9.7 Symptoms, function, illness burden, and trajectory
In our review of the theoretical model to underpin factors influencing place
of death in non malignant conditions, we considered that aspects of non
malignant conditions such as symptoms, function, illness burden and
trajectory which might be very different from cancer (see the proposed
theoretical model Figure 2). Some studies did provide some limited
evidence on these factors (see Figure 28, Figure 29, Figure 30, Figure 31).
Figure 28. Association of activity of daily living (ADL) impairment with home
death
There was evidence that activities of daily living impairment was not
associated with home death (see Figure 28), although one study did
suggest that the most severely impaired levels of function were associated
with home death (102) - in this study higher ADL scores equated to better
function. Note that the study by Muramatsu reported no significance for
number of ADL limitations, although actual 95% confidence intervals were
not given.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 104
Figure 29. Association of cognitive impairment with home death
Only two studies addressed cognitive impairment (102, 117), and showed
conflicting findings, which may reflect their differing approaches to
categorisation (detailed in Figure 29).
Figure 30. Association of ‘expected death’ with home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 105
Figure 31. Association of trajectory of illness with home death
It may be that there is a more complex interaction between trajectory of
functional decline, physical symptoms, cognitive impairment and trajectory,
which is not easily captured, or indeed much measured, in relation to place
of death. The two studies which address whether death is ‘expected’ or not,
and the trajectory of illness (117, 125) show a rather complex picture;
illness between 1 and 30 days duration is associated with reduced chance of
home death, while very short illness (< 1 day) and more prolonged decline
over 5 months (called ‘rapid’) are associated with increased chance of home
death.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 106
10 Results: factors associated with hospital death
Most of the evidence relates to the general advanced non-cancer
population, although there was a small amount of evidence specific to the
COPD population. We address the general population first, and will discuss
the COPD-specific evidence later.
10.1 Factors operating in the general non-cancer population
In the general non-cancer population, we found evidence for 25 factors
associated with hospital death (see Table 17). Interpretation of this
evidence was also complex. There was high strength and consistent
evidence for only two factors associated with hospital death: marital status
(single/divorced) and gender. Somewhat surprisingly, being single or
widowed is associated with decreased likelihood of hospital death. Seven
studies reported on the effects of gender, and of these six were high
quality. Six studies found gender to have no effect on likelihood of hospital
death, with one study showing men to be significantly more likely to die in
hospital. Other factors require more detailed interpretation, and will be
addressed in turn below.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 100
Table 17. Evidence on factors associated with hospital death in the general non cancer population
Type of factor Factor Strength of evidence
Consistency Direction
Personal and demographic factors
Age Gender Ethnicity: Non-white Marital status: Single Widowed Divorced Having a partner / spouse Education Preference Income
Moderate High Moderate High High Moderate Moderate Moderate Low Low
67% (4/6) 86% (6/7) 100% (3/3) 100% (3/3) 100% (3/3) 67% (2/3) 100% (2/2) 50% (2/4) 50% (2/4)
- (1/1)
- (1/1)
Increasing age ▼hospital death No significant effect All non-white ethnic groups ▲hospital death Single - ▼hospital death Widowed - ▼hospital death Divorced - ▼hospital death No significant effect More education - ▼hospital death More education – no significant difference Strong preference for out of hospital death – no significant effect No significant effect
Disease-related factors
Diagnosis: CHF Heart/circulaty disease COPD Dementia Nephritis Stroke Co-morbidity
Low Low Low Low Low Low Low Low Low Low
- (1/1) - (1/1)
- (1/1) - (1/1)
- (1/1) - (1/1)
- (1/1)
- (1/1)
- (1/1)
- (1/1)
CHF compared to cancer - ▲ hospital death Heart/circulatory disease compared to cancer – ▲ hospital death COPD compared to other - ▲ hospital death COPD compared to cancer - ▲hospital death Alzheimers dementia compared to cancer - ▼hospital death Dementia compared to other - ▼hospital death compared to death in a LTC facility Dementia compared to other - ▲hospital death compared to death at home (with or without home care) Nephritis compared to cancer - ▲hospital death Stroke compared to cancer - ▲hospital death Increasing co-morbidity – ▲hospital death whether compared to home (with or without home care) or LTC facility
Environmental factors – social support
Living situation: Living alone Single or living in a couple Living in a residential home Home care situation: Formal care No formal or informal care Informal care
Low Low Low Low Low Low
- (1/1) - (1/1)
- (1/1)
- (1/1) - (1/1) - (1/1)
Living alone compared to other – no significant effect Single or living in a couple compared to living with other relatives - no significant effect Living in a residential home - ▼hospital death No significant effect No significant effect Informal care compared to both formal and informal care - ▲hospital death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 108
Environmental factors – health and social care input
Hospital bed availability Palliative care provision Nursing home bed availability Primary care physician availability Physician availability (not specified whether primary care or specialist) Specialist physician availability
Moderate Moderate Low Low Low Low Low Low Low
67% (2/3) 100% (2/2) 33% (1/3) 33% (1/3) 33% (1/3) 50% (1/2) 50% (1/2)
- (1/1)
- (1/1)
Increased hospital bed availability - ▲hospital death Increased medicare hospice expenditure or increased hospice enrolees per 1000 - ▼hospital death Increased number nursing home beds per 1000 - ▲hospital death Increased number nursing home residents / 1000 - ▼hospital death Increased percentage residing in nursing homes – no significant effect Increased number primary care MDs per 100000 – no significant effect Increased percentage primary care physicians - ▲hospital death Increased number physicians per capita – no significant effect Increased number specialist physicians per 100000 - ▲hospital death
Environmental factors – macrosocial factors
Population age Population density Population ethnicity Population education Population religious membership Socioeconomics: % of families in poverty Income ratio Social deprivation % recent immigrants
Low Low Low Low Low Low Low Low Low
50% (1/2) 50% (1/2)
- (1/1)
- (1/1)
- (1/1)
- (1/1)
- (1/1) - (1/1) - (1/1) - (1/1)
Increasing population age - ▼hospital death Increasing population age – no significant effect No significant effect Increasing % black or Hispanic in population – ▲hospital death Increasing education - ▼hospital death Increasing % with religious membership - ▼hospital death Increasing % families in poverty - ▼hospital death Increasing income ratio - ▲hospital death Increasing social deprivation - ▲hospital death Increasing % recent immigrants - ▲hospital death
Symptoms/ illness burden/ trajectory
Cognitive decline Functional status Prognosis predictor Trajectory of illness
Low Moderate Low Low
- (1/1)
100% (3/3)
- (1/1)
- (1/1)
No significant effect No significant effect Apache III score – no significant effect No significant effect
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 102
10.2 Personal and demographic factors
Figure 32. Association of age with hospital death
Six studies report the association of age with hospital death. Of these, five
were of high quality (91, 98, 103, 104, 108) and one was of medium quality
(87). Four of the six studies reported significant findings at the 95 % level,
and indicate that increasing age is associated with decreased likelihood of
hospital death (87, 98, 103, 104). The two studies which reported a non-
significant association between age and likelihood of hospital death showed
a similar trend.
Figure 33. Association of gender with hospital death
Seven studies reported the association between gender and hospital death.
Of these, six were of high quality, and one was of medium quality. Six of
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 110
seven showed no significant association between gender and likelihood of
hospital death (87, 91, 98, 103, 108, 124). No confidence intervals were
given for Wachterman, but the data are reported as non significant. One
study (104) reported a significant effect, that male gender is associated
with death in hospital, but the effect size was small (OR 1.07).
Figure 34. Association of ethnicity with hospital death
Three studies from the USA reported on the association between ethnicity
and death in hospital. All three studies show a significant association
between non-white ethnicity and hospital death. In two of the studies (103,
104), several different ethnicities were compared to white. Strikingly, all
ethnicities were significantly more likely to die in hospital than white.
Although all three of the studies showed non-whites were significantly more
likely to die in hospital, because one of the papers (87) is of medium
quality, this is classed as a moderate association.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 111
Figure 35. Association of marital status with hospital death
Five studies reported on the association between marital status and hospital
death. Three of these, all from the USA, compared being single, divorced or
widowed with married. In all three studies being single or widowed was
significantly associated with decreased likelihood of hospital death,
providing high strength of evidence. For decedents who were divorced,
Gruneir and Hansen found a significantly decreased likelihood of dying in
hospital (103, 104), whereas Wachterman found a non-significant trend
towards decreased likelihood of dying in hospital (no CI given, but data
reported as non-significant) (124), therefore providing moderate strength of
evidence that being divorced is associated with decreased likelihood of
dying in hospital.
Two studies reported on the association between the presence of a partner
((108), The Netherlands) or living spouse ((98), Japan) with hospital death.
Both studies were of high quality and reported no significant association.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 112
Figure 36. Association of education with hospital death
Four studies reported on the association between education and hospital
death. Two high quality studies, both from the USA, showed a significant
association between increased education and lower likelihood of hospital
death (103, 104). A further USA study (87), which was of medium quality,
showed a similar trend although this was non-significant. Klinkenberg
compared decedents who had lower vocational education with those who
had less education in a high quality study (108). They found no significant
difference, although the trend was towards greater education being
associated with increased likelihood of hospital death.
Taken together, these studies suggest that increased education is
associated with decreased likelihood of hospital death, although the
inconsistency of their conclusions and variation in the reference groups
used, make it difficult to draw definitive conclusions. In addition, one study
(104) compared hospital with both home and nursing home, in contrast to a
direct comparison between just two settings (hospital and home) in other
studies (103, 108).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 113
Figure 37. Association of preference for place of death with hospital death
In a medium quality study from the USA, Pritchard et al found that a strong
preference for out of hospital death was not significantly associated with
likelihood of dying in hospital (87). However, in this study, patients were
asked about their hypothetical rather than actual preferences; the question
was related to a possible rather than an actual situation.
One study reported the association between personal income and likelihood
of dying in hospital (87). This medium quality study found no significant
association between income and hospital death. Note that the effect of
population income on death in hospital is reported below.
Figure 38. Association of personal income with hospital death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 114
10.3 Disease-related factors
As described above, we have sought evidence for the association of dying in
hospital with the following disease categories:
Chronic Heart Failure (CHF)
Chronic Obstructive Pulmonary Disease (COPD)
Dementia
Stroke
End-stage Chronic Kidney Disease (ESKD)
Long-term Neurological Conditions (LTNC)
Studies used a wide variety of reference groups – to achieve some level of
comparability, we report diagnostic categories which were related to cancer
and other conditions in general, rather than specific alternative diagnoses.
Figure 39. Association of CHF (including heart disease) with hospital death
A study from the USA reported on the association of heart disease with
hospital death in the general non cancer population (103). In this study,
both CHF and heart disease were associated with significantly higher
likelihood of dying in hospital when compared to cancer (OR 1.56 and 2.32
respectively). This single study provides low strength evidence that heart
disease is associated with increased likelihood of dying in hospital than
cancer.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 115
Figure 40. Association of COPD with hospital death
Two high quality studies reported on the association between a diagnosis of
COPD and likelihood of dying in hospital. Ramon et al found that a diagnosis
of COPD was significantly associated with increased odds of death in
hospital when compared to other diagnoses (OR 1.72) (91). Similarly,
Gruneir et al found that a diagnosis of COPD was significantly associated
with increased odds of dying in hospital when compared with diagnosis of
cancer (OR 2.57) (103).
Figure 41. Association of dementia with hospital death
Two North American studies reported on the association between diagnosis
of dementia and death in hospital. Gruneir et al compared diagnosis of
Alzheimers Disease with cancer, and showed that a diagnosis of Alzheimers
disease is significantly associated with decreased likelihood of death in
hospital compared with death at home, although the effect size was small
(OR 0.93) (103). Motiwala et al compared diagnosis of dementia with
‘other’, and looked at the likelihood of dying in hospital compared to several
other settings: long term care (LTC) facilities, home with home care, and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 116
home without home care (116). They found that patients with dementia
were significantly less likely to die in hospital than in a LTC facility (OR
0.35), but significantly more likely to die in hospital than at home, either
with or without home care (ORs 1.36 and 2.14 respectively).
Figure 42. Association of renal disease with hospital death
One study reported the impact of renal disease on death in hospital. In this
high quality study, a diagnosis of nephritis was significantly and strongly
associated with increased likelihood of death in hospital compared to
diagnosis of cancer (OR 5.42) (103).
Figure 43. Association of stroke (cerebrovascular disease) with hospital death
One study reported the association of death in hospital with a diagnosis of
cerebrovascular disease. In this high quality study, diagnosis of
cerebrovascular disease was significantly associated with increased odds of
death in hospital compared to diagnosis of cancer, and the effect size was
large (OR 5.45) (103).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 117
Figure 44. Association of comorbidity with hospital death
One study examined the association between co-morbidity and death in
hospital compared to death in a long term care facility, death at home
without home care, and death at home with home care. In this high quality
study, increasing number of Charlson co-morbidities was significantly
associated with increased likelihood of hospital death irrespective of the
comparator setting (116). The effect size was strongest when compared to
death at home without home care (OR 1.77).
10.4 Environmental factors – social support
Figure 45. Association of living situation with hospital death
Two studies reported on the effect of living situation on likelihood of hospital
death. Pritchard et al found that living alone had no significant effect on
odds of hospital death in a medium quality study from the USA (87). A
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 118
Japanese study from the same year found that decedents who were either
single or living in a couple were more likely to die in hospital than those
who were living with other family, although again this result was not
significant at the 95 % level (98).
Figure 46. Association of place of residence with hospital death
A Spanish study examined the effect of living in a residential home on
likelihood of hospital death. In this high quality study, there was a
significantly reduced chance of dying in hospital if the decedent lived in a
residential home compared to any other place of residence (91).
Figure 47. Association of carer situation with hospital death
One study from The Netherlands examined the influence of carer situation
on likelihood of hospital death (108). This high quality study compared no
formal or informal care, formal care alone, and informal care alone, to both
formal and informal care. There was a significant association between
decedents who received informal care only and increased likelihood of
hospital death (OR 3.68) compared to those who received both formal and
informal care. Receipt of formal care alone had a non-significant effect on
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 119
likelihood of dying at home. Somewhat surprisingly, decedents who
received neither formal nor informal care did not have a significantly
increased likelihood of dying in hospital.
10.5 Environmental factors – health and social care input
Figure 48. Association of hospital bed availability with hospital death
Three North American studies examined the impact of hospital bed
availability on death in hospital. Two of these showed a significant but
modest increase in likelihood of hospital death with increasing hospital bed
availability (104, 116), (ORs 1.15, 1.26 respectively). A third study showed
no significant effect (OR 1.00)(103). Together these studies provide
moderate strength evidence that increased hospital bed availability is
associated with increased likelihood of hospital death.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 120
Figure 49. Association of health care provision with hospital death
Three studies reported the effect of nursing home beds / residents on
likelihood of hospital death. Hansen et al found that an increasing
proportion of nursing home residents in the population was significantly
associated with a decreased likelihood of hospital death, although the effect
size was small (OR 0.99) (104). Conversely, Gruneir et al found that an
increasing number of nursing home beds was significantly associated with
increased likelihood of hospital death, albeit with a similarly small effect size
(OR 1.01) (103). Both studies demonstrate a small effect size in the context
of a very large study size; this should be borne in mind when considering
the apparent contradiction between these data. A third study which
examined the impact of the proportion of nursing home residents on
hospital death found no significant association (87).
Two studies considered the effect of primary care physician availability on
likelihood of hospital death. Hansen et al found a small but significant
increase in the likelihood of hospital death with increasing percentage of
primary care physicians (OR 1.02) (104), whilst Pritchard et al found no
significant association between the number of primary care physicians per
100000 and likelihood of hospital death (87).
In the same study (87) there was a significant positive association between
the number of specialist physicians and likelihood of hospital death. Hansen
et al reported no significant association between an increasing number of
physicians per 100000 and death in hospital (OR 1.00), although it is
unclear whether this statistic refers to primary care, specialist, or both)
(104).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 121
Figure 50. Association of palliative care provision with hospital death
Two American studies reported the association between palliative care provision
and likelihood of hospital death. Hansen et al found a small but significant
decreased likelihood of hospital death with increasing numbers of hospice
enrolees (OR 0.99) (104). Pritchard et al found a large significant decrease
in the likelihood of hospital death with increasing Medicare hospice
expenditure per beneficiary (87). Together, these studies provide low
strength of evidence that increasing palliative care provision is associated
with decreased likelihood of hospital death.
10.6 Environmental factors – macrosocial factors
Figure 51. Association of population age with hospital death
Two papers described the association of population age with hospital death.
Hansen et al found a small but significant inverse relationship between the
percentage of the population over 65 and likelihood of hospital death (OR
0.97) (104). Gruneir et al examined the same factors and found no
significant effect on likelihood of hospital death (103).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 122
Figure 52. Association of population density with hospital death
Hansen et al examined the effect of population density on likelihood of
hospital death (104). In this high quality study from the USA, there was no
effect of population density on likelihood of dying in hospital (OR 1.00).
Figure 53. Association of population ethnicity with hospital death
In addition to individual patient ethnicity (reported above), Hansen et al
examined the effect of population ethnicity on odds of hospital death. They
reported a significantly increased likelihood of hospital death in populations
with greater percentages of both black and Hispanic ethnicities (104).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 123
Figure 54. Association of population education with hospital death
Gruneir examined the effect of population education on hospital death. They
reported a small but significant increase in likelihood of hospital death in
populations with greater percentages of adults without high school
education (OR 1.03) (103).
Figure 55. Association of population religious membership with hospital death
Hansen et al examined the effect of religious membership: they reported
little effect of higher proportion of religious membership in the population
on the likelihood of hospital death (OR 0.997).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 124
Figure 56. Association of socioeconomic factors with hospital death
Two North American studies reported on the impact of socioeconomic
factors on likelihood of hospital death. Gruneir et al found that populations
where the income ratio (average household income for top fifth of
population / average household income for bottom fifth of population) was
greater had significantly higher likelihood of dying in hospital (OR 1.09)
(103). Motiwala et al found that populations with higher percentages of
recent immigrants had a small but significant increase in the odds of
hospital death (OR 1.02) (116). Similarly, greater social deprivation was
associated with increased likelihood of dying in hospital than at home,
either with or without home care. Conversely, Gruneir et al found that
communities with greater percentages of families in poverty had a
significant reduction in likelihood of hospital death, although the effect size
was small (OR 0.98) (103).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 125
10.7 Symptoms, function, illness burden, and
trajectory
Figure 57. Association of cognitive impairment with hospital death
One study examined the influence of cognitive decline on likelihood of
hospitalisation, and found no significant effect (108).
Figure 58. Association of functional status with hospital death
Three studies examined the relationship between functional status and
hospital death. All three studies showed a trend towards higher levels of
disability being associated with reduced odds of dying in hospital, but none
were significant at the 95 % level (87, 91, 108). Of the studies, two were of
high quality and one was medium quality. Together, these studies provide
moderate strength of evidence that poor functional status does not affect
likelihood of death in hospital.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 126
Figure 59. Association of prognosis predictor with hospital death
Pritchard et al found no significant association between Apache III score (a
prognostic score) and likelihood of hospital death (87).
Figure 60. Association of illness trajectory with hospital death
Only one study investigated the relationship between duration of illness and
death in hospital. Babazono et al found no significant effect of greater
duration of illness on likelihood of death in hospital (98).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 127
Figure 61. Association of geographical region with hospital death
Four studies considered the effect of geographical location on hospital death
(87, 91, 104, 108). All four studies showed significant effects, with Pritchard
et al demonstrating a more than seven-fold difference in the likelihood of
hospital death between study sites (87).
10.8 Factors specific to the COPD population
The only disease in which we identified evidence on hospital death specific
to a disease population was COPD. Three studies undertook multi-variate
analysis on this topic in COPD populations (96, 97, 111), although most
relate to COPD patients with intensive care admission, respiratory failure, or
needing intubation, so care needs to be taken in relating these to a wider
community-based COPD population.
We identified 10 factors related to hospital death for the COPD population.
These are grouped by type of factor in Table 18.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 121
Table 18. Evidence on factors associated with hospital death in the COPD population
Type of factor Factor Strength of evidence
Consistency Direction
Personal and demographic factors
Age
Low
- (1/1)
Increasing age -▲hospital death
Disease-related factors
Albumin level Sepsis Co-morbidity Previous intubation Post intubation PH level Medication (long term oral steroids) Disease prognosis Trajectory Ventilation
Low Low Moderate Low Low Low Low Moderate Low Low
- (1/1)
100% (2/2) 100% (2/2) 50% (1/2) 50% (1/2)
- (1/1) 66% (2/3)
- (1/1)
- (1/1)
Low albumin levels - ▲hospital death Sepsis - no significant effect Increased number of organ failures -▲hospital death Increased number of co-morbid conditions - no significant effect Previous intubation ▲ hospital death Low post intubation pH▲ hospital death Long term steroids - no significant effect Increased severity of illness (APACHE II) score ▲hospital death Increased hospital stay ▲hospital death Invasive mechanical ventilation - no significant effect
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 122
Figure 62. Association of age with hospital death
One study from Singapore (97) examines the association between age and
hospital death with particular reference to patients with acute respiratory
failure attributable to COPD. It is high quality and shows a small but
significant association of older age with increased chances of dying in a
hospital setting for COPD patients (OR 1.3).
Figure 63. Association of albumin level with hospital death
The relationship between serum albumin and hospital mortality was
explored by one high quality study, which reports a significant association
between higher albumin levels (hypoalbuminemia) and reduced likelihood of
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 130
hospital death (OR 0.8) (97); COPD patients with lower levels of albumin
appear more likely to die in hospital (OR 1.25).
Figure 64. Association of sepsis with hospital death
The association of sepsis with hospital death was analysed by two studies,
one high quality (96) and based in the USA, and another medium quality
from China (111). Both studies report a small and non-significant effect,
which constitutes moderate evidence that there is no association between
sepsis and hospital death in COPD patients.
Figure 65. Association of comorbidity with hospital death
Two studies report an association of co-morbidity with hospital death in
COPD patients (96, 111). Note the x-axis has been changed for this graph,
from the ‘standard’ in other graphs. One North American study provides
high quality evidence that the probability for hospital death increases more
than five-fold for patients with a higher number of organ failures when
compared to those with none (96). A similar observation in terms of the
positive correlation between co-morbidities and increased chances of
hospital death is made by another study, from China, which reports medium
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 131
quality evidence that patients with multi-organ dysfunction syndrome are
six times more likely to die in hospital compared to those without the
condition (111). The same study reports an inverse relationship between
hospital death and co-morbidity, this time measured by the Charlson Co-
morbidity index, where each increase in unit of severity predicts lower
probability of hospital death (111). However, this latter association is of
relatively small effect (OR 0.5) and non-significant.
Figure 66. Association of intubation with hospital death
Only one study, of high quality, investigates the association between
previous intubation and hospital death for COPD patients (97). The research
group found that a previous history of intubation was a predictor of higher
probability for hospital death by a small, but significant effect (OR 1.3,
p<0.04).
Figure 67. Association of post intubation PH with hospital death
One study, of moderate quality, reports the relationship between post
intubation PH and hospital death (111). The results suggest a significant
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 132
and strong inverse correlation, that is, the lower the pH levels post
intubation, the higher the probability of hospital death.
One study, of high quality, evaluates possible association between
medication and hospital death, more particularly that patients receiving long
term oral steroids have significantly higher chances of dying in hospital than
those who are not (OR 121) (97). The study suggests that the poor adverse
effect profile of steroids may lead to higher chances of hospital death, but
found no significant effect with very wide confidence intervals (3 to 5,280).
Figure 68. Association of prognosis with hospital death
The correlation between disease prognosis and hospital death for COPD
patients was examined by two high quality studies (97) and one medium
quality study (111). All three studies are consistent in using the APACHE II
scoring system (acute physiology and chronic health evaluation measure, to
predict prognosis) to measure the severity of illness, which facilitates
comparison between the results. In all three cases, greater severity of
illness is significantly associated with a higher likelihood of hospital death,
although this varies in terms of the strength of effect. The evidence from
North America shows the smallest effect of disease prognosis on hospital
death (OR 1.046), which increases to a stronger two-fold effect in the study
from Singapore (OR 2), and to more than three-fold in the Chinese context
(OR 3.128). As a whole, the three studies constitute moderate evidence
that the likelihood of hospital death for COPD patients increases with illness
severity.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 133
Figure 69. Association of trajectory with hospital death
Only one study reports the relationship between the duration of hospital
stay and hospital death (97). More days spent in hospital by COPD patients
are associated with greater chances of dying in hospital, however the effect
is small (OR 1.2) and the evidence is weak.
Figure 70. Association of ventilation with hospital death
Only one study examines the association between invasive mechanical
ventilation and hospital death for COPD patients and reports that this type
of intervention is not associated with hospital death (96).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 134
11 Results: factors associated with hospice death
There was relatively little evidence on factors associated with death in
hospice. Only three studies considered this, and all are from the US. All
relate to death in inpatient hospice beds, in the specific context of the US
model of hospice care, however only one study reported factors associated
with hospice death for a population which approximates to a ‘general’
population – a study of older people (65 years plus) in a community based
service (102). The remaining two studies identified among hospice
patients only who was more or less likely to have an inpatient hospice
death, and the evidence needs to be considered in this context.
11.1 Factors operating in the general non cancer population
Because of the limited evidence, factors are not tabulated as before.
Instead, we report first the evidence derived from a study among 620 older
people (65 years plus) in a community based service (102). This study
looked at only two factors associated with inpatient hospice (compared to
hospital or home death); diagnosis and place of residence. Cancer was
found to be associated with inpatient hospice death (OR 14.48) but with
very wide 95% confidence intervals (3.55 – 59.14). Place of residence, in
this study across Connecticut, was strongly associated with inpatient
hospice death (OR 7.53) again with wide confidence intervals (3.63 –
15.58) (102); having a geographic location close to the inpatient hospice
facility increased the likelihood of inpatient hospice death.
11.2 Factors specific to those already referred to palliative care
The remaining two studies only reflect hospice death for those already referred
into palliative care (‘hospice programmes’ in the US model9), and so provide
evidence about likelihood of inpatient hospice death in this specific group of
people.
9 There are marked differences in the models of palliative care between US and other countries, and
correspondingly in how terms such as ‘palliative care’ and ‘hospice’ are used. We have adopted a UK
perspective, but tried to apply it consistently in our interpretation of the evidence.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 135
Figure 71. Association of age with inpatient hospice death
The relationship between age and inpatient hospice death for those in hospice
programmes is reported by two North American studies (107, 114), both of high
quality. Both indicate that there is no association between age (overall) and
hospice death. However, both studies also indicate that being over 85 years
(rather than capturing age across the whole age spectrum) is associated with
lower chance of hospice death. This effect is small when hospice is compared to
home with hospice care (OR 0.94) (114), but increases when hospice is
compared to nursing home (OR 0.54) (107).
Figure 72. Association of gender with inpatient hospice death
Both studies also report a small reduction in likelihood of a hospice death for
women (OR between 0.88 – 0.94) (107, 114).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 136
Figure 73. Association of ethnicity with inpatient hospice death
The evidence on ethnicity is confusing, because of the variations in how ethnicity
is classified, and the variations in comparison between places of death. Johnson
et al show differences in the older Hispanic population versus the older white
population, according to place of death (Hispanics in hospice care have a greater
chance of hospice versus nursing home death, but a lower chance of hospice
versus home death) (107). Differences in Miller’s study, where different ethnic
groups are compared to the white population, are small and non significant,
except for the ‘other non Latino, non-white’ group who have lower odds of
hospice death compared to the white population (OR 0.82).
Figure 74. Association of diagnosis with inpatient hospice death
Only one study evaluates the effect of diagnosis on hospice death, again in those
already referred to hospice, indicating that dementia is associated with reduced
chance of hospice death (114). There is no evidence as to the impact of co-
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 137
morbidity on likelihood of hospice death, although the same study does consider
the impact of ‘debility’ as a diagnosis (see Figure 74).
Figure 75. Association of symptoms (pain) with inpatient hospice death
The study by Miller reports that the symptom of pain among those already
known to hospice is associated with greater likelihood of hospice death (OR
range from 1.10 for mild pain, to 1.67 for severe pain, and 2.28 if pain score is
‘missing’ – this group were assessed as having highest intensity of need) (114).
Figure 76. Association of living situation with inpatient hospice death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 138
Two studies (107, 114) also report the association between living situation and
inpatient hospice death for this specific ‘hospice-referred’ population, and this
evidence reflects the complex interactions of place, carers, and living situation.
The term “living situation” is used here to capture both the setting of care and
the availability or absence of a carer. Johnson et al (107) make a specific
distinction between the types of carers involved: child, or relative other than a
spouse; but report that having no caregiver or a caregiver other than spouse,
reduces the odds of dying in hospice versus nursing home.
On the other hand, having no caregiver increases the odds of hospice rather than
home death. Miller et al (114) report that patients who live in nursing homes are
considerably less likely to die in hospice when compared to patients living in
another setting without a carer (OR 0.07). According to the same study, patients
who live with a caregiver in a setting other than a nursing home have reduced
chances of hospice death in comparison to people living alone (OR 0.51).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 139
12 Results: factors associated with nursing home death
There was evidence on three different aspects of deaths within nursing
homes. First, six studies provided evidence on the likelihood of nursing
home death for the general non-cancer population (99, 116, 117, 124, 125,
129). Second, one study reviewed the likelihood of nursing home death
specifically among people with dementia (106). And third, six further
studies specifically addressed the likelihood of nursing home (as opposed to
hospital) death for those already resident in nursing homes (95, 109, 110,
112, 113, 123). We address these different areas in turn.
12.1 Factors operating in the general non-cancer population
We identified 18 factors associated with nursing home death for the general
population with advanced non cancer (see Table 19).
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 133
Table 19. Evidence on factors associated with nursing home death in the general non-cancer population
Type of factor Factor Strength of evidence
Consistency Direction
Personal and demographic factors
Age Gender Ethnicity: Non-white Marital status: Single Widowed Education
Moderate High Moderate Moderate Moderate High
67% (2/3) 75% (3/4) 66% (2/3) 100% (2/2) 66% (2/3) 100% (3/3)
Increasing age over 75 year ▲nursing home death Female gender ▲nursing home death Non-white ethnic groups ▼nursing home death Single versus married ▲nursing home death Widowed versus married ▲nursing home death No significant effect
Disease-related factors
Diagnosis: CHF COPD Dementia Nephritis Stroke Co-morbidity
Low Low Low Low High Low Low Low Low
50% (1/2) 50% (1/2) 50% (1/2) 50% (1/2)
100% (3/4)
- (1/1)
- (1/1) - (1/1)
- (1/1)
CHF compared to cancer - ▲ nursing home death CHF compared to cancer – no significant effect COPD compared to cancer - ▲ nursing home death COPD compared to cancer –▼nursing home death Dementia compared to cancer or other – ▲ nursing home death Nephritis compared to cancer – ▲ nursing home death Stroke compared to cancer - ▲ nursing home death Stroke compared to cancer – no significant effect Higher co-morbidity - ▲ nursing home death
Environmental factors – social support
Home care situation: Had caregiver
Low
- (1/1)
Presence of caregiver –▼nursing home death
Environmental factors – macrosocial factors
Socioeconomics: Social deprivation (regional) Income disparity (between poorest and richest in region)
Low Low
- (1/1) - (1/1)
Increased social deprivation – ▲ nursing home death Increasing income disparity – ▼ nursing home death
Symptoms/ illness burden/ trajectory
Functional status Trajectory
Low Low
- (1/1)
- (1/1)
Worse functional status over year before death - ▲ nursing home death Decline over five months to death compared to no decline - ▲ nursing home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 141
12.2 Personal and demographic factors
Several personal and demographic factors in the general non-cancer
population related to nursing home death were studied.
Figure 77. Association of age with nursing home death
Figure 77 is plotted on an extended x axis, as compared with other figures,
to accommodate the range of evidence. Over 75 years, there is moderate
strength evidence that older age is associated with increased likelihood of
nursing home death (125, 129). This ranges from being twice as likely for
those 75-84 compared to 65-74 years (125), up to more than four times as
likely in those 75-84 years compared to those < 65 years (129). In the
very old (95 years plus), Gruneir et al report odds ratio of 12.82 for dying
in a nursing home versus home (129).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 142
Figure 78. Association of gender with nursing home death
Three studies also report a small but significant association between being
female and dying in a nursing home (117, 125, 129). There is some
inconsistency in the evidence, with (124) reporting a small opposite effect.
Figure 79. Association of ethnicity with nursing home death
The evidence on ethnicity is more complex, but in general ethnic minority
groups have a reduced likelihood of nursing home death (117, 125, 129).
This is consistent across studies, although these are all US based studies,
and reflect the US context specifically.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 143
Figure 80. Association of marital status with nursing home death
Marital status is often categorised differently across studies, making
synthesis of findings difficult. Two studies depicted in Figure 80 indicate
that being single or widowed is associated with greater likelihood of nursing
home death, with a two to three-fold likelihood of nursing home death
(124, 129). However, work by Weitzen et al does not confirm this,
suggesting no significant effect (125).
Figure 81. Association of education with nursing home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 144
Evidence is more consistent in relation to the effect of education on nursing
home death; with two studies showing no definite effect, with only a small
significant association of reduced nursing home death with education in the
work by Gruneir et al (129).
12.3 Disease-related factors
Figure 82. Association of dementia with nursing home death
There is more convincing evidence on the association of dementia with
nursing home death – note this graph extends the x axis beyond our
convention of 0.0 to 5.0. Two studies are consistent in demonstrating
increased likelihood of nursing home death for the general non-cancer
population, with about 4 fold likelihood of nursing home death compared to
cancer (129), or to other conditions (116), increasing to over nine-fold if
nursing home death is compared to home death without home care (116).
Figure 83. Association of stroke with nursing home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 145
Three studies assess the association of stroke with nursing home death
(99, 125, 129). It is difficult to compare the study by Bell and colleagues,
since the reference group is coronary heart disease, but the remaining two
studies show conflicting results.
Figure 84. Association of CHF with nursing home death
Those studies which categorise CHF separately suggest an association
between CHF (versus cancer) and nursing home death, albeit small (125,
129); this does not reach significance in one of the studies (125).
Figure 85. Association of COPD with nursing home death
Evidence about COPD is conflicting; two studies report the association
between COPD and nursing home death with opposite findings. This may
reflect different comparator groups; Gruneir compares nursing home with
home (129), while (125) compares nursing home with hospital.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 146
Figure 86. Association of ESKD with nursing home death
Only one study reports the relationship between ESKD and nursing home
death for the general non-cancer population, and this study only reports
‘nephritis’ rather than ESKD deaths per se (129). However, it shows a
marked increase in likelihood of nursing home death for those with
nephritis (versus cancer).
12.4 Symptoms, function, illness burden, and trajectory
Figure 87. Association of co-morbidity with nursing home death
Only one study reports the relationship between co-morbidity and nursing
home death, with a small but significant association demonstrated between
number of co-morbid conditions and nursing home death (116).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 147
Figure 88. Association of functional limitations with nursing home death
Weitzen et al report that functional limitations in the year before death are
associated with nursing home death (OR 2.93) (125); this relates to a
comparison of the likelihood of nursing home versus hospital death.
Figure 89. Association of trajectory with nursing home death
The same study also assessed speed of decline, and the association of this
with nursing home death; rapid decline over the five months prior to death,
in contrast to no particular decline, was strongly associated with nursing
home death (125).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 148
12.5 Environmental factors – social support
Figure 90. Association of living situation with nursing home death
Weitzen et al are also the only study to report the association of caregiver
presence or absence with odds of nursing home death; having a caregiver
markedly reduces the chance of nursing home death (odds ratio 0.23)
(125).
12.6 Environmental factors – health and social care input
Figure 91. Association of PC provision with nursing home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 149
There is no consistent evidence about health care provision and nursing
home death; although some studies do report this (116, 129), the way
different studies have treated each variable (either categorised differently
or used as continuous or categorical variables), and the different reference
groups make useful synthesis of evidence difficult. Just one study reports
the association between receiving palliative care10 and likelihood of nursing
home death (125), with the finding of no significant association.
12.7 Environmental factors – macrosocial factors
Figure 92. Association of socioeconomic status with nursing home death
The evidence with respect to socioeconomic status is complex. Work by
Motiwala et al in Canada shows that social deprivation is associated with
increased liklelihood of nursing home death, although with a small effect
(OR 1.18 or 1.30, depending on comparator group – home without
homecare or home with homecare, respectively) (116). This is based on
social deprivation index for an area or region. Gruneir however looked at
different meaasures; income disparity and proportion of families below the
poverty line (129). The latter showed little effect, while income disparity
was associated with reduced likelihood of nursing home death. The authors
postulate that states with greater income disparity may have more older
residents unable to afford nursing home care (129).
10
‘Hospice care’ in US terms.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 150
12.8 Factors specific to dementia populations
Table 20. Evidence on factors associated with nursing home death in the
dementia population
One study looked at factors influencing place of death especially in the
dementia population, making a comparison across five European countries
(106). Table 20 summarizes this evidence. This study identified four further
factors related to dying in a nursing home which were specifically relevant
to the dementia population. This study compared nursing home deaths to
hospital deaths, and related specifically to those over 65 years.
Type of factor Factor Strength of evidence Consistency Direction
Personal and demographic factors
Age Gender
Low Low
- (1/1)
- (1/1)
Increasing age - ▲nursing home death Female gender - ▲nursing home death
Environmental factors – health and social care input
Bed availability NH beds Hospital beds
Low Low
- (1/1) - (1/1)
Increased NH beds - ▲nursing home death Increased hospital beds - ▼nursing home death
Environmental factors – macro-social
Country of residence
Low
- (1/1)
Residence in Netherlands compared to UK - ▲nursing home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 151
Figure 93. Association of age with nursing home death in those who die with
dementia ≥ 65 years
As for the general non-cancer population, older age is associated with
increased likelihood of death in a nursing home (106).
Figure 94. Association of gender with nursing home death in those who die
with dementia ≥ 65 years
Again as for the general non-cancer population, being female is associated
with increased likelihood of death in a nursing home, although this is much
more marked in the dementia population (OR 1.94) (106).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 152
Figure 95. Association of country of residence with nursing home death in
those who die with dementia ≥ 65 years
Country of residence showed marked association with likelihood of death in
a nursing home, with highest likelihood of nursing home (versus hospital)
death in the Netherlands (OR 15.99), but also markedly high likelihood of
this in Belgium (OR 4.68) (106). Netherlands also had increased likelihood
of nursing home (versus home) death (OR 1.25) compared to England, with
a lesser likelihood of nursing home (versus home) death in both Scotland
(OR 0.68) and Belgium (OR 0.39) (not charted).
Figure 96. Association of bed availability with nursing home death in those
who die with dementia ≥ 65 years
The same study reported nursing home and hospital bed availability as
being associated with nursing home death across the five countries (106);
increased nursing home bed availability is associated with increased
likelihood of nursing home death (OR 0.88 for every additional bed/1,000
population), while increased hospital bed availability is associated with
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 153
reduced likelihood of nursing home death (OR 1.17 for every additional
bed/1,000 population).
12.9 Factors specific to nursing home populations
Lastly, we identified factors in the nursing home population where there
was evidence about their relationship with place of death. All of these relate
to the odds of dying in nursing home versus hospital. This evidence comes
from two Japanese studies; they relied on nursing home records (123), and
a national nursing home survey of the last five deaths (95), rather than
direct patient or proxy data. Since they are both from the same country,
generalisability is also hard to assess.
Table 21. Evidence on factors associated with nursing home death among
nursing home residents
Type of factor Factor Strength of evidence Consistency Direction
Personal and demographic factors
Age Gender Marital status Being unmarried Preference for nursing home
Moderate Moderate Low Moderate
100% (2/2)
100% (2/2)
- (1/1) 100% (2/2)
Increasing age - no significant effect Female gender - no significant effect Being unmarried - no significant effect Preference for nursing home care (individual or their family) - ▲nursing home death
Disease-related factors
Diagnosis CHF
Low
- (1/1)
CHF compared to pneumonia - ▲nursing home death
Environmental factors – health and social input
Acceptance by NH of resident’s NH preference NH ‘do not transfer’ policy Hospitalisation in three months before death Doctor face to face presence in NH out of hours Number of beds: NH beds
Low Low Low Moderate Low
- (1/1)
- (1/1)
- (1/1) 100% (2/2)
- (1/1)
Acceptance - ▲nursing home death Policy in place - ▲nursing home death Hospitalisation - ▼nursing home death Doctor presence out of hours -▲nursing home death increased NH beds – no significant effect
Symptoms/ illness burden /trajectory
Cognitive impairment Functional status Length NH stay
Low Low Low Low
- (1/1)
50% (1/2) 50% (1/2)
- (1/1)
Severe cognitive impairment - ▲nursing home death Severe dependence - ▲nursing home death Bedridden – no significant effect Length of NH stay – no significant effect
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 154
Figure 97. Association of age with nursing home death in the nursing home
population
Two studies, both from Japan, report age (as a continuous variable) in
relation to place of death, showing no association between older age and
nursing home or hospital death (95, 123).
Figure 98. Association of gender with nursing home death in the nursing
home population
The same two studies show no significant association between gender and
nursing home or hospital death (95, 123).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 155
Figure 99. Association of marital status with nursing home death in the
nursing home population
Only one study reports marital status, again with no significant association
between marital status and nursing home or hospital death (123).
Figure 100. Association of individual care preferences with nursing home death
in the nursing home population
The Japanese studies also study the preferences of the individual and
family for place of care. In this graph, the x axis has been extended
markedly beyond the ‘standard’ (5.0) length in other graphs. Both studies
report a significant association between preference for nursing home death
and death there (95, 123). In one study, this is the person’s own
preference (OR 16.3) (95), whereas the other reports the preference of a
family decision maker (OR 3.95) (123). Conversely, preference for hospital
death is associated with reduced chance of nursing home death (95).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 156
Figure 101. Association of acceptance by home of resident’s choice with
nursing home death in the nursing home population
Similarly, acceptance by the nursing home of the individual’s preference for
nursing home death is associated with much increased chance of nursing
home death (95). (Again, this graph is extended).
Figure 102. Association of diagnosis with nursing home death in the nursing
home population
Only one study assesses the effect of diagnosis (95), and reports increased
likelihood of nursing home death among those with CHF (OR 4.6) and
cancer (OR 2.2), in relation to pneumonia.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 157
Figure 103. Association of cognitive status with nursing home death in the
nursing home population
The same study reports severe cognitive impairment as increasing the
likelihood of nursing home death (OR 1.4) (95).
Figure 104. Association of functional status with nursing home death in the
nursing home population
Two studies (95, 123) report increased functional limitation as being
associated with nursing home death, although this does not reach
significance in one study (123).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 158
Figure 105. Association of hospitalisation with nursing home death in the
nursing home population
Hospitalisation in the three months prior to death is reported as having a
reduced association with nursing home death (OR 0.68) (95).
Figure 106. Association of length of nursing home stay with nursing home
death in the nursing home population
In contrast, length of stay in the three months prior to death is reported as
having no association with nursing home death; the range of length of stay
of nursing home residents is not reported (95).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 159
Figure 107. Association of healthcare provision with nursing home death in the
nursing home population
Two further studies look at measures of healthcare provision and how these
relate to nursing home death. As evident from the graph (see Figure 107),
greater availability of medical and nursing support, especially out of hours,
increases the likelihood of nursing home death (95, 123). This is both
significant and with a large effect (x-axis extended).
Figure 108. Association of number of nursing home beds with nursing home
death in the nursing home population
The number of nursing home beds was not shown to have a significant
association with nursing home deaths (95, 123).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 160
Figure 109. Association of nursing home transfer policy with nursing home
death in the nursing home population
Finally, nursing home policy on transfers is associated with nursing home
deaths; as expected, a ‘do not transfer’ policy (reference ‘no policy’) is
associated with increased likelihood of nursing home death (95).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 161
13 Results: qualitative evidence
55 studies reported qualitative evidence, plus one mixed methods
(qualitative and quantitative) study. This qualitative evidence covered
various aspects of preferences in relation to place of care and death,
aspects relating to actual place of care and death, and transitions. Given
that the main focus of the review is non malignant conditions, we present
this evidence by disease group (CHF, COPD, LTNC, ESKD) with the
exception of dementia and stroke. Specific evidence on stroke and
dementia is presented in the context of the evidence on older people, since
there were considerable overlap of qualitative findings between older
people, those with dementia, and those with stroke.
Note that we touch on areas beyond the remit of this review, but this is
primarily to provide the context of the qualitative research being
undertaken, not to extend the findings of the review beyond the research
questions.
13.1 Chronic heart failure
The nature of the evidence: There were ten qualitative studies
investigating concerns and experience relating to people with chronic heart
failure (CHF) (49, 130-138, 181). There was five from the UK (49, 133,
134, 136, 138, 181), two from Canada (131, 132), and one each from
Sweden (130), the USA (135), and New Zealand (137).
The quality of these studies was appraised and is shown in APPENDIX 5.
The studies were of varying quality, with the lowest quality score 27 out of
40. Weaker studies tended to be the smaller qualitative component of
mixed method studies, or very small studies with less transferability.
The focus of the studies was mainly on the experience of living with heart
failure, from the patient’s view point (49, 130, 131, 133, 135, 137, 138,
181). This experience was related in particular to prognosis and
communication (131) and the experience of being a burden for family
(135). In addition, one study explored carers’ views about dying of heart
failure (135, 136). Patterns of decline and comparing experience with lung
cancer and other types of cancer (49, 135, 138) were also explored, and
one study reviewed preferences in the context of the dominant ‘cancer’
model of end of life care (181).
The majority of the qualitative studies collected their data with interviews
(ten) and or focus group discussions (two). There was one mixed method
study using a questionnaire plus interviews and one using a case study
approach. Analysis, when indicated, either used a grounded theory (131,
133, 137) or phenomenological (130) approach, or qualitative descriptive
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 162
analysis (49, 134-136). Three studies used longitudinal or serial in-depth
interviews with two or three interviews with participants (49, 134, 137,
138).
Context: The studies investigated issues related to home, or hospital or
nursing home. None were hospice based studies.
Home care: Horne and Payne explored the experiences of people living
with severe heart failure and aimed to identify their needs, finding that
people’s problems mainly related to activities of daily living and fatigue
(133). None of the patients in their study had been referred for specialist
palliative care. In two papers, Murray et al explored patterns of decline for
people living with heart failure at home (49, 134). Brannstrom explored
patients’ experiences of lung cancer with severe heart failure at home
(130). Similarly, Waterworth and Jorgensen explored the experiences of
older people living with heart failure at home or in residential care (137).
Hospital care: Johnson et al explored the experiences of living with severe
heart failure and in particular the experience of being a burden on family
(135). Caldwell explored the preferences of patients with advanced heart
failure regarding communication about their prognosis and the implications
of communication preferences, in a hospital setting (131). In contrast to
the experience of patients living with cancer, CHF patients saw hospital as
place for survival or safety. Despite this, carers expressed dissatisfaction if
they perceived futile treatments had been given at the end of life (136)
Preferences: Preferences regarding end of life care were explored in
several studies. Patients wanted to learn about their prognosis and its
implications at a time of optimal functioning and not when their capacity for
end of life decision making was diminished. Patients identified that they
wanted to be told the truth but that this should be balanced with hope
(131). On a similar theme, Desharnais et al found there was little
concordance between patient and professional views on end of life decision
making (132). The longitudinal studies show that patients’ expectations of
health professionals increased as time progressed (137).
Place of care: The majority of studies were related to experiences of care
at home. The studies provided insight into several aspects that were
experienced or evaluated as appropriate care at the end of life for people
living with and dying from heart failure. Issues included: facing death (49,
133) and feelings of being a burden (133, 135). Barriers to accessing
information and social services were identified (133). A unique aspect of
living with an illness with an uncertain trajectory of decline was ‘knocking
on deaths door although surviving’ (130). Living with heart failure was
perceived as a rollercoaster with an ongoing oscillation between ups and
downs. The will to survive was strong (130). When comparing home and
hospital care, home was generally perceived as better (136).
Place of death: One study explored preferences about place of death
(181), and did this in specific relation to the dominant cancer model of
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 163
‘open awareness’ of death. Patients with CHF were influenced by thoughts
about the nature of death, with most expressing a preference for a sudden
and unaware death, sometimes linking this with home death (181). The
idea of personal autonomy about place of death seemed somewhat alien to
many, with their considerations and concerns about their family equally
important as concerns about themselves.
Transitions: The patterns of decline in heart failure were perceived as
different to other illnesses such as lung cancer (49, 135). Social and
psychological decline tracked the physical decline while spiritual distress
exhibited background fluctuations. Moreover transition was not perceived
by patients as a linear process moving from one phase to another (137);
experiences illustrated the complexity of transition in peoples’ lives as well
as the challenges of managing these. This fluctuation was perceived as a
rollercoaster with professionals providing a ‘safety belt’ (130).
Only one study explored carers views of the experience of heart failure
(136). Most carers found discussion about end of life with their family
member prior to death difficult. Carers were sometimes dissatisfied with
care that their loved one had received, and this tended to be focused on
hospital care particularly if carers believed futile treatments have been
given. Deaths that occurred in the home were more likely to be perceived
as good.
13.2 Chronic obstructive pulmonary disease
The nature of the evidence: There were 12 qualitative studies
investigating concerns relating to people with chronic respiratory disease
(75, 139-149); four from the USA (75, 141, 145, 148), four from the UK
(139, 140, 143, 146), three from Canada (144, 147, 149) and one from the
Netherlands (142). The quality of the evidence is shown in APPENDIX 5.
The majority of studies were high quality, scoring more than 30 out of 40.
Weaker studies tended to be the smaller qualitative component of mixed
method studies.
The focus of the studies was mainly on living with or dying from advanced
COPD, including symptoms and care needs (141-143, 145, 147, 149) or
caring for people with advanced COPD (140, 144), as well as preferences
for site of treatment or care provided at end of life (75, 139, 148), and
access to care (146). Not all the studies were exclusively related to COPD.
For instance, 2 studies (144, 145) compared COPD with other illness such
as cancer.
There were ten studies which collected their data with interviews and/or
focus group discussions, and two mixed method studies using a
questionnaire plus interviews. Analysis, when indicated, was either using
grounded theory (75, 140, 145, 148) or a Framework approach (143). Two
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 164
studies used longitudinal or serial in-depth interviews with two or three
interviews with participants (141, 147).
Context: The studies investigated issues related to home, or hospital or
nursing home; none were hospice-based studies. Two studies did not state
their setting (141, 145).
Home care: Borgsteede et al (142) undertook a study exploring the
aspects valued by patients and GPs in end of life care at home. They found
that the aspect of availability or attention of the GP was prominent in the
patients’ interviews. Interestingly they found no difference in the aspects of
care mentioned by patients of GPs trained in end of life care compared to
their counterparts who were not trained.
Two studies related the impact of symptom experiences by patients with
COPD in their last year of life. Elkington (143) found that the major
symptom reported by carers was breathlessness, which impaired their
loved ones mobility and contributed to them being housebound. Anxiety,
depression and panic were also associated with breathlessness. Similarly,
Hall et al (149) found that dying of COPD meant ‘suffocating’ to patients.
Shipman (146) also found that all the patients with COPD in their study
reported severe breathlessness. Care needs of older people living with
COPD at home (147) were for self reliance and independence, through
adaption to suitable health and living constraints. The predominant theme
was that all participants wanted to maintain their independence, which
required considerable adaption (147).
Those studies that compared perceptions of care of patients with COPD to
people living with other illnesses, such as cancer, found that even in
primary care teams committed to the delivery of palliative care in end stage
COPD, those with COPD were less likely to receive full and easily
understood information and to be aware that they were dying. In addition,
they were less likely to receive district nursing care (139).
From the carers viewpoint, Gysels and Higginson (140) found that several
key issues affected care in a negative way; these were uncertainty, carers
own health problems, an ‘imploded world’, negative reactions from the
outside, person loss, and acute exacerbations of illness. On the other hand
resources that carers drew on were acceptance, self care, availability of
support, and feeling that caring is a shared responsibility. They also found
that breathlessness was particularly challenging, with carers indicating that
they did not have the resources to deal with this symptom.
Hospital care: Goodridge et al (144) found that patients with COPD
experienced an overall quality of dying in the ICU that is similar to those
who died from other causes, in spite of a unique and significant
impediments to end of life care for this group of patients.
Preferences: Preferences regarding end of life care were variable.
However in contrast to other conditions, there was no indication that
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 165
preferences tended changed according to stage of illness (75, 148),
although the evidence on this is limited. These studies explored people with
COPD in both acute and terminal stages.
When comparing home and hospital, Fried (148) found that that patients
were sceptical of expanded home care to treat acute illness, since they
perceived home care as low intensity and low frequency of care. They
associated hospital with the greatest chance of survival. If the sites proved
equal in terms of survival, then they preferred home because of the
freedom from constraints and the associated comfort of home. Those who
did prefer hospital largely found home a lonely and frightening place to be
sick. Views were shaped by social support, self reliance, religion, and past
illness experience. However, when exploring preferences for home or
hospital in terminal care, Fried (75) also found that preferences were
shaped by concerns about being a burden and the family’s ability to provide
care. Concerns about long term care resulted in preferences for a nursing
home when choice was not constrained to either home or hospital.
Place of care: The studies provided insight into several aspects that were
experienced or evaluated as appropriate care at the end of life for people
living with and dying from COPD. Issues included desire to be with family
members (147) and availability of support (140); concerns about being a
burden (75), the need for information (139), providing continuity (141);
the availability of GP’s (142) and receiving care from one’s own GP. As well
as this, communication and information (139), and a desire for a place
associated with survival (148) were important. In addition, patients who
were afraid of suffocating while dying (149) preferred to be in hospital
surrounded by family and friends. Improving assistance and promoting
independence were seen as important to prevent additional suffering and
reduce exacerbations requiring hospitalisation (147).
Place of death: No studies were primarily concerned with place of death.
Transitions: There were two papers (145) (146) which specifically related
to transitions. In one study, Reinke et al focused on participants’
experience of transitions. They defined transitions as ‘experiences that
patients and family members viewed as milestones in the evolution of their
illnesses and therapies’. They found transitions were concerned primarily
with new or different treatments and having no more treatments available.
Themes unique to patients with COPD were activity limitations due to
functional decline and initiation of oxygen therapy. A theme that was
unique to the clinicians was acute exacerbation of illness or hospitalization.
In a further study of factors influencing contact with general practitioners
(and hence potential transitions during exacerbations) (146) patients
contacted general practices for routine, urgent and emergency care, but
factors other than need influenced patterns of help-seeking. Some patients,
for instance, identified that they avoided bothering the doctor or found
travelling to the surgery too difficult; ease of access or relationship with the
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 166
GP drove contacts as much as need, and sometimes delayed seeking help
(146).
13.3 Long term neurological conditions
The nature of the evidence: There were six qualitative studies
investigating concerns and experience relating to people with progressive
advanced neurological disease. There were three from the UK (150-152),
and one from Sweden (153), Netherlands (154), and Australia(155). The
quality of the evidence is reported in APPENDIX 5. The studies scored
consistently high, reflecting their methodological rigour and usefulness for
practice.
Progressive neurological disease represented in this qualitative synthesis
are multiple sclerosis (MS) (150, 154), motor neurone disease (MND) (151-
153), and muscular dystrophy (155). The focus of the studies was mainly
on people’s experience of living with the advanced stage of progressive
disease. The studies had differing foci, such as how people accommodate
to their illness in the advanced stage of MS (154); how people cope with
MS in the advanced stage of their illness (150), and the differing views of
patients and carers managing at home with ALS (152, 153). In addition,
one study explored the experience of MND, in order to influence policy and
practice (151). Finally, the potential role for palliative care services for
people with muscular dystrophy was examined (155). The studies were
either from the carer perspective (152, 155) or both patients and carers
and (150, 151, 153, 154). None were solely from the patients’ perspective.
One study incorporated patients, carer and health professional views (151).
The majority of the qualitative studies collected their data through face to
face interviews (five). There was one mixed method study using interviews
with a subset from a larger correlational design using patient/carer dyads.
Context: All the studies investigated issues related to home. None were
primarily concerned with hospital or hospice based care.
Home care: Boeije et al explored the concept of biographical disruption in
relation to MS, offering insights into the way people with advanced MS
accommodate to their illness (154). They found that MS was an exhausting
illness for patients and family members with far reaching psychosocial
consequences for both patients and family members, with implications for
the care of people at home in the advanced stages of their illness. In
contrast, Bolmsjo et al found differences between patients and carers about
how people with ALS were cared for at home(153). The differences were
related to how they both perceive their needs, and how they view, judge
and evaluate the illness; as well as the process of the illness. They suggest
patients and their close family members should be viewed as individuals
with their own preferences. Caregivers, in particular, have specific needs
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 167
for support and information. The need for better information and
communication for people affected by advanced MND was also
recommended in the findings of Hughes et al. Nolan et al identified the
importance of shared decision making related to end of life decisions for
people with ALS dying at home (152). Edmonds et al found that people
living with advanced MS at home requires specific emotional support, in
order to deal with the physical and psychosocial losses associated with their
illness (150).
Preferences: Patients and carers views were shown to differ in relation to
perceptions and preferences about care (153). For instance, patients
perceived information about their illness was sufficient although carers
judged the information received was insufficient. Perceptions about the
future were perceived differently with relatives preferring not to think about
the future whereas patients were preparing for death. Preferences
regarding advanced directives were also perceived differently for carers and
patients (152). Carers also reported a lack of coordination in care and
access to skilled and competent carers (155). In addition patients indicated
that they were unsure about service entitlements and accessing services
(151). This recurring inconsistency should be noted and specific concerns
for both people with advanced neurological illness and their family
members assessed and managed appropriately.
Place of care: The majority of studies were related to experiences of care
at home. The studies provided insight into several aspects that were
experienced or evaluated as appropriate care at the end of life for people
living with and dying from advanced neurological disease. Patients and
carers interpreted MS as an all-encompassing illness and emphasized the
process of giving up everything (154).
Place of death: No studies were primarily concerned with place of death.
Transitions: People severely affected by MS identified concerns with loss
and change which evolved throughout the course of their illness and are as
much an issue for those severely affected as for those just diagnosed
(150). This may be a characteristic unique to those suffering advanced
progressive neurological disease, since these are illnesses which are most
specifically characterised by both physical and psychosocial loss of function
and impairment (154).
In addition, people living with advanced MS at home accommodate to their
illness by either by putting their lives back together, adapting and learning
new skills, or conversely retreating or refusing to consider MS as a part of
their lives. The unpredictable course of MS makes it challenging to give new
direction to the life course (154). The heavy demands placed on the
biographical work of people with advanced MS can be explained by the fact
that the illness touches upon all concepts of body as well as shattering the
connection between them. This again may be a finding specific to living
with and dying from advanced neurological disease.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 168
13.4 End stage kidney disease
The nature of the evidence: There were five qualitative studies about
end stage renal disease (156-160). Three studies came from New Zealand
(158-160), one from Australia (157) and one from the USA (156). Four
studies focused on patient experiences (157-160) and one study examined
the views of patients, carers and health professionals (156).
Types of studies: Three studies from the same author described the
characteristic attitudes of patients towards treatment regimen among a
group living on home haemodialysis (158-160). Patients experimented with
the use of dialysis to better accommodate it to their lifestyles and to
minimize symptoms caused by the treatment which diverted from the
prescribed ways of treatment. They did not communicate these practices to
their physicians.
Another study explored daily activities and challenges of home
haemodialysis for four older patients (157). This found that bringing the
treatment home facilitated the development of independent self-
management, for which spousal and professional partnerships were
essential. Patients described a successful relationship with health
professionals as a changed relationship where they received less direct care
and more support and advice.
Berzoff et al reported on the first part of The Renal Palliative Care Initiative,
a demonstration project that conducted research aimed at integrating
nephrology and palliative medicine (156). They held six focus group
discussions with patients, staff, carers and bereaved carers about their
experiences and thoughts regarding the development of Renal Supportive
Care Team (a multidisciplinary team designed to provide education and
support to severely ill dialysis patients and their families). Participants
agreed that there needed to be greater education of both patients and
families regarding all aspects of the disease process, open communication,
on-going support between patients, families, and the staff, continuity of
care, pain control, and assistance with advance care planning.
The studies explored living with end stage kidney disease from the patients
perspective in four of the studies (157-160) and from the perspective of all
involved in the caring process in one study (156). The main conclusion was
that little is known about care needs of patients with this condition and that
palliative care for this patient group is underdeveloped. Communication,
support, education, and continuity of care are all issues that need to be
better addressed.
Transitions: Two studies informed about important transitions in the
illness trajectory of end-stage kidney disease (156, 160). Polaschek et al
described patients’ experiences when they started with treatment on
dialysis, which was initiated because of troubling symptoms. A process of
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 169
normalization was experienced although they had concerns about
dependency and chronicity (160).
Berzoff et al found that patients and carers had different views about when
to introduce end of life issues (156). Carers thought that hospice and
advance care planning needed to be discussed as early as possible while
patients often needed to come to terms with the uncertainty of their illness
in a gradual way. A supportive care team first need to assess patients and
relatives readiness to receive information, and this needs to be done in a
culturally competent way.
Preferences and place of care or death: There was no information in
the identified studies specifically on preferences for care, or on place of
death.
13.5 Older people, including stroke and dementia
The nature of the evidence: There were 23 studies investigating
concerns relating to older people. Most of these came from the USA (75,
76, 161-172), four from the UK (173-176), two from Japan (177, 178), two
from Canada (179, 180), and one from Taiwan (182).
Among these studies, seven focused on people with dementia (161, 165,
166, 168, 174, 176, 178) and another group was directed to older people
suffering from different chronic illnesses (75, 76, 162-164, 167, 170-173,
175, 177, 179, 180, 182). One study had a majority of dementia patients
but also included patients with Parkinsons Disease (169). One study
compared a group of patients with and without dementia (167). The studies
focusing on dementia were conducted with family members who were
informal carers, one study also included bereaved carers (178), and one
study combined informal and professional carers (174). There was one
study with bereaved carers (176) and one with hospice staff (168).
In the studies on older people, the participants were patients with chronic
illness receiving treatment from an institution (75, 76, 170, 177). One
study focused specifically on patients affected by stroke (175). Patients
were combined with other study participants in three studies (175, 179,
182). Two studies used health professionals perspectives (172, 179). Two
studies sought the views of managers of care organisations (171, 173). The
studies worked with carers in seven of the studies (162, 164, 167, 175,
179, 180, 182), of which three were bereaved carers (162, 164, 180).
The studies with dementia patients were all in the advanced stages of
illness. The majority of studies concerned with older people also had
advanced conditions. One study specified their subjects as ‘ill patients, but
well enough to come to clinic’ (170), and in three other studies the stage of
illness was not mentioned (177, 179, 182).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 170
There were fifteen qualitative studies which collected their data with
interviews and or focus group discussions. Two studies used ethnographic
methods with participant observation, in combination with the other
methods of interviews and focus group discussions (180) and follow-up
surveys (172). There were eight mixed method studies using a
questionnaire plus interviews (168), chart review and interviews (167), and
three studies were conducted with questionnaires, of which the open
questions were analysed with qualitative approaches (173, 176, 182). One
study was a secondary analysis of data obtained in a larger intervention
study where interview data of a sub-sample were combined with
quantitative data (169).
Types of studies: The studies investigated issues related to a few areas:
1. Hospice care, 2. home-based care, 3. entry into a care home, 4. decision
making regarding the end of life, and 5. relocation.
Hospice care: Two studies on dementia patients were concerned with
access to hospices in the USA (162, 168). McCarty et al (168) aimed to
investigate what promotes and prevents involvement of individuals with
dementia in a hospice program. They found that facilitators for utilization of
hospice for Alzheimers patients were the availability of in-patient or
residential hospice facilities, and alternative services for patients who are
not eligible by Medicare criteria. An important barrier to access was the
prognosis of a six month survival prognosis required by Medicare
regulations, due to the uncertain trajectory of dementia. This study also
discussed financial concerns of families and institutional barriers, such as
the lack of expertise regarding dementia and misunderstandings regarding
the purpose of palliative care. Access to hospice care in the USA is not only
problematic for dementia patients but also, to a lesser extent, for older
people in general. The physician who acts as gatekeeper is often the person
who initiates the idea of hospice where families could benefit from palliative
care. However, the physician has little knowledge about what hospice care
has to offer and patients learn this more often from hospice representatives
or experience previously unexpected benefits in the brief time the patient is
finally admitted.
The two studies of Waldrop and Kirkendall (171) and Waldrop and Rinfrette
(172) provide information about transition issues relating to hospice in the
USA. The first study (171) explored the interrelationship between location
(rural or urban) and the type of agency (hospice or CHHA-hospice) and
described how it influences the experiences of older persons and their
carers at the end of life. It showed that the challenges with transitions
resulted from Medicare guidelines that homecare and hospice care are
separate and mutually exclusive. The second study (172), while enquiring
about hospice professionals views on the appropriate timing for and
communication about hospice, clarified once again that flexibility in
admissions is required depending on the circumstances of patients and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 171
families and that the current structure of hospice care can lead to
unnecessary problems regarding transitions.
Older hospice patients from minority ethnic groups are likely to underuse
hospice. Chung et al targeted caregivers of this group to examine the level
of knowledge about hospice and how it influenced enrolment decisions and
the use of hospice services after enrolment (163). They found that better
knowledge increased access to hospice care and that it may enable some
patients to enrol early enough to benefit from comprehensive care.
Home care: Evans et al (164) undertook a study with caregivers of
deceased hospice patients to gain an understanding of the reasons why
patients transfer from home-based palliative care to acute care, and what
their experiences were. They found that patients who transferred to
facilities while receiving home hospice care transferred because of an acute
medical event, an uncontrolled symptom, imminent death, or inability to
safely provide care at home. All caregivers expressed a strong preference
for care at home, but they saw other concerns, such as symptom control,
safety, quality and also quantity of life as more important than staying at
home.
One evaluation of a home-based palliative care programmes was an urban
programme to meet unmet needs for groups of older people likely to be
disadvantaged, such as African-American people and dementia patients
(167). This evaluation from the United States showed that quality end of
life care can be provided for dementia patients within the structure of a
home palliative care program (167). Another study also addressed whether
care at home is possible for patients with dementia patients and found that
specialist support was needed with a key person to coordinate care and out
of hours services to avoid hospital admission (176). A caregiving
intervention study focused specifically on a subsample of participants who
decided during the intervention to institutionalize their relative with
Alzheimer or Parkinsons Disease. Some carers did not attain higher levels
of caregiving skills needed when patients conditions worsened over time,
often due to their own health limitations (169).
Stadjuhar et al (180) explored the variations in and factors influencing
family members’ decisions for palliative home care for older persons. They
found that these were often based on uninformed, or indifferent decisions.
(An ‘indifferent’ decision was characterized by situations where caregivers
felt they had little choice in the decision for home care.) Those people who
made decisions that were the result of open discussion in the family and
who were motivated by keeping the person in an environment where
‘normal’ life happened, were eventually best able to cope.
Brown et al (179) identified barriers and facilitators to independence as
experienced by older people with chronic health problems as they
interacted with services. They found that apart from service accessibility,
the attributes and attitudes of people were important. Specific attitudes,
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 172
such as ‘fighting back’ to challenges (as adopted by older people) and
positive attitudes towards aging (as adopted by professionals) were crucial.
Entry into care homes: Two studies addressed issues regarding entry into
a care home (173, 182). The study from the UK described the
characteristics of end of life care in care homes from the home manager’s
perspective informing the development of appropriate end of life care in the
homes (173). There was a range of conditions present in the care homes,
for which they provided care at the end of life with 15% of people dying
from cancer and 85% from other conditions, and this had implications for
the type of care that needed to be provided. A longer-term understanding
of the end of life period needs to be adopted for an older person at a care
home, and this could include all the time they live in the care home. The
other study explored the decision-making process of elderly people who
entered nursing homes and that of their carers/key families in Taiwan
(182). The decisions were taken mainly within a family context. Care was
mostly provided at home by the family in Taiwan, but there were
circumstances that placement in a nursing home became inevitable.
Considerable anxiety existed about the quality of care in nursing homes.
End of life decision making: Waldrop & Kirkendal’s study (171) elicited
descriptions from managers and staff of home and hospice organizations of
local and organizational factors that shape end of life care provision. There
were differences between rural and urban areas in end of life care provision
due to geographical, market forces, programming issues and fee-
structures. This had implications for social work practice. A further study,
by Gessert et al (166), explored differences in attitudes towards the end of
life between rural and urban families regarding their cognitively impaired,
institutionalized loved ones, and this had implications for the way decisions
were made. As rural participants were more accepting of death they were
less in favour of interventions that prolong dying. Urban families on the
other hand saw their role in ensuring that their relative was not abandoned
or neglected and some insisted on aggressive medical care in advanced
dementia.
Black et al (161) undertook a study to find out how surrogate decision
makers for advanced dementia patients developed an understanding of
patient end of life preferences and wishes. They found some intrinsic and
extrinsic catalysts for discussions and advance directives, and one led to
the other which led to greater specificity. Also Forbes et al (165) addressed
families’ decision making processes regarding end of life treatments for
nursing home residents with dementia. Nursing home placement was the
most difficult decision ever made, only done when all other options were
exhausted. Its strongest theme was the emotional burden of end of life
decision making, on issues they had little knowledge about. Decisions were
made ad hoc, without the necessary planning or consideration. Wakunami
et al (178) showed the complex process by which Japanese family
members come to accept a patient’s condition as terminal and can only
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 173
then proceed to make well-considered decisions regarding the end of life
phase.
One study provided insight in patients’ and family members’ experiences of
acute stroke and their preferences for end of life care (175). The study
highlighted their unmet needs for care and support. As stroke was not
generally associated with the end of life, patients care consisted of active
acute care and rehabilitation, although they could have benefited from
palliative care. This also impacted on people’s problems with the shift to a
more supportive approach when the patient’s situation deteriorated.
Four studies investigated the wishes and attitudes of older people regarding
the end of life (75) (76, 170, 177). Hattori et al (177) found that the
wishes of older people who were hospitalized were varied, depending on
several factors. The main influences were patients’ ability to make
decisions, and the family; and wishes changed, even in the short time
during interview. Vig et al (170) examined the relation between actual
preferences and patients values and they found a heterogeneity of views.
They cautioned not to infer end of life preferences from patients’ values but
to develop a thorough understanding of wishes and the underlying reasons.
Fried et al (75) focused more specifically on perceptions of home and
hospital care and found that all respondents preferred the site associated
with greatest survival. Fried et al (76) later reported that one third of
patients changed their preference when asked about their preference in the
event of a terminal illness. The desire to be with family members and
concerns about burden to family and their ability to provide necessary care
then started to play a role.
Relocation: Lee et al (174) documented the process of relocation of
elderly dementia patients from a long stay psychiatric ward to a purpose-
built nursing home resulting from the UK mixed economy of welfare
measures where many services, including continuing care for older people
are moved to the independent sector. This can have serious consequences
for confused elderly people who need to adapt to a completely new
environment. Careful preparation was needed in collaboration with staff to
negotiate the new working conditions and to be able to work in a different
culture with a new set of values.
13.5.1 Outcomes: place of care and death, preferences, and
transitions
The studies were ordered according to their main objective: there were
nine studies that focused on place of care, five on preferences, and nine on
transitions. No studies addressed place of death. However, most of the
studies provided information about several of these areas of interest, as
these were often related.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 174
Preferences: Preferences regarding end of life care were found to be
highly variable across older patients with chronic health problems (75, 76,
170, 177) and to change according to circumstances (76, 164, 170, 177).
They were also different to the stage of illness, as preferences changed
according to whether terminal care was needed (76). A preference for
home care at the end of life could be found in eight studies (164, 165, 167,
176, 177, 180-182). This is however a complex issue, as this wish is
associated with values, emotions, and dependent on circumstances.
Attempting to reflect this preference by taking into account the number of
studies that present these data does not reflect the detailed and complex
qualifications these studies reveal regarding this preference. The ideal of
home care also appeared through negative evidence of the guilt felt by
families when home care was not possible (165, 177, 180, 182) or its
motivation to avoid institutional care (176, 180).
The wish for home care was balanced with the conditions older people lived
in. The studies showed that the ill person had to have someone who was
willing and able to take on the caring role at home (76, 170). It depended
on whether the older people felt safe at home and whether they did not
perceive their care as too heavy a burden for the informal carers(76).The
preference for home care appeared to be less important than the goals of
care that patients and families aimed to accomplish, consistent with the
values they held (164).
Place of care: The studies provided insight into several aspects that were
experienced or evaluated as appropriate care at the end of life for older
people. A recurrent theme was continuity regarding care. Services that
were provided by several institutions were difficult to obtain in case of
home-bound older people (167). The different skills or support available,
spread over separate places and professions, needed adequate information
giving and called for coordination by a key person who could provide advice
(175, 179). System rigidity and the structure of care that was not
consistent with patients and families wishes and purposes worsened
experiences at the end of life (171, 179). Coordination of care reduced the
workload for informal carers considerably and provided valued time for
carers to spend with patients (162).
Communication was another key ingredient for good care. Understanding
needed to be built on early in the disease trajectory between the different
people involved, the patient, family and health professionals (175), and
ideally before transitions into institutions were made (173). This was
especially crucial in the case of a diagnosis of dementia, and discussions as
well as the making of advance directives were necessary so that the carer
could later act as a proxy decision maker for the patient (161, 170, 178,
179). Shared collaborative decision making was recommended (76, 170,
172, 177, 178), also to relieve the proxy of the emotional burden of difficult
end of life decisions (165).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 175
The promotion of independence for older people in the community was
found to enhance the quality of life for older people (179, 182) and the
availability of a variety of settings that accommodated older patients needs
was important (76).
Place of death: Vig et al (170) only identified a few consistent themes in
patients concerns about death. A good death was one without pain, quick,
without suffering or knowledge of impending death. The opposites were
identified in descriptions of good deaths. However there was more
heterogeneity in patients’ views about what constituted a good or a bad
death. Participants also did not have a consistent preference for the
location of death. Fried et al (76) presented more patterned results on
patients’ preference for the place of death in case of terminal illness. Carers
expressed the wish to ensure patients experienced a peaceful and dignified
death (165, 177).
Death would preferably take place in familiar surroundings where people
knew the patient and were aware of their wishes. Waldrop et al (172) found
that the time patients generally receive hospice care, is too short for staff
to get to know a patient. Hospital admittance when patients are dying is
also not conducive to good deaths and this can be avoided by specialist
out-of-hours advice and support.
Transitions: Lee et al (174) showed that transfers in place of care have
potential negative effects for older patients and that these could be life
threatening. Transitions in care are usually seen as undesirable, inevitable
events, such as the placement of an older relative in a nursing home which
often goes together with negative feelings as guilt or failure to provide good
care (165, 169, 177, 182). However, transfers are actively orchestrated by
carers when they are considered the best care option in line with the goals
of care, and then they are used as alternative options that exceed the
existing possibilities in the usual place of care (163, 164).
Transitions need to be smooth (167, 168). This implies early referrals to
places or types of care that are attuned to the specific condition of the
person (167). Smooth transitions also mean that they are timely (172),
personalised (164), with previous goals and relationships continued (167,
171) and in collaborative agreement between those who are able to direct
the desired care (178). A negative example is provided by Waldrop &
Rinfrette (172) where bureaucratic structures form barriers to good end of
life care in the USA by separating home care facilities from hospice care,
which excludes patients either from one or the other. This then results in
ruptures or breaks in transitions, rather than natural and smooth
transitions in care.
A lack of knowledge and awareness by those affected and health
professionals, about the end of life care needs of older people, patients with
dementia or stroke, lead to missed opportunities to coming to terms with
the prospects of living with chronic illness or approaching death and result
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 176
in difficult transitions (165, 166, 168, 175, 178). A high level of knowledge
about hospice care increased access to hospice and positively influenced
care after enrolment (163).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 177
14 Results: transitions in care towards end of life
Table 22. The definition of ‘transition’, as adopted for this review
The qualitative evidence presented in the previous chapter identified some
of the important transitions reported by patients with advanced illness and
their carers towards end of life, such as:
Social and psychological decline, alongside physical deterioration (135)
Transitions when new or different treatments commenced, or into a more
palliative phase when no further active treatments were available (145)
Entry into care homes or hospital (173, 182)
Other forms of relocation, especially for older people with advanced
conditions, including from hospital into nursing home (174), from home
into nursing or residential care (165, 169, 177, 182), or into hospice care
(172)
Those with ESKD also experienced the specific transition of starting
dialysis (156, 160)
In contrast to the qualitative evidence, the quantitative evidence on
transitions focused almost exclusively on transitions in settings, rather than
transitions in the focus of care. We have therefore grouped the evidence
according to specific types of transition in setting:
Transition into hospital
Transition into nursing homes
Transition back to home
Transition into hospice
50 included papers undertook multi-variate analysis of factors associated
with transitions towards the end of life, and are summarized in Table 23.
“a change in the setting or place of care, a change in the focus of
care, or a distinct change/movement in the patient’s emotional or
psychological journey”
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 171
Table 23. Included papers on transitions towards end of life in non malignant conditions
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Ahmed. 2003
USA retrospective cohort study
Medicare beneficiaries medical records
Regional CHF patients discharged from hospital
65+ 79 (7.5) CHF 908 4 logistic regression
50% medium
Albert. 1999:
USA prospective cohort study
longitudinal participant data and hospital records
acute hospital
Subjects from the Washington Heights-Inwood Columbia Aging Project living in the community
65+ Not given all diagnoses: Alzheimer’s disease 20% other 80%
1034 6 logistic regression
36% medium
Anderson. 1993
USA prospective cohort study
Data from dialysis centres
Regional ESRD patients living in Nursing homes that were receiving dialysis
17+ 65.5 (14.2) ESKD 228 3 linear regression
36% medium
Arling. 2010 USA retrospective cohort study
secondary data from minimum data set
nursing home
nursing home residents
not specified 77.3 (12.7) all diagnoses: Cancer 14% Dementia 16% Depression 24% Diabetes 26% End stage disease 3%
24648 12 logistic regression
71% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 179
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Arora. 2000 USA prospective cohort study
Mecial records and dialysis centre database.
Medical unit providing Dialysis
ESRD patients newly starting dialysis (1st 3 months of starting ) and ESRD patients hainv g had at least 3months of dialysis
18+ 56 ESKD 100% Heart Disease 64% CHF 70%, Cancer 7.5% Other 39%
128 8 logistic regression
64% medium
Bain. 2009 USA Retrospective case-comparison study
Data from a National hospice pharmacy provider
Hospices served by the national pharmacy provider
CHF hospice patients compared to Cancer hospice patients
all ages 85.0(SD 9.2) CHF 27% Cancer 73%
42982 8 logistic regression
71%high
Becker. 2009
USA retrospective cohort study
Medicaid data of nursing home residents
Regional Medicaid nursing home provision and hospitalisations
nursing home residents
all ages 77.3 (13.8) all diagnoses: Alzheimers disease 15.1% Other dementias 26.4% Other 23.6%
72251 17 logistic regression
79% high
Biola. 2010 USA cross-sectional regional survey
after-death interview with caregivers
27 nursing homes and 85 residential care/assisted living setting in 4 states
both family and staff caregivers
All ages/non specified
86.6 (9.9) all diagnoses: overall proportions not given
654 (caregivers) 327 LTC residents
17 logistic regression
44% medium
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 180
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Bradley. 2000
USA cross-sectional regional survey
proxy data from questionnaires completed by hospital physicians
Acute hospitals
Physicians that had admitted at least 5 patients to the hospices in the previous year
All ages/non specified
Not given Not applicable 0 5 logistic regression
64% medium
Brody. 2010 USA case-control study
Hospital medical records
acute hospital
Patients saw the inpatient palliative care team (PCT). Matched cohort of patients that did not have PCT input.
18+ 68.5 all diagnoses: Cancer 36.3% circulatory 10.8%, respiratory 6.7% digestive system 8.0% other 38.2%
722 9 linear regression
86% high
Casarett. 2001
USA retrospective cohort study
hospice program providers databases
hospice All admissions to one hospice program over a 2 year period
<65 26.6% 65-79 36.7% >79 36.7%
Not given all diagnoses: 76% Cancer, 24% Non cancer (unspecified)
1691 88 linear regression *RR,HR and CI given
64%medium
Casarett. 2002
USA retrospective cohort study
medical records Veterans administration medical center
Patients referred to a palliative care clinic
18+ 68.2 (41-88)
all diagnoses: Cancer 85%, COPD 2% CHF 1% Dementia 8% Stroke 1% ESKD 1% other 2%
100 5 linear regression ** HR and CI given
43% medium
Chen & Narsavage 2006
Taiwan prospective cohort study
longitudinal participant data
Regional Patients hospitalised with COPD as the primary or secondary diagnosis
All ages/non specified 65yrs<81.4%
Not given COPD 145 9 logistic regression
86% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 181
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Dai YT. 2002 Taiwan prospective cohort study
longitudinal participant data
Regional Patients discharged from hospitals with progressive chronic diseases or chronic impairments following an acute episode
≤ 64 23.1%, > 65 76.9%
71 (SD not given)
all diagnoses: Stroke 31.7%, traumatic brain injury 8.9% and hip fracture 22.3%
350 8 linear regression
57% medium
Factor. 2003 USA retrospective longitudinal cohort study
longitudinal participant data from the PSYCLOPS(PSychosis and CLOzapine in PD study)
Regional population dataset
Patients with Parkinsons disease enrolled in a clinical trial
All ages/non specified
71.3(SD 8.2) Parkinson’s Disease 100% Dementia 56%
59 2 logistic regression
57%medium
Fillenbaum. 2001
USA prospective cohort study
Secondary analysis of research data from CERAD(Consortium to establish a Registry for Alzheimer's disease)
national population dataset
Alzheimer's disease patients living at home or in institutions
All ages/non specified
<75 54% ≥75 46%
Alzheimer's disease 420 4 logistic regression
57%medium
Fried. 1995: USA retrospective cohort study
Infection control data and medical records
single nursing home
nursing home residents who had an episode of pneumonia
All ages/non specified
88.8 (range 72-101)
Dementia 17% 316 2 logistic regression
71% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 182
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Fried. 1997 USA retrospective cohort study
administrative data plus nursing home records
Long Term Care Facilities
Nursing Home residents that had had no transitions outside of the facility during the first 6 months residence
65+ 83(7.3) all diagnoses: CHF 56%, Cancer 57%, respiratory disorder 61%, dementia 42%
3782 9 logistic regression
71%high
Gessert. 2006
USA retrospective cohort study
Medicare and Medicaid administrative data
Long Term Care Facilities in rural and urban settings
Nursing Home residents with severe dementia
65+ 86.7(7.7) all diagnoses: Dementia 100% Alzheimer’s disease 31% Stroke 32.7%
3710 4 logistic regression
79% high
Gielen B. 2010
Belgium cross-sectional national survey
Administrative data of one sickness fund
national population dataset
Patients without cancer who were sickness fund members that died 2005-6
Age>40 41-59 8.8% 60-69 10% 70-79 25% 80-89 36% 90+ 18%
all diagnoses: Cancer 27.5%, non cancer 72.5% (non specified)
29575 1 linear regression
64%medium
Givens. 2010 USA retrospective cohort study
Medicare claims data
national population dataset
CHF medicare beneficiaries with no previous hospice admissions
All ages/non specified
79.9 CHF 98258 14 logistic regression
86%high
Haupt 1993 Germany prospective cohort study
longitudinal participant data obtained from outpatient departments
Regional Alzheimer's disease patients living at home followed for 1 year
All ages/non specified
73 (56-89) Alzheimer's disease 66 7 linear regression
71%high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 183
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Hauptman. 2007
USA cross-sectional regional survey
Data from the Acute Decompensated Heart Failure Registry
regional acute hospitals
CHF patients discharged to hospice or community
18+ Not given all diagnoses: CHF 100% Cancer 12.7% Hypertension 66.5%, Stroke 20.8% Diabetes 37.9% COPD 36% other 83.1%
182898 8 linear regression
50% medium
Hebert. 2001
Canada prospective cohort study
Longitudinal proxy data from caregivers interviewed
Country Dementia patients living in the community
All ages/non specified
Not given Dementia 326 15 logistic regression
71% high
Krumholz. 1997
USA retrospective cohort study
Medicare database regional
regional CHF patients discharged from hospitals
65+ 65-74 30% 75-84 43% >85 25%
CHF 17448 7 logistic regression
79% high
Kwak 2008
USA retrospective cohort study
Medicaid claims data and death registration data
Medicare and Medicaid nursing home provision
nursing home residents that died between 2000-2
65+ 85.9 (8.1) all diagnoses: cancer 5.4% dementia 14.9% heart disease 36.7% other 43.1%
30765 4 logistic regression
79% high
Lieve Van den Block. 2007
Belgium retrospective cohort study
death registration data from national network of general practitioners.
Country All patients that died over a given period
All ages Not given all diagnoses: Cancer 39%, cardiovascular disease 26%, nervous system diseases 4%. other 30%
319 4 logistic regression
79% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 184
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Lim. 2009
Singapore retrospective randomised case-control study
Once fortnightly clinical reviews and multidisciplinary meetings
acute hospital
Patients being considered for discharge from a community rehab hospital
All ages/non specified
<60 (10%), 61-74 (22%) >75(68%)
all diagnoses: CVA 46% other 54%
300 10 logistic regression
71% high
Marengoni. 2008
Italy Prospective cross-sectional survey
Inpatient assessment including medical, functional and neuropsychological assessments.
acute hospital
Patients discharged from a hospital
65+ 78.5 (7.2) all diagnoses: heart diseases 63%, hypertension55% lung diseases 34% cerebrovascular diseases 31% other 78%
830 10 logistic regression
79% high
Mehr. 1997 USA Prospective cross-sectional survey
national nursing home provider data
National group of nursing homes
nursing home residents
65+ Not given all diagnoses: cancer 34.1% dementia 24.1% cardiovascular disease 67.5% cerebrovascular disease 36.6% COPD 37.4% hypertension 37.4% other 61.3%
5895 11 logistic regression
79% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 185
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Menec 2009
Canada retrospective cohort study
administrative data plus nursing home and hospital databases
long term care facilities
all long term care residents from participating facilities
≤ 74 9.2% 75-84 9.5% 85-94 48.2% 95+ 13.1%
All ages/non specified
all diagnoses: frailty trajectory 41.5% organ failure trajectory 39.7% terminal illness trajectory 10.1% sudden death or other trajectory 8.6%
2379 7 logistic regression
86% high
Miller. 2001 USA retrospective cohort study regional
Medicare claims data and minimum data set assessments
long term care facilities
nursing home residents, two groups one enrolled in the hospice(9202) and (27500) not.
All ages/non specified
83.7(9.3) all diagnoses: Cancer 59%, COPD 18%, CHF 36%, Dementia 16% other 25%
36702 10 logistic regression
79% high
Mitchell. 2004
USA retrospective cohort study
Minimum data sets for long term care facilities and community dwelling patients
regional population dataset
AD patients from nursing homes and the community (terminal care patients)
All ages/non specified
83.2 Dementia 3020 5 logistic regression
79% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 186
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Morcillo. 2005
Spain Randomised control trial
Single-blinded patient assessments performed in clinic, via medical records and telephone consultations
acute hospital
CHF patients enrolled in an intervention to reduce readmissions and healthcare costs
65+ 77.7(5.85) CHF 70 5 logistic regression
70% high
Murray 2006
USA retrospective cohort study national
United States Renal, Medicare and Medicaid databases
national population dataset
ESRD- Dialysis patients that died between 2001-2
All ages/non specified
73.4 (11.0) ESKD 115239 9 logistic regression
71% high
Nihtila. 2007 Finland prospective cohort study national
Population registration database of socio-demographic, including health and welfare data
national population dataset
People over 65yrs living in the community
65+ 72.6(6.1) all diagnoses: Dementia 55%, Parkinson’s disease 30% Hip fracture 32%
108474 18 logistic regression**HR and CI alone
86% high
Parashos 2002.
USA Historical cohort study
medical records regional Parkinson’s disease patients and matched non-sufferers
All ages/non specified
(48-88)(median73)
Parkinson’s Disease 50%
178 28 logistic regression * RR+CI alone
71% high
Qartararo. 1991
Australia Randomised control trial
longitudinal participant data
regional Nursing home applicants in a regional sample
All ages/non specified
81.6 (7.6) all diagnoses: Stroke 22.3% Dementia 34%, Cardiorespiratory disease 31% other 27.4%
296 4 logistic regression
70% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 187
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Rich. 1995 USA Randomised control trial
longitudinal participant data
acute hospital
CHF patients recently admitted under that diagnosis
70+ 79.25(6) CHF 282 (142 treatment, 140 control)
5 logistic regression
75% high
Rockwood. 1996
Canada cross-sectional national survey
longitudinal participant data
Country Community dwelling and institutionalised older adults
65+ Not given all diagnoses: hypertension 38.15%; Heart disease 26.3%, CVA14.9%, Parkinson’s disease 3.6%, other 86.65%
10371 11 logistic regression
86% high
Rosenberg. 2005
USA prospective cohort study
longitudinal participant data
Regional Residents from assisted living facilities
All ages/non specified
85.7 (8.3) all diagnoses: Hip fracture 9.9%, Chronic pain 1.6%
192 5 logistic regression
86% high
Rosenwax. 2008
Australia retrospective cohort study
death registrations and linked health care databases
regional population dataset
deaths from Alzheimer's disease compared with other causes
75+ 75-84 35.2% >85 64.8%
all diagnoses: Alzheimer’s disease 3.7%
15903 4 logistic regression
71% high
Rundek. 2000
USA prospective cohort study
longitudinal participant data from NOMASS(Northern Manhattan Stroke Study) - Secondary analysis
regional CVA patients discharged to various locations
40+ 70(12) all diagnoses: CVA 100%, cardiac disease 43%, cognitive impairment 21%, motor deficit 78%
893 10 logistic regression
86% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 188
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Salive. 1993 USA prospective cohort study
longitudinal participant data derived from epidemiologic studies
regional population dataset
Community dwelling older adults
65+ Not given all diagnoses: proportions not given
4074 7 logistic regression
79% high
Smith and Stevens. 2009
USA retrospective cohort study
Medical and billing records
Regional Patients discharged from hospital (2006)
60+ Not given all diagnoses: proportions not given
6006 11 logistic regression
79% high
Smith. 2001 USA prospective cohort study
longitudinal participant data from Dementia Patient registry
Region Dementia patients and matched cognitively un-impaired persons
65+ 80(6.2) Dementia 100%,
985 6 logistic regression
64% medium
Tsuchihashi 2001
Japan prospective cohort study
Medical records Regional hospitals
CHF patients discharged from hospital
All ages/non specified
69 (14) all diagnoses: CHF 100%, Diabetes 25% Stroke 14% Renal failure 11% Hypertension 41%
230 6 logistic regression
79% high
Wheelock. 2003
USA retrospective cohort study
Huntington Disease Patient database
International - Largest population 68.8% from USA
Huntington Disease patients living in nursing facilities and in communities
All ages/non specified
50 Huntington Disease
3070 3 logistic regression **HR and CI alone
57% medium
Williams. 2006
USA prospective longitudinal study
longitudinal participant data from clinical assessments
regional population dataset
AD and LBD patients assessed for mortality and time to Nursing home
All ages/non specified
Not given all diagnoses: Alzheimer’s disease 80%, Lewy Body Dementia 20%
315 2 logistic regression **HR and CI alone
79% high
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 189
Paper
Country
Design
Data collection
Study setting or context
Participants
Target age group
Mean age, yrs (SD) (proportions if mean not provided)
Disease group
N
No of factors
Type of analysis
Quality score
Yaffe. 2002 USA Randomised control trial
longitudinal participant data from Medicare records and clinical consultations
regional population dataset
Dementia patients living in the community
All ages/non specified
78.9(7.8) Dementia 5788 5 logistic regression
85% high
TOTAL PARTICIPANTS:
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 183
This quantitative evidence should be interpreted in the context of the
broader conception of transitions, as defined above (see Table 22), and also
bearing in mind a patient-centred perspective on transitions. For instance,
some patients (such as people severely affected by MS) experience loss and
change which evolves throughout their illness, rather than at a specific
transition point (150). Patients also report that they experience transition,
not as a linear or clearly defined process, (137) but as a complex and rather
iterative process (130).
14.1 Transition into hospital
Six studies report the relationship between age and transition into hospital
(113, 183-187).
Figure 110. Association between age and transition into hospital
Figure 110 illustrates the evidence on the association between age and
transition into hospital from home (185, 187), and nursing home (113, 183,
184). One study assessed whether palliative care provision in a nursing
home setting changed hospital admission (184), and a further study from
Belgium (186) assessed hospital admission from all settings. Although there
is no apparent relationship between age in general and hospital admission
(184-187), two studies which compare the odds of hospital admission in
younger residents as compared to older age residents, specifically among
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 191
nursing home residents (113, 183) identify a significant difference; this is
pronounced when comparison is made of those < 85 with those > 95 years
(183).
Figure 111. Association between functional status and transition into hospital
Three studies report the relationship between functional status and hospital
admission (113, 183, 188); one in relation to people with advanced disease
recently discharged from hospital and their odds of re-admission (188) and two
with respect to the re-admission of long term nursing home residents (113)
(183). No significant associations were demonstrated, with the exception of one
study (113) which showed association between good functional status (Level 1
care is independent or minimally dependent) and transition into hospital, when
compared to likelihood of admission for those with poor functional status and
high dependency (level 4 is maximal dependence) (see Figure 111).
14.2 Transition into nursing home
Six studies provide evidence on age and transition into nursing home care (189-
194). Care needs to be taken in interpreting these findings, since the studies
relate to markedly different populations, ranging from a general population (192,
194), a general hospitalised population (190), an older hospitalised population
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 192
(193), and specific disease populations: work by Factor et al relates to with
people with severe Parkinson’s disease and psychotic features for instance (189),
and Ahmed’s study relates to hospitalised heart failure patients (191).
Figure 112. Association between age and transition into nursing home
There is inconsistency of findings, although in general, age as a continuous
variable across all age groups has little association with nursing home admission,
while categorisation of age to compare older categories (>65, by Ramdek et al,
and > 75 by Lim) with younger ages tend to show a significant effect.
The association of dementia and transition into nursing home has been studied
by several authors; we have grouped these studies with those papers which
report cognitive impairment without specific diagnosis (see (see Figure 113
overleaf). This evidence relates to transition into nursing home from hospital
(190, 193, 194) and from home (192, 195, 196).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 193
Figure 113. Association between dementia or cognitive impairment and
transition into nursing home
There is high grade evidence that, regardless of whether the transition is from
hospital or home, dementia is strongly associated with nursing home admission,
with odds ratios between 2.62 and 29.1. The evidence is similar for cognitive
impairment, although only two studies report this, with wide confidence intervals
(but still significance at the 95% level) (see Figure 113).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 194
Figure 114. Association between pre-admission living situation and transition
into nursing home
Three studies report association between living alone prior to admission to
nursing home, and transition into nursing home care (190, 193, 196). These are
consistent in demonstrating significant association between living alone and
nursing home admission, although one study reports by ethnic group (196), with
no significance for those of white ethnic background, in contrast to those of black
ethnic background.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 195
Figure 115. Association between length of hospital stay and transition from
hospital into nursing home
Two out of three studies demonstrate an association between longer hospital
stay and increased likelihood of nursing home admission (191, 194, 197). Again,
these studies are consistent in demonstrating that longer hospital stay is
associated with nursing home admission, and the size of the odds ratio in part
may reflect whether length of stay has been treated as a continuous variable,
with small increase in odds per extra day of stay (OR 1.21) (191), or categorised
into longer versus shorter stays, and correspondingly larger odds ratios (OR
2.06-2.43) (194, 197).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 196
Figure 116. Association between pre-admission social support and transition
from hospital into nursing home
Four studies evaluate whether pre-admission social support is related to nursing
home transition (192, 193, 195, 198). Three studies are consistent in
demonstrating the association of a lack of caregiver at home with greater
likelihood of nursing home admission, although with differences in the size of this
association (OR varies from 1.3 – 8.3) (192, 195, 198). In the fourth study (193)
there is no significant association; this study looked at older people in a specific
geriatric hospital ward in Italy.
14.3 Transition back home
There is a limited amount of evidence on the transition back home from nursing
home (80) (199), and from hospice (200). These studies provide evidence that
age is not associated with increased likelihood of discharge home from these
settings (see Figure 117 overleaf), since all have non-significant findings in
relation to age for this transition.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 197
Figure 117. Association between age and transition back to home
14.4 Transition to palliative care
There is limited evidence on transition into palliative care, with only five
studies providing evidence on this (109, 197, 201, 202) (203). All these
studies are from the US, and so relate to the US understanding of hospice;
i.e. palliative care programmes which are largely, but not exclusively,
community based – some have hospice in-patient beds, although this is
often distinct from palliative care in a hospital setting. One study relates to
palliative care provision in a hospital context, and the effect of this on
discharge to hospice programmes (197). One further study also relates
specifically to the ESKD population (203).
Together, these studies provide information on two factors and their
relationship with transition into palliative care programmes; age and
ethnicity.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 198
Figure 118. Association between age and transition into palliative care
programmes
Four studies report age and transition in palliative care. This evidence
relates to likelihood of transfer into palliative care programmes while in the
community (202, 203), in hospital (201), and nursing home (109). Kwak et
al report no effect of age, whereas Givens report a small but significant
effect for age, as do Hauptman using ten-year increments to categorise age
(201). Murray, in the dialysis population, reports a marked effect of older
age (with reference to those <45 years) (203).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 199
Figure 119. Association between ethnicity and transition into palliative care
programmes
Finally, four studies report ethnicity and transition into palliative care
programmes. This are consistent in demonstrating a reduced likelihood of
transition into palliative care for those with black ethnicity, regardless of
whether this relates a heart failure population (202), those with ESKD
(203), or more general nursing home (109) or hospital populations (197).
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 200
15 Discussion
15.1 What is the prevalence of a home death preference?
In advanced non malignant conditions, weighted mean prevalence of
preference for home death is 42-48%, derived either direct from patients
themselves (weighted mean prevalence 42%, range 36-75%), or from
professional report (weighted mean prevalence 48%, range 30-69%). Data
from patient and professionals produced very similar weighted mean
prevalence and range, whereas families or close carers generally reported
home death prevalence at a higher level and with greater range (weighted
mean prevalence 66%, range 25-97%). Evidence derived from records gave
a higher prevalence still (weighted mean prevalence 79%, range 68-82%).
For this reasons, we have not pooled evidence from all four sources.
15.2 How does this preference vary by diagnosis?
There was little indication of variation in prevalence of preference for home
death by diagnosis, although insufficient disease-specific evidence to
elucidate this with much detail; only in ESKD was evidence sufficient to
calculate a weighted mean prevalence specific to ESKD of 41.7% (range
36.1-77.0%); this compares with a prevalence of 50% from the single
study of CHF. All other patient-reported studies were of mixed non
malignant diagnoses and evidence on preferences was not reported
separately by diagnosis, making interpretation of any variance by diagnosis
impossible. Again, we have not pooled evidence on preference for home
The limited evidence available indicates that, in advanced non
malignant conditions, the prevalence of patients’ preference
for home death as reported by patients themselves or their
professional carers is 42 - 48%. This is at the lower end of the
range reported among cancer patients.
There is insufficient evidence to understand how
preference for home death varies across different non
malignant conditions.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 201
death derived from sources other than patient report, because of the wide
variation in prevalence from different sources which we identified.
This prevalence of preference for home death in non malignant conditions is
generally lower than that reported by cancer populations or among the
general public, where home death preference is reported to range from
between 49-100% (16). It is also of interest that where disease-specific
evidence exists, such as for ESKD, it is similar prevalence to that in all non
malignant conditions. Evidence could not be broken down by age, because
of insufficient variation in the ages of study population, and no reporting of
preferences in relation to age. It was also not possible to further analyse
according to stage of disease (early or late stage), or co-morbidity because
of limitations in the evidence.
15.3 What determines preferences for place of care and place of death?
There is surprisingly little research conducted about patients’ preferences
for place of care and even the studies that have been conducted provide
only a very limited picture. Only two studies interviewed patients directly,
whereas the other studies presented less reliable data either from proxies
or from medical records. Studies providing data on patients’ preferences for
place of care from proxies and from medical records only reported the views
of nursing home residents. Preferences for place of care vary in type and
location. Only the study by Fried et al reported home versus hospital as
locations in a hypothesized scenario of patients who recently have been
hospitalized because of their disease (75). Here more than half of the
patients chose hospital as preferred place of care. Two studies from the US
related preference for place of care to hospice enrollment (74). As the term
‘hospice’ reflects different concepts in the US and the UK, hospice has to be
seen as type of care rather than location in this context. The studies
Determinants of preferences for place of care and death differ between
individuals and across non malignant conditions. However, some
patterns emerge; those with chronic heart failure may want to express
their preferences early at a time when they can deal with and plan for
optimal outcomes. In COPD, preference for hospital care may relate to
a sense of safety or be thought to offer better chance of survival.
Those with long term neurological conditions often wish to plan ahead,
amid profound concern about burden on their family carers. Among
older people, preferences for place of care and death are complex,
highly dependent on circumstances, and may change over time. Carer
burden is of major concern in determining preferences across
conditions.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 202
reporting nursing home residents’ views focus on referral to hospital or
discharge home. Four of the six studies on preferences for place of care also
assessed factors influencing patients’ preferences. However, only one of
these studies reports multivariate analysis for a very specific group of
patients, namely patients with end stage renal disease on peritoneal dialysis
in nursing homes (78).
The qualitative evidence provided some insights into what determines place
of care and death, and there were notable differences between different non
malignant conditions. Some of those with CHF wanted to learn about
adverse prognosis and its implications early, and to express their
preferences when they could deal with and act on relevant information in
the context of deteriorating or fluctuating health (131). However, ‘open’
awareness of approaching death, with acknowledgement of deterioration
and decline, may not correspond to the preferences and perspectives of all
those with CHF (181); older people with CHF studied by Gott et al did not
necessarily want open awareness of death, preceded by acknowledgement
of the likelihood of dying, but instead preferred a sudden, unaware’ death,
despite the difficulties this might raise for their families (181). In COPD,
there was evidence that preferences did not change over time or by stage
of illness, with some indication that timing of information and exploration of
preferences was less critical (75, 148) and did not correspond to a planned
trajectory or open awarenss, but instead provided a sense of ‘chaos
narrative’ and lack of progressive steps in illness in COPD, as described by
Pinnock and colleagues (204). There is also evidence that COPD patients
perceived home care as low intensity and low frequency care (148), and
relate hospital to best chance of survival; home was only preferred if it
offered similar survival odds. Some COPD patients also expressly preferred
hospital – largely because home was a lonely and frightening place to be ill.
Social support, self reliance, religion, and past illness experience all impact
on the preferences of CHF and COPD patients. But the major driver,
especially as illness progressed, was concern about being a burden to family
and carers (75); this may also drive changes in preference for place of care
and death as illness advances, although longitudinal study of preferences
remains infrequent, so there is little evidence to understand this better.
Those with long term neurological conditions may have different drivers of
their preferences about place of care and death (153). Patients with LTNCs
themselves often want to plan ahead (152), and lack of coordination of and
access to care at home are important considerations (151) (155).
Inconsistency of preferences between patients and carers is an important
feature, however, reported from several sources, and this may impact on
preference for place of care and death. Concern for family may influence
individual’s preferences at least as much as considerations of autonomy and
choice (181).
Among older people, preferences for place of care and death are highly
variable (75, 76, 170, 177) and change in response to different
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 203
circumstances (76, 164, 170, 177). They also differ according to stage of
illness, with some evidence of a shift to prefer hospital as a ‘safe place, or a
place less burdensome for family (76). Preference for place of care and
death emerged as a complex issue bound up with values and emotions and
highly dependent on circumstances. Although home care was often the
ideal, especially to avoid institutional care (176, 180), patient anxiety about
burden on their family and the corresponding guilt felt by families when
home care was not possible (165, 177, 180, 182) was a major factor in
determining preferences.
Sustainable and realistic home care (with adequate supporting services)
was important, but the goals of care which patients and families aimed for,
consistent with their values, was also a key consideration (164).
None of the identified quantitative evidence on preferences reported factors
that influenced patients’ preferences, so it was not possible to integrate
qualitative and quantitative evidence. Evidence on general practitioner
awareness of preferences (92, 93) indicates that treatment goals (palliative
rather than curative), presence of psychosocial or spiritual concerns,
location at home/nursing home/hospice (rather than hospital). And
increased recent contacts are all associated with better GP awareness of
preferences for place of care and death.
15.4 What factors are associated with actual place of death?
Factors with high strength and consistent quantitative evidence are grouped
according to the theoretical model, and presented in Figure 120 with
indication of the direction of effect (for and against home death, hospital or
nursing home death). Factors with high or moderate strength evidence are
presented on the subsequent page in Figure 121.
Factors associated with place of death are detailed in Figure
120 and Figure 121.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 197
Figure 120. Model of the factors affecting place of death in non malignant conditions (high strength evidence only)
Disease-related factors:
Age: no effect on home death Primary disease: heart disease (not CHF) ▲
home death COPD ▲ home death dementia ▲ home death dementia ▲ nursing home
death Associated co-morbidities: increased co-morbidity ▼home
death
Environmental factors
Health and social care input: Service provision: increased hospital bed availability ▼ home
death
Social support:
Personal and demographic factors: Gender: no effect hospital death female ▲nursing
home death Education: no effect nursing home death Marital status: single or widowed ▼hospital death
Place of care and place of death
Specific symptoms. illness burden and trajectories: Degree of impaired mobility/disability (severe functional impairment ▲ home death)
Macro-social factors:
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 198
Figure 121. Model of the factors affecting place of death in non malignant conditions (high & moderate strength evidence)
Disease-related factors:
Age: no effect on home death older age▼ hospital death age > 75 ▲NH death
Primary disease: Heart disease (not CHF) ▲
home death CHF no effect on home death COPD ▲ home death dementia ▲ home death dementia ▲ NH death
Associated co-morbidities: co-morbidity ▼home death
Environmental factors
Health and social care input: Service provision: increased hospital bed availability ▼ home
death increased hospital bed availability ▲ hospital
death palliative care provision ▼ hospital death
Social support: Living alone ▼ home
death
Personal and demographic factors: Gender: no effect home or hospital death female ▲NH death
Ethnicity: black ▼ home death non white ▲ hospital death non white ▼ NH death
Education: no effect nursing home death further education ▲ home death further education ▼ hospital death
Marital status: single/widowed ▼hospital death single/widowed ▲ NH death married ▲ home death
partner no effect hospital death divorced ▼hospital death
Socioeconomic status: higher household income ▲home death
Place of care and place of death
Specific symptoms. illness burden and trajectories: Degree of impaired mobility/disability severe functional impairment ▲
home death functional status - no effect on hospital death
Macro-social factors
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 199
As evident from the tables, there are some factors with high strength and
consistent quantitative evidence. Most are personal, demographic or disease
related factors. With respect to home death, diagnosis and co-morbidity
have a major effect; heart disease in general (although not CHF
specifically), COPD and dementia are all associated with increased likelihood
of home death. (The contrast between heart disease as an overall category
and CHF specifically may in part be accounted for by sudden death from
ischaemic heart disease, which may often occur at home). Co-morbidity is
also associated with reduced likelihood of home death. The odds ratios for
these high strength factors in relation to home death are plotted in Figure
122 below.
Figure 122. Factors with high strength evidence of association with home death
in non malignant conditions
It is clear from Figure 122 that heart disease in general (not CHF) and
dementia are the factors with greatest weight of association with home
death. Co-morbidity is consistently associated with reduced home death,
and hospital bed availability has a small association.
The factors with strong evidence of association with nursing home death are
plotted in Figure 123. In the general non-cancer population, being female or
having dementia increases likelihood of nursing home death.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 207
Figure 123. Factors with high strength evidence of association with nursing
home death in non malignant conditions
Although in the theoretical conceptualisation of factors affecting home
death, a number of specific symptom, illness burden and trajectory-related
factors were considered, almost no high strength evidence on these was
present in published studies. This may in part reflect the relatively recent
introduction of these areas (especially trajectory) into non cancer research
streams. It may also reflect the challenges of studying how these impact
upon place of death. There was also a dearth of high strength evidence in
relation to health and social care input; only hospital bed availability is
underpinned by evidence. Increased hospital bed availability reduces the
likelihood of home death. This made it impossible to reflect in any detail on
how health and social care resources influence actual place of death.
When moderate strength evidence is also included in the model (see Figure
121), a number of factors are seen to operate across more than one
setting. For instance, the same factor is sometimes associated with
increased likelihood of hospital death and reduced likelihood of home death.
Age, ethnicity, education, marital status, and hospital bed availability all fall
into this category:
Older age is associated with reduced likelihood of hospital death and
increased likelihood of nursing home death
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 208
Non white ethnicity is associated with increased likelihood of
hospital death, and decreased likelihood of home and nursing home
death.
Further education is associated with increased likelihood of home
death, and decreased likelihood of hospital death.
Being single or widowed is associated with greater likelihood of
nursing home death and reduced likelihood of hospital death
Hospital bed availability is associated with reduced likelihood of
home death and greater likelihood of hospital death
In addition, other factors emerge; living alone lessens the likelihood of
home death, and higher household income increases the likelihood of home
death.
Some factors should be interpreted with caution; the evidence on ethnicity
largely comes from the US, and may not be generalisable to other
countries. Nevertheless, there is some consistency of effect across settings,
with factors tending to operate in similar directions. Type of disease
(especially dementia), functional status, marital and educational status, and
living situation emerge as important influences on place of death. Age is
important above 75 years, and hospital bed availability (although a small
effect) is very consistent.
The evidence reports some factors specific to certain populations; in nursing
home populations, there are additional moderate strength factors which
operate. The effects of increasing age and female gender are lost, in age
probably because there is less distinction between ages within this
population – most are older. Preference for nursing home as place of death
is also a factor, as is the availability of face to face out of hours medical and
nursing support.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 209
15.5 What are the key transitions in care at the end
of life and what influences these transitions
The key transitions reported by patients with advanced illness and their
carers towards end of life, and largely identified from the qualitative
evidence, are:
Social and psychological decline, alongside physical deterioration
(135)
Transitions when new or different treatments commenced, or into a
more palliative phase when no further active treatments were
available (145)
Entry into care homes or hospital (173, 182)
Other forms of relocation, especially for older people with advanced
conditions, including from hospital into nursing home (174), from
home into nursing or residential care (165, 169, 177, 182), or into
hospice care (172)
For ESKD patients, the specific transitions onto or off dialysis (156,
160)
The quantitative evidence also provided evidence on transitions back to
home, from hospital or nursing home, and into palliative care programmes.
Synthesizing this evidence was challenging, because few of the studies were
directly comparable. It is especially difficult to compare evidence on
transitions between settings which comes from different countries, because
of the differing health systems involved. The type of services and how each
is used within a health system varies enormously between countries, and
interpretation of evidence should therefore be cautious.
However, there is some moderate and low strength evidence existing for
different transitions. For instance, the transition into hospital from nursing
home is more likely in younger nursing home residents, who are more
functionally able. This is consistent with the evidence on place of death;
nursing home death is more likely in the older and those more functionally
impaired.
There is high strength evidence that transition into nursing home is much
more likely in those cognitively impaired or with dementia; again, this is a
pronounced effect, and consistent with the evidence on place of death.
Nursing home admission is also more likely in those who are living alone
prior to their admission, or those who lack any informal carer support, and
this is consistent with the moderate evidence of association between living
alone and reduced likelihood of home death. A new finding is that longer
hospital stay is associated with increased odds of transition into nursing
home, although this evidence is moderate, rather than strong.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 210
In the UK there has recently been some emphasis on transfer out of
hospital towards the end of life, sometimes to home to die, but also to care
homes in the last weeks or months of life. There was no evidence in the
literature which reflects this shift, or reported in detail the duration of stay
in the care home prior to death. This may be an important factor in whether
a care home is perceived as ‘home’ or not, and it would be particularly
valuable to have evidence about this, especially about trends in the duration
of care home stay up to death over time.
15.6 The work of the Management Fellow within this review
One of the particular strengths of this review was the work of the
Management Fellow attached to the project, who made considerable
contributions to the project in a number of areas:
Integration into the research team – introducing a new perspective
on undertaking a systematic review which had a more clinical and
health service management orientation
Major contributions to the proposed theoretical modelling
Close liaison with management and policy makers, in order to
explore how best to implement findings – this component is still
developing, because of the timing of outputs from the project
(subsequent dissemination to policy makers and mangers to follow)
Critique of the emerging findings, to help increase relevance for
management and policy makers
Through the case studies, through illumination of evidence (see
below for examples of case study integration)
15.6.1 Communication, planning ahead and enabling smooth
transitions
Our work with the Management Fellow associated with this project, in
particular reflecting on the case studies undertaken, helped bring home the
reality of the rather abstract evidence in terms of the actual illness
pathways experienced by individuals and their families. The resonance was
especially apparent in the qualitative evidence. For instance, from the
qualitative evidence, communication was identified as a key ingredient for
good care, with understanding of prognosis, preferences and priorities early
in the disease trajectory between the different people involved, the patient,
family and health professionals (175), and ideally before transitions into
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 211
other settings were made (173). This was also an important element in the
case studies, with lack of sufficiently early (or sometimes any) advance care
planning, poor communication between different services (especially out of
hours providers), and fragmented care reported within all twelve case
studies.
15.6.2 Disease specific issues
Although only selected case studies were conducted by the Management
Fellow, there was again resonance in the concerns raised, especially in the
qualitative evidence, and the case studies. For example, in the evidence on
COPD, patients’ concerns included a desire to be with family members (147)
and availability of support (140); concerns about being a burden (75), the
need for information (139), providing continuity (141); the availability of
GP’s (142) and receiving care from one’s own GP. As well as this,
communication and information (139), and a desire for a place associated
with survival (148) were important. In addition, patients who were afraid of
suffocating while dying (149) preferred to be in hospital surrounded by
family and friends. Improving assistance and promoting independence were
seen as important to prevent additional suffering and reduce exacerbations
requiring hospitalisation (147). In the case studies, these issues also
emerged as dominant; continuity of care, sufficient information, the
availability of health care professionals at home and not feeling a burden to
carers were important factors for this group of patients to consider.
Breathlessness coupled with anxiety, and the presence of social support,
were important factors in determining where the patient wanted to be at
the end of their life. As identified in the evidence, the fear of suffocation at
the end of life contributed to preference to be in hospital in the case studies
of COPD too.
15.7 Strengths of this review
This review makes several key contributions to the existing knowledge on
preferences for place of care and death, actual place of death and factors
which influence this, and transitions towards end of life. It was a challenging
task to identify, critically appraise, and synthesize the rather fragmented
evidence on preferences regarding place of care and place of death for non-
cancer patients, and the factors influencing place of death.
Our extended model of factors which influence place of care and death for
patients with non-cancer diseases constitutes an original point of
contribution to the current body of knowledge and clinical practice. This
model was based on an existing model of factors influencing place of death
in cancer patients (15), to which important and new elements have been
added, underpinned by detailed consultation with clinical experts in key
non-cancer disease areas. This yielded an extended model of the potential
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 212
factors which influence place of care and death for those with non malignant
conditions, that incorporates contributions from the chronic illness literature
(61-64), including experiential and societal perspectives pertinent to non
malignant disease, as well as expert opinion. This approach ensured a more
objective assessment of the review findings, and informs consideration of
gaps in the evidence.
It is worth reflecting on the methods used in this systematic review. We
adopted a narrative synthesis approach – this is an approach to synthesis of
findings from multiple studies that relies primarily on the use of words and
text to summarise and explain the findings of the synthesis. Narrative
synthesis is suitable for the synthesis of studies where meta-analysis is not
possible or for the integration of findings from both quantitative and
qualitative studies (as in this review). It does not rest on an authoritative
body of knowledge or on reliable and rigorous techniques developed over
time and therefore it has been critiqued as being prone to bias. However,
guidance has been developed on the use of narrative synthesis (206) and
when it is used within the framework and systematic procedures of
systematic reviewing and verified within a team of reviewers this can be a
valid approach to addressing the problem of integrating complex and
diverse evidence into a useable format for policy makers (206). Narrative
synthesis was the approach with which we used to synthesize the
qualitative evidence on the level of findings and discussion. This was
selected as the most appropriate way to integrate the findings as the focus
of the studies were too diverse to be investigated with, for example, the
meta-ethnographic approach.
The joint working between the review team and the management fellow
constitutes an additional point of original contribution. The post was held by
a senior member of staff from St. Christopher’s hospice in London,
representing the perspective of a key potential setting of care and death for
people with advanced illness. The role involved close liaison with
management and policy makers, exploring how best to implement findings
and share insights from the literature evidence; critique of the emerging
results, and highlighted effort to increase awareness about our findings to
management and policy makers. The case studies developed by the
management fellow helped fortify the evidence that transpired through the
literature review. There was notable resonance between the qualitative
evidence of the literature review and the case studies particularly in terms
of communication between different services (e.g. out of hours providers),
identified as a key component of good care; understanding early in the
disease trajectory between the different people involved, including the
patient, family and health professionals (ideally before transitions into other
settings); lack of sufficiently early advance care planning, fragmented care
and others.
Finally, with its magnitude and span of non-cancer conditions, this review
has drawn on extensive joint working between a wide range of experts, and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 213
presents both the framework to conceptualise and the evidence to underpin
a challenging area both for research and for clinical practice.
15.8 Limitations
15.8.1 Identifying the evidence
This review has several limitations. The first one relates to the
thoroughness of our literature search, and more concretely to its
sensitivity11 (high) and specificity12 (low). In our scoping searches, we found
the evidence was very widely disseminated, across a wide range of general
medical, palliative, geriatric, and disease specific journals (see for instance
the diversity of journals from which our included studies come; APPENDIX
3). This provided a major challenge for our identification of evidence. In our
search strategy, we made every effort to capture the maximum number of
high quality studies of place of care/death preferences for the six selected
non-cancer diseases (high sensitivity), and chose a strategy of mid range
precision; but even that led to only 238 included studies out of the 10,350
titles/abstracts screened. We therefore overcame the challenge of
identifying evidence through our primary search by introducing detailed
citation and reference searching of included studies. This was successful,
with 559 further references identified this way, but time-consuming,
resource intensive, and this overall approach could militate against
identifying those studies less frequently referenced by other authors. Whilst
we achieved high disease-specificity in the identified literature, we are
aware that we could have omitted some literature on place of care/death
preferences which are not specific to any particular disease.
15.8.2 Heterogeneity of evidence
The considerable heterogeneity of the studies identified through this
review and the challenges that this involves in terms of synthesis of the
results, constitute our second limitation. Studies were heterogeneous in
three main areas: study design; study population and setting; and methods
of data collection. In terms of study design, the literature spanned the full
range of qualitative methodologies, to analysis of death registration data
(with or without linkage to other routine health databases), secondary
11 Sensitivity for a given topic is defined as the proportion of high quality studies
that are retrieved for that topic 205. Wilczynski NL, Haynes RB. EMBASE search
strategies achieved high sensitivity and specificity for retrieving methodologically
sound systematic reviews. J Clin Epidemiol. 2007 Jan;60(1):29-33. 12
Specificity is defined as the proportion of low quality articles not retrieved through the search 205. Ibid.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 214
analyses of routinely-collected data from provider organisations, post-death
interviews with bereaved family members, and surveys. The identified
literature ranged widely also in terms of study population and setting.
Several studies used complete national or cross-national datasets, while
others were conducted in a single hospital, nursing home, or community
setting, drawing on small population cohorts.
The studies identified in this review also originate from a range of countries
and regions across the world, such as the USA, Japan, Canada, Australia,
Belgium, the Netherlands, England, Scotland, Wales, Singapore, China,
Spain and Arabia (see Table 24 and Table 14). Despite the diversity of
countries from which evidence derived, evidence from the USA dominated
the review. It is not clear how generalisable evidence from one specific
health system is, particularly in relation to models of palliative care, and in
terms of ecological and population-based factors. Both the diversity of
methods and the diversity of health systems from which evidence derives,
presented a major challenge for data synthesis and interpretation of our
findings across health policy settings. The diverse range of data sources
also constituted a limitation particularly in terms of elucidating patient
preferences. There are well documented differences between the views of
patients and those of proxies (family and health professionals). It is also
debatable as to how reliable (past) medical records are as a method to
capture preferences.
Another key challenge in this review was in the categorisation of place of
death. There was marked heterogeneity in approach, within the included
studies, both to the outcome of interest and to the potential factors
affecting it. For example, some studies classify place of death into a binary
variable (e.g. home deaths v deaths elsewhere, hospital deaths v deaths
elsewhere), whereas others categorise it into several variables and
sometimes in parallel (e.g. hospice death v nursing home death, compared
with hospice death v home death). Our approach towards tackling some of
the challenges for synthesis that this heterogeneity brings, was to group
papers which treat place of death as a binary variable (e.g. home v all other
places) together with papers which categorise place of death into several
variables (e.g. home v nursing home, home v hospice). However, because
of this approach, it is not possible to assume that an increase in one factor
associated with greater likelihood of home death necessarily means that a
reduction in that same factor is associated with greater likelihood of hospital
deaths. The comparator (reference) may be another setting different to
hospital. If a factor, however, operates in both categories (is associated
with likelihood of home death and likelihood of hospital death), this adds to
the validity of the evidence.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 215
15.8.3 Quality and extent of the evidence
The quality of the evidence was variable; in particular, the number of
studies which undertook multi-variate analysis was relatively small. In
addition, few studies reported on aspects of transitions, place of care and
place of death which are important in the clinically-driven understanding of
these phenomena. Although demographic factors are reasonably well
studied, perhaps because they are easy to measure, other factors are rarely
studied. Major deficiencies in the evidence have emerged, including on:
Preferences and how these influence actual place of death
Disease related factors such as illness burden, trajectory of illness,
and specific symptoms
Environmental factors, such as social support, health and social care
provision
Macro-social factors
Social support, and health and social care provision are particularly
important to study, because these may be among the factors which can
most readily be modified to improve outcomes, and to make best use of
resources. There is some indication from this review, for instance, that
geographic factors may have a marked effect on place of death, although
the strength of this evidence is weak; this may well be related to variations
in provision of health and social care.
Evidence from different countries is also very uneven, with relatively few
analyses of existing large, linked datasets from countries other than the US.
With some notable exceptions, few European countries have undertaken
this work.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 216
16 Conclusions
This systematic literature review demonstrates clearly that there are major
gaps in the evidence to inform policy and services in relation to end of life
care for those with non-cancer conditions.
Practice is often extrapolated from models of cancer care, and these may
not be optimal. For example, the conception of home as preferred place of
end of life care, with ‘open’ awareness of approaching death and planning of
health care to accommodate decline, is derived largely from cancer care,
and (while very appropriate for some) may not accommodate the
preferences of some of those with advanced non-cancer conditions,
especially those who have a more unpredictable course of illness.
It is also clear that those with advanced non-cancer conditions have
complex and variable experiences, and widely differing preferences, which
serves to emphasise the need for greater patient- and family- centeredness
of care.
The studies identified in this review originate from a range of countries and
regions across the world, with evidence predominantly from the USA. It is
not clear how generalisable evidence from one specific health system is,
particularly in relation to models of palliative and end of life care, and in
terms of ecological and population-based factors. Both the diversity of
methods and the diversity of health systems from which evidence derives,
presented a major challenge for data synthesis and interpretation of our
findings across health policy settings. Our conclusions should be read with
this very much in mind.
This complex systematic review of the evidence has identified that, among
those with advanced non malignant disease:
Prevalence of patients’ preference for home death is 42-48%, lower
than previously reported among cancer patients.
Determinants of preferences for place of care and death differ
markedly between individuals and across non malignant conditions.
Among older people, preferences for place of care and death are
complex, highly dependent on circumstances, and may change over
time.
Across all conditions, considerations of carer/family burden are a
major influence on the preferences of those with advanced disease,
and this resonates with evidence on the factors which precipitate
transition into hospital or nursing home care, and likelihood of death
in those places.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 217
For those with chronic heart failure and chronic obstructive
pulmonary disease:
o the lack of a clearly predictable pattern of illness has a marked
impact on awareness of deterioration and subsequent perspectives
about place of care and death
o preferences may not always be consistent with a desire for ‘open’
awareness of death
Those who prefer hospital care may relate hospital to a sense of
safety and/or the perception of better chance of survival.
Those with long term neurological conditions may prefer to plan
ahead, amid profound concern about burden on their family carers.
Dementia is associated with greater odds of home and nursing home
death, while co-morbidity is associated with reduced odds of home
death.
Other factors associated with increased likelihood of home death in
non malignant conditions include being married, having further
education, and higher household income.
Living alone, the absence of an informal carer, and cognitive
impairment or dementia all reduce the likelihood of home death,
increase the likelihood of transition into nursing home care, and
increase likelihood of nursing home death.
Health and social care input are associated with place of death, but
the evidence on this is limited; what evidence there is indicates that
increased palliative care provision is associated with reduced
hospital deaths.
Increased hospital bed availability also has a consistent, if small,
association with place of death - increased availability of hospital
beds is associated with reduced home deaths and increased hospital
deaths.
A well founded theoretical model of the factors affecting place of death has
been developed, and the evidence relating to place of death conceptualised
in this context. This model brings together what evidence exists, but also
highlights the pronounced gaps in the evidence. There is generally very
limited evidence in all areas, and this is compounded by the limitations of
applying evidence from one country to another, where health systems may
be very different. Focused and high quality research, using multi-variate
analyses, is particularly needed on:
Preferences and how these influence actual place of death
Disease related factors such as illness burden, trajectory of illness,
and specific symptoms
Environmental factors, such as social support, health and social care
provision
Macro-social factors
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 218
Analysis of national or sub-national pre-existing linked datasets to
understand these factors, plus targeted original research, including
longitudinal qualitative, quantitative and mixed methods, will be crucial to
advancing this area of practice and research. Comparative research
between countries is also important, especially when health system and
service factors are studied.
The demographic changes of an ageing UK population and increased
numbers of deaths make this work imperative. In this context, social
support, and health and social care provision are particularly important to
study, because these may be among the factors which can most readily be
modified to improve patient outcomes, and to make best use of resources.
16.1.1 Practice and policy implications
The evidence on the factors influencing place of death in non malignant
conditions is complex and inter-related. However, clear implications for
practice and policy emerge from this synthesis of the evidence.
1. A preference for home death (while still common among non cancer
patients) is less prevalent than for cancer patients, so attention
should be given to achieving preferred place of care and death in non
malignant conditions, wherever that place is.
2. For those with non-cancer conditions, the presence or absence of a
family or informal carer is a key component in achieving home death.
Effective and sustained carer support, especially in context of longer
illnesses trajectories, is likely to increase home death rates.
3. Minority ethnic groups, and those with lower socio-economic status
achieve lower rates of home death; best practice initiatives need to
target this imbalance, while still taking preferred place into account.
4. For older people, advance care planning is especially important, as
their preferences are complex, highly dependent on circumstances,
and may well evolve over time. The advance care planning they
require therefore needs skilled professionals who can re-visit
preferences and planning sensitively and frequently in response to
change.
5. In addition, older people are more likely to die in care homes,
especially if there is cognitive impairment or dementia. Resources
need to be targeted both at improving care home deaths and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 219
ensuring the opportunity for home death among older people
(especially those who live alone) is adequately supported, when
desired and feasible.
6. For those with end-stage heart or respiratory disease, the conception
of an ‘open’ awareness of approaching death and forward planning of
health care to accommodate decline (derived largely from models of
care in cancer), is not always appropriate. Again, skilled advance care
planning is important, to work across the range of awareness, and to
respond to rapidly changing circumstances.
7. In those conditions with longer trajectories, for instance some long
term neurological conditions, advance planning may be welcomed
and home death can be supported even when there is marked
functional impairment.
8. The increase in the ageing population, with correspondingly higher
levels of co-morbidity, will likely mitigate against home deaths and in
favour of hospital deaths, unless innovative approaches can be
developed to support those with complex co-morbidities in the
community.
9. Increased resources for the delivery of palliative care early in the
illness trajectory across non malignant conditions will likely reduce
hospital deaths.
16.1.2 Research implications
There are major gaps in the current evidence to inform practice and policy,
and research is particularly needed:
10.To provide insights into how preferences for place of care and death
among those with advanced non malignant conditions evolve over time and with advancing illness, and what factors shape these
preferences.
11.To understand how duration and trajectory of illness affect transitions
in place of care, and place of death.
12.To determine how health and social care provision in non cancer conditions can influence place of care and death.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 220
References
1. National Kidney F. K/DOQI clinical practice guidelines for chronic
kidney disease: evaluation, classification, and stratification. Am J Kidney
Dis. 2002;39(2 (Suppl 1)):S1-S266.
2. Department of Health. National Service Framework for long term
(neurological) conditions. London: Department of Health 2005.
3. Gomes B, Higginson IJ. Where people die (1974-2030); Past trends,
future projections, and implications for care. Palliat Med. 2008 Jan; 22(1):
33-41.
4. Scitovsky AA. "The high cost of dying": what do the data show?
1984. Milbank Q. 2005;83(4):825-41.
5. Office of the Actuary at the Centers for M, Medicaid S. Highlights of
National Spending of Health Goods and Services for 2005. USA: Centers for
Medicare and Medicaid Services 2007.
6. Polder JJ, Barendregt JJ, van Oers H. Health care costs in the last
year of life - the Dutch experience. Soc Sci Med. 2006 Oct;63(7):1720-31.
7. Office for National S. Total UK Health Expenditure 1997-2002.
London, UK: ONS 2007.
8. Hatziandreu E, Archontakis F, Daly A. The potential cost savings of
greater use of home- and hospice-based end of life care in England.
London, UK: National Audit Office 2008.
9. Bloom BS, Kissick PD. Home and hospital cost of terminal illness.
Med Care. 1980 May;18(5):560-4.
10. Beckford M. Quality of life more important than length of time left.
Daily Telegraph. 2011 26 March 2011.
11. Department of Health. End of Life Care Strategy. London:
Department of Health 2008.
12. Department of Health - Next Stage Review. High Quality Care for
All: Our Journey So Far. London UK: Department of Health 2009.
13. Department of Health. Building on the best: Choice, responsiveness
and equity in the NHS. London, UK: Department of Health 2003.
14. Donabedian A. The seven pillars of quality. ArchPatholLab Med.
1990;114(11):1115-8.
15. Gomes B, Higginson IJ. Factors influencing death at home in
terminally ill patients with cancer: Systematic review. British Medical
Journal. 2006(7540):515-8.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 221
16. Higginson IJ, Sen-Gupta GJ. Place of care in advanced cancer: a
qualitative systematic literature review of patient preferences. Journal of
Palliative Medicine. 2000;3(3):287-300.
17. Lock A, Higginson I. Patterns and predictors of place of cancer death
for the oldest old. BMC Palliat Care. 2005 Oct 8;4:6.
18. World Health Organisation. The Solid Facts: Palliative Care: WHO
2004.
19. Gomes B, Gysels M, Higginson IJ. What is the latest evidence on
preferences for place of care and place of death? (Abstract). Palliat Med.
2008;22:465-6.
20. Office for National Statistics. Mortality Statistics Series DH1 no 38.
London: Office for National Statistics 2007.
21. Harris S, Ho D, Verne J. Deaths from Alzheimers disease, dementia
and senility in England: National End of Life Intelligence Network, 2010.
22. Addington-Hall JM, Karlsen S. Age is not the crucial factor in
determining how the palliative care needs of people who die from cancer
differ from those of people who die from other causes. J Palliat Care.
1999;15(4):13-9.
23. Agar M, Currow DC, Shelby-James TM, Plummer J, Sanderson C,
Abernethy AP. Preference for place of care and place of death in palliative
care: are these different questions? PalliatMed. 2008;22(7):787-95.
24. Seymour J, Payne S, Chapman A, Holloway M. Hospice or home?
Expectations of end-of-life care among white and Chinese older people in
the UK. Sociology of Health & Illness29(6):872-90. 2007.
25. Tang ST, Liu TW, Lai MS, McCorkle R. Discrepancy in the
preferences of place of death between terminally ill cancer patients and
their primary family caregivers in Taiwan. Soc Sci Med. 2005;61(7):1560-6.
26. Ward C. The need for palliative care in the management of heart
failure. Heart. 2002;87(3):294-8.
27. Gibbs JS, McCoy AS, Gibbs LM, Rogers AE, Addington-Hall JM. Living
with and dying from heart failure: the role of palliative care. [Review] [27
refs]. Heart88 Suppl 2:ii36-9. 2002.
28. Lewis C, Stephens B. Improving palliative care provision for patients
with heart failure. Br J Nurs. 2005;14(10):563-7.
29. Goodridge D. People with chronic obstructive pulmonary disease at
the end of life: a review of the literature. [Review] [69 refs]. International
Journal of Palliative Nursing, 2006; 12(8):390-6.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 222
30. Seamark DA, Seamark CJ, Halpin DM. Palliative care in chronic
obstructive pulmonary disease: a review for clinicians. Journal of the Royal
Society of Medicine, 2007; 100(5):225-33.
31. Hardin KA, Meyers F, Louie S. Integrating palliative care in severe
chronic obstructive lung disease. COPD: Journal of Chronic Obstructive
Pulmonary Disease. 2008;5(4):207-20.
32. Birch D, Draper J. A critical literature review exploring the
challenges of delivering effective palliative care to older people with
dementia. J Clin Nurs. 2008;17(9):1144-63.
33. Sampson EL, Ritchie CW, Lai R, Raven PW, Blanchard MR. A
systematic review of the scientific evidence for the efficacy of a palliative
care approach in advanced dementia. Int Psychogeriatr. 2005;17(1):31-40.
34. Downs M, Small N, Froggatt K. Explanatory models of dementia:
links to end-of-life care. International Journal of Palliative Nursing.
2006;12(5):209-13.
35. Roger KS. A literature review of palliative care, end of life, and
dementia. Palliative & Supportive Care. 2006;4(3):295-303.
36. Oliver DP, Porock D, Zweig S. End-of-life care in U.S. nursing
homes: a review of the evidence. Journal of the American Medical Directors
Association, 2004; 5(3):147-55.
37. Rensbergen G, Nawrot TS, Van Hecke E, Nemery B. Where do the
elderly die? The impact of nursing home utilisation on the place of death.
Observations from a mortality cohort study in Flanders. BMC Public Health.
2006;6:178.
38. Kapo J, Morrison LJ, Liao S. Palliative care for the older adult.
Journal of Palliative Medicine. 2007;10(1):185-209.
39. Lunney JR, Lynn J, Foley DJ, Lipson S, Guralnik JM. Patterns of
functional decline at the end of life. JAMA. [10.1001/jama.289.18.2387 doi
;289/18/2387 pii]. 2003;289(18):2387-92.
40. Cosgriff JA, Pisani M, Bradley EH, O'Leary JR, Fried TR. The
association between treatment preferences and trajectories of care at the
end-of-life. J Gen Intern Med. 2007;22(11):1566-71.
41. Murtagh FE, Preston M, Higginson I. Patterns of dying: palliative
care for non-malignant disease. ClinMed. 2004;4(1):39-44.
42. Coventry PA, Grande GE, Richards DA, Todd CJ. Prediction of
appropriate timing of palliative care for older adults with non-malignant life-
threatening disease: a systematic review. Age & Ageing 34(3):218-27.
2005.
43. Gold Standards Framework Programme. The Gold Standards
Framework Prognostic Indicator Guidance. 2006.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 223
44. Fox E, Landrum-McNiff K, Zhong Z, Dawson NV, Wu AW, Lynn J.
Evaluation of prognostic criteria for determining hospice eligibility in
patients with advanced lung, heart, or liver disease. SUPPORT Investigators.
Study to Understand Prognoses and Preferences for Outcomes and Risks of
Treatments.[see comment]. JAMA 282(17):1638-45. 1999.
45. Elkington H, White P, Higgs R, Pettinari CJ. GPs' views of discussions
of prognosis in severe COPD. Family Practice, 2001(4):440-4.
46. Curtis JR, Engelberg RA, Wenrich MD, Au DH. Communication about
palliative care for patients with chronic obstructive pulmonary disease. J
Palliat Care. 2005;21(3):157-64.
47. Edmonds P, Karlsen S, Khan S, Addington-Hall J. A comparison of
the palliative care needs of patients dying from chronic respiratory diseases
and lung cancer. PalliatMed. 2001;15(4):287-95.
48. Traue DC, Ross JR. Palliative care in non-malignant diseases. J
RSocMed. 2005;98(11):503-6.
49. Gysels M, Higginson IJ, White P, Barclay S, Worth A, Murray S, et
al. Scoping exercise on generalist services for adults at the end of life:
research, knowledge, policy and future research needs: Report 2: The
literature scoping. London: NCCSDO 2007.
50. Luddington L, Cox S, Higginson I, Livesley B. The need for palliative
care for patients with non-cancer diseases: a review of the evidence. Int J
Palliat Nurs. 2001 May;7(5):221-6.
51. Grande GE, McKerral A, Addington-Hall JM, Todd CJ. Place of death
and use of health services in the last year of life. J PalliatCare.
2003;19(4):263-70.
52. Cohen J, Bilsen J, Hooft P, Deboosere P, van der WG, Deliens L.
Dying at home or in an institution using death certificates to explore the
factors associated with place of death. Health Policy. 2006;78(2-3):319-29.
53. Elkington H, White P, Addington-Hall J, Higgs R, Edmonds P. The
healthcare needs of Chronic obstructive pulmonary disease patients in the
last year of life. Palliative Medicine, 2005; 19(6):485-91.
54. Murtagh FE, Addington-Hall J, Edmonds P, Donohoe P, Carey I,
Jenkins K, et al. Symptoms in the month before death for stage 5 chronic
kidney disease patients managed without dialysis. JPain Symptom Manage.
[S0885-3924(10)00391-X pii ;10.1016/j.jpainsymman.2010.01.021 doi].
2010;40(3):342-52.
55. Solano JP, Gomes B, Higginson IJ. A comparison of symptom
prevalence in far advanced cancer, AIDS, heart disease, chronic obstructive
pulmonary disease and renal disease. Journal of Pain and Symptom
Management, 2006; 31(1):58-69.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 224
56. Saleem T, Leigh PN, Higginson IJ. Symptom prevalence among
people affected by advanced and progressive neurological conditions - a
systematic review. J Palliat Care. 2007;23(4):291-9.
57. Currow DC, Abernethy AP, Fazekas BS. Specialist palliative care
needs of whole populations: A feasibility study using a novel approach.
Palliat Med. 2004;18(3):239-47.
58. Lewin K, Heider F, Heider G. Principles of topological psychology. 1st
ed. New York: McGraw Hill; 1986.
59. Moos RH. Context and coping: toward a unifying conceptual
framework. Am J Community Psychol. 1984 Feb;12(1):5-36.
60. Moos RH, Brennan PL, Schutte KK, Moos BS. Older adults' coping
with negative life events: common processes of managing health,
interpersonal, and financial/work stressors. Int J Aging Hum Dev.
2006;62(1):39-59.
61. Bury M. Chronic illness as biographical disruption. Sociol Health Illn.
1982;4(2):167-82.
62. Bury M. Health care and chronic illness: a sociological view. Health
Care Anal. 1994;2(3):240-3.
63. Bury M. The sociology of chronic illness: a review of research and
prospects. Sociol Health Illn. 2001;13:451-68.
64. Field D, Kelly MP. Chronic illness and physical disability. In: Taylor
S, editor. Sociology of health and health care. Malden, MA: Blackwell
Publishing; 2007. p. 137-58.
65. Khan KS. Undertaking Systematic Reviews of Research on
Effectiveness: CRD's guidelines for those carrying out or commissioning
reviews. University of York: NHS Centre for Reviews and Dissemination;
2001.
66. Lawson B, Burge FI, Critchley P, McIntyre P. Factors associated with
multiple transitions in care during the end of life following enrollment in a
comprehensive palliative care program. BMC Palliat Care. 2006;5:4.
67. Reviews NHSCf, Dissemination. Systematic reviews: CRD's guidance
for undertaking reviews in healthcare. York: NHS Centre for Reviews and
Dissemination, University of York 2009.
68. Edwards A, Elwyn G, Hood K, Rollnick S. Judging the 'weight of
evidence' in systematic reviews: introducing rigour into the qualitative
overview stage by assessing Signal and Noise. J Eval Clin Pract. 2000
May;6(2):177-84.
69. Higginson IJ, Finlay I, Goodwin DM, Cook AM, Hood K, Edwards AG,
et al. Do hospital-based palliative teams improve care for patients or
families at the end of life? J Pain Symptom Manage. 2002;23(2):96-106.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 225
70. Hawker S, Payne S, Kerr C, Hardey M, Powell J. Appraising the
evidence: reviewing disparate data systematically. QualHealth Res.
2002;12(9):1284-99.
71. Hoogendoorn WE, van Poppel MN, Bongers PM, Koes BW, Bouter
LM. Systematic review of psychosocial factors at work and private life as
risk factors for back pain. Spine (Phila Pa 1976). 2000 Aug 15;25(16):2114-
25.
72. Ebell MH, Siwek J, Weiss BD, Woolf SH, Susman J, Ewigman B, et
al. Strength of recommendation taxonomy (SORT): a patient-centered
approach to grading evidence in the medical literature. J Am Board Fam
Pract. 2004 Jan-Feb;17(1):59-67.
73. Fried TR, van Doorn C, O'Leary JR, Tinetti ME, Drickamer MA. Older
persons' preferences for site of terminal care. Ann Intern Med. 1999 Jul
20;131(2):109-12.
74. Casarett D, Van Ness PH, O'Leary JR, Fried TR. Are patient
preferences for life-sustaining treatment really a barrier to hospice
enrollment for older adults with serious illness? J Am Geriatr Soc. 2006
Mar;54(3):472-8.
75. Fried. Older people preferences for home vs hospital care in
treatment of acute illness. Arch Int Med. 2000;160:1501-6.
76. Fried. Older Persons Preferences for Site of Terminal Care. Annals of
Internal Medicine. 1999;131(2):109-12.
77. Biola H, Sloane PD, Williams CS, Daaleman TP, Zimmerman S.
Preferences Versus Practice: Life-Sustaining Treatments in Last Months of
Life in Long-Term Care. Journal of the American Medical Directors
Association. 2010;11(1):42-51.
78. Anderson J, Sikorski I, Finucane T. Advance Care Planning by or on
Behalf of Peritoneal Dialysis Patients in Long-Term Care. American Journal
of Kidney Diseases. 2006;48(1):122-7.
79. Triplett P, Black BS, Phillips H, Richardson Fahrendorf S, Schwartz J,
Angelino AF, et al. Content of advance directives for individuals with
advanced dementia. J Aging Health. 2008 Aug;20(5):583-96.
80. Arling G, Kane RL, Cooke V, Lewis T. Targeting Residents for
Transitions from Nursing Home to Community. Health Services Research.
2010;45(3):691-711.
81. Al-Jahdali. Advance care planning preferences among dialysis
patients and factoris influencing their decisions. Saudi J Kidney Dis Transpl.
2009;20(2):232-9.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 226
82. Davison SN. End-of-life care preferences and needs: perceptions of
patients with chronic kidney disease. Clin J Am Soc Nephrol. 2010
Feb;5(2):195-204.
83. Formiga F, Chivite D, Ortega C, Casas S, Ramon JM, Pujol R. End-
of-life preferences in elderly patients admitted for heart failure. QJM. 2004
Dec;97(12):803-8.
84. Stajduhar KI, Allan DE, Cohen SR, Heyland DK. Preferences for
location of death of seriously ill hospitalized patients: perspectives from
Canadian patients and their family caregivers. Palliat Med. 2008
Jan;22(1):85-8.
85. Casarett D. Improving the Use of Hospice Services in Nursing
Homes: A Randomized Controlled Trial. JAMA: The Journal of the American
Medical Association. 2005;294(2):211-7.
86. Groth-Juncker A, McCusker J. Where do elderly patients prefer to
die? Place of death and patient characteristics of 100 elderly patients under
the care of a home health care team. J Am Geriatr Soc. 1983
Aug;31(8):457-61.
87. Pritchard RS, Fisher ES, Teno JM, Sharp SM, Reding DJ, Knaus WA,
et al. Influence of patient preferences and local health system
characteristics on the place of death. SUPPORT Investigators. Study to
Understand Prognoses and Preferences for Risks and Outcomes of
Treatment. J Am Geriatr Soc. 1998 Oct;46(10):1242-50.
88. Cohen LM, Germain MJ, Woods AL, Mirot A, Burleson JA. The family
perspective of ESRD deaths. American Journal of Kidney Diseases.
2005;45(1):154-61.
89. Tolle SW, Tilden VP, Rosenfeld AG, Hickman SE. Family reports of
barriers to optimal care of the dying. Nurs Res. 2000 Nov-Dec;49(6):310-7.
90. Curtis. A Measure of the Quality of Dying and Death: Initial
Validation using after-death Interviews with Family Members. JPSM.
2002;24(1 ):17-31.
91. Ramon JM. Place of death of elderly persons in Catalonia(article in
Spanish). Rev Clin Esp. 2006;206(11):549-55.
92. Abarshi E, Onwuteaka-Philipsen B, Donker G, Echteld M, Van den
Block L, Deliens L. General Practitioner Awareness of Preferred Place of
Death and Correlates of Dying in a Preferred Place: A Nationwide Mortality
Follow-Back Study in The Netherlands. Journal of Pain and Symptom
Management. 2009;38(4):568-77.
93. Meeussen K, Van den Block L, Bossuyt N, Bilsen J, Echteld M, Van
Casteren V, et al. GPs' awareness of patients' preference for place of death.
Brit J Gen Pract. 2009 Sep;59(566):665-70.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 227
94. Van den Block L, Deschepper R, Drieskens K, Bauwens S, Bilsen J,
Bossuyt N, et al. Hospitalisations at the end of life: using a sentinel
surveillance network to study hospital use and associated patient, disease
and healthcare factors. BMC Health Services Research. 2007;7(1):69.
95. Shinoda-Tagawa T, Ikegami N. Resident and facility characteristics
associated with the site of death among Japanese nursing home residents.
Age Ageing. 2005 Sep;34(5):515-8.
96. Afessa. Prognostic factors, clinical course, and hospital outcome of
COPD admitted to ICU for acute respiratory failure. Crit Care Med.
2002;30(7):1610-5.
97. Ai-Ping C. In-Hospital and 5-Year Mortality of Patients Treated in the
ICU for Acute Exacerbation of COPD: A Retrospective Study. Chest.
2005;128(2):518-24.
98. Babazono. Health policy in transition: terminal care and site of
death in Japan. J Health Serv Res Policy. 1998;3(2):77-81.
99. Bell CL, Davis J, Harrigan RC, Somogyi-Zalud E, Tanabe MKG,
Masaki KH. Factors associated with place of death for elderly Japanese-
American men: the Honolulu Heart Program and Honolulu-Asia Aging Study.
J Am Geriatr Soc. 2009;57(4):714-8.
100. Clifford. Where people die in Victoria. The Medical Journal of
Australia. 1991;155 446-56.
101. Enguidanos S, Yip J, Wilber K. Ethnic variation in site of death of
older adults dually eligible for Medicaid and Medicare. J Am Geriatr Soc.
2005 Aug;53(8):1411-6.
102. Fried TR, Pollack DM, Drickamer MA, Tinetti ME. Who dies at home?
Determinants of site of death for community-based long-term care patients.
J Am Geriatr Soc. 1999 Jan;47(1):25-9.
103. Gruneir A, Mor V, Weitzen S, Truchil R, Teno J, Roy J. Where people
die: a multilevel approach to understanding influences on site of death in
America. Med Care Res Rev. 2007 Aug;64(4):351-78.
104. Hansen SM, Tolle SW, Martin DP. Factors associated with lower
rates of in-hospital death. J Palliat Med. 2002 Oct;5(5):677-85.
105. Haydar ZR, Lowe AJ, Kahveci KL, Weatherford W, Finucane T.
Differences in end-of-life preferences between congestive heart failure and
dementia in a medical house calls program. J Am Geriatr Soc.
2004;52(5):736-40.
106. Houttekier D, Cohen J, Bilsen J, Addington-Hall J, Onwuteaka-
Philipsen BD, Deliens L. Place of death of older persons with dementia. A
study in five European countries. J Am Geriatr Soc. 2010 Apr;58(4):751-6.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 228
107. Johnson KS, Kuchibhatala M, Sloane RJ, Tanis D, Galanos AN,
Tulsky JA. Ethnic differences in the place of death of elderly hospice
enrollees. J Am Geriatr Soc. 2005 Dec;53(12):2209-15.
108. Klinkenberg M, Visser G, van Groenou MI, van der Wal G, Deeg DJ,
Willems DL. The last 3 months of life: care, transitions and the place of
death of older people. Health Soc Care Community. 2005 Sep;13(5):420-
30.
109. Kwak J, Haley WE, Chiriboga DA. Racial differences in hospice use
and in-hospital death among Medicare and Medicaid dual-eligible nursing
home residents. Gerontologist. 2008 Feb;48(1):32-41.
110. Levy CR, Fish R, Kramer AM. Site of death in the hospital versus
nursing home of Medicare skilled nursing facility residents admitted under
Medicare's Part A Benefit. J Am Geriatr Soc. 2004 Aug;52(8):1247-54.
111. Liu H, Zhang TT, Ye J. Analysis of risk factors for hospital mortality
in patients with chronic obstructive pulmonary diseases requiring invasive
mechanical ventilation. Chin Med J (Engl). 2007 Feb 20;120(4):287-93.
112. McGregor MJ, Tate RB, Ronald LA, McGrail KM. Variation in site of
death among nursing home residents in British Columbia, Canada. J Palliat
Med. 2007 Oct;10(5):1128-36.
113. Menec VH, Nowicki S, Blandford A, Veselyuk D. Hospitalizations at
the end of life among long-term care residents. J Gerontol A Biol Sci Med
Sci. 2009 Mar;64(3):395-402.
114. Miller. How does timing of hospice referral influence hospice care in
the last days of life. JAGS. 2003;51:798-806.
115. Mitchell. A national study of the location of death of older persons
with dementia. JAGS. 2005;53:299-305.
116. Motiwala SS, Croxford R, Guerriere DN, Coyte PC. Predictors of
place of death for seniors in Ontario: a population-based cohort analysis.
Can J Aging. 2006 Winter;25(4):363-71.
117. Muramatsu N, Hoyem RL, Yin H, Campbell RT. Place of death among
older Americans: does state spending on home- and community-based
services promote home death? Med Care. 2008 Aug;46(8):829-38.
118. Pritchard RS, Fisher ES, Teno JM, Sharp SM, Reding DJ, Knaus WA,
et al. Influence of patient preferences and local health system
characteristics on the place of death. SUPPORT Investigators. Study to
Understand Prognoses and Preferences for Risks and Outcomes of
Treatment. J Am Geriatr Soc. 1998;46(10):1242-50.
119. Rosenwax L, McNamara B, Zilkens R. A population-based
retrospective cohort study comparing care for Western Australians with and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 229
without Alzheimer's disease in the last year of life. Health Soc Care
Community. 2009 Feb;17(1):36-44.
120. Sauvaget C, Tsuji I, Li JH, Hosokawa T, Fukao A, Hisamichi S.
Factors affecting death at home in Japan. Tohoku J Exp Med. 1996
Oct;180(2):87-98.
121. Shega JW, Hougham GW, Stocking CB, Cox-Hayley D, Sachs GA.
Patients dying with dementia: experience at the end of life and impact of
hospice care. J Pain Symptom Manage. 2008 May;35(5):499-507.
122. Silveira MJ, Copeland LA, Feudtner C. Likelihood of home death
associated with local rates of home birth: influence of local area healthcare
preferences on site of death. Am J Public Health. 2006;96(7):1243-8.
123. Takezako Y, Tamiya N, Kajii E. The nursing home versus the
hospital as the place of dying for nursing home residents in Japan. Health
Policy. 2007 May;81(2-3):280-8.
124. Wachterman MW, Sommers BD. The impact of gender and marital
status on end-of-life care: evidence from the National Mortality Follow-Back
Survey. J Palliat Med. 2006 Apr;9(2):343-52.
125. Weitzen S, Teno JM, Fennell M, Mor V. Factors associated with site
of death: a national study of where people die. Med Care. 2003
Feb;41(2):323-35.
126. Yang L, Sakamoto N, Marui E. A study of home deaths in Japan from
1951 to 2002. BMC Palliative Care. 2006;5(1):2.
127. Yasamura. Factors relating to place of death of Japanese people
from a small town in a rural area. Ageing Clin Exp Res. 2000;12:449-54.
128. Silveira MJ, Copeland LA, Feudtner C. Likelihood of home death
associated with local rates of home birth: influence of local area healthcare
preferences on site of death. Am J Public Health. 2006 Jul;96(7):1243-8.
129. Gruneir A, Mor V, Weitzen S, Truchil R, Teno J, Roy J. Where people
die: a multilevel approach to understanding influences on site of death in
America. Med Care Res Rev. 2007 Aug;64(4):351-78.
130. Brannstrom M, Ekman I, Norberg A, Boman K, Strandberg G. Living
with severe chronic heart failure in palliative advanced home care. Eur J
Cardiovasc Nurs. 2006 Dec;5(4):295-302.
131. Caldwell PH, Arthur HM, Demers C. Preferences of patients with
heart failure for prognosis communication. Can J Cardiol. 2007;23(10):791-
6.
132. DesHarnais S, Carter RE, Hennessy W, Kurent JE, Carter C. Lack of
concordance between physician and patient: reports on end-of-life care
discussions. J Palliat Med. 2007 Jun;10(3):728-40.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 230
133. Horne G, Payne S. Removing the boundaries: palliative care for
patients with heart failure. Palliat Med. 2004 May;18(4):291-6.
134. Murray SA, Boyd K, Kendall M, Worth A, Benton TF, Clausen H.
Dying of lung cancer or cardiac failure: prospective qualitative interview
study of patients and their carers in the community. BMJ. 2002 Oct
26;325(7370):929.
135. Johnson JO, Sulmasy DP, Nolan MT. Patients' Experiences of Being a
Burden on Family in Terminal Illness. J Hosp Palliat Nurs. 2007
Sep;9(5):264-9.
136. Small N, Barnes S, Gott M, Payne S, Parker C, Seamark D, et al.
Dying, death and bereavement: a qualitative study of the views of carers of
people with heart failure in the UK. BMC Palliat Care. 2009;8:6.
137. Waterworth S, Jorgensen D. It's not just about heart failure--voices
of older people in transition to dependence and death. Health Soc Care
Community. 2010 Mar;18(2):199-207.
138. Murray SA, Kendall M, Grant E, Boyd K, Barclay S, Sheikh A.
Patterns of social, psychological, and spiritual decline toward the end of life
in lung cancer and heart failure. J Pain Symptom Manage. 2007;34(4):393-
402.
139. Exley C, Field D, Jones L, Stokes T. Palliative care in the community
for cancer and end-stage cardiorespiratory disease: the views of patients,
lay-carers and health care professionals. Palliat Med. 2005 Jan;19(1):76-83.
140. Gysels MH, Higginson IJ. Caring for a person in advanced illness and
suffering from breathlessness at home: threats and resources. Palliat
Support Care. 2009 Jun;7(2):153-62.
141. Back AL, Young JP, McCown E, Engelberg RA, Vig EK, Reinke LF, et
al. Abandonment at the end of life from patient, caregiver, nurse, and
physician perspectives: loss of continuity and lack of closure. Arch Intern
Med. 2009 Mar 9;169(5):474-9.
142. Borgsteede SD, Graafland-Riedstra C, Deliens L, Francke AL, van
Eijk JT, Willems DL. Good end-of-life care according to patients and their
GPs. Br J Gen Pract. 2006 Jan;56(522):20-6.
143. Elkington H, White P, Addington-Hall J, Higgs R, Pettinari C. The last
year of life of COPD: a qualitative study of symptoms and services. Respir
Med. 2004 May;98(5):439-45.
144. Goodridge D, Duggleby W, Gjevre J, Rennie D. Exploring the quality
of dying of patients with chronic obstructive pulmonary disease in the
intensive care unit: a mixed methods study. Nurs Crit Care. 2009 Mar-
Apr;14(2):51-60.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 231
145. Reinke LF, Engelberg RA, Shannon SE, Wenrich MD, Vig EK, Back
AL, et al. Transitions regarding palliative and end-of-life care in severe
chronic obstructive pulmonary disease or advanced cancer: themes
identified by patients, families, and clinicians. J Palliat Med. 2008
May;11(4):601-9.
146. Shipman C, White S, Gysels M, White P. Access to care in advanced
COPD: factors that influence contact with general practice services. Prim
Care Respir J. 2009 Dec;18(4):273-8.
147. Wilson DM, Ross C, Goodridge D, Davis P, Landreville A, Roebuck K.
The care needs of community-dwelling seniors suffering from advanced
chronic obstructive pulmonary disease. Can J Aging. 2008
Winter;27(4):347-57.
148. Fried TR, van DC, Tinetti ME, Drickamer MA. Older persons'
preferences for site of treatment in acute illness. J Gen Intern Med.
1998;13(8):522-7.
149. Hall S, Legault A, Cote J. Dying means suffocating: perceptions of
people living with severe COPD facing the end of life. Int J Palliat Nurs.
2010 Sep;16(9):451-7.
150. Edmonds P, Vivat B, Burman R, Silber E, Higginson IJ. Los and
change: Experiences of people severely affected by multiple sclerosis.
Palliative Medicine, 2007; 21(2):101-7.
151. Hughes RA, Sinha A, Higginson I, Down K, Leigh PN. Living with
motor neurone disease: lives, experiences of services and suggestions for
change. Health Soc Care Community. 2005 Jan;13(1):64-74.
152. Nolan MT, Kub J, Hughes MT, Terry PB, Astrow AB, Carbo CA, et al.
Family health care decision making and self-efficacy with patients with ALS
at the end of life. Palliat Support Care. 2008 Sep;6(3):273-80.
153. Bolmsjo I, Hermeren G. Interviews with patients, family, and
caregivers in amyotrophic lateral sclerosis: comparing needs. J Palliat Care.
2001 Winter;17(4):236-40.
154. Boeije HR, Duijnstee MS, Grypdonck MH, Pool A. Encountering the
downward phase: biographical work in people with multiple sclerosis living
at home. Soc Sci Med. 2002 Sep;55(6):881-93.
155. Parker D, Maddocks I, Stern LM. The role of palliative care in
advanced muscular dystrophy and spinal muscular atrophy. J Paediatr Child
Health. 1999 Jun;35(3):245-50.
156. Berzoff J, Swantkowski J, Cohen LM. Developing a renal supportive
care team from the voices of patients, families, and palliative care staff.
Palliative & Supportive Care. 2008;6(2):133-9.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 232
157. Namiki S, Rowe J, Cooke M. Living with home-based haemodialysis:
insights from older people. J Clin Nurs. 2010 Feb;19(3-4):547-55.
158. Polaschek N. Living on dialysis: concerns of clients in a renal
setting. Journal of Advanced Nursing. 2003;41(1):44-52.
159. Polaschek N. Client attitudes towards home dialysis therapy. J Ren
Care. 2007 Jan-Mar;33(1):20-4.
160. Polaschek N. 'Doing dialysis at home': client attitudes towards renal
therapy. J Clin Nurs. 2007 Mar;16(3A):51-8.
161. Black BS, Fogarty LA, Phillips H, Finucane T, Loreck DJ, Baker A, et
al. Surrogate decision makers' understanding of dementia patients' prior
wishes for end-of-life care. J Aging Health. 2009 Jun;21(4):627-50.
162. Casarett DJ, Crowley RL, Hirschman KB. How should clinicians
describe hospice to patients and families? J Am Geriatr Soc. 2004
Nov;52(11):1923-8.
163. Chung K, Essex EL, Samson L. Does caregiver knowledge matter for
hospice enrollment and beyond? Pilot study of minority hospice patients.
Am J Hosp Palliat Care. 2009 Jun-Jul;26(3):165-71.
164. Evans WG, Cutson TM, Steinhauser KE, Tulsky JA. Is there no place
like home? Caregivers recall reasons for and experience upon transfer from
home hospice to inpatient facilities. J Palliat Med. 2006 Feb;9(1):100-10.
165. Forbes S, Bern-Klug M, Gessert C. End-of-life decision making for
nursing home residents with dementia. J Nurs Scholarsh. 2000;32(3):251-
8.
166. Gessert CE, Elliott BA, Peden-McAlpine C. Family decision making
for nursing home residents with dementia: rural-urban differences. Journal
of Rural Health. 2006;22(1):1-8.
167. Holley AP, Gorawara-Bhat R, Dale W, Hemmerich J, Cox-Hayley D.
Palliative Access Through Care at Home: experiences with an urban,
geriatric home palliative care program. J Am Geriatr Soc. 2009
Oct;57(10):1925-31.
168. McCarty CE, Volicer L. Hospice access for individuals with dementia.
Am J Alzheimers Dis Other Demen. 2009 Dec-2010 Jan;24(6):476-85.
169. McLennon SM, Habermann B, Davis LL. Deciding to institutionalize:
why do family members cease caregiving at home? J Neurosci Nurs. 2010
Apr;42(2):95-103.
170. Vig EK, Davenport NA, Pearlman RA. Good deaths, bad deaths, and
preferences for the end of life: a qualitative study of geriatric outpatients. J
Am Geriatr Soc. 2002 Sep;50(9):1541-8.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 233
171. Waldrop D, Kirkendall AM. Rural-urban differences in end-of-life
care: implications for practice. Soc Work Health Care. 2010 Mar;49(3):263-
89.
172. Waldrop DP, Rinfrette ES. Making the transition to hospice:
exploring hospice professionals' perspectives. Death Stud. 2009
Jul;33(6):557-80.
173. Froggatt K, Payne S. A survey of end-of-life care in care homes:
issues of definition and practice. Health Soc Care Community. 2006
Jul;14(4):341-8.
174. Lee S. Relocating elderly people and nursing staff from the NHS to
the independent sector. J Adv Nurs. 1998 Oct;28(4):859-64.
175. Payne S, Burton C, Addington-Hall J, Jones A. End-of-life issues in
acute stroke care: a qualitative study of the experiences and preferences of
patients and families. Palliat Med. 2010 Mar;24(2):146-53.
176. Treloar A, Crugel M, Adamis D. Palliative and end of life care of
dementia at home is feasible and rewarding: results from the Hope for
Home study. Dementia. 2009;8(3):335-47.
177. Hattori A, Masuda Y, Fetters MD, Uemura K, Mogi N, Kuzuya M. A
qualitative exploration of elderly patients' preferences for end-of-life care.
Japan Med Assoc J. 2005;48(8):388-97.
178. Wakunami M, Kawabata H, Murakami M, Maezawa M. Families'
acceptance of near death: a qualitative study of the process for introducing
end-of-life care. Geriatr Gerontol Int. 2009 Jun;9(2):140-7.
179. Brown JB, McWilliam CL, Mai V. Barriers and facilitators to seniors'
independence. Perceptions of seniors, caregivers, and health care providers.
Can Fam Physician. 1997 Mar;43:469-75.
180. Stajduhar KI, Davies B. Variations in and factors influencing family
members' decisions for palliative home care. Palliat Med. 2005
Jan;19(1):21-32.
181. Gott M, Small N, Barnes S et al. Older people's views of a good
death in heart failure: implications for palliative care provision. Social
Science & Medicine, 2008; 67: 1113-21.
182. Liu LF, Tinker A. Admission to nursing homes in Taiwan. Soc Policy
Adm. 2003;37(4):376-94.
183. Fried TR, Mor V. Frailty and hospitalization of long-term stay nursing
home residents. J Am Geriatr Soc. 1997 Mar;45(3):265-9.
184. Miller SC, Gozalo P, Mor V. Hospice enrollment and hospitalization of
dying nursing home patients. Am J Med. 2001 Jul;111(1):38-44.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 234
185. Krumholz HM, Parent EM, Tu N, Vaccarino V, Wang Y, Radford MJ,
et al. Readmission after hospitalization for congestive heart failure among
Medicare beneficiaries. Arch Intern Med. 1997 Jan 13;157(1):99-104.
186. Gielen B, Remacle A, Mertens R. Patterns of health care use and
expenditure during the last 6 months of life in Belgium: differences between
age categories in cancer and non-cancer patients. Health Policy. 2010
Sep;97(1):53-61.
187. Albert. Hospitalisation and Alzheimer's disease: results from a
community-based study. Journal of Gerontology. 1999;54A (5):M267-M71.
188. Dai. Unplanned hospital readmission and its predictors in patients
with chronic conditions. J Formos Med Assoc. 2002;101 (11 ):779-85.
189. Factor SA, Feustel PJ, Friedman JH, Comella CL, Goetz CG, Kurlan
R, et al. Longitudinal outcome of Parkinson's disease patients with
psychosis. Neurology. 2003 Jun 10;60(11):1756-61.
190. Rundek T, Mast H, Hartmann A, Boden-Albala B, Lennihan L, Lin IF,
et al. Predictors of resource use after acute hospitalization: the Northern
Manhattan Stroke Study. Neurology. 2000 Oct 24;55(8):1180-7.
191. Ahmed A. Predictors of nursing home admission for older adults
hospitalized with heart failure. Archives of Gerontology and Geriatrics.
2003;36(2):117-26.
192. Rockwood K, Stolee P, McDowell I. Factors associated with
institutionalization of older people in Canada: testing a multifactorial
definition of frailty. J Am Geriatr Soc. 1996 May;44(5):578-82.
193. Marengoni A, Aguero-Torres H, Timpini A, Cossi S, Fratiglioni L.
Rehabilitation and nursing home admission after hospitalization in acute
geriatric patients. J Am Med Dir Assoc. 2008 May;9(4):265-70.
194. Lim E. Predictors of Nursing Home Admission: A Social Work
Perspective. Australian Social Work. 2009;62(1):90-8.
195. Quartararo M, O'Neill TJ, Tang C, MacMaster M. Assessing the
residential care needs of nursing home applicants. Aust J Public Health.
1991 Sep;15(3):222-7.
196. Salive ME, Collins KS, Foley DJ, George LK. Predictors of nursing
home admission in a biracial population. Am J Public Health. 1993
Dec;83(12):1765-7.
197. Brody AA, Ciemins E, Newman J, Harrington C. The effects of an
inpatient palliative care team on discharge disposition. J Palliat Med. 2010
May;13(5):541-8.
198. Smith ER, Stevens AB. Predictors of discharges to a nursing home in
a hospital-based cohort. J Am Med Dir Assoc. 2009 Nov;10(9):623-9.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 235
199. Mehr DR, Williams BC, Fries BE. Predicting discharge outcomes of
VA nursing home residents. J Aging Health. 1997 May;9(2):244-65.
200. Bain KT, Maxwell TL, Strassels SA, Whellan DJ. Hospice use among
patients with heart failure. Am Heart J. 2009 Jul;158(1):118-25.
201. Hauptman PJ, Goodlin SJ, Lopatin M, Costanzo MR, Fonarow GC,
Yancy CW. Characteristics of patients hospitalized with acute
decompensated heart failure who are referred for hospice care. Arch Intern
Med. 2007 Oct 8;167(18):1990-7.
202. Givens JL, Tjia J, Zhou C, Emanuel E, Ash AS. Racial and ethnic
differences in hospice use among patients with heart failure. Arch Intern
Med. 2010 Mar 8;170(5):427-32.
203. Murray AM, Arko C, Chen SC, Gilbertson DT, Moss AH. Use of
hospice in the United States dialysis population. Clin J Am Soc Nephrol.
2006 Nov;1(6):1248-55.
204. Pinnock H, Kendall M, Murray SA, Worth A, Levack P, Porter M, et
al. Living and dying with severe chronic obstructive pulmonary disease:
multi-perspective longitudinal qualitative study. BMJ. 2011;342:d142.
205. Wilczynski NL, Haynes RB. EMBASE search strategies achieved high
sensitivity and specificity for retrieving methodologically sound systematic
reviews. J Clin Epidemiol. 2007 Jan;60(1):29-33.
206. Pope C, Mays N, Popay J. How can we synthesize qualitative and
quantitative evidence for healthcare policy-makers and managers? Healthc
Manage Forum. 2006 Spring;19(1):27-31.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 228
Appendix 1: Consultation participants
Table 24. Participants in the consultation to develop the theoretical model
Title Surname Name Position Area of expertise and input into the theoretical model
Prof Banerjee Sube Professor of Mental Health and Ageing, Institute of Psychiatry and Clinical
Director MHOA, South London and Maudsley NHS Foundation Trust Head, Centre for Innovation and Evaluation in Mental Health
Dementia
Dr. Bausewein Claudia Cicely Saunders Foundation Research Fellow Breathlessness, COPD
Mrs Blackwell Kara Head of Nursing for Cancer and Palliative Care, King’s College NHS Trust
Transitions
Dr Burman Rachel Consultant in Palliative Care/Hon Senior Lecturer, King’s College NHS Trust
Long Term Neurological Conditions
Ms Gomes Barbara Research Associate, KCL Factors affecting place of death; analysis; research; systematic reviews
Dr Gysels Marjolein Senior Research Fellow, Associate Research Professor Qualitative research; Breathlessness; Chronic Obstructive Pulmonary Disease (COPD)
Mrs. Hansford Penny Director of Nursing, St. Christopher’s Hospice; Management Fellow with the project
Dementia; Various aspects of palliative care
Prof. Higginson Irene J Head of Department, Professor of Palliative Care and Policy, Honorary Consultant King's College Hospital
All aspects of palliative care
Dr Jackson Diana Senior Research Fellow, King’s College London Long Term Neurological Conditions
Dr Johnston
Bridget Senior Research Fellow, School of Nursing and Midwifery, College of Medicine, Dentistry and Nursing, University of Dundee
Health policy, Palliative care nursing perspective
Prof Kalra Lalit Professor of Stroke Medicine, King’s College London The acute treatment of patients with stroke and their rehabilitation; prevention of stroke and reducing risks
Prof Murray Scott St Columba's Hospice Chair of Primary Palliative Care, Primary Palliative Care Research Group, Centre for Population Health Sciences, University of Edinburgh
All aspects of palliative care, and special interest across all non malignant conditions
Dr Murtagh Fliss Consultant and Clinical Senior Lecturer in Palliative Care, Systematic Review Project lead, KCL
End Stage Chronic Kidney Disease; All aspects of palliative care
Dr Petkova Hristina Research Associate, KCL Qualitative Research; Health care policy
Mrs Sam Emily Deputy Director of Policy Development The National Council for Palliative Care, NCPC
Parkinson’s Disease; Health care policy
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 237
Mrs Shipman Cathy Senior Research Fellow, KCL Palliative care in primary care; patient experiences and preferences; medical sociology, transition to community settings; Qualitative Research; Health care policy
Dr Simon Steffen Research Fellow, King's College London Palliative care for non-malignant conditions; Stroke; Chronic Heart Failure, Breathlessness
Mrs Shepherd Kate Nurse, King’s College NHS Trust End Stage Kidney Disease
Mrs Stacpoole Min Southwark Primary Care Trust, Dementia, Palliative care nursing
Prof Turner-Stokes
Lynne Herbert Dunhill Chair of Academic Rehabilitation Long Term Neurological Conditions
Dr Williams David Research Associate, King’s College London Replacement (respite) care among carers of people with long term neurological condition
In addition the Management Fellow organised two focus groups of relevant clinicians for discussion and feedback.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 230
Appendix 2: Final search strategy
The terms used in the final search strategy are presented in Table 25:
Table 25. Terms used for the final search strategy (these were refined and
adapted to each electronic database)
CRITERIA APPLICABLE ACROSS CONDITIONS
(1) Advanced
disease terms
and their
synonyms
palliative care OR terminal care OR terminally ill
OR hospice OR end-of-life OR (advanced adj3
(disease or condition or illness)) OR
(progressive adj3 (disease or condition or
illness))
(2) Factors
influencing the
outcomes
demography OR population OR birth OR (stage
or phase) of disease OR co-morbidity OR
(quality of life) OR (daily activity) OR (functional
status) OR prognosis OR (life expectancy) OR
(health service) OR facilities OR rehabilitation
OR respite OR access OR cost OR (resource
allocation) OR compensation OR (social care) OR
caregiver OR relative OR family OR support OR
spiritual OR ethnicity OR culture OR
hospitalization OR (social environment) OR rural
OR urban OR attitude OR preference OR choice
OR priority Or factor OR determinant OR
predictor OR effect OR influence OR cause OR
(transition or transfer or change or referral or
move) OR (communication or language) OR
(minority or group*) OR (advance care plan*)
OR ACP OR (living will) OR (clinical pathway)
(3) Outcomes (place adj3 (care or caring or death or dying))
OR (site adj3 (care or caring or death or dying))
OR (location adj3 (care or caring or death or
dying)) OR (setting adj3 (care or caring or death
or dying)) OR (change adj3 (care or place or
site or location or home or hospital or hospice or
setting)) OR (transition adj3 (care or place or
site or location or home or hospital or hospice or
setting)) OR (admission adj3 (care or place or
site or location or home or hospital or hospice or
setting)) OR (admission adj3 (care or place or
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 239
site or location or home or hospital or hospice or
setting)) OR (home or house or flat or residence
or dwelling) OR (nursing home) OR (care home)
OR (day care) OR facility
SPECIFIC DISEASE TERMS
(4) COPD (chronic obstructive pulmonary disease) OR
chronic obstructive lung disease) OR COPD
(5) CHF chronic heart failure OR chronic cardiac failure
OR congestive heart failure OR CHF OR CCF
(6) Dementia Dementia OR alzheimer’s disease
(7) ESKD Chronic kidney failure OR Chronic Kidney
Disease OR CKD OR chronic renal failure OR end
stage kidney failure OR end stage renal failure
(8) LTNC amyotrophic lateral sclerosis OR motor neuron
disease OR ALS OR MND OR multiple sclerosis
OR Parkinson’s disease OR multiple system
atrophy OR progressive supranuclear palsy OR
Huntington’s chorea OR Huntington$
(9) Stroke Stroke OR cerebrovascular disease OR brain
ischemia OR brain hemorrhage OR brain
infarction
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 240
Appendix 3: All included studies
1. Aaltonen M, Forma L, Rissanen P, Raitanen J, Jylhä M. Transitions in health and social service system at the end of life. European Journal of Ageing.
2010;7(2):91-100.
2. Aarsland D, Larsen JP, Tandberg E, Laake K. Predictors of nursing home
placement in Parkinson's disease: a population-based, prospective study. JAGS. 2000;48(8):938-42.
3. Abarshi E, Onwuteaka-Philipsen B, Donker G, Echteld M, Van den Block L,
Deliens L. General Practitioner Awareness of Preferred Place of Death and Correlates of Dying in a Preferred Place: A Nationwide Mortality Follow-Back
Study in The Netherlands. Journal of Pain and Symptom Management. 2009;38(4):568-77.
4. Ackermann RJ, Kemle KA. Death in a Nursing Home with active medical
management. Annals of Long-Term Care. 1999;7(8):313-9.
5. Addington-Hall JM, Lay M, Altmann D, McCarthy M. Community care for
stroke patients in the last year of life: results of a national retrospective survey of surviving family friends and officials. Health and Social Care in the Community. 1998;6(2):112-9.
6. Afessa B, Morales IJ, Scanlon P, Peters, SG. Prognostic factors, clinical course, and hospital outcome of COPD admitted to ICU for acute respiratory
failure. Crit Care Med. 2002;30(7):1610-5.
7. Agraharker M, Barclay C, Agraharkar A. Staff-assisted home hemodialysis in debilitated or terminally ill patients. International Urology and Nephrology.
2002;33:139-44.
8. Ahmed A, Allman RM, DeLong JF. Predictors of nursing home admission for
older adults hospitalized with heart failure. Archives of Gerontology and Geriatrics. 2003;36(2):117-26.
9. Aimonino N, Molaschi M, Salerno D, Roglia D. The home hospitalisation of
frail elderly patietns with advanced dementia. Arch Gerontol Geriatr Suppl. 2001;suppl 7 19-23.
10. Aimonino N, Tibaldi V, Leff B, Scarafiotti C, Marinello R, Zanocchi M, et al. Substitutive "Hospital at Home" Versus Inpatient Care for Elderly Patients with Exacerbations of Chronic Obstructive Pulmonary Disease: A Prospective
Randomized, Controlled Trial. Journal of the American Geriatrics Society. 2008;56(3):493-500.
11. Ai-Ping C. In-Hospital and 5-Year Mortality of Patients Treated in the ICU for Acute Exacerbation of COPD: A Retrospective Study. Chest. 2005;128(2):518-24.
12. Akiyama A, Numata K, Mikami H. Factors enabling home death of the elderly in an institution specializing in home medical care: Analysis of
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 241
apprehension of the bereaved family. Geriatrics & Gerontology International. 2008;8(2):73-9.
13. Albert SM, Costa R, Merchant C, Small S, Jenders RA, Stern Y. Hospitalisation and Alzheimer's disease: results from a community-based study. Journal of Gerontology. 1999;54A (5):M267-M71.
14. Al-Jahdali HH, Bahroon S, Babgi Y, Tanmin H, Al-Ghamdi SM, Al-Sayyari AA. Advance care planning preferences among dialysis patients and factoris
influencing their decisions. Saudi J Kidney Dis Transpl. 2009;20(2):232-9.
15. Anderson JE, Kraus J, Sturgeon D. Incidence, prevalence, and outcomes of end-stage renal disease patients placed in nursing homes. American Journal of
Kidney Diseases. 1993;21(6):619-27.
16. Anderson J, Sikorski I, Finucane T. Advance Care Planning by or on Behalf
of Peritoneal Dialysis Patients in Long-Term Care. American Journal of Kidney Diseases. 2006;48(1):122-7.
17. Annema C, Luttik M-L, Jaarsma T. Reasons for readmission in heart
failure: Perspectives of patients, caregivers, cardiologists, and heart failure nurses. Heart & Lung: The Journal of Acute and Critical Care. 2009;38(5):427-
34.
18. Arling G, Kane RL, Cooke V, Lewis T. Targeting Residents for Transitions from Nursing Home to Community. Health Services Research. 2010;45(3):691-
711.
19. Arora P, Kausz AT, Obrador GT, Ruthazer R, et al. Hospital utilization
among chronic dialysis patients. Journal of the American Society of Nephrology. 2000;11:740-6.
20. Ashcraft A, Owen D, Feng D. A Comparison of Cognitive and Functional
Care Differences in Four Long-term Care Settings. Journal of the American Medical Directors Association. 2006;7(2):96-101.
21. Au DH, Udris EM, Fihn SD et al. Differences in Health Care Utilization at the End of life among patients with COPD and lung cancer. Arch Int Med.
2006;166 326 - 31.
22. Babazono A, Weiner J, Hamada H et al. Health policy in transition: terminal care and site of death in Japan. J Health Serv Res Policy. 1998;3(2):77-
81.
23. Back AL, Young JP, McCown E, Engelberg RA, Vig EK, Reinke LF, et al.
Abandonment at the end of life from patient, caregiver, nurse, and physician perspectives: loss of continuity and lack of closure. Arch Intern Med. 2009 Mar 9;169(5):474-9.
24. Bain KT, Maxwell TL, Strassels SA, Whellan DJ. Hospice use among patients with heart failure. Am Heart J. 2009 Jul;158(1):118-25.
25. Banaszak-Hall J, Fendrick AM, Foster NL et al. Predicting nursing home admission: estimates from a 7-year followup of a nationally representative sample of older Americans. Alzheimer Dis Assoc Disord. 2004;18 (2):83-8.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 242
26. Becker MA, Boaz TL, Andel R et al. Predictors of preventable nursing home hospitalizations: the role of mental disorders and dementia. Am J Geriatr
Psychiatry. 2009;18(6):475-82.
27. Bekelman D, Black B, Shore A, Kasper J, Rabins P. Hospice Care in a Cohort of Elders with Dementia and Mild Cognitive Impairment. Journal of Pain
and Symptom Management. 2005;30(3):208-14.
28. Bell CL, Davis J, Harrigan RC, Somogyi-Zalud E, Tanabe MKG, Masaki KH.
Factors Associated with Place of Death for Elderly Japanese-American Men: The Honolulu Heart Program and Honolulu-Asia Aging Study. Journal of the American Geriatrics Society. 2009;57(4):714-8.
29. Bercovitz A, Gruber-Baldini AL, Burton LC, Hebel JR. Healthcare Utilization of Nursing Home Residents: Comparison Between Decedents and Survivors.
Journal of the American Geriatrics Society. 2005;53(12):2069-75.
30. Bergmann K, Foster EM, Justice AW, Matthews V. Management of the demented elderly patient in the community. British Journal of Psychiatry.
1978;132 441-9.
31. Berzoff J, Swantkowski J, Cohen LM. Developing a renal supportive care
team from the voices of patients, families, and palliative care staff. Palliat Support Care. 2008 Jun;6(2):133-9.
32. Biola H, Sloane PD, Williams CS, Daaleman TP, Zimmerman S. Preferences
Versus Practice: Life-Sustaining Treatments in Last Months of Life in Long-Term Care. Journal of the American Medical Directors Association. 2010;11(1):42-51.
33. Black BS, Fogarty LA, Phillips H, Finucane T, Loreck DJ, Baker A, et al. Surrogate decision makers' understanding of dementia patients' prior wishes for end-of-life care. J Aging Health. 2009 Jun;21(4):627-50.
34. Bly, Kissick. Hospice care for patients living alone: results of a demonstration program. The Hospice Journal. 1994;9(4):9-20.
35. Boeije HR, Duijnstee MS, Grypdonck MH, Pool A. Encountering the downward phase: biographical work in people with multiple sclerosis living at
home. Soc Sci Med. 2002 Sep;55(6):881-93.
36. Bolmsjo I, Hermeren G. Interviews with patients, family, and caregivers in amyotrophic lateral sclerosis: comparing needs. J Palliat Care. 2001
Winter;17(4):236-40.
37. Borgsteede SD, Graafland-Riedstra C, Deliens L, Francke AL, van Eijk JT,
Willems DL. Good end-of-life care according to patients and their GPs. Br J Gen Pract. 2006 Jan;56(522):20-6.
38. Bourbeau J, Julien M, Maltais F, et al. Reduction of Hospital Utilization in
Patients with COPD. Arch Int Med 2003;163:585-91.
39. Bradley EH, Fried TR, Kasl SV, Cicchetti DV, Johnson-Hurzeler R, Horwitz
SM. Referral of terminally ill patients for hospice: frequency and correlates. J Palliat Care. 2000 Winter;16(4):20-6.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 243
40. Bradley WG, Anderson F, Bromberg M, Gutmann L, Harati Y, Ross M, et al. Current management of ALS: comparison of the ALS CARE Database and the
AAN Practice Parameter. The American Academy of Neurology. Neurology. 2001 Aug 14;57(3):500-4.
41. Brannstrom M, Ekman I, Norberg A, Boman K, Strandberg G. Living with
severe chronic heart failure in palliative advanced home care. Eur J Cardiovasc Nurs. 2006 Dec;5(4):295-302.
42. Brazil K, Bedard M, Willison K. Factors associated with home death for individuals who receive home support services: a retrospective cohort study. BMC Palliat Care. 2002 Mar 25;1(1):2.
43. Brock DW, Foley DJ. Demography and epidemiology of dying in the US with emphasis on deaths of older persons. A good dying: shaping health care for
the last months of life. 1998.
44. Brock D, Foley DJ, Salive ME. Hospital and Nursing Home Use in the Last Three Months of Life. Journal of Aging and Health. 1996;8(3):307-19.
45. Brody AA, Ciemins E, Newman J, Harrington C. The effects of an inpatient palliative care team on discharge disposition. J Palliat Med. 2010 May;13(5):541-
8.
46. Brown JB, McWilliam CL, Mai V. Barriers and facilitators to seniors' independence. Perceptions of seniors, caregivers, and health care providers. Can
Fam Physician. 1997 Mar;43:469-75.
47. Brumley RD, Enguidanos S, Cherin DA. Effectiveness of a home-based
palliative care program for end-of-life. J Palliat Med. 2003 Oct;6(5):715-24.
48. Buechner JS. Trends and patterns in place of death, 1989-2000. Med Health R I. 2002 Sep;85(9):289-90.
49. Burt J, Shipman C, Richardson A, Ream E, Addington-Hall J. The experiences of older adults in the community dying from cancer and non-cancer
causes: a national survey of bereaved relatives. Age Ageing. 2010 Jan;39(1):86-91.
50. Burton LC, German PS, Gruber-Baldini AL, Hebel JR, Zimmerman S, Magaziner J. Medical care for nursing home residents: differences by dementia status. Epidemiology of Dementia in Nursing Homes Research Group. J Am
Geriatr Soc. 2001 Feb;49(2):142-7.
51. Busse R, Krauth C, Wagner HP et al. Hausärtzliche Betreuung und
Therapie von Finalkranken - eine Längsschnittstudie. Gesundheitswesen. 1997;59:231-5.
52. Caldwell PH, Arthur HM, Demers C. Preferences of patients with heart
failure for prognosis communication. Can J Cardiol. 2007 Aug;23(10):791-6.
53. Cantin B, Rothuisen LE, Buclin T, Pereira J, Mazzocato C. Referrals of
cancer versus non-cancer patients to a palliative care consult team: do they differ? J Palliat Care. 2009 Summer;25(2):92-9.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 244
54. Carson RC, Juszczak M, Davenport A, Burns A. Is maximum conservative management an equivalent treatment option to dialysis for elderly patients with
significant comorbid disease? Clin J Am Soc Nephrol. 2009 Oct;4(10):1611-9.
55. Casarett DJ, Marenberg ME, Karlawish JHT. Predictors of withdrawal from hospice. Journal of Palliative Medicine. 2001;4 (4):491-7.
56. Casarett DJ, Hirschmann KB, Coffey JF, Pierre L. Does a palliative care clinic have a role in improving end-of-life care? Results of a pilot program.
Journal of Palliative Medicine. 2002;5(3):387-96.
57. Casarett D Karlawish J, Morales K, et al. Improving the Use of Hospice Services in Nursing Homes: A Randomized Controlled Trial. JAMA: The Journal of
the American Medical Association. 2005;294(2):211-7.
58. Casarett D, Van Ness PH, O'Leary JR, Fried TR. Are patient preferences for
life-sustaining treatment really a barrier to hospice enrollment for older adults with serious illness? J Am Geriatr Soc. 2006 Mar;54(3):472-8.
59. Casarett DJ, Crowley RL, Hirschman KB. How should clinicians describe
hospice to patients and families? J Am Geriatr Soc. 2004 Nov;52(11):1923-8.
60. Casarett DJ, Hirschman KB, Crowley R, Galbraith LD, Leo M. Caregivers'
satisfaction with hospice care in the last 24 hours of life. Am J Hosp Palliat Care. 2003 May-Jun;20(3):205-10.
61. Chaudri MB, Kinnear WJ, Jefferson D. Patterns of mortality in patients with
motor neurone disease. Acta Neurol Scand. 2003 Jan;107(1):50-3.
62. Chen YJ, Narsavage GL. Factors Related to Chronic Obstructive Pulmonary
Disease Readmission in Taiwan. Western Journal of Nursing Research. 2006;28(1):105-24.
63. Christakis NA, Escarce JJ. Survival of Medicare patients after enrollment in
hospice programs. The New England Journal of Medicine. 1996;335:172-8.
64. Chung K, Essex EL, Samson L. Does caregiver knowledge matter for
hospice enrollment and beyond? Pilot study of minority hospice patients. Am J Hosp Palliat Care. 2009 Jun-Jul;26(3):165-71.
65. Clifford CA, Jolley DJ, Giles GG. Where people die in Victoria. The Medical Journal of Australia. 1991;155 446-56.
66. Cohen CA, Gold DP, Shulman KI, et al. Factors determining the decision to
institutionalize dementing individuals: a prospective study. The Gerontologist. 1993;33 (6 ):714 - 20.
67. Cohen LM, Germain MJ, Poppel DM, et al. Dying well after discontinuing the life-support treatment of dialysis. Archives of Internal Medicine. 2000;160:2513-8.
68. Cohen J, Bilsen J, Hooft P, Deboosere P, Wal G, Deliens L. Dying at home or in an institutionUsing death certificates to explore the factors associated with
place of death. Health Policy. 2006;78(2-3):319-29.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 245
69. Cohen LM, Germain MJ, Woods AL, Mirot A, Burleson JA. The family perspective of ESRD deaths. American Journal of Kidney Diseases.
2005;45(1):154-61.
70. Connor S, Elwert F, Spence C, Christakis N. Geographic Variation in Hospice Use in the United States in 2002. Journal of Pain and Symptom
Management. 2007;34(3):277-85.
71. Connor S, Elwert F, Spence C, Christakis N. Racial disparity in hospice use
in the United States in 2002. Palliative Medicine. 2008;22(3):205-13.
72. Currow DC, Burns CM, Abernethy AP. Place of death for people with non-cancer and cancer illness in South Australia: a population-based survey. Journal
of Palliative Care. 2008;24(3):144-50.
73. Curtis JR, Patrick DL, Engleberg RA, et al. A Measure of the Quality of
Dying and Death: Initial Validation using after-death Interviews with Family Members. JPSM. 2002;24(1 ):17-31.
74. Dai YT, Wu SC, Weng R. Unplanned hospital readmission and its predictors
in patients with chronic conditions. J Formos Med Assoc. 2002;101 (11 ):779-85.
75. Davison SN. End-of-life care preferences and needs: perceptions of patients with chronic kidney disease. Clin J Am Soc Nephrol. 2010 Feb;5(2):195-204.
76. DeCourtney CA, Jones K, Merriman MP, Heavener N, Branch PK. Establishing a culturally sensitive palliative care program in rural Alaska Native
American communities. J Palliat Med. 2003 Jun;6(3):501-10.
77. DesHarnais S, Carter RE, Hennessy W, Kurent JE, Carter C. Lack of concordance between physician and patient: reports on end-of-life care
discussions. J Palliat Med. 2007 Jun;10(3):728-40.
78. Dubinsky R, Chen J, Lai SM. Trends in hospital utilization and outcome for
patients with ALS Analysis of a large U.S. cohort. Neurology. 2006;67:777-80.
79. Dubinsky RM. No going home for hospitalized Huntington's disease
patients. Movement Disorders. 2005;20(10):1316-22.
80. duPreez AE, Smith MA, Liou JI, Frytak JR, Finch MD, Cleary JF, et al. Predictors of hospice utilization among acute stroke patients who died within
thirty days. J Palliat Med. 2008 Nov;11(9):1249-57.
81. Edes TE, Lindbloom EJ, Deal JL, Madsen RW. Improving care at lower cost
for end-stage heart and lung disease: integrating end of life planning with home care. Mo Med. 2006 Mar-Apr;103(2):146-51.
82. Edmonds P, Vivat B, Burman R, Silber E, Higginson IJ. Loss and change:
experiences of people severely affected by multiple sclerosis. Palliat Med. 2007 Mar;21(2):101-7.
83. Elkington H, White P, Addington-Hall J, Higgs R, Edmonds P. The healthcare needs of chronic obstructive pulmonary disease patients in the last year of life. Palliat Med. 2005 Sep;19(6):485-91.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 246
84. Elkington H, White P, Addington-Hall J, Higgs R, Pettinari C. The last year of life of COPD: a qualitative study of symptoms and services. Respir Med. 2004
May;98(5):439-45.
85. Emanuel EJ, Ash A, Yu W, Gazelle G, Levinsky NG, Saynina O, et al. Managed care, hospice use, site of death, and medical expenditures in the last
year of life. Arch Intern Med. 2002 Aug 12-26;162(15):1722-8.
86. Enguidanos S, Yip J, Wilber K. Ethnic variation in site of death of older
adults dually eligible for Medicaid and Medicare. J Am Geriatr Soc. 2005 Aug;53(8):1411-6.
87. Enguidanos SM, Cherin D, Brumley R. Home-based palliative care study:
site of death, and costs of medical care for patients with congestive heart failure, chronic obstructive pulmonary disease, and cancer. J Soc Work End Life Palliat
Care. 2005;1(3):37-56.
88. Evans WG, Cutson TM, Steinhauser KE, Tulsky JA. Is there no place like home? Caregivers recall reasons for and experience upon transfer from home
hospice to inpatient facilities. J Palliat Med. 2006 Feb;9(1):100-10.
89. Exley C, Field D, Jones L, Stokes T. Palliative care in the community for
cancer and end-stage cardiorespiratory disease: the views of patients, lay-carers and health care professionals. Palliat Med. 2005 Jan;19(1):76-83.
90. Fabris F, Molaschi M, Aimonino N, Ponzetto M, Maero B, Tibaldi V, et al.
Home care for demented subjects: new models of care and home-care allowance. Arch Gerontol Geriatr Suppl. 2004(9):155-62.
91. Factor SA, Feustel PJ, Friedman JH, Comella CL, Goetz CG, Kurlan R, et al. Longitudinal outcome of Parkinson's disease patients with psychosis. Neurology. 2003 Jun 10;60(11):1756-61.
92. Fillenbaum G, Heyman A, Peterson BL, Pieper CF, Weiman AL. Use and cost of hospitalization of patients with AD by stage and living arrangement:
CERAD XXI. Neurology. 2001 Jan 23;56(2):201-6.
93. Fleming J, Zhao J, Farquhar M, Brayne C, Barclay S. Place of death for the
'oldest old': > or =85-year-olds in the CC75C population-based cohort. Br J Gen Pract. 2010 Apr;60(573):171-9.
94. Flory J, Yinong YX, Gurol I, Levinsky N, Ash A, Emanuel E. Place of death:
U.S. trends since 1980. Health Aff (Millwood). 2004 May-Jun;23(3):194-200.
95. Forbes S, Bern-Klug M, Gessert C. End-of-life decision making for nursing
home residents with dementia. J Nurs Scholarsh. 2000;32(3):251-8.
96. Formiga F, Chivite D, Ortega C, Casas S, Ramon JM, Pujol R. End-of-life preferences in elderly patients admitted for heart failure. QJM. 2004
Dec;97(12):803-8.
97. Fried TR. Older Persons Preferences for Site of Terminal Care. Annals of
Internal Medicine. 1999;131(2):109-12.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 247
98. Fried TR, van Doorn C, O’Leary JR, et al. Older people preferences for home vs hospital care in treatment of acute illness. Arch Int Med.
2000;160:1501-6.
99. Fried TR, Gillick MR. Medical decision-making in the last six months of life: choices about limitation of care. J Am Geriatr Soc. 1994 Mar;42(3):303-7.
100. Fried TR, Gillick MR, Lipsitz LA. Whether to transfer? Factors associated with hospitalization and outcome of elderly long-term care patients with
pneumonia. J Gen Intern Med. 1995 May;10(5):246-50.
101. Fried TR, Mor V. Frailty and hospitalization of long-term stay nursing home residents. J Am Geriatr Soc. 1997 Mar;45(3):265-9.
102. Fried TR, Pollack DM, Drickamer MA, Tinetti ME. Who dies at home? Determinants of site of death for community-based long-term care patients. J
Am Geriatr Soc. 1999 Jan;47(1):25-9.
103. Fried TR, van Doorn C, Tinetti ME, Drickamer MA. Older persons' preferences for site of treatment in acute illness. J Gen Intern Med. 1998
Aug;13(8):522-7.
104. Froggatt K, Payne S. A survey of end-of-life care in care homes: issues of
definition and practice. Health Soc Care Community. 2006 Jul;14(4):341-8.
105. Fromme EK, Bascom PB, Smith MD, Tolle SW, Hanson L, Hickam DH, et al. Survival, mortality, and location of death for patients seen by a hospital-based
palliative care team. J Palliat Med. 2006 Aug;9(4):903-11.
106. Ganzini L, Johnston WS, Silveira MJ. The final month of life in patients with
ALS. Neurology. 2002 Aug 13;59(3):428-31.
107. Gessert CE, Elliott BA, Peden-McAlpine C. Family decision making for nursing home residents with dementia: rural-urban differences. Journal of Rural
Health. 2006;22(1):1-8.
108. Gessert CE, Haller IV, Kane RL, Degenholtz H. Rural-urban differences in
medical care for nursing home residents with severe dementia at the end of life. J Am Geriatr Soc. 2006 Aug;54(8):1199-205.
109. Gielen B, Remacle A, Mertens R. Patterns of health care use and expenditure during the last 6 months of life in Belgium: differences between age categories in cancer and non-cancer patients. Health Policy. 2010 Sep;97(1):53-
61.
110. Givens JL, Tjia J, Zhou C, Emanuel E, Ash AS. Racial and ethnic differences
in hospice use among patients with heart failure. Arch Intern Med. 2010 Mar 8;170(5):427-32.
111. Goetz CG, Stebbins GT. Risk factors for nursing home placement in
advanced Parkinson's disease. Neurology. 1993 Nov;43(11):2227-9.
112. Gomes B, Higginson IJ. Where people die (1974--2030): past trends,
future projections and implications for care. Palliat Med. 2008 Jan;22(1):33-41.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 248
113. Goodridge D, Duggleby W, Gjevre J, Rennie D. Exploring the quality of dying of patients with chronic obstructive pulmonary disease in the intensive care
unit: a mixed methods study. Nurs Crit Care. 2009 Mar-Apr;14(2):51-60.
114. Goodridge D, Lawson J, Duggleby W, Marciniuk D, Rennie D, Stang M. Health care utilization of patients with chronic obstructive pulmonary disease and
lung cancer in the last 12 months of life. Respir Med. 2008 Jun;102(6):885-91.
115. Gravil JH, Al-Rawas OA, Cotton MM, Flanigan U, Irwin A, Stevenson RD.
Home treatment of exacerbations of chronic obstructive pulmonary disease by an acute respiratory assessment service. Lancet. 1998 Jun 20;351(9119):1853-5.
116. Grbich C, Maddocks I, Parker D, Brown M, et al. Palliative care in aged
care facilities for residents with a non-cancer disease: results of a survey of aged care facilities in South Australia. Australian journal of ageing. 2005;24 (2):108-
13.
117. Groth-Juncker A, McCusker J. Where do elderly patients prefer to die? Place of death and patient characteristics of 100 elderly patients under the care
of a home health care team. J Am Geriatr Soc. 1983 Aug;31(8):457-61.
118. Gruneir A, Mor V, Weitzen S, Truchil R, Teno J, Roy J. Where people die: a
multilevel approach to understanding influences on site of death in America. Med Care Res Rev. 2007 Aug;64(4):351-78.
119. Gysels MH, Higginson IJ. Caring for a person in advanced illness and
suffering from breathlessness at home: threats and resources. Palliat Support Care. 2009 Jun;7(2):153-62.
119. Haggerty MC, Stockdale-Woolley R, Nair S. Respi-Care. An innovative home care program for the patient with chronic obstructive pulmonary disease. Chest. 1991 Sep;100(3):607-12.
120. Hall S, Legault A, Cote J. Dying means suffocating: perceptions of people living with severe COPD facing the end of life. Int J Palliat Nurs. 2010
Sep;16(9):451-7.
121. Haller IV, Gessert CE. Utilization of medical services at the end of life in
older adults with cognitive impairment: focus on outliers. J Palliat Med. 2007 Apr;10(2):400-7.
122. Hamner JB, Ellison KJ. Predictors of hospital readmission after discharge in
patients with congestive heart failure. Heart Lung. 2005 Jul-Aug;34(4):231-9.
123. Han B, Tiggle RB, Remsburg RE. Characteristics of Patients Receiving
Hospice Care at Home Versus in Nursing Homes: Results From the National Home and Hospice Care Survey and the National Nursing Home Survey. American Journal of Hospice and Palliative Medicine. 2007;24(6):479-86.
124. Hanrahan P, Luchins DJ. Access to hospice programmes in end stage dementia: A national survey of hospice programmes. Journal of the American
Geriatrics Society. 1995;43:56-9.
125. Hansen SM, Tolle SW, Martin DP. Factors associated with lower rates of in-hospital death. J Palliat Med. 2002 Oct;5(5):677-85.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 249
126. Hanson LC, Earp JA, Garrett J, Menon M, Danis M. Community physicians who provide terminal care. Arch Intern Med. 1999 May 24;159(10):1133-8.
127. Hanyu H, Sato T, Hirao K, Kanetaka H, Sakurai H, Iwamoto T. Differences in clinical course between dementia with Lewy bodies and Alzheimer's disease. Eur J Neurol. 2009 Feb;16(2):212-7.
128. Happ MB, Capezuti E, Strumpf NE, Wagner L, Cunningham S, Evans L, et al. Advance care planning and end-of-life care for hospitalized nursing home
residents. J Am Geriatr Soc. 2002 May;50(5):829-35.
129. Hattori A, Masuda Y, Fetters MD, Uemura K, Mogi N, Kuzuya M. A qualitative exploration of elderly patients' preferences for end-of-life care. Japan
Med Assoc J. 2005;48(8):388-97.
130. Haupt M, Kurz A. Predictors of Nursing Home placement in patients with
Alzheimers disease. International Journal of Geriatric psychiatry. 1993;8:741-6.
131. Hauptman PJ, Goodlin SJ, Lopatin M, Costanzo MR, Fonarow GC, Yancy CW. Characteristics of patients hospitalized with acute decompensated heart
failure who are referred for hospice care. Arch Intern Med. 2007 Oct 8;167(18):1990-7.
132. Haydar ZR, Lowe AJ, Kahveci KL, Weatherford W, Finucane T. Differences in end-of-life preferences between congestive heart failure and dementia in a medical house calls program. J Am Geriatr Soc. 2004 May;52(5):736-40.
133. Hayley DC. Not ready for hospice: Characteristics of patients in a pre-hospice program. American Journal of Hospice and Palliative Medicine.
2001;18(6):377-82.
134. Hebert R, Dubois MF, Wolfson C, et al. Factors associated with long term institutionalisation of older people with dementia: data from the canadian study
of health and aging. Journal of Gerontology. 2001;56a(11):M693-9.
135. Ho D, S H, Verne J, W S. Deaths from Renal Diseases in England 2001 to
2008. In: Programme NEoLC, editor. Bristol: National End of Life Care Intelligence Network; 2010.
136. Holley AP, Gorawara-Bhat R, Dale W, Hemmerich J, Cox-Hayley D. Palliative Access Through Care at Home: experiences with an urban, geriatric home palliative care program. J Am Geriatr Soc. 2009 Oct;57(10):1925-31.
137. Holloway RG, Ladwig S, Robb J, Kelly A, Nielsen E, Quill TE. Palliative care consultations in hospitalized stroke patients. J Palliat Med. 2010 Apr;13(4):407-
12.
138. Hope T, Keene J, Gedling K, Fairburn CG, Jacoby R. Predictors of institutionalization for people with dementia living at home with a carer. Int J
Geriatr Psychiatry. 1998 Oct;13(10):682-90.
139. Horne G, Payne S. Removing the boundaries: palliative care for patients
with heart failure. Palliat Med. 2004 May;18(4):291-6.
140. Horttana BM, Ahlstrom G, Fahlstrom G. Patterns of and reasons for relocation in dementia care. Geriatr Nurs. 2007 May-Jun;28(3):193-200.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 250
141. Houttekier D, Cohen J, Bilsen J, Addington-Hall J, Onwuteaka-Philipsen B, Deliens L. Place of death in metropolitan regions: metropolitan versus non-
metropolitan variation in place of death in Belgium, The Netherlands and England. Health Place. 2010 Jan;16(1):132-9.
142. Houttekier D, Cohen J, Bilsen J, Addington-Hall J, Onwuteaka-Philipsen
BD, Deliens L. Place of death of older persons with dementia. A study in five European countries. J Am Geriatr Soc. 2010 Apr;58(4):751-6.
143. Houttekier D, Cohen J, Bilsen J, Deboosere P, Verduyckt P, Deliens L. Determinants of the place of death in the Brussels metropolitan region. J Pain Symptom Manage. 2009 Jun;37(6):996-1005.
144. Hughes RA, Sinha A, Higginson I, Down K, Leigh PN. Living with motor neurone disease: lives, experiences of services and suggestions for change.
Health Soc Care Community. 2005 Jan;13(1):64-74.
145. Iwashyna TJ, Zhang JX, Christakis NA. Disease-specific patterns of hospice and related healthcare use in an incidence cohort of seriously ill elderly patients.
J Palliat Med. 2002 Aug;5(4):531-8.
146. Jakobsson E, Bergh I, Gaston-Johansson F, Stolt CM, Ohlen J. The turning
point: Clinical identification of dying and reorientation of care. J Palliat Med. 2006 Dec;9(6):1348-58.
147. Jakobsson E, Bergh I, Ohlen J, Oden A, Gaston-Johansson F. Utilization of
health-care services at the end-of-life. Health Policy. 2007 Aug;82(3):276-87.
148. Jakobsson E, Johnsson T, Persson LO, Gaston-Johansson F. End-of-life in a
Swedish population: demographics, social conditions and characteristics of places of death. Scand J Caring Sci. 2006 Mar;20(1):10-7.
149. Johnson JO, Sulmasy DP, Nolan MT. Patients' Experiences of Being a
Burden on Family in Terminal Illness. J Hosp Palliat Nurs. 2007 Sep;9(5):264-9.
150. Johnson KS, Kuchibhatala M, Sloane RJ, Tanis D, Galanos AN, Tulsky JA.
Ethnic differences in the place of death of elderly hospice enrollees. J Am Geriatr Soc. 2005 Dec;53(12):2209-15.
151. Johnson KS, Kuchibhatla M, Tanis D, Tulsky JA. Racial differences in the growth of noncancer diagnoses among hospice enrollees. J Pain Symptom Manage. 2007 Sep;34(3):286-93.
152. Katz BP, Zdeb MS, Therriault GD. Where people die. Public Health Rep. 1979 Nov-Dec;94(6):522-7.
153. Khan NA, Palepu A, Norena M, Ayas N, Wong H, Chittock D, et al. Differences in hospital mortality among critically ill patients of Asian, Native Indian, and European descent. Chest. 2008 Dec;134(6):1217-22.
154. Kim EY, Cho E, June KJ. Factors influencing use of home care and nursing homes. Journal of Advanced Nursing. 2006;54(4):511-7.
155. Kinnunen T, Saynajakangas O, Keistinen T. The COPD-induced hospitalization burden from first admission to death. Respir Med. 2007 Feb;101(2):294-9.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 251
156. Knopman DS, Berg JD, Thomas R, Grundman M, Thal LJ, Sano M. Nursing home placement is related to dementia progression: experience from a clinical
trial. Alzheimer's Disease Cooperative Study. Neurology. 1999 Mar 10;52(4):714-8.
157. Kronman AC, Ash AS, Freund KM, Hanchate A, Emanuel EJ. Can primary
care visits reduce hospital utilization among Medicare beneficiaries at the end of life? J Gen Intern Med. 2008 Sep;23(9):1330-5.
158. Krumholz HM, Parent EM, Tu N, Vaccarino V, Wang Y, Radford MJ, et al. Readmission after hospitalization for congestive heart failure among Medicare beneficiaries. Arch Intern Med. 1997 Jan 13;157(1):99-104.
159. Kutner JS, Meyer SA, Beaty BL, Kassner CT, Nowels DE, Beehler C. Outcomes and characteristics of patients discharged alive from hospice. J Am
Geriatr Soc. 2004 Aug;52(8):1337-42.
160. Kwak J, Haley WE, Chiriboga DA. Racial differences in hospice use and in-hospital death among Medicare and Medicaid dual-eligible nursing home
residents. Gerontologist. 2008 Feb;48(1):32-41.
161. Lamberg JL, Person CJ, Kiely DK, Mitchell SL. Decisions to hospitalize
nursing home residents dying with advanced dementia. J Am Geriatr Soc. 2005 Aug;53(8):1396-401.
162. Lane Mj, Davis DR, Cornman CB et al. Location of death as an indicator of
end-of-life costs for the person with dementia. Am J Alzheimers Disease. 1998;13(4):208-10.
163. Lechtzin N, Wiener CM, Clawson L, Chaudhry V, Diette GB. Hospitalization in amyotrophic lateral sclerosis: causes, costs, and outcomes. Neurology. 2001 Mar 27;56(6):753-7.
164. Lee S. Relocating elderly people and nursing staff from the NHS to the independent sector. J Adv Nurs. 1998 Oct;28(4):859-64.
165. Levy CR, Fish R, Kramer AM. Site of death in the hospital versus nursing home of Medicare skilled nursing facility residents admitted under Medicare's Part
A Benefit. J Am Geriatr Soc. 2004 Aug;52(8):1247-54.
166. Lieberman MA, Kramer JH. Factors affecting decisions to institutionalize demented elderly. Gerontologist. 1991 Jun;31(3):371-4.
167. Lim E. Predictors of Nursing Home Admission: A Social Work Perspective. Australian Social Work. 2009;62(1):90-8.
168. Liu H, Zhang TT, Ye J. Analysis of risk factors for hospital mortality in patients with chronic obstructive pulmonary diseases requiring invasive mechanical ventilation. Chin Med J (Engl). 2007 Feb 20;120(4):287-93.
169. Liu LF, Tinker A. Admission to nursing homes in Taiwan. Soc Policy Adm. 2003;37(4):376-94.
170. Lynn J, Teno JM, Phillips RS, Wu AW, Desbiens N, Harrold J, et al. Perceptions by family members of the dying experience of older and seriously ill patients. SUPPORT Investigators. Study to Understand Prognoses and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 252
Preferences for Outcomes and Risks of Treatments. Ann Intern Med. 1997 Jan 15;126(2):97-106.
171. Madigan EA, Curet OL. A data mining approach in home healthcare: outcomes and service use. BMC Health Serv Res. 2006;6:18.
172. Mandler RN, Anderson FA, Jr., Miller RG, Clawson L, Cudkowicz M, Del
Bene M. The ALS Patient Care Database: insights into end-of-life care in ALS. Amyotroph Lateral Scler Other Motor Neuron Disord. 2001 Dec;2(4):203-8.
173. Marengoni A, Aguero-Torres H, Timpini A, Cossi S, Fratiglioni L. Rehabilitation and nursing home admission after hospitalization in acute geriatric patients. J Am Med Dir Assoc. 2008 May;9(4):265-70.
174. Martikainen P, Moustgaard H, Murphy M, Einio EK, Koskinen S, Martelin T, et al. Gender, living arrangements, and social circumstances as determinants of
entry into and exit from long-term institutional care at older ages: a 6-year follow-up study of older Finns. Gerontologist. 2009 Feb;49(1):34-45.
175. McCarthy M, Hall JA, Ley M. Communication and choice in dying from
heart disease. J R Soc Med. 1997 Mar;90(3):128-31.
176. McCarty CE, Volicer L. Hospice access for individuals with dementia. Am J
Alzheimers Dis Other Demen. 2009 Dec-2010 Jan;24(6):476-85.
177. McGregor MJ, Tate RB, Ronald LA, McGrail KM. Variation in site of death among nursing home residents in British Columbia, Canada. J Palliat Med. 2007
Oct;10(5):1128-36.
178. McKinley RK, Stokes T, Exley C, Field D. Care of people dying with
malignant and cardiorespiratory disease in general practice. Br J Gen Pract. 2004 Dec;54(509):909-13.
179. McLennon SM, Habermann B, Davis LL. Deciding to institutionalize: why do
family members cease caregiving at home? J Neurosci Nurs. 2010 Apr;42(2):95-103.
180. Meeussen K, Van den Block L, Bossuyt N, Bilsen J, Echteld M, Van Casteren V, et al. GPs' awareness of patients' preference for place of death. Brit J
Gen Pract. 2009 Sep;59(566):665-70.
181. Mehr DR, Williams BC, Fries BE. Predicting discharge outcomes of VA nursing home residents. J Aging Health. 1997 May;9(2):244-65.
182. Menec VH, Nowicki S, Blandford A, Veselyuk D. Hospitalizations at the end of life among long-term care residents. J Gerontol A Biol Sci Med Sci. 2009
Mar;64(3):395-402.
183. Gott M, Small N, Barnes S et al. Older people's views of a good death in heart failure: implications for palliative care provision. Social Science &
Medicine, 2008; 67: 1113-21.
184. Miller SC, Kinzbrunner B, Pettit P, Williams JR. How does timing of hospice
referral influence hospice care in the last days of life. JAGS. 2003;51:798-806.
185. Miller SC, Gozalo P, Mor V. Hospice enrollment and hospitalization of dying nursing home patients. Am J Med. 2001 Jul;111(1):38-44.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 253
186. Miller SC, Weitzen S, Kinzbrunner B. Factors associated with the high prevalence of short hospice stays. J Palliat Med. 2003 Oct;6(5):725-36.
187. Mitchell SL, Teno JM, Miller SC, Mor V. A national study of the location of death of older persons with dementia. JAGS. 2005;53:299-305.
188. Mitchell SL, Kiely DK, Miller SC, Connor SR, Spence C, Teno JM. Hospice
care for patients with dementia. J Pain Symptom Manage. 2007 Jul;34(1):7-16.
189. Mitchell SL, Morris JN, Park PS, Fries BE. Terminal care for persons with
advanced dementia in the nursing home and home care settings. J Palliat Med. 2004 Dec;7(6):808-16.
190. Morcillo C, Valderas J, Aguado O, Delas J, Sort D, Pujadas R, et al.
Evaluation of a Home-Based Intervention in Heart Failure Patients. Results of a Randomized Study. Revista Espanola de Cardiologia. 2005;58(6):618-25.
191. Motiwala SS, Croxford R, Guerriere DN, Coyte PC. Predictors of place of death for seniors in Ontario: a population-based cohort analysis. Can J Aging. 2006 Winter;25(4):363-71.
192. Moylan T, Roberts M, Murray S. Medical needs and survival of NHS continuing care residents. Scott Med J. 2008 Aug;53(3):21-3.
193. Muramatsu N, Hoyem RL, Yin H, Campbell RT. Place of death among older Americans: does state spending on home- and community-based services promote home death? Med Care. 2008 Aug;46(8):829-38.
194. Murray AM, Arko C, Chen SC, Gilbertson DT, Moss AH. Use of hospice in the United States dialysis population. Clin J Am Soc Nephrol. 2006
Nov;1(6):1248-55.
195. Murray SA, Boyd K, Kendall M, Worth A, Benton TF, Clausen H. Dying of lung cancer or cardiac failure: prospective qualitative interview study of patients
and their carers in the community. BMJ. 2002 Oct 26;325(7370):929.
196. Murray SA, Kendall M, Grant E, Boyd K, Barclay S, Sheikh A. Patterns of
social, psychological, and spiritual decline toward the end of life in lung cancer and heart failure. J Pain Symptom Manage. 2007 Oct;34(4):393-402.
197. Namiki S, Rowe J, Cooke M. Living with home-based haemodialysis: insights from older people. J Clin Nurs. 2010 Feb;19(3-4):547-55.
198. Nanda A, Bourbonniere M, Wetle T, Teno J. Home care in the last year of
life: family member perceptions of unmet need associated with last place of care. J Am Med Dir Assoc. 2010 Jan;11(1):21-5.
199. Naylor MD, Brooten D, Campbell R, Jacobsen BS, Mezey MD, Pauly MV, et al. Comprehensive discharge planning and home follow-up of hospitalized elders: a randomized clinical trial. JAMA. 1999 Feb 17;281(7):613-20.
200. Neudert C, Oliver D, Wasner M, Borasio GD. The course of the terminal phase in patients with amyotrophic lateral sclerosis. J Neurol. 2001
Jul;248(7):612-6.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 254
201. Neumann PJ, Araki SS, Arcelus A, Longo A, Papadopoulos G, Kosik KS, et al. Measuring Alzheimer's disease progression with transition probabilities:
estimates from CERAD. Neurology. 2001 Sep 25;57(6):957-64.
202. Nicolas L, Tranchant L, Neuder Y, et al. Hospitalisation a domicile (HAD) et gériatrie: le reseau insuffisance cardiaque. Le revue de Gériatrie. 2004;29(10).
203. Nihtila EK, Martikainen PT, Koskinen SV, Reunanen AR, Noro AM, Hakkinen UT. Chronic conditions and the risk of long-term institutionalization among older
people. Eur J Public Health. 2008 Feb;18(1):77-84.
204. Nolan MT, Kub J, Hughes MT, Terry PB, Astrow AB, Carbo CA, et al. Family health care decision making and self-efficacy with patients with ALS at the end of
life. Palliat Support Care. 2008 Sep;6(3):273-80.
205. Oliver D. The quality of care and symptom control--the effects on the
terminal phase of ALS/MND. J Neurol Sci. 1996 Aug;139 Suppl:134-6.
206. Österlind J, Hansebo G, Lindqvist R, Ternestedt B-M. Moving on a roundabout at the end of life—What counts?Waiting times for transfer to
sheltered accommodation for older people in Sweden. Health Policy. 2009;91(2):183-8.
207. Ouslander JG, Lamb G, Perloe M, Givens JH, Kluge L, Rutland T, et al. Potentially avoidable hospitalizations of nursing home residents: frequency, causes, and costs: [see editorial comments by Drs. Jean F. Wyman and William
R. Hazzard, pp 760-761]. J Am Geriatr Soc. 2010 Apr;58(4):627-35.
208. Parashos SA, Maraganore DM, O'Brien PC, Rocca WA. Medical services
utilization and prognosis in Parkinson disease: a population-based study. Mayo Clin Proc. 2002 Sep;77(9):918-25.
209. Parker D, Maddocks I, Stern LM. The role of palliative care in advanced
muscular dystrophy and spinal muscular atrophy. J Paediatr Child Health. 1999 Jun;35(3):245-50.
210. Payne S, Burton C, Addington-Hall J, Jones A. End-of-life issues in acute stroke care: a qualitative study of the experiences and preferences of patients
and families. Palliat Med. 2010 Mar;24(2):146-53.
211. Pekmezaris R, Breuer L, Zaballero A, Wolf-Klein G, Jadoon E, D'Olimpio JT, et al. Predictors of site of death of end-of-life patients: the importance of
specificity in advance directives. J Palliat Med. 2004 Feb;7(1):9-17.
212. Polaschek N. Living on dialysis: concerns of clients in a renal setting. J Adv
Nurs. 2003 Jan;41(1):44-52.
213. Polaschek N. Client attitudes towards home dialysis therapy. J Ren Care. 2007 Jan-Mar;33(1):20-4.
214. Polaschek N. 'Doing dialysis at home': client attitudes towards renal therapy. J Clin Nurs. 2007 Mar;16(3A):51-8.
215. Pritchard RS, Fisher ES, Teno JM, Sharp SM, Reding DJ, Knaus WA, et al. Influence of patient preferences and local health system characteristics on the place of death. SUPPORT Investigators. Study to Understand Prognoses and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 255
Preferences for Risks and Outcomes of Treatment. J Am Geriatr Soc. 1998 Oct;46(10):1242-50.
216. Quartararo M, O'Neill TJ, Tang C, MacMaster M. Assessing the residential care needs of nursing home applicants. Aust J Public Health. 1991 Sep;15(3):222-7.
217. Ramon JM. Place of death of elderly persons in Catalonia(article in Spanish). Rev Clin Esp. 2006;206(11):549-55.
218. Reinke LF, Engelberg RA, Shannon SE, Wenrich MD, Vig EK, Back AL, et al. Transitions regarding palliative and end-of-life care in severe chronic obstructive pulmonary disease or advanced cancer: themes identified by patients, families,
and clinicians. J Palliat Med. 2008 May;11(4):601-9.
219. Retsinas J, Garrity P. Going home: analysis of nursing home discharges.
Gerontologist. 1986 Aug;26(4):431-6.
220. Rich MW, Beckham V, Wittenberg C, Leven CL, Freedland KE, Carney RM. A multidisciplinary intervention to prevent the readmission of elderly patients
with congestive heart failure. N Engl J Med. 1995 Nov 2;333(18):1190-5.
221. Rockwood K, Stolee P, McDowell I. Factors associated with
institutionalization of older people in Canada: testing a multifactorial definition of frailty. J Am Geriatr Soc. 1996 May;44(5):578-82.
222. Rosenberg PB, Mielke MM, Samus QM, Rosenblatt A, Baker A, Brandt J, et
al. Transition to nursing home from assisted living is not associated with dementia or dementia-related problem behaviors. J Am Med Dir Assoc. 2006
Feb;7(2):73-8.
223. Rosenwax L, McNamara B, Zilkens R. A population-based retrospective cohort study comparing care for Western Australians with and without
Alzheimer's disease in the last year of life. Health Soc Care Community. 2009 Feb;17(1):36-44.
224. Rundek T, Mast H, Hartmann A, Boden-Albala B, Lennihan L, Lin IF, et al. Predictors of resource use after acute hospitalization: the Northern Manhattan
Stroke Study. Neurology. 2000 Oct 24;55(8):1180-7.
225. Ruth K, Pring A, Verne J. Variations in Place of Death in England: Inequalities or appropriate consequences of age, gender and cause of death? In:
Programme NEoLC, editor. Bristol: National End of Life Care Intelligence Network; 2010.
226. Salive ME, Collins KS, Foley DJ, George LK. Predictors of nursing home admission in a biracial population. Am J Public Health. 1993 Dec;83(12):1765-7.
227. Sauvaget C, Tsuji I, Li JH, Hosokawa T, Fukao A, Hisamichi S. Factors
affecting death at home in Japan. Tohoku J Exp Med. 1996 Oct;180(2):87-98.
228. Schamp R, Tenkku L. Managed death in a PACE: pathways in present and
advance directives. J Am Med Dir Assoc. 2006 Jul;7(6):339-44.
229. Schwarz KA. Predictors of early hospital readmissions of older adults who are functionally impaired. J Gerontol Nurs. 2000 Jun;26(6):29-36.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 256
230. Sedeno MF, Nault D, Hamd DH, Bourbeau J. A self-management education program including an action plan for acute COPD exacerbations. COPD. 2009
Oct;6(5):352-8.
231. Setoguchi S, Glynn RJ, Stedman M, Flavell CM, Levin R, Stevenson LW. Hospice, opiates, and acute care service use among the elderly before death
from heart failure or cancer. Am Heart J. 2010 Jul;160(1):139-44.
232. Sharma G, Kuo YF, Freeman JL, Zhang DD, Goodwin JS. Outpatient follow-
up visit and 30-day emergency department visit and readmission in patients hospitalized for chronic obstructive pulmonary disease. Arch Intern Med. 2010 Oct 11;170(18):1664-70.
233. Shega JW, Hougham GW, Stocking CB, Cox-Hayley D, Sachs GA. Patients dying with dementia: experience at the end of life and impact of hospice care. J
Pain Symptom Manage. 2008 May;35(5):499-507.
234. Shinoda-Tagawa T, Ikegami N. Resident and facility characteristics associated with the site of death among Japanese nursing home residents. Age
Ageing. 2005 Sep;34(5):515-8.
235. Shipman C, White S, Gysels M, White P. Access to care in advanced COPD:
factors that influence contact with general practice services. Prim Care Respir J. 2009 Dec;18(4):273-8.
236. Silveira MJ, Copeland LA, Feudtner C. Likelihood of home death associated
with local rates of home birth: influence of local area healthcare preferences on site of death. Am J Public Health. 2006 Jul;96(7):1243-8.
237. Sloane PD, Zimmerman S, Williams CS, Hanson LC. Dying with dementia in long-term care. Gerontologist. 2008 Dec;48(6):741-51.
238. Small N, Barnes S, Gott M, Payne S, Parker C, Seamark D, et al. Dying,
death and bereavement: a qualitative study of the views of carers of people with heart failure in the UK. BMC Palliat Care. 2009;8:6.
239. Smith C, Da Silva-Gane M, Chandna S, Warwicker P, Greenwood R, Farrington K. Choosing not to dialyse: evaluation of planned non-dialytic
management in a cohort of patients with end-stage renal failure. Nephron Clin Pract. 2003;95(2):c40-6.
240. Smith ER, Stevens AB. Predictors of discharges to a nursing home in a
hospital-based cohort. J Am Med Dir Assoc. 2009 Nov;10(9):623-9.
241. Smith GE, O'Brien PC, Ivnik RJ, Kokmen E, Tangalos EG. Prospective
analysis of risk factors for nursing home placement of dementia patients. Neurology. 2001 Oct 23;57(8):1467-73.
242. Smith WR, Kellerman A, Brown JS. The impact of nursing home transfer
policies at the end of life on a public acute care hospital. J Am Geriatr Soc. 1995 Sep;43(9):1052-7.
243. Snell K, Pennington S, Lee M, Walker R. The place of death in Parkinson's disease. Age Ageing. 2009 Sep;38(5):617-9.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 257
244. Solloway M, LaFrance S, Bakitas M, Gerken M. A chart review of seven hundred eighty-two deaths in hospitals, nursing homes, and hospice/home care.
J Palliat Med. 2005 Aug;8(4):789-96.
245. Spataro R, Lo Re M, Piccoli T, Piccoli F, La Bella V. Causes and place of death in Italian patients with amyotrophic lateral sclerosis. Acta Neurologica
Scandinavica. 2010;122(3):217-23.
246. Stajduhar KI, Allan DE, Cohen SR, Heyland DK. Preferences for location of
death of seriously ill hospitalized patients: perspectives from Canadian patients and their family caregivers. Palliat Med. 2008 Jan;22(1):85-8.
247. Stajduhar KI, Davies B. Variations in and factors influencing family
members' decisions for palliative home care. Palliat Med. 2005 Jan;19(1):21-32.
248. Steinhauser KE, Christakis NA, Clipp EC, McNeilly M, McIntyre L, Tulsky JA.
Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA. 2000 Nov 15;284(19):2476-82.
249. Stewart S, Marley JE, Horowitz JD. Effects of a multidisciplinary, home-
based intervention on unplanned readmissions and survival among patients with chronic congestive heart failure: a randomised controlled study. Lancet. 1999
Sep 25;354(9184):1077-83.
250. Stuart B, D'Onofrio CN, Boatman S, Feigelman G. CHOICES: promoting early access to end-of-life care through home-based transition management. J
Palliat Med. 2003 Aug;6(4):671-83.
251. Takezako Y, Tamiya N, Kajii E. The nursing home versus the hospital as
the place of dying for nursing home residents in Japan. Health Policy. 2007 May;81(2-3):280-8.
252. Temkin-Greener H, Mukamel DB. Predicting place of death in the program
of all-inclusive care for the elderly (PACE): participant versus program characteristics. J Am Geriatr Soc. 2002 Jan;50(1):125-35.
253. Teno JM, Clarridge BR, Casey V, Welch LC, Wetle T, Shield R, et al. Family perspectives on end-of-life care at the last place of care. JAMA. 2004 Jan
7;291(1):88-93.
254. Teno JM, Mitchell SL, Skinner J, Kuo S, Fisher E, Intrator O, et al. Churning: the association between health care transitions and feeding tube
insertion for nursing home residents with advanced cognitive impairment. J Palliat Med. 2009 Apr;12(4):359-62.
255. Te-Wierik. Nursing home admission a study among elderly applicants. Tijdschr Gerontol Geriatr. 1991;22:209-15.
256. Thompson DR, Roebuck A, Stewart S. Effects of a nurse-led, clinic and
home-based intervention on recurrent hospital use in chronic heart failure. Eur J Heart Fail. 2005 Mar 16;7(3):377-84.
257. Tibaldi V, Isaia G, Scarafiotti C, Gariglio F, Zanocchi M, Bo M, et al. Hospital at home for elderly patients with acute decompensation of chronic heart failure: a prospective randomized controlled trial. Arch Intern Med. 2009 Sep
28;169(17):1569-75.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 258
258. Tolle SW, Tilden VP, Rosenfeld AG, Hickman SE. Family reports of barriers to optimal care of the dying. Nurs Res. 2000 Nov-Dec;49(6):310-7.
259. Treloar A, Crugel M, Adamis D. Palliative and end of life care of dementia at home is feasible and rewarding: results from the Hope for Home study. Dementia. 2009;8(3):335-47.
260. Tresch DD, Simpson WM, Jr., Burton JR. Relationship of long-term and acute-care facilities. The problem of patient transfer and continuity of care. J Am
Geriatr Soc. 1985 Dec;33(12):819-26.
261. Triplett P, Black BS, Phillips H, Richardson Fahrendorf S, Schwartz J, Angelino AF, et al. Content of advance directives for individuals with advanced
dementia. J Aging Health. 2008 Aug;20(5):583-96.
262. Tsuchihashi M, Tsutsui H, Kodama K, Kasagi F, Setoguchi S, Mohr M, et al.
Medical and socioenvironmental predictors of hospital readmission in patients with congestive heart failure. Am Heart J. 2001 Oct;142(4):E7.
263. Van den Block L, Deschepper R, Drieskens K, Bauwens S, Bilsen J, Bossuyt
N, et al. Hospitalisations at the end of life: using a sentinel surveillance network to study hospital use and associated patient, disease and healthcare factors. BMC
Health Services Research. 2007;7(1):69.
264. van der Velden LFJ, Francke AL, Hingstman L, Willems DL. Dying from cancer or other chronic diseases in the Netherlands: ten-year trends derived
from death certificate data. BMC Palliative Care. 2009;8(1):4.
265. Vasudev A, Palmer T, Thomas A, Burn D, Barker W. Factors predicting
discharge of Huntington's disease patients from a neuropsychiatry unit. Int Psychogeriatr. 2010 May;22(3):489-92.
266. Vig EK, Davenport NA, Pearlman RA. Good deaths, bad deaths, and
preferences for the end of life: a qualitative study of geriatric outpatients. J Am Geriatr Soc. 2002 Sep;50(9):1541-8.
267. Vinson JM, Rich MW, Sperry JC, Shah AS, McNamara T. Early readmission of elderly patients with congestive heart failure. J Am Geriatr Soc. 1990
Dec;38(12):1290-5.
268. Virnig BA, Kind S, McBean M, Fisher E. Geographic variation in hospice use prior to death. J Am Geriatr Soc. 2000 Sep;48(9):1117-25.
269. Volicer L, Hurley AC, Blasi ZV. Characteristics of dementia end-of-life care across care settings. Am J Hosp Palliat Care. 2003 May-Jun;20(3):191-200.
270. Wachterman MW, Sommers BD. The impact of gender and marital status on end-of-life care: evidence from the National Mortality Follow-Back Survey. J Palliat Med. 2006 Apr;9(2):343-52.
271. Wakunami M, Kawabata H, Murakami M, Maezawa M. Families' acceptance of near death: a qualitative study of the process for introducing end-of-life care.
Geriatr Gerontol Int. 2009 Jun;9(2):140-7.
272. Waldrop D, Kirkendall AM. Rural-urban differences in end-of-life care: implications for practice. Soc Work Health Care. 2010 Mar;49(3):263-89.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 259
273. Waldrop DP, Rinfrette ES. Making the transition to hospice: exploring hospice professionals' perspectives. Death Stud. 2009 Jul;33(6):557-80.
274. Waterworth S, Jorgensen D. It's not just about heart failure--voices of older people in transition to dependence and death. Health Soc Care Community. 2010 Mar;18(2):199-207.
275. Weinberger. Does increased access to primary care reduce hospital Readmissions? The New England Journal of Medicine. 1996;334 (22).
276. Weissert WC, Scanlon WJ. Determinants of nursing home discharge status. Med Care. 1985 Apr;23(4):333-43.
277. Weitzen S, Teno JM, Fennell M, Mor V. Factors associated with site of
death: a national study of where people die. Med Care. 2003 Feb;41(2):323-35.
278. Wennberg JE, Fisher ES, Stukel TA, Skinner JS, Sharp SM, Bronner KK.
Use of hospitals, physician visits, and hospice care during last six months of life among cohorts loyal to highly respected hospitals in the United States. BMJ. 2004 Mar 13;328(7440):607.
279. Wheelock VL, Tempkin T, Marder K, Nance M, Myers RH, Zhao H, et al. Predictors of nursing home placement in Huntington disease. Neurology. 2003
Mar 25;60(6):998-1001.
280. Williams MM, Xiong C, Morris JC, Galvin JE. Survival and mortality differences between dementia with Lewy bodies vs Alzheimer disease. Neurology.
2006 Dec 12;67(11):1935-41.
281. Wilson DM. The duration and degree of end-of-life dependency of home
care clients and hospital inpatients. Appl Nurs Res. 2002 May;15(2):81-6.
282. Wilson DM, Northcott HC, Truman CD, Smith SL, Anderson MC, Fainsinger RL, et al. Location of death in Canada. A comparison of 20th-century hospital and
nonhospital locations of death and corresponding population trends. Eval Health Prof. 2001 Dec;24(4):385-403.
283. Wilson DM, Ross C, Goodridge D, Davis P, Landreville A, Roebuck K. The care needs of community-dwelling seniors suffering from advanced chronic
obstructive pulmonary disease. Can J Aging. 2008 Winter;27(4):347-57.
284. Wilson RS, McCann JJ, Li Y, Aggarwal NT, Gilley DW, Evans DA. Nursing home placement, day care use, and cognitive decline in Alzheimer's disease. Am
J Psychiatry. 2007 Jun;164(6):910-5.
285. Wong CF, McCarthy M, Howse ML, Williams PS. Factors affecting survival
in advanced chronic kidney disease patients who choose not to receive dialysis. Ren Fail. 2007;29(6):653-9.
286. Yaffe K. Patient and Caregiver Characteristics and Nursing Home
Placement in Patients With Dementia. JAMA: The Journal of the American Medical Association. 2002;287(16):2090-7.
287. Yang L, Sakamoto N, Marui E. A study of home deaths in Japan from 1951 to 2002. BMC Palliative Care. 2006;5(1):2.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 260
288. Yasamura. Factors relating to place of death of Japanese people from a small town in a rural area. Ageing Clin Exp Res. 2000;12:449-54.
289. Young AJ, Rogers A, Addington-Hall JM. The quality and adequacy of care received at home in the last 3 months of life by people who died following a stroke: a retrospective survey of surviving family and friends using the Views of
Informal Carers Evaluation of Services questionnaire. Health Soc Care Community. 2008 Jul;16(4):419-28.
290. Ziegert K, Hogstedt B, Fridlund B, Lidell E. Time distribution factors of hospital and home care among chronic haemodialysis patients. EDTNA ERCA J. 2004 Jan-Mar;30(1):19-22.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 253
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 262
Appendix 4: Excluded papers Author(s) Year Title Journal Vol Issue Start
page End
page Reason for exclusion
Acorn M 2008 In-home palliative care increased patient satisfaction and reduced use and costs of medical services
Evidence-Based Nursing
11 1 22 22 Exclude no original data
Agraharkar M, Patlovany M, Henry S, Bonds B
2003 Promoting use of home dialysis Advances in peritoneal dialysis Conference on Peritoneal Dialysis
19 163 167 Exclude no outcome of interest
Ahluwalia SC, Fried TR, 2009 Physician factors associated with outpatient palliative care referral
Palliative Medicine 23 7 608 615 Exclude no outcome of interest
Ahmad S, O'Mahony MS 2005 Where older people die: A retrospective population-based study
QJM - Monthly Journal of the Association of Physicians
98 12 865 870 Exclude not original study
Akazawa Y, Satoh H, Takiguchi K, Sekizawa K, Yamashita YT
2002 Nursing discharge summaries of deceased patients with respiratory diseases. [References]
International Journal of Nursing Practice
8 5 282 285 Exclude no outcome of interest
Almagro P, Calbo E, Ochoa de Echagen A, Barreiro B, Quintana S, Heredia JL, Garau J
2002 Mortality after hospitalization for COPD Chest 121 5 1441 1448 Exclude no outcome of interest
Anderson JE, Sturgeon D, Lindsay J, Schiller A
1990 Use of continuous ambulatory peritoneal dialysis in a nursing home: Patient characteristics, technique success, and survival predictors
American Journal of Kidney Diseases
16 2 137 141 Exclude no outcome of interest
Anderson MA, Hanson KS, DeVilder NW, Helms LB
1996 Hospital readmissions during home care: A pilot study
Journal of Community Health Nursing
13 1 1 12 Exclude, not eligible population, not advanced disease
Andersson FL, Svensson K, Gerhardsson D
2006 Hospital use for COPD patients during the last few years of their life
Respiratory Medicine 100 8 1436 1441 Exclude no outcome of interest
Aupperle PM, MacPhee ER, Strozeski JE, Finn M, Heath JM
2004 Hospice use for the patient with advanced Alzheimer's disease: The role of the geriatric psychiatrist
American Journal of Hospice and Palliative Medicine
21 6 427 437 Exclude no outcome of interest
Aylin P, Majeed FA, Cook DG 1996 Home visiting by general practitioners in England and Wales
British Medical Journal 313 7051 207 210 Exclude no outcome of interest
Balcells E, Anto JM, Gea J, Gomez FP, Rodr¡guez E, Marin A, Ferrer A, de Batlle J, Farrero E, Benet M, Orozco-Levi M, Ferrer J, Agusto AG, Gíldiz JB, Belda J, Garcia-Aymerich J
2009 Characteristics of patients admitted for the first time for COPD exacerbation
Respiratory Medicine 103 9 1293 1302 Exclude not advanced disease
Balinsky W, Rehman S 1984 Home health care: a comparative analysis of hospital-based and community-based agency patients
Home health care services quarterly
5 1 45 60 Exclude not advanced disease
Barker WH, Zimmer JG, Hall WJ, Ruff BC, Freundlich CB, Eggert GM
1994 Rates, patterns, causes, and costs of hospitalization of nursing home residents: A population-based study
American Journal of Public Health
84 10 1615 1620 Exclude not advanced disease
Barnato AE, Labor RE, Freeborne NE, Jayes RL, Campbell DE, Lynn J
2005 Qualitative analysis of Medicare claims in the last 3 years of life: A pilot study
Journal of the American Geriatrics Society
53 1 66 73 Exclude not appropriate population
Barry PP, Crescenzi CA, Radovsky L, Kern DC, Steel K
1988 Why elderly patients refuse hospitalization Journal of the American Geriatrics Society
36 5 419 424 Exclude not advanced disease
Bartolomeo N, Trerotoli P, Moretti A, Serio G
2008 A markov model to evaluate hospital readmission
BMC Medical Research Methodology
8 Exclude no outcome of interest
Bass DM, Pestello FP, Garland TN 1984 Experiences with home hospice care: Determinants of place of death
Death Education 8 4 199 222 Exclude not eligible population
Behnke M, Taube C, Kirsten D, Lehnigk B, JÂrres RA, Magnussen H
2000 Home-based exercise is capable of preserving hospital-based improvements in severe chronic obstructive pulmonary disease
Respiratory Medicine 94 12 1184 1191 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 263
Bell CL, Somogyi-Zalud E, Masaki KH 2010 Factors Associated with Congruence Between Preferred and Actual Place of Death
Journal of Pain and Symptom Management
39 3 591 604 Exclude not original research
Bergman H, Clarfield AM 1991 Appropriateness of patient transfer from a nursing home to an acute-care hospital: A study of emergency room visits and hospital admissions
Journal of the American Geriatrics Society
39 12 1164 1168 Exclude not eligible population
Bixby MB, Konick-McMahon J, McKenna CG
2000 Applying the transitional care model to elderly patients with heart failure
Journal of Cardiovascular Nursing
14 3 53 63 Exclude not appropriate design only 4 case studies
Black B. 2009 Understanding Dementia Patients’ Prior Wishes for End-of-Life Care
Southern Online Journal of Nursing Research
Exclude not appropriate design
Black BS, Rabins PV, German PS 1999 Predictors of nursing home placement among elderly public housing residents
Gerontologist 39 5 559 568 Exclude no outcome of interest
Bleeker F, Kruschinski C, Breull A, Berndt M, Hummers-Pradier E
2007 Characteristics of palliative care patients in general practice. [German]
Zeitschrift fur Allgemeinmedizin
83 12 477 482 Exclude no outcome of interest
Borrayo EA, Salmon JR, Polivka L, Dunlop BD
2002 Utilization across the continuum of long-term care services
Gerontologist 42 5 603 612 Exclude not eligible population
Branch LG, Jette AM 1982 A prospective study of long-term care institutionalization among the aged
American Journal of Public Health
72 12 1373 1379 Exclude not eligible population
Brazil K, Bedard M, Krueger P, Abernathy T, Lohfeld L, Willison K
2005 Service preferences among family caregivers of the terminally ill
Journal of Palliative Medicine 8 1 69 78 Exclude no outcome of interest
Brodaty H, McGilchrist C, Harris L, Peters KE
1993 Time until institutionalization and death in patients with dementia: Role of caregiver training and risk factors
Archives of Neurology 50 6 643 650 Exclude not eligible population
Brunker CP 2008 Challenges of non cancer patients transitioning to hospice
Home Health Care Management & Practice
20 5 400 403 Exclude not original article
Buchanan JL, Murkofsky RL, O'Malley AJ, Karon SL, Zimmerman D, Caudry DJ, Marcantonio ER
2006 Nursing home capabilities and decisions to hospitalize: A survey of medical directors and directors of nursing
Journal of the American Geriatrics Society
54 3 458 465 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 264
Buchanan RJ, Wang S, Huang C, Simpson P, Manyam BV
2002 Analyses of nursing home residents with Parkinson's disease using the minimum data set
Parkinsonism and Related Disorders
8 5 369 380 Exclude no outcome of interest
Carey EC, Covinsky KE, Lui L, Eng C, Sands LP, Walter LC
2008 Prediction of mortality in community-living frail elderly people with long-term care needs
Journal of the American Geriatrics Society
56 1 68 75 Exclude no outcome of interest
Carey HB, Chorney W, Pherson K, Finkelstein FO, Kliger AS
2001 Continuous peritoneal dialysis and the extended care facility
American Journal of Kidney Diseases
37 3 580 587 Exclude no outcome of interest
Carter MW, Porell FW 2005 Vulnerable populations at risk of potentially avoidable hospitalizations: The case of nursing home residents with Alzheimer's disease
American Journal of Alzheimer's Disease and other Dementias
20 6 349 358 Exclude not advanced disease
Casarett D, Crowley R, Stevenson C, Xie S, Teno J
2005 Making difficult decisions about hospice enrollment: What do patients and families want to know?
Journal of the American Geriatrics Society
53 2 249 254 Exclude no outcome of interest
Casarett DJ, Hirschman KB, Henry MR 2001 Does hospice have a role in nursing home care at the end of life?
Journal of the American Geriatrics Society
49 11 1493 1498 Exclude no outcome of interest
Catalano C, Goodship THJ, Graham KA, Marino C, Brown AL, Tapson JS, Ward MK, Wilkinson R
1996 Withdrawal of renal replacement therapy in Newcastle upon Tyne: 1964-1993
Nephrology Dialysis Transplantation
11 1 133 139 Exclude no outcome of interest
Chahine LM, Malik B, Davis M 2008 Palliative care needs of patients with neurologic or neurosurgical conditions
European Journal of Neurology 15 12 1265 1272 Exclude no outcome of interest
Charlton RC 1991 Attitudes towards care of the dying: A questionnaire survey of general practice attenders
Family Practice 8 4 356 359 Exclude not eligible population
Chau PH, Kwok T, Woo J, Chan F, Hui E, Chan KC
2010 Disagreement in preference for residential care between family caregivers and elders is greater among cognitively impaired elders group than cognitively intact elders group
International Journal of Geriatric Psychiatry
25 1 46 54 Exclude not advanced disease
Chen LK, Peng LN, Lin MH, Lai HY, Hwang SJ, Lan CF
2010 Predicting mortality of older residents in long-term care facilities: Comorbidity or care problems?
Journal of the American Medical Directors Association
11 8 567 571 Exclude no outcome of interest
Chen WT, Wang SJ, Lu SR, Fuh JL 2002 Which level of care is preferred for end-stage dementia? Survey of Taiwanese caregivers
Journal of Geriatric Psychiatry & Neurology
15 1 16 19 Exclude no outcome of interest
Christakis NA 1994 Timing of referral of terminally ill patients to an outpatient hospice
Journal of General Internal Medicine
9 6 314 320 Exclude no outcome of interest
Christakis NA, Escarce JJ 1996 Survival of Medicare patients after enrollment in hospice programs
New England Journal of Medicine
335 3 172 178 Exclude no outcome of interest
Christakis NA, Iwashyna TJ, Zhang JX 2002 Care after the Onset of Serious Illness: A Novel Claims-Based Dataset Exploiting Substantial Cross-Set Linkages to Study End-of-Life Care
Journal of Palliative Medicine 5 4 515 529 Exclude no outcome of interest
Ciemins EL, Stuart B, Gerber R, Newman J, Bauman M
2006 An evaluation of the Advanced Illness Management (AIM) program: Increasing hospice utilization in the San Francisco Bay area
Journal of Palliative Medicine 9 6 1401 1411 Exclude no outcome of interest, no information on diagnosis and population can not be defined
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 265
Claessens MT, Lynn J, Zhong Z, Desbiens NA, Phillips RS, Wu AW, Harrell FE Jr., Connors AFJ
2000 Dying with lung cancer or chronic obstructive pulmonary disease: Insights from SUPPORT
Journal of the American Geriatrics Society
48 5 Suppl S146 S153 Exclude no outcome of interest
Clarke M, Bratan T, Kulkarni S, Jones R 2007 The impact of remote patient monitoring in managing silent myocardial infarction in a residential home setting
Anadolu Kardiyoloji Dergisi 7 SUPPL. 1
186 188 Exclude not appropriate design only 3 case studies
Coehlo DP, Hooker K, Bowman S 2007 Institutional placement of persons with dementia: What predicts occurrence and timing?
Journal of Family Nursing 13 2 253 277 Exclude not advanced disease
Cogen R, Patterson B, Chavin S, Cogen J, Landsberg L, Posner J
1992 Surrogate decision-maker preferences for medical care of severely demented nursing home patients
Archives of Internal Medicine 152 9 1885 1888 Exclude no outcome of interest
Cohen J, Bilsen J, Miccinesi G, LÂfmark R, Addington-Hall J, Kaasa S, Norup M, Van Der Wal G, Deliens L
2007 Using death certificate data to study place of death in 9 European countries: Opportunities and weaknesses
BMC Public Health 7 Exclude no original data
Colerick EJ, George LK 1986 Predictors of institutionalization among caregivers of patients with Alzheimer's disease
Journal of the American Geriatrics Society
34 7 493 498 Exclude not appropriate population
Collins C, Ogle K 1994 Patterns of predeath service use by dementia patients with a family caregiver
Journal of the American Geriatrics Society
42 7 719 722 Exclude no outcome of interest
Condelius A, Edberg A, Hallberg IR, Jakobsson U
2010 Utilization of medical healthcare among people receiving long-term care at home or in special accommodation
Scandinavian Journal of Caring Sciences
24 2 404 413 Exclude not advanced disease
Connolly S, O'Reilly D, 2009 Variation in care home admission across areas of Northern Ireland
Age and Ageing 38 4 461 465 Exclude not advanced disease
Davidson PM, Paull G;Introna K, Cockburn J, Davis JM, Rees D, Gorman D, Magann L, Lafferty M, Dracup K
2004 Integrated, Collaborative Palliative Care in Heart Failure The St. George Heart Failure Service Experience 1999–2002
Journal of Cardiovascular Nursing
19 1 68 75 Exclude not appropriate design
Duthie J, Chesson R 1996 Physiotherapy in private nursing homes Physiotherapy 82 10 566 572 Exclude no outcome of interest
Eaker ED, Vierkant RA, Mickel SF 2002 Predictors of nursing home admission and/or death in incident Alzheimer's disease and other dementia cases compared to controls: A population-based study
Journal of Clinical Epidemiology
55 5 462 468 Exclude not advanced disease
Ellershaw J 2007 Care of the dying: What a differene an LCP makes!
Palliative Medicine 21 5 365 368 Exclude no outcome of interest
Emanuel LL, Von Gunten CF, Ferris FD 2000 Gaps in end-of-life care Archives of Family Medicine 9 10 1176 1180 Exclude not original study
Engel SE, Kiely DK, Mitchell SL 2006 Satisfaction with end-of-life care for nursing home residents with advanced dementia
Journal of the American Geriatrics Society
54 10 1567 1572 Exclude no outcome of interest
Enguidanos S, Chambers J 2008 In-home palliative care increased patient satisfaction and reduced use and costs of medical services: Commentary
Evidence-Based Medicine 13 1 19 Exclude not original study
Finlayson M 2002 Changes predicting long-term care use among the oldest-old
Gerontologist 42 4 443 453 Exclude not advanced disease
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 266
Finn JC, Flicker L, Mackenzie E, Jacobs IG, Fatovich DM, Drummond S, Harris M, Holman DCDJ, Sprivulis P
2006 Interface between residential aged care facilities and a teaching hospital emergency department in Western Australia
Medical Journal of Australia 184 9 432 435 Exclude not advanced disease
Forbat L, Service KP 2005 Who cares? Contextual layers in end-of-life care for people with intellectual disability and dementia
Dementia 4 3 413 431 Exclude no outcome of interest
Forma L, Rissanen P, Noro A, Raitanen J, Jylhñ M
2007 Health and social service use among old people in the last 2 years of life
European Journal of Ageing 4 3 145 154 Exclude no outcome of interest
Formiga F, Espel E, Chivite D, Pujol R 2002 Dying from heart failure in hospital: Palliative decision making analysis
Heart 88 2 187 Exclude no outcome of interest
Fox E, Landrum-McNiff K, Zhong Z, Dawson NV, Wu AW, Lynn J
1999 Evaluation of prognostic criteria for determining hospice eligibility in patients with advanced lung, heart, or liver disease
Journal of the American Medical Association
282 17 1638 1645 Exclude no outcome of interest
Freeborne N, Lynn J, Desbiens NA, Phillips RS, Hamel MB, Covinsky KE, Lynn J
2000 Insights about dying from the SUPPORT project Journal of the American Geriatrics Society
48 5 SUPPL.
Exclude no outcome of interest
Freedman VA, Berkman LF, Rapp SR, Ostfeld AM
1994 Family networks: Predictors of nursing home entry
American Journal of Public Health
84 5 843 845 Exclude not appropriate population
Freiman MP, Murtaugh CM, Shine KI 1995 Interactions: Between hospital and nursing home use
Public Health Reports 110 5 546 554 Exclude no outcome of interest
Torres H, Von Strauss E, Viitanen M, Winblad B, Fratiglioni L
2001 Institutionalization in the elderly: The role of chronic diseases and dementia. Cross-sectional and longitudinal data from a population-based study
Journal of Clinical Epidemiology
54 8 795 801 Exclude not advanced disease
Galanos AN, Hays JC, Moore JD, Poppe A
2004 Where do continuing care retirement community residents die? [1]
Journal of the American Geriatrics Society
52 8 1401 1402 Exclude not original study, letter to editor
Garcia-Garcia G, Briseno-Renteria G, Luquin-Arellan VH, Gao Z, Gill J, Tonelli M
2007 Survival among patients with kidney failure in Jalisco, Mexico
Journal of the American Society of Nephrology
18 6 1922 1927 Exclude no outcome of interest
Gibson R, 1973 Supporting the patient in the home British Medical Journal 1 5844 35 36 Exclude no outcome of interest
Gilmartin M, Celli BR, Bach JR, Hill NS, Sortor- Leger, Hoffman, Fletcher, Make B, Stoller JK
1994 Transition from the intensive care unit to home: Patient selection and discharge planning
Respiratory Care 39 5 456 480 Exclude no original data
Gonsalkorale M 2005 Palliative care in Parkinson's disease CME Journal Geriatric Medicine 7 1 22 28 Exclude no outcome of interest
Goodman C, Evans C, Wilcock J, Froggatt K, Drennan V, Sampson E, Blanchard M, Bissett M, Iliffe S
2010 End of life care for community dwelling older people with dementia: An integrated review
International Journal of Geriatric Psychiatry
25 4 329 337 Exclude no outcome of interest, context paper
Gott M, Seymour J, Bellamy G, Clark D, Ahmedzai S
2004 Older people's views about home as a place of care at the end of life
Palliative Medicine 18 5 460 467 Exclude not eligible population
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 267
Goy ER, Carter J, Ganzini L 2008 Neurologic disease at the end of life: caregiver descriptions of Parkinson disease and amyotrophic lateral sclerosis
Journal of Palliative Medicine 11 4 548 554 Exclude no outcome of interest
Goy ER, Carter JH, Ganzini L 2008 Jun
Needs and experiences of care-givers for family members dying with Parkinson disease
Journal of Palliative Care 24 2 69 75 Exclude no outcome of interest
Grande GE, Todd CJ, Barclay SIG, Farquhar MC
1999 Does hospital at home for palliative care facilitate death at home? Randomised controlled trial
British Medical Journal 319 7223 1472 1475 Exclude not eligible population
Greene VL, Ondrich JI 1990 Risk factors for nursing home admissions and exits: A discrete-time hazard function approach
Journals of Gerontology 45 6 Exclude not advanced disease
Gruneir A, Mor V, Weitzen S, Truchil R, Teno J, Roy J
2007 Where people die: A multilevel approach to understanding influences on site of death in America
Medical Care Research and Review
64 4 351 378 Exclude no outcome of interest, context paper
Gruneir A, Mor V, Weitzen S, Truchil R, Teno J, Roy J
2007 Where people die: A multilevel approach to understanding influences on site of death in America
Medical Care Research and Review
64 4 351 378 Exclude not original research
Grunfeld E, Glossop R, McDowell I, Danbrook C
1997 Caring for elderly people at home: The consequences to caregivers
CMAJ 157 8 1101 1105 Exclude no outcome of interest
Hain D 2010 Place of Death by PCT by month Eastern Region Public Health Observatory Exclude not eligible population
Han B, Remsburg RE, Iwashyna TJ 2006 Differences in hospice use between black and white patients during the period 1992 through 2000
Medical Care 44 8 731 737 Exclude no outcome of interest
Han B, Tiggle RB, Remsburg RE 2008 Characteristics of patients receiving hospice care at home versus in nursing homes: Results from the National Home and Hospice Care Survey and the National Nursing Home Survey
American Journal of Hospice and Palliative Medicine
24 6 479 486 Exclude no outcome of interest
Hanratty B, Jacoby A, Whitehead M 2008 Socioeconomic differences in service use, payment and receipt of illness-related benefits in the last year of life: Findings from the British Household Panel Survey
Palliative Medicine 22 3 248 255 Exclude not eligible population
Hansen-Flaschen J 2004 Chronic obstructive pulmonary disease: the last year of life
Respiratory care 49 1 90 97 Exclude not original study
Happ MB, Naylor MD, Roe-Prior P. 1997 Factors contributing to rehospitalization of elderly patients with heart failure
The Journal of Cardiovascular Nursing
11 4 75 84 Exclude not advanced disease
Haupt BJ 2003 Characteristics of hospice care discharges and their length of service: United States, 2000
Vital and health statistics Series 13, Data from the National Health Survey
154 1 36 Exclude no outcome of interest
Hernandez C, Casas A, Escarrabill J, et al
2003 Home hospitalisation of exacerbated chronic obstructive pulmonary disease patients
European Respiratory Journal 21 1 58 67 Exclude not advanced disease
Heyman A, Peterson B, Fillenbaum G, Pieper C
1997 Predictors of time to institutionalization of patients with Alzheimer's disease: The CERAD experience, Part XVII
Neurology 48 5 1304 1309 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 268
Hicks F, Corcoran G 1993 Should hospices offer respite admissions to patients with motor neurone disease?
Palliative Medicine 7 2 145 150 Exclude no outcome of interest
Hirahara S 2004 [The present state and subject of home hospice for patients with non-cancer disorders]. [Japanese]
Gan to Kagaku Ryoho [Japanese Journal of Cancer & Chemotherapy]
31 Suppl 8 Exclude no outcome of interest
Hirakawa Y, Masuda Y, Kuzuya M, Iguchi A, Asahi T, Uemura K
2006 Home end-of-life care for advanced dementia vs advanced cancer elderly patients: Dying elderly at home project
Japanese Journal of Geriatrics 43 3 355 360 Exclude no outcome of interest
Hirakawa Y, Masuda Y, Kuzuya M, Iguchi A, Uemura K,
2006 How elderly people die of nonmalignant pulmonary disease at home
Japan Medical Association Journal
49 3 106 111 Exclude no outcome of interest
Hirakawa Y, Masuda Y, Kuzuya M, Kimata T, Iguchi A, Uemura K
2006 Symptom experience and patterns of end-of-life - Home care for elderly patients with cancer vs. those without cancer in Japan
Japan Medical Association Journal
49 07-Aug 243 250 Exclude no outcome of interest
Hirakawa Y, Masuda Y, Kuzuya M, Kimata T, Iguchi A, Uemura K
2006 End-of-life experience of demented elderly patients at home: Findings from DEATH project
Psychogeriatrics 6 2 60 67 Exclude no outcome of interest
Hirakawa Y, Masuda Y, Uemura K, Kuzuya M, Kimata T, Iguchi A
2006 End-of-life care at group homes for patients with dementia in Japan: findings from an analysis of policy-related differences
Archives of Gerontology & Geriatrics
42 3 233 245 Exclude no outcome of interest
Hoenig H, Sloane R, Horner RD, Zolkewitz M, Reker D
2001 Differences in rehabilitation services and outcomes among stroke patients cared for in Veterans Hospitals
Health Services Research 35 6 1293 1318 Exclude not eligible population
Howarth G, Willison KB 1995 Preventing crises in palliative care in the home. Role of family physicians and nurses
Canadian Family Physician 41 439 445 Exclude no outcome of interest
Hu WY, Chiu TY, Cheng YR, Chuang RB, Chen CY
2004 Why Taiwanese hospice patients want to stay in hospital: Health-care professionals' beliefs and solutions
Supportive Care in Cancer 12 5 285 292 Exclude not eligible population (cancer)
Huang YC, Huang SJ, Ko WJ 2009 Going home to die from surgical intensive care units
Intensive Care Medicine 35 5 810 815 Exclude no outcome of interest
Huber DL, McClelland E 2003 Patient preferences and discharge planning transitions
Journal of Professional Nursing 19 4 204 210 Exclude no outcome of interest
Hunt RW, Bond MJ, Groth RK, King PM 1991 Place of death in South Australia. Patterns from 1910 to 1987
Medical Journal of Australia 155 8 549 553 Exclude not eligible population (cancer)
Husebo BS, Husebo S 2005 Nursing homes as arenas of terminal care: Practical aspects. [Norwegian]
Tidsskrift for den Norske Laegeforening
125 10 1352 1354 Exclude no original data
Iecovich E, Carmel S, Bachner YG 2009 Where they want to die: Correlates of elderly persons' preferences for death site
Social Work in Public Health 24 6 527 542 Exclude not advanced disease
Inglis SC, Pearson S, Treen S, Gallasch T, Horowitz JD, Stewart S
2006 Extending the horizon in chronic heart failure: Effects of multidisciplinary, home-based intervention relative to usual care
Circulation 114 23 2466 2473 Exclude not advanced disease
Iwashyna TJ, Chang VW, Zhang JX, Christakis NA
2002 The lack of effect of market structure on hospice use
Health Services Research 37 6 1531 1551 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 269
Johnson KS, Kuchibhatla M, Tulsky JA 2008 What explains racial differences in the use of advance directives and attitudes toward hospice care?
Journal of the American Geriatrics Society
56 10 1953 1958 Exclude not appropriate population
Jordhoy MS, Fayers P, Saltnes T, hlner-Elmqvist M, Jannert M, Kaasa
2000 A palliative-care intervention and death at home: A cluster randomised trial
Lancet 356 9233 888 893 Exclude not eligible population
Joshi K, Guthmann R, Kishman C 2006 How do we decide when a patient with nonmalignant disease is eligible for hospice care?
Journal of Family Practice 55 6 525 529 Exclude no original data
Kohnlein P, Kobler A, Raubold S, Worrell M, Kurt A, Gdynia HJ, Sperfeld AD, Ludolph AC
2008 Palliative care and circumstances of dying in German ALS patients using non-invasive ventilation
Amyotrophic Lateral Sclerosis 9 2 91 98 Exclude no outcome of interest
Kao HFS, Travis SS, Acton GJ, Talerico KA,
2004 Relocation to a long-term care facility: Working with patients and families before, during, and after
Journal of Psychosocial Nursing and Mental Health Services
42 3 10 16 Exclude no original data
Kapo J, Harrold J, Carroll JT, Rickerson E, Casarett D
2005 Are we referring patients to hospice too late? Patients' and families' opinions
Journal of Palliative Medicine 8 3 521 527 Exclude no outcome of interest
Kapo J, Macmoran H, Casarett D 2005 'Lost to follow-up': Ethnic disparities in continuity of hospice care at the end of life
Journal of Palliative Medicine 8 3 603 608 Exclude not appropriate population
Kavanagh S, Knapp M 1998 The impact on general practitioners of the changing balance of care for elderly people living in institutions
British Medical Journal 317 7154 322 327 Exclude not advanced disease
Kind AJH, Smith MA, Liou JI, Pandhi N, Frytak JR, Finch MD
2010 Discharge Destination's Effect on Bounce-Back Risk in Black, White, and Hispanic Acute Ischemic Stroke Patients
Archives of Physical Medicine and Rehabilitation
91 2 189 195 Exclude not advanced disease
Kite S, Jones K, Tookman A 1999 Specialist palliative care and patients with noncancer diagnoses: The experience of a service
Palliative Medicine 13 6 477 484 Exclude no outcome of interest
Kovach CR 1998 Nursing home dementia care units. Providing a continuum of care rather than aging in place
Journal of Gerontological Nursing
24 4 30 36 Exclude no outcome of interest
Kralik D, Anderson B 2008 Differences in home-based palliative care service utilisation of people with cancer and non-cancer conditions
Journal of Clinical Nursing 17 11C 429 435 Exclude no outcome of interest
Krivickas LS, Shockley L, Mitsumoto H 1997 Home care of patients with amyotrophic lateral sclerosis (ALS)
Journal of the Neurological Sciences
152 SUPPL. 1
Exclude no outcome of interest
Lagman RL, Walsh D, Kunkle C, LeGrand SB, Davis MP
2006 Deaths in an Academic Medical Center Journal of Palliative Medicine 9 6 1260 1263 Exclude no population of interest
Lagoe RJ, Noetscher CM, Murphy MP 2001 Hospital readmission: predicting the risk Journal of Nursing Care Quality 15 4 69 83 Exclude no outcome of interest
Landi F, Lattanzio F, Gambassi G, Zuccal+á G, Sgadari A, Panfilo M, Ruffilli MP, Bernabei R
1999 A model for integrated home care of frail older patients: The Silver Network project
Aging - Clinical and Experimental Research
11 4 262 272 Exclude not advanced disease
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 270
Landi F, Onder G, Cesari M, Zamboni V, Russo A, Barillaro C, Bernabei R
2006 Functional decline in frail community-dwelling stroke patients
European Journal of Neurology 13 1 17 23 Exclude no outcome of interest
Leovi-ì M 2009 Analysis of long term care in the context of social and health services in social institutional facilities in Slovakia
Central European Journal of Public Health
17 3 128 132 Exclude no outcome of interest
Lee T, Kovner CT, Mezey MD, Ko IS 2001 Factors influencing long-term home care utilization by the older population: Implications for targeting
Public Health Nursing 18 6 443 449 Exclude no outcome of interest
Leff B, Burton L, Mader SL, Naughton B, Burl J, Inouye SK, Greenough III WB, Guido S, Langston C, Frick KD, Steinwachs D, Burton JR
2005 Hospital at home: Feasibility and outcomes of a program to provide hospital-level care at home for acutely III older patients
Annals of Internal Medicine 143 11 Exclude not advanced disease
Levenson JW, McCarthy EP, Lynn J, Davis RB, Phillips RS
2000 The last six months of life for patients with congestive heart failure
Journal of the American Geriatrics Society
48 5 SUPPL.
S101 S109 Exclude no outcome of interest
Lewis BE, O'Mara P, Pezzella S 1999 The development and implementation of a disease management program in a managed care setting
Annals of Long Term Care 7 6 226 231 Exclude not advanced disease
Lewis L 1988 Housing people with HIV dementia AIDS Patient Care 2 3 35 37 Exclude non original research
Lezovic M 2009 Analysis of the structure of services provided in the healthcare facilities in long term care in Slovakia
Bratislava Medical Journal 110 11 701 704 Exclude no outcome of interest
Liu LF, Tinker A 2001 Factors associated with nursing home entry for older people in Taiwan, Republic of China
Journal of Interprofessional Care
15 3 245 255 Exclude no outcome of interest
London MR, McSkimming S, Drew N, Quinn C, Carney B
2005 Evaluation of a comprehensive, adaptable, life-affirming, longitudinal (CALL) palliative care project
Journal of Palliative Medicine 8 6 1214 1225 Exclude no outcome of interest
Louis ED, Henchcliffe C, Bateman BT, Schumacher C
2007 Young-onset Parkinson's disease: Hospital utilization and medical comorbidity in a nationwide survey
Neuroepidemiology 29 01-Feb 39 43 Exclude no outcome of interest
Lubin S 1992 Palliative care--could your patient have been managed at home?
Journal of palliative care 8 2 18 22 Exclude no outcome of interest
Luchins DJ, Hanrahan P 1993 What is appropriate health care for end-stage dementia?
Journal of the American Geriatrics Society
41 1 25 30 Exclude no outcome of interest
Luchins DJ, Hanrahan P, Murphy K 1997 Criteria for enrolling dementia patients in hospice
Journal of the American Geriatrics Society
45 9 1054 1059 Exclude no outcome of interest
Ludke RL, Smucker DR 2007 Racial differences in the willingness to use hospice services
Journal of Palliative Medicine 10 6 1329 1337 Exclude no outcome of interest
Lunney JR, Lynn J, Hogan C 2002 Profiles of older medicare decedents Journal of the American Geriatrics Society
50 6 1108 1112 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 271
Lynn J, Harrell F, Cohn F, Wagner D, Connors AF
1997 Prognoses of seriously ill hospitalized patients on the days before death: Implications for patient care and public policy
New Horizons: Science and Practice of Acute Medicine
5 1 56 61 Exclude no outcome of interest
MacGregor MS, Agar JWM, Blagg CR 2006 Home haemodialysis - International trends and variation
Nephrology Dialysis Transplantation
21 7 1934 1945 Exclude no outcome of interest
Madigan E 2008 People with heart failure and home health care resource use and outcomes
Journal of Clinical Nursing 17 7B 253 259 Exclude not eligible population
Maessen M, Veldink JH, Onwuteaka-Philipsen BD, de Vries JM, Wokke JH, G, LH
2009 Trends and determinants of end-of-life practices in ALS in the Netherlands
Neurology 73 12 954 961 Exclude no outcome of interest
Marek KD, Rantz MJ 2000 Aging in place: a new model for long-term care Nursing administration quarterly
24 3 1 11 Exclude no outcome of interest, context paper
Martikainen P, Nihtil+ñ E, Moustgaard H
2008 The effects of socioeconomic status and health on transitions in living arrangements and mortality: A longitudinal analysis of elderly finnish men and women from 1997 to 2002
Journals of Gerontology - Series B Psychological Sciences and Social Sciences
63 2 Exclude no outcome of interest
Mazzocato C, Michel-Nemitz J, Anwar D, Michel P
2010 The last days of dying stroke patients referred to a palliative care consult team in an acute hospital
European Journal of Neurology 17 1 73 77 Exclude no outcome of interest
McCarthy EP, Pencina MJ, Kelly-Hayes M, Evans JC, Oberacker EJ, D'Agostino S, Burns RB, Murabito JM
2008 Advance care planning and health care preferences of community-dwelling elders: The Framingham heart study
Journals of Gerontology - Series A Biological Sciences and Medical Sciences
63 9 951 959 Exclude not appropriate population
McCarthy M, ddington-Hall J, Altmann D
1997 The experience of dying with dementia: A retrospective study
International Journal of Geriatric Psychiatry
12 3 404 409 Exclude no outcome of interest
McCluskey L, Houseman G 2004 Medicare Hospice Referral Criteria for Patients with Amyotrophic Lateral Sclerosis: A Need for Improvement
Journal of Palliative Medicine 7 1 47 53 Exclude no outcome of interest
McDonald GJ 1981 A home care program for patients with chronic lung disease
Nursing Clinics of North America
16 2 259 273 Exclude no original data
McGhee SM, Schooling CM, Wong LC, Leung GM, Ho LM, Thomas GN, Ho DS, Lam TH, Hedley AJ
2008 Does smoking affect hospital use before death? A comparison of ever- and never-smokers in the last years of life
Medical Care 46 6 614 619 Exclude no outcome of interest
McKeown A, Agar R, Gambles M, Ellershaw JE, Hugel H
2008 Renal failure and specialist palliative care: an assessment of current referral practice
International Journal of Palliative Nursing
14 9 454 458 Exclude no outcome of interest
McWhinney IR, Bass MJ, Orr V 1995 Factors associated with location of death (home or hospital) of patients referred to a palliative care team
CMAJ 152 3 361 367 Exclude not appropriate population, < 10% non-cancer patients, only relevant group is MND with 2.3%
Medici V, Rossaro L, Wegelin JA, Kamboj A, Nakai J, Fisher K, Meyers FJ
2008 The utility of the model for end-stage liver disease score: A reliable guide for liver transplant candidacy and, for select patients, simultaneous hospice referral
Liver Transplantation 14 8 1100 1106 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 272
Meehan T, Robertson S, Vermeer C 2001 The impact of relocation on elderly patients with mental illness
The Australian and New Zealand journal of mental health nursing
10 4 236 242 Exclude not advanced disease
Menecier P, Debacker J, Ravier A, Arezes C, Menecier-Ossia L, Lenoir C, Guillermet M
2002 Death in medical facilities for elders Revue de Geriatrie 27 8 629 634 Excluded no outcome of interest
Miller EA, Rosenheck RA 2006 Risk of nursing home admission in association with mental illness nationally in the Department of Veterans Affairs
Medical Care 44 4 343 351 Exclude not advanced disease
Miller SC 2004 Hospice care in nursing homes: Is site of care associated with visit volume?
Journal of the American Geriatrics Society
52 8 1331 1336 Exclude no outcome of interest
Miller SC, Intrator O, Gozalo P, Roy J, Barber J, Mor V
2004 Government expenditures at the end of life for short- and long-stay nursing home residents: Differences by hospice enrollment status
Journal of the American Geriatrics Society
52 8 1284 1292 Exclude no outcome of interest
Miller SC, Weitzen S, Kinzbrunner B 2003 Factors Associated with the High Prevalence of Short Hospice Stays
Journal of Palliative Medicine 6 5 725 736 Exclude no outcome of interest
Mitchell SL, Kiely DK, Jones RN, Prigerson H, Volicer L, Teno JM
2006 Advanced dementia research in the nursing home: The CASCADE study
Alzheimer Disease and Associated Disorders
20 3 166 175 Exclude no outcome of interest
Mitchell SL, Teno JM, Intrator O, Feng Z, Mor V
2007 Decisions to forgo hospitalization in advanced dementia: A nationwide study
Journal of the American Geriatrics Society
55 3 432 438 Exclude no outcome of interest
Mitty EL, 2004 Assisted living: Aging in place and palliative care Geriatric Nursing 25 3 149 156+163
Exclude no outcome of interest
Moinpour CM, Polissar L, Conrad DA 1990 Factors associated with length of stay in hospice Medical Care 28 4 363 368 Exclude no outcome of interest
Montgomery RJV, Kosloski K 1994 A longitudinal analysis of nursing home placement for dependent elders cared for by spouses vs adult children
Journals of Gerontology 49 2 Exclude not advanced disease
Moss AH 2001 Shared decision making in dialysis: A new clinical practice guideline to assist with dialysis-related ethics consultations
Journal of Clinical Ethics 12 4 406 414 Exclude no outcome of interest
Munday D, Dale J, Murray S 2007 Choice and place of death: Individual preferences, uncertainty, and the availability of care
Journal of the Royal Society of Medicine
100 5 211 215 Exclude not appropriate design, a case study
Munroe CA, Sirdofsky MD, Kuru T, Anderson ED
2007 End-of-life decision making in 42 patients with amyotrophic lateral sclerosis
Respiratory Care 52 8 996 999 Exclude no outcome of interest
Murray LM, Laditka SB 2010 Care transitions by older adults from nursing homes to hospitals: Implications for long-term care practice, geriatrics education, and research
Journal of the American Medical Directors Association
11 4 231 238 Exclude no outcome of interest, context paper
Nakanishi M, Honda T 2009 Processes of decision making and end-of-life care for patients with dementia in group homes in Japan
Archives of Gerontology and Geriatrics
48 3 296 299 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 273
Namba R 2006 [Palliative care at the end of life for intractable neurological diseases]. [Japanese]
Gan to Kagaku Ryoho [Japanese Journal of Cancer & Chemotherapy]
33 Suppl 42 Exclude not appropriate study design case studies
Nihtila E, Martikainen P 2007 Household income and other socio-economic determinants of long-term institutional care among older adults in Finland
Population Studies 61 3 299 314 Exclude not eligible population
O'Brien T, Kelly M, Saunders C 1992 Motor neurone disease: A hospice perspective British Medical Journal 304 6825 471 473 Exclude no outcome of interest
Owen JE, Goode KT, Haley WE 2001 End of life care and reactions to death in African-American and white family caregivers of relatives with Alzheimer's disease
Omega: Journal of Death & Dying
43 4 349 361 Exclude no outcome of interest
Partridge MR, Khatri A, Sutton L, Welham S, Ahmedzai SH
2009 Palliative care services for those with chronic lung disease
Chronic Respiratory Disease 6 1 13 17 Exclude no outcome of interest
Patil SP, Krishnan JA, Lechtzin N, Diette GB
2003 In-hospital mortality following acute exacerbations of chronic obstructive pulmonary disease
Archives of Internal Medicine 163 10 1180 1186 Exclude no outcome of interest
Patti P, Amble K, Flory M 2010 Placement, relocation and end of life issues in aging adults with and without Down's syndrome: A retrospective study
Journal of Intellectual Disability Research
54 6 538 546 Exclude not eligible population
Polissar L, Severson RK, Brown NK 1987 Factors affecting place of death in Washington State, 1968-1981
Journal of Community Health 12 1 40 55 Exclude no outcome of interest
Pooler J, Yates A, Ellison S 2007 Caring for patients dying at home from heart failure: a new way of working
International Journal of Palliative Nursing
13 6 266 271 Exclude not appropriate design, a case study
Porock D, Oliver DP, Zweig S, Rantz M, Mehr D, Madsen R, Petroski G
2005 Predicting death in the nursing home: development and validation of the 6-month Minimum Data Set mortality risk index
Journals of Gerontology Series A-Biological Sciences & Medical Sciences
60 4 491 498 Exclude no outcome of interest
Porter B, Henry SR, Gray WK, Walker RW
2010 Care requirements of a prevalent population of people with idiopathic Parkinson's disease
Age and Ageing 39 1 57 61 Exclude not eligible population
Ratner E, Norlander L, McSteen K 2001 Death at home following a targeted advance-care planning process at home: The kitchen table discussion
Journal of the American Geriatrics Society
49 6 778 781 Exclude not eligible population
Reddy NC, Korbet SM, Wozniak JA, Floramo SL, Lewis EJ
2007 Staff-assisted nursing home haemodialysis: patient characteristics and outcomes
Nephrology Dialysis Transplantation
22 5 1399 1406 Exclude no outcome of interest
Reilly RB, Teasdale TA, McCullough LB 1994 Projecting patients' preferences from living wills: An invalid strategy for management of dementia with life-threatening illness
Journal of the American Geriatrics Society
42 9 997 1003 Exclude not advanced disease
Rhodes RL, Teno JM, Connor SR 2007 African American Bereaved Family Members' Perceptions of the Quality of Hospice Care: Lessened Disparities, But Opportunities to Improve Remain
Journal of Pain and Symptom Management
34 5 472 479 Exclude no outcome of interest
Rhodes RL, Teno JM, Welch LC 2006 Access to hospice for African Americans: Are they informed about the option of hospice?
Journal of Palliative Medicine 9 2 268 272 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 274
Rich A, Ellershaw J, Ahmad R 2001 Palliative care involvement in patients stopping haemodialysis
Palliative Medicine 15 6 513 514 Exclude no outcome of interest
Riesenberg D 2000 Hospital care of patients with dementia Journal of the American Medical Association
284 1 87 89 Exclude no original data
Roberts JC, Kjellstrand CM 1988 Choosing death. Withdrawal from chronic dialysis without medical reason
Acta Medica Scandinavica 223 2 181 186 Exclude no outcome of interest
Roberts SE, Goldacre MJ 2003 Case fatality rates after admission to hospital with stroke: Linked database study
British Medical Journal 326 7382 193 194 Exclude no outcome of interest
Robertson C, Warrington J, Eagles JM 1993 Relocation mortality in dementia: The effects of a new hospital
International Journal of Geriatric Psychiatry
8 6 521 525 Exclude no outcome of interest
Rocker GM, Dodek PM, Heyland DK 2008 Toward optimal end-of-life care for patients with advanced chronic obstructive pulmonary disease: Insights from a multicentre study
Canadian Respiratory Journal 15 5 249 254 Exclude no outcome of interest
Rosenfeld K, Rasmussen J 2003 Palliative Care Management: A Veterans Administration Demonstration Project
Journal of Palliative Medicine 6 5 831 839 Exclude no outcome of interest
Sampson WI 1977 Dying at home Journal of the American Medical Association
238 22 2405 2406 Exclude no original data
Saphir A 1999 Hospice lengths of stay dwindling Modern healthcare 29 12 40 Exclude no outcome of interest
Schrag D, Xu F, Hanger M, Elkin E, Bickell NA, Bach PB
2006 Fragmentation of care for frequently hospitalized urban residents
Medical Care 44 6 560 567 Exclude not advanced disease
Schwarz KA, Elman CS 2003 Identification of factors predictive of hospital readmissions for patients with heart failure
Heart and Lung: Journal of Acute and Critical Care
32 2 88 99 Exclude not advanced disease
Scocco P, Rapattonoi M, Fantoni G 2006 Nursing home institutionalization: A source of eustress or distress for the elderly?
International Journal of Geriatric Psychiatry
21 3 281 287 Exclude no outcome of interest
Scott S, Pace V 2009 The first 50 patients: a brief report on the initial findings from the Palliative Care in Dementia Project
Dementia 8 3 435 441 Exclude no outcome of interest
Seamark DA, Ryan M, Smallwood N, Gilbert J
2002 Deaths from heart failure in general practice: implications for palliative care
Palliative Medicine 16 6 495 498 Exclude no outcome of interest
Seneff MG, Wagner DP, Wagner RP, Zimmerman JE, Knaus WA
1995 Hospital and 1-year survival of patients admitted to intensive care units with acute exacerbation of chronic obstructive pulmonary disease
Journal of the American Medical Association
274 23 1852 1857 Exclude no outcome of interest
Sergi-Swinehart P 1985 Hospice home care: How to get patients home and help them stay there
Seminars in Oncology 12 4 461 465 Exclude not appropriate study design descriptive study
Shah SH, Gambles M, Jack B, Ellershaw J
2005 The Liverpool Care Pathway: Its impact on improving the care of the dying (multiple letters) [8]
Age and Ageing 34 2 197 199 Exclude no original data
Shega JW, Levin A, Hougham GW, Cox-Hayley D, Luchins D, Hanrahan P, Stocking C, Sachs GA
2003 Palliative Excellence in Alzheimer Care Efforts (PEACE): A program description
Journal of Palliative Medicine 6 2 315 320 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 275
Shepperd S, Harwood D, Jenkinson C, Gray A, Vessey M, Morgan P
1998 Randomised controlled trial comparing hospital at home care with inpatient hospital care. I: Three month follow up of health outcomes
British Medical Journal 316 7147 1786 1791 Exclude no outcome of interest
Shyu YIL, Lee HC 2002 Predictors of nursing home placement and home nursing services utilization by elderly patients after hospital discharge in Taiwan
Journal of Advanced Nursing 38 4 398 406 Exclude not eligible population
Silver MI 1999 Providing dialysis services for patients in a skilled nursing facility
Nephrology news & issues 13 10 14 19 Exclude no outcome of interest
Simard J 1999 Making a positive difference in the lives of nursing home residents with Alzheimer disease: the lifestyle approach
Alzheimer Disease & Associated Disorders
13 Suppl 72 Exclude no outcome of interest
Singer Y, Bachner YG, Shvartzman P, Carmel S
2005 Home death - The caregivers' experiences Journal of Pain and Symptom Management
30 1 70 74 Exclude not appropriate population
Sit JWH, Wong TKS, Clinton M, Li LSW, Fong YM
2004 Stroke care in the home: The impact of social support on the general health of family caregivers
Journal of Clinical Nursing 13 7 816 824 Exclude no outcome of interest
Sloane PD, Zimmerman S, Hanson L, Mitchell CM, Riedel-Leo C, Custis-Buie V
2003 End-of-Life Care in Assisted Living and Related Residential Care Settings: Comparison with Nursing Homes
Journal of the American Geriatrics Society
51 11 1587 1594 Exclude no outcome of interest
Smith GE, Kokmen E, O'Brien PC 2000 Risk factors for nursing home placement in a population-based dementia cohort
Journal of the American Geriatrics Society
48 5 519 525 Exclude not advanced disease
Smith LN, Craig LE, Weir CJ, McAlpine CH
2008 The evidence-base for stroke education in care homes
Nurse Education Today 28 7 829 840 Exclude no outcome of interest
Smith MA, Frytak JR, Liou JI, Finch MD 2005 Rehospitalization and survival for stroke patients in managed care and traditional medicare plans
Medical Care 43 9 902 910 Exclude not advanced disease
Soler JJ, Sínchez L, Latorre M, Alamar J, Romín P, Perpií M
2001 The impact of COPD on hospital resources: The specific burden of COPD patients with high rates of hospitalization
Archivos de Bronconeumologia 37 9 375 381 Exclude no outcome of interest
Spitznagel MB, Tremont G, Davis JD, Foster SM
2006 Psychosocial predictors of dementia caregiver desire to institutionalize: Caregiver, care recipient, and family relationship factors
Journal of Geriatric Psychiatry and Neurology
19 1 16 20 Exclude not advanced disease and not appropriate population (caregivers)
Steele LL, Mills B, Hardin SR, Hussey LC 2005 The quality of life of hospice patients: Patient and provider perceptions. [References]
American Journal of Hospice & Palliative Medicine
22 2 95 110 Exclude no outcome of interest
Steinbach U 1992 Social networks, institutionalization, and mortality among elderly people in the United States
Journals of Gerontology 47 4 Exclude not advanced disease
Steinhauser KE, Clipp EC, McNeilly M, Christakis NA, McIntyre LM, Tulsky JA
2000 In search of a good death: Observations of patients, families, and providers
Annals of Internal Medicine 132 10 825 832 Exclude no outcome of interest
Strahan GW 1994 An overview of home health and hospice care patients: preliminary data from the 1993 National Home and Hospice Care Survey
Advance data 256 1 12 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 276
Stuart P, Knott D 2008 Communication in end-of-life cardiac care 1: difficult issues
Nursing Times 104 10 26 27 Exclude no original data
Stuart P, Knott D 2008 Communication in end-of-life cardiac care 2: skills
Nursing Times 104 11 26 27 Exclude no original data
Teno JM, Shu JE, Casarett D, Spence C, Rhodes R, Connor S
2007 Timing of Referral to Hospice and Quality of Care: Length of Stay and Bereaved Family Members' Perceptions of the Timing of Hospice Referral
Journal of Pain and Symptom Management
34 2 120 125 Exclude no outcome of interest
Teno JM, Weitzen S, Fennell ML, Mor V 2001 Dying trajectory in the last year of life: does cancer trajectory fit other diseases?
Journal of Palliative Medicine 4 4 457 464 Exclude no outcome of interest
Tiernan E, O'Connor M, O'Siorain L, Kearney M
2002 A prospective study of preferred versus actual place of death among patients referred to a palliative care home-care service
Irish Medical Journal 95 8 232 235 Exclude no outcome of interest
Tison F, Barberger-Gateau P, Dubroca B, Henry P, Dartigues JF
1997 Dependency in Parkinson's disease: A population-based survey in nondemented elderly subjects
Movement Disorders 12 6 910 915 Exclude not advanced disease
Tomiak M, Berthelot JM, Guimond E, Mustard CA
2000 Factors associated with nursing-home entry for elders in Manitoba, Canada
Journals of Gerontology - Series A Biological Sciences and Medical Sciences
55 5 Exclude not advanced disease
Tong EM, Nissenson AR, 2002 Dialysis in nursing homes Seminars in Dialysis 15 2 103 106 Exclude no original data
Trerotoli P, Bartolomeo N, Moretti AM, Serio G
2008 Hospitalisation for COPD in Puglia: The role of hospital discharge database to estimate prevalence and incidence
Monaldi Archives for Chest Disease - Pulmonary Series
69 3 94 106 Exclude no outcome of interest
Tsuji I, Whalen S, Finucane TE 1995 Predictors of nursing home placement in community-based long-term care
Journal of the American Geriatrics Society
43 7 761 766 Exclude not advanced disease
Tyrer F, Exley C 2005 Receiving care at home at end of life: Characteristics of patients receiving Hospice at Home care
Family Practice 22 6 644 646 Exclude not appropriate population
Vieregge P, Kortke D, Meyer-Bornsen C 1994 Medical and social care in elderly Parkinsonian patients
Zeitschrift fur Gerontologie 27 4 260 269 Exclude no outcome of interest
Von Bonsdorff M, Rantanen T, Laukkanen P, Suutama T, Heikkinen E
2006 Mobility limitations and cognitive deficits as predictors of institutionalization among community-dwelling older people
Gerontology 52 6 359 365 Exclude not advanced disease
Wang T, Izatt S, Dalglish C, Bargman J, Jassal S, Vas S, Oreopoulos D
2003 Peritoneal dialysis in a nursing home: Limited survival expectations [2]
Clinical Nephrology 60 5 373 374 Exclude no outcome of interest
Wang T, Izatt S, Dalglish C, Jassal SV, Bargman J, Vas S, Tziviskou E, Oreopoulos D
2002 Peritoneal dialysis in the nursing home International Urology and Nephrology
34 3 405 408 Exclude no outcome of interest
Wellard SJ, Street AF 1999 Family issues in home-based care International journal of nursing practice
5 3 132 136 Exclude no outcome of interest
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 277
Wijkstra PJ, Van Altena R, Kraan J, Otten V, Postma DS, Koeter GH
1994 Quality of life in patients with chronic obstructive pulmonary disease improves after rehabilitation at home
European Respiratory Journal 7 2 269 273 Exclude no outcome of interest
Wilkes E 1973 Where to die British Medical Journal 1 5844 32 33 Exclude no original data
Williams A 2002 Changing geographies of care: Employing the concept of therapeutic landscapes as a framework in examining home space
Social Science and Medicine 55 1 141 154 Exclude no outcome of interest, context paper
Williams BC, Phillips EK, Torner JC, Irvine AA
1990 Predicting utilization of home health resources. Important data from routinely collected information
Medical care 28 5 379 391 Exclude no outcome of interest
Wilson DM 2000 End-of-life care preferences of Canadian senior citizens with caregiving experience
Journal of Advanced Nursing 31 6 1416 1421 Exclude not eligible population
Wilson DM, Truman CD, Thomas R, Fainsinger R, Kovacs-Burns K, Froggatt K, Justice C
2009 The rapidly changing location of death in Canada, 1994-2004
Social Science and Medicine 68 10 1752 1758 Exclude no outcome of interest
Wimo A, R+Ânnb+ñck E, Larsson B, Eriksson T, Eriksson IB, Thorslund M
1999 'Misplacement' of elderly people in the caring organisation: Reasons and alternatives
Archives of Gerontology and Geriatrics
28 3 227 237 Exclude not advanced disease
Windisch W, Petermann F, Laier-Groeneveld G, Fischer S, Criee CP
1997 [Quality of life in home ventilation]. [German] Medizinische Klinik 92 Suppl 100 Exclude no outcome of interest
Winn P 2004 Hospice and palliative care: The last year in review
Annals of Long-Term Care 12 12 17 23 Exclude no original data
Wolff JL, Kasper JD, Shore AD 2008 Long-term care preferences among older adults: A moving target?
Journal of Aging and Social Policy
20 2 182 200 Exclude not appropriate population
Wolinsky FD, Callahan CM, Fitzgerald JF, Johnson RJ
1992 The risk of nursing home placement and subsequent death among older adults
Journals of Gerontology 47 4 Exclude not eligible population
Woo J, Ho SC, Lau J, Yuen YK 1994 Age and marital status are major factors associated with institutionalisation in elderly Hong Kong Chinese
Journal of Epidemiology and Community Health
48 3 306 309 Exclude not advanced disease
Woodford H, Walker R 2005 Emergency hospital admissions in idiopathic Parkinson's disease
Movement Disorders 20 9 1104 1108 Exclude not advanced disease
Yang S, Tan KL, Devanand A, Fook-Chong S, Eng P
2004 Acute exacerbation of COPD requiring admission to the intensive care unit
Respirology 9 4 543 549 Exclude not advanced disease
Zambroski CH, Moser DK, Roser LP, Heo S, Chung ML
2005 Patients with heart failure who die in hospice American Heart Journal 149 3 558 564 Exclude no outcome of interest
Zerzan J, Stearns S, Hanson L 2000 Access to palliative care and hospice in nursing homes
Journal of the American Medical Association
284 19 2489 2494 Exclude not original research, overview article
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 270
Appendix 5: Quality score for qualitative studies
Table 26. The qualitative studies’ methodological quality scores (according to the method developed by Hawker et al.
2002)
Study Abstract Introduction and aims
Method and data
Sampling Analysis Ethics Bias Results Transferability, generalisability
Implications and usefulness
TOTAL
Berzoff 2008
3 4 4 3 4 1 2 4 3 4 32
Black 2009
4 4 4 4 4 3 2 4 4 4 37
Boeije 2002
4 4 4 4 4 1 3 4 3 3 34
Bolmsjo 2001
4 4 4 4 4 4 2 4 2 2 34
Borgsteede 2006
4 4 3 4 3 1 2 3 3 2 29
Brannstrom 2006
4 4 2 4 2 3 2 4 3 2 33
Brown 1997
4 4 4 4 4 1 1 4 4 4 34
Caldwell 2007
4 4 4 4 4 1 1 4 4 4 37
Cassarett 2004
4 4 4 4 4 1 4 4 4 4 37
Chung 2009
3 4 4 4 3 2 3 4 4 4 35
Decourtney 2003
3 3 3 3 1 1 1 1 3 3 22
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 279
Study Abstract Introduction and aims
Method and data
Sampling Analysis Ethics Bias Results Transferability, generalisability
Implications and usefulness
TOTAL
Desharnais 2007
4 4 4 4 4 4 3 3 3 3 36
Edmonds 2007
4 4 4 4 4 3 4 3 3 4 37
Evans 2006
4 4 4 3 4 2 2 4 4 4 35
Evans 2006
4 4 4 3 4 2 2 4 4 4 31
Exley 2005
4 4 3 4 3 4 3 4 3 3 35
Forbes 2000
4 4 4 4 4 4 4 4 4 4 40
Fried 1998
4 4 4 4 4 4 2 3 4 4 37
Fried 1999
4 4 3 4 3 2 1 4 4 4 33
Frogatt 2006
4 4 3 4 2 1 1 4 3 4 30
Gessert 2006
4 4 4 4 4 3 3 4 4 4 38
Goodridge 2009
4 4 4 3 2 2 2 2 2 2 27
Gott 2008
4 4 4 3 4 4 3 4 4 3 37
Gysels 2009
4 4 4 4 4 4 4 4 2 4 38
Hattori 2005
4 4 4 4 4 4 3 4 4 3 38
Holley 2009
3 3 3 4 4 4 4 4 4 4 37
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 280
Study Abstract Introduction and aims
Method and data
Sampling Analysis Ethics Bias Results Transferability, generalisability
Implications and usefulness
TOTAL
Hughes 2005
4 4 4 4 4 4 4 4 3 4 39
Lee 1997
3 3 4 4 3 2 1 4 4 4 32
Liu 2003
2 4 3 4 2 1 1 3 4 4 28
McCarty 2009
2 3 2 3 2 1 1 2 2 3 21
McLennon 2010
3 4 4 4 2 3 1 4 4 4 33
Murray 2007
4 4 3 3 3 2 2 4 3 4 30
Namiki 2009
4 4 4 4 4 2 1 4 4 4 35
Nolan 2008
4 4 4 4 4 4 3 3 3 3 36
Parker 1999
4 4 3 3 3 4 2 4 3 3 33
Payne 2010
3 4 4 4 4 2 2 4 3 4 34
Polaschek 2002
2 4 3 3 2 1 3 4 3 3 28
Polaschek 2007
3 2 2 2 1 1 1 3 2 2 19
Polaschek 2007
3 4 3 2 3 4 1 4 2 4 30
Reinke 2008
4 4 3 4 4 4 2 2 2 4 33
Shipman 2009
4 4 4 4 3 4 3 4 3 4 37
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 281
Study Abstract Introduction and aims
Method and data
Sampling Analysis Ethics Bias Results Transferability, generalisability
Implications and usefulness
TOTAL
Small 2009
4 4 4 3 4 3 4 3 4 4 37
Stadjuhar 2005
3 4 4 4 4 1 4 4 4 3 35
Treloar 2010
2 3 2 4 2 1 1 4 4 4 27
Vig 2002
4 4 4 4 4 3 4 4 4 3 38
Waldrop 2009
2 4 4 4 4 3 3 4 4 3 35
Waldrop 2010
3 4 4 4 4 1 1 4 4 4 33
Wakunami 2009
4 3 4 3 4 2 1 2 2 2 27
Waterworth 2009
4 4 4 4 4 4 4 3 3 3 37
Wilson 2008
4 4 4 4 4 4 2 4 2 2 34
Each area was rated on a 4 point scale from 1 (very poor) to 4 (good). Minimum total score = 10; maximum total score
= 40.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 282
Appendix 6: Data extraction for qualitative studies
Table 27. Data Extraction Table CHF
Author/ Year/ country
Theoretical perspective/ methods
Study aim Participants Setting Main concepts/themes Recommendations
Brannstrom et al 2006 Sweden (130)
Phenomenology- hermeneutic Narrative interviews
The aim of the study was to illuminate the meaning of living with severe CHF in palliative advanced home care through patients’ narratives
4 patients Home 4 main themes Being aware that one’s life hangs by a fine thread Struggling to cope with one’s unpredictable deteriorated body Struggling with isolation Being positively dependent on receiving care that facilitates life at home
Meaning of living with severe CHF in palliative advanced home care is on one hand, being aware of one’s imminent death, on the other hand, making it through the downs i.e. surviving life-threatening conditions, breed confidence in also surviving the current down. Being constructively dependent on palliative advanced home care facilitates everyday life at home
Caldwell et al 2007 Canada (131)
Grounded theory qualitative study
To identify the preferences of patients with advanced HF regarding communication about their prognosis and its implications
20 patients Heart function clinic
The following four main themes about patient preferences were identified: level of wellness – patients wanted to learn about their prognosis and its implications at a time of optimal cognitive function, and not when their capacity for EOL decision making was diminished; opportunity to be informed – patients preferred physicians to initiate discussions about prognosis at the time of diagnosis; tell the truth – there was a strong preference for physicians to disclose prognostic possibilities, treatments and outcomes associated with HF, including the possibilities of deterioration and death; and maintain hope – there was a need for truth to be balanced with hope. Hope for quality of life, symptom control and control over EOL decisions were important to participants.
The findings suggested that communication about prognosis between patients and physicians may be difficult and deferred. Preferences identified by patients offer guidance to physicians in planning and initiating dialogue about prognosis.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 283
Desharnais et al 2007 USA (132)
interviews Pilot study quantitatively analysed – multivariate analysis
To study the effectiveness of patient physician communications regarding health care choices at the end of life. We studied communications occurring between physicians and their patients who had either terminal cancer or congestive heart failure, with less than 6 months to live.
22 physicians and 71 of their (matched) patients
Home patients- Office physicians
Subjects provided paired responses to questions regarding their conversations related to end-of-life care, including resources, attitudes, and preferences. We calculated the concordance of patient and physician reports about these discussions. We examined the physicians' and the patients' agreement on the patient's diagnosis, and on whether a variety of care options were discussed. Both bivariate and multivariate models were used. Results: As a whole, the concordance scores were poor; however, concordance varied across domains of issues discussed. Patients with less education had significantly lower concordance score
We have identified domains in which the physicians and patients may be least effective in discussing end-of-life care options. Findings may help in designing interventions to improve communication, especially for patients with less education.
Gott et al 2008 UK (181)
Qualitative design Semi structured interviews
This study aimed to explore the palliative needs of older people with heart failure, including how these were consistent with the model of a ‘good death’ which underpins palliative care delivery
40 patients Home Patients reported varied preferences in relation to home death. Home death preference was regarded as an opportunity for ‘sudden’ death more than a deliberate and planned choice, while others considered it as a more pleasant place than hospital. Preferences were linked to type of death; a sudden, unaware death was preferred by some, and an ‘aware’ death more important for others. Family stress and burden was a consideration, while being with family at the end was important – some specifically feared being alone to die.
“Open” awareness of death is problematic for this population, and concepts of autonomy in this respect were also alien to most. For many, family concerns are at least as important as personal preferences, and this needs consideration; the ‘cancer’ model should not dominate.
Horne and Payne 2004 UK (133)
Qualitative design Semi structured interviews
This study aimed to explore the experiences of patients with severe heart failure and identify their needs for palliative care
20 patients Home Patients reported difficulties in walking, extreme fatigue and problems managing daily activities. Having to rely on family, friends and neighbours was common, causing feelings of being a burden, loneliness and isolation. Patients talked about dying as well as their fears and frustrations in living with heart failure. Barriers to accessing information and social services were identified. None of these patients had been referred to specialist palliative care services.
Patients' experiences were similar to those of patients living with advanced cancer and yet they received little support. Comprehensive routine assessment of the palliative care needs of patients living with severe heart failure is recommended
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 284
Johnson et al 2007 USA (135)
Case study Interviews
The purpose of this case study was to explore that concern in-depth in three persons with different terminal illnesses
3 people 1 cancer 1 ALS 1 heart failure
Content analysis of their interviews resulted in four themes: managing the burden, spirituality, supportive relationships, and planning for the future. Themes contained specific categories of thoughts, feelings, and actions related to fear of being a burden. These themes should be explored in greater depth in future larger studies of persons with terminal illness.
Patients who have terminal illness with a rapid, gradual, or unpredictable rate of decline share common concerns about being a burden on their families. Nurses and other healthcare providers should assess terminally ill patients to determine how they manage the transition from being independent to relying on family for their care. Findings from this study have suggested that patients use various strategies to make this transition easier and manage the fear of burdening their families. Strategies included accepting the support of friends and willing caregivers, maintaining spiritual beliefs, and making plans for their future care needs.
Murray et al 2002 UK (134)
Qualitative interviews Focus groups
Aim: To compare the illness trajectories, needs, and service use of patients with cancer and those with advanced non-malignant disease.
20 patients with inoperable lung cancer and 20 patients with advanced cardiac failure and their main informal and professional carers
Home 219 qualitative interviews were carried out. Patients with cardiac failure had a different illness trajectory from the more linear and predictable course of patients with lung cancer. Patients with cardiac failure also had less information about and poorer understanding of their condition and prognosis and were less involved in decision making. The prime concern of patients with lung cancer and their carers was facing death. Frustration, progressive losses, social isolation, and the stress of balancing and monitoring a complex medication regimen dominated the lives of patients with cardiac failure. More health and social services including financial benefits were available to those with lung cancer, although they were not always used effectively. Cardiac patients received less health, social, and palliative care services, and care was often poorly coordinated.
Conclusions: Care for people with advanced progressive illnesses is currently prioritised by diagnosis rather than need. End of life care for patients with advanced cardiac failure and other non-malignant diseases should be proactive and designed to meet their specific needs
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 285
Murray et al 2007 UK (138)
Qualitative Longitudinal interviews
Aim to identify and compare changes in the psychological, social, and spiritual needs of people with end-stage disease during their last year of life by synthesizing data from two longitudinal, qualitative, in-depth interview studies investigating the experiences and needs of people with advanced illnesses
48 patients with advanced lung cancer (n = 24) and heart failure (n =24) who gave a total of 112 in-depth interviews
Home In lung cancer, the social trajectory mirrored physical decline, while psychological and spiritual well-being decreased together at four key transitions: diagnosis, discharge after treatment, disease progression, and the terminal stage. In advanced heart failure, social and psychological decline both tended to track the physical decline, while spiritual distress exhibited background fluctuations.
Holistic end-of-life care needs to encompass all these dimensions. An appreciation of common patterns of social, psychological, and spiritual well-being may assist clinicians as they discuss the likely course of events with patients and carers and try to minimize distress as the disease progresses.
Small et al 2009 UK (136)
Longitudinal qualitative
The aim:to assess carers' views on end of life care, the circumstances of the death and bereavement experiences.
20 interviews were conducted with bereaved carers of older people with heart failure (HF) who had been participating in a longitudinal study. Carers were approached in writing 3 months after the death
Home Findings were grouped into three time periods: prior to death; the death itself and bereavement. Most carers found discussions about end of life with their family member prior to death difficult. Dissatisfaction with the manner of the death was focused around hospital care, particularly what they believed to be futile treatments. In contrast deaths in the home were considered 'good'. Carers adopted a range of coping strategies to deal with grief including 'using their faith' and 'busying themselves' with practicalities. There was some satisfaction with services accessed during the bereavement period although only a small number had taken up counselling
The findings suggest that an absence of discussion about end of life care wishes with family members or health professionals is a barrier to advance care planning. Carers' perceptions about prioritising making the dying person comfortable can be in conflict with doctors' decisions to treat. Whilst carers report a range of strategies adopted in response to bereavement there is a need for continued support for vulnerable carers after the death of the person with HF.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 286
Waterworth and Jorgensen 2009 New Zealand (137)
Longitudinal qualitative study using General Inductive approach was used. Participants were interviewed every 3 months for a 12-month period
The aims were to explore the experiences of older people living with heart failure and their transitions from independence to dependence and for some death
25 patients Home or Residential care
The findings showed that transition was not a simple linear process with the older person moving from one phase to another; instead their experiences illustrated the complexity of transitions they faced and what helped them to manage these. The older people in this study illustrated the importance of trust in health professionals and believed they would receive good care. Their fears revealed concerns about being a burden as they deteriorate and becoming more dependent
Understanding the complex issues related to transition to dependence can provide health professionals with a framework for assessment and approaches to providing the support required.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 287
Table 28. Data Extraction Table COPD
Author/ Year/ country
Theoretical perspective/ methods
Study aim Participants Setting Main concepts/themes Recommendations
Back et al 2009 USA (141)
Longitudinal qualitative study Semis- structured interviews
Data for this analysis were drawn from a qualitative study that examined how patients, family caregivers, physicians, and nurses talk about hope in the context of providing or receiving information about a life-limiting illness
31 physicians who identified 55 patients with incurable cancer or advanced chronic obstructive pulmonary disease, 36 family caregivers, and 25 nurses
Not stated 2 themes describing abandonment of patients and their families: before death, abandonment worries related to loss of continuity between patient and physician; at the time of death or after, abandonment experiences related to lack of closure for patients and families.
The professional value of non abandonment at the end of life consists of 2 different elements: (1) providing continuity, of both expertise and the patient-physician relationship; and (2) facilitating closure of an important therapeutic relationship. Framing this professional value as continuity and closure could promote the development of interventions to improve this aspect of end-of-life care
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 288
Borgsteede et al 2006 Netherlands (142)
Qualitative Semi structured interviews 20 GPs 30 Patients
To explore the aspects valued by both patients and GPs in end-of-life care at home, and to reflect upon the results in the context of future developments in primary care.
20 GPs 30 Patients .
Patients’ home
4 main themes Availability of the GP for home visits and after office hours; medical competence and cooperation with other professional;, attention, and continuity of care. The aspect of attention was more prominent in the patient interviews. There was no indication of any differences in the aspects that were mentioned by (patients of) GPs who were trained in end-of-life care and their counterparts who were not trained.
The study s raises some questions about the quality of future end-of-life care at home. Developments such as out of hours services not provided by local GP’s are a challenge for general practice to find a way to both organise a modern primary care system, and to continue to provide good end-of-life care at home according to the valued aspects: availability of the GP for home visits and after office hours, medical competence, attention, and continuity of care.
Elkington et al 2004 UK (143)
Qualitative in-depth interviews framework analysis
To assess the symptoms experienced and their impact on patients’ lives in the last year of life of COPD, and to assess patients’ access to and contact with health services
25 Bereaved carers 19 Home 6 Research Dept
The average age of death was 77.4 years. The majority of patients died in hospital. The major symptom reported by the carers was breathlessness which impaired the deceased's mobility and contributed to their being housebound. Anxiety and panic were also associated with breathlessness. Depression was reported. Oxygen, though beneficial, was seen to impose lifestyle restrictions due to increasing dependence on it. Some patients only health care contact was through repeat prescriptions from their GP whereas three had regular follow up by a respiratory nurse specialist who linked community and secondary care. Overall, follow-up, systematic review or structured care were uncommon.
Breathlessness causes major disability to patients with COPD in the last year of life. The expertise of palliative care in treating breathlessness may be valuable in these patients many of whom lacked regular health service contact in the year before death. Patients who are housebound with high levels of morbidity require community health services. Respiratory nurse specialists were rarely involved in the patients’ care and may provide a link between the GP, the chest physician and the palliative care team.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 289
Exley et al 2005 UK (139)
Qualitative semi structured interviews
This study explored the views of health professionals, patients and their carers about care provided at the end of life.
50 interviews - relating to 29 patients and carers – usually dyads and bereaved carers (7) 11 patients and 5 bereaved carers (cancer) and 16 patients and 2 bereaved carers
Home Differences between the care of people with cancer and those with end-stage cardiorespiratory disease were found in four main areas: management and progression of disease, communication and information, health care in the community and awareness of dying. The research shows that even in PHCTs (primary health care teams) committed to the delivery of palliative care, people dying with end-stage cardiorespiratory disease are less likely than those with cancer to receive full, and easily understood, information, to be aware that they are dying or to receive district nursing care
There is an increasing call for palliative care to be extended to all, but further work is needed to develop appropriate packages of care for those dying with conditions other than cancer.
Fried et al. 1998 USA (148)
Qualitative In-depth interviews Constant comparative analysis
To explore how older persons form preferences for site of medical care by investigating their perceptions of home and hospital care
29 persons who had been hospitalized with CHF, COPD and pneumonia and receiving home care services
Home and hospital
Respondents who initially thought of home care as low-intensity and frequency care were skeptical of expanded home care to treat acute illness. Regardless of their opinions of home or and hospital, all respondents preferred the site associated with greatest survival. If the sites provided equal survival, preferred the home because of freedom from constraints of the hospital and the comfort of home. Those preferring hos pital found home a lonely and frightening place to be sick. Views were shaped by social support, self-reliance, religion, past illness experiences.
Fried et al. 1999 USA (76)
Mixed methods Cross-sectional quantitative and qualitative interviews
To describe older persons’ preferences for home or hospital as the site of terminal care and to explore potential reasons for their preferences.
Community-dwelling persons, 65 years or older, recently hospitalized with CHF, COPD, or pneumonia and not selected according to life expectancy; 246 patients participated in quantitative and 29 in qualitative interviews.
Home, hospital, Nusing home
In quantitative interviews, 118 patients (48%) preferred terminal care in the hospital, 106 (43%) preferred home, and 22 (9%) did not know. One third changed their preference when asked about their preference in the event of a non-terminal illness. Reasons for preference identified during qualitative interviews included the desire to be with family members and concerns about burden to family and their ability to provide necessary care. Concern about long-term care needs resulted in preference for a nursing home when choice was not constrained to home and hospital.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 290
Goodridge et al 2009 Canada (144)
Mixed Methods Qualitative focus groups
The aims of this study were (a) to examine critical care clinician perspectives on the quality of dying of patients with COPD and (b) to compare nurse ratings of the quality of dying and death between patients with COPD with those who died from other illnesses in critical care settings.
Three focus groups provided data describing the EOL care provided to patients with COPD dying in the intensive care unit (ICU). Patients Total sample 103 patients
Hospital Three primary themes emerged from the qualitative data are as follows: managing difficult symptoms, questioning the appropriateness of care and establishing care priorities
Attention to the management of dyspnoea, anxiety and treatment decision-making are priority concerns when providing EOL care in the ICU to patients with COPD
Gysels & Higginson 2009 (140)
In depth interviews Analysis grounded Theory
The objective was to investigate the caring experience of carers for patients with an advanced progressive illness (chronic obstructive pulmonary disease [COPD], heart failure, cancer, or motor neuron diseases [MND]), who suffer from breathlessness.
15 carers Home Several key issues affected caring in a positive or a negative way. The threats to caring were uncertainty, carers' own health problems, an imploded world, negative reactions from outside, person loss, and acute exacerbations. Resources that carers drew on were acceptance, self-care, availability of support, feeling that caring is a shared responsibility with the patient, and "getting on with" caring in case of emergencies. Breathlessness was particularly challenging, and carers did not have any strategies to relieve the symptom. They were ill prepared for acute exacerbations.
Carers need to be included in opportunities for support provision in advanced illness. Negotiated involvement of a health professional could buffer the heavy responsibilities related to home care. They could provide problem-solving skills and build on the resources that carers draw on in response to what they experience as most threatening to their caring role
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 291
Hall et al 2010 Canada (149)
Descriptive exploratory study
The purpose of this research was to describe the perceptions of people living with severe chronic obstructive pulmonary disease (COPD) with respect to the end of life.
6 patients Hospital The analysis yielded four themes that reflect the perceptions of participants with respect to the end of life, namely: living and seeing oneself decline, living and preparing to die, dying of COPD means suffocating, and dying in hospital surrounded by family and friends. What emerges from the study is that persons living with severe COPD wish to die without suffocating, in hospital, surrounded by family and friends, all the while hoping to go on living.
This study contributes to a more comprehensive understanding of the end-of-life experience. It shows the importance of accompanying these persons properly towards the end of life and at the moment of dying. The study proposes a series of avenues for future research and makes recommendations for practice.
Reinke et al 2008 USA (145)
Qualitative Grounded theory
This study explored transitions, defined as experiences that patients and family members viewed as milestones in the evolution of their illnesses and therapies, and compare these perceptions with the perspectives of the patient's physician and nurse to provide insights about communication concerning EOL care.
31 Physicians 55 patients 36 family members 25 Nurses
Not stated Six themes were identified regarding participants' experiences with transitions. Themes that defined transitions among both patients with COPD and those with cancer included: new or different treatments and no more treatments available. Themes unique to patients with COPD were activity limitations due to functional decline and initiation of oxygen therapy. One theme unique to clinicians was acute exacerbation of illness or hospitalization.
This study identified differences in the meaning of transitions for patients versus clinicians and for patients with COPD versus those with cancer. These findings may offer clinicians the opportunity to provide a more patient-centred approach to communication about end-of-life care by acknowledging and addressing transitions in palliative care from the perspective of the patient and family.
Shipman 2009 UK (146)
Exploratory qualitative In-depth interviews
To explore factors that influence the use of general practice services by people with advanced COPD.
16 Patients Home All patients reported severe breathlessness. They contacted general practices for routine, urgent and emergency care. Contact was influenced by perceptions of ease of access; quality of relationship with their general practitioner (GP), and perceived disease severity and threat. Some patients wanted to avoid bothering the doctor or found travelling to the surgery too difficult.
Factors other than need influenced patterns of health service use. Expectations of difficulty in access, and poor relationships with their GP, may have delayed help-seeking in severe acute exacerbations.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 292
Wilson et al 2008 Canada (147)
Ethnographic qualitative study Longitudinal indepth interviews
This study was undertaken to determine the care needs of Canadian seniors living at home with advanced chronic obstructive pulmonary disease (COPD).
12 community-dwelling seniors interviewed 3 times in their homes over an eight-month period in 2006.
Home 3 themes emerged, each with concrete care needs: (a) self-reliance and independence through adaptation, (b) stable health through maintenance, and (c) living with constraints. The predominant theme was that all participants wanted to maintain their independence. This required considerable adaptation, as well as assistance from others. Ensuring and improving assistance is important to prevent additional suffering and reduce exacerbations requiring hospitalization, a particularly important aim given the high and rising incidence of advanced COPD.
A number of other insightful findings reveal the significance of learning directly from the persons who live with chronic illnesses about their lives.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 293
Table 29. Data Extraction Table Long Term Neurological Conditions
Author/ Year/ country
Theoretical perspective/ methods
Study aim Participants Setting Main concepts/themes Recommendations
Boeije et al 2002 Netherlands (154)
Qualitative Semi structured interviews Biographical construction- constant comparative analysis
The aim of this study is to ascertain how people in the advanced stages of MS accommodate to their illness. It examines how MS induces biographical work in terms of body biographical time and conceptions of self (BBC). It then goes on to describe what this biographical work involves in terms of the four different types of work and touches on the differences between patients.
22 people with MS and 21 family caregi vers
Home The continuity of biography is at risk since body and performance failures lead to the loss of salient aspects of self. Participants interpret MS as an all-encompassing illness and emphasize the process of having to give up everything. Four case stories are described to demonstrate the complex intertwining of the biographical processes and to show the range in biographical accommodation. Some patients are capable of putting their lives back together again, while others retreat or do not consider MS a part of their lives. The unpredictable course of MS makes it impossible to give new direction to the life.
This study offers insights into the way people with advanced MS accommodate to their illness. The body of data as a whole confirms MS as a physically exhausting disease with far-reaching psychosocial consequences for patients and family members alike. A variety of examples of serious bodily failure and accompanying performance failures were reported. The loss of salient aspects of self was also illustrated. The empirical data presented illustrate some of the general and already known features of progressive illnesses, and some of the more specific points that relate to MS. As bodily failure increases, performances have to be eliminated altogether and it appears to be hard and sometimes even impossible to find new, fulfilling activities.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 294
Bolmsjo et al 2001 Sweden (153)
Qualitative Interviews Descriptive analysis Kvale 1996
This study challenges the assumption that people with amyotrophic lateral sclerosis (ALS) and those who care for them at home have a similar view the disease.
8 patients 8 carers
Home and neurology clinic
By interviewing ALS patients and close relatives to ALS patients, and by comparing the quotes from the interviews, some differences between these two groups are revealed. The results of the .present study show that ALS patients and close relatives of ALS patients differ in at least five respects: they perceive their needs, and they view, judge, and evaluate the disease, as well as the process of the disease, in different ways
Furthermore, the study suggests that the patients and their close relatives should be viewed as individuals with their own preferences. Moreover, close relatives need someone in whom to confide, and caregivers have specific needs for support and information.
Edmonds et al 2007 UK (150)
Qualitative semi structured interviews
This study aimed to explore important issues for people severely affected by multiple sclerosis (MS)
23 people with MS and 17 informal carers
Home Personal issues in relation to loss and change, particularly in terms of losses of or changes in physical abilities, including maintaining mobility, independence, relationships and social role were raised commonly in response to an open-ended question about what issues were important in living with MS. Coping with MS requires individuals to deal with the losses and changes brought about by their illness. Our study suggests that even patients who have had MS for many years and are now severely affected continue to experience loss and change.
We recommend that attention be given to emotional support which specifically addresses three main areas of dealing with loss and change for people that are severely affected - physical issues, independence and relationships. Palliative care providers may have expertise in managing loss that could be useful for these patients in partnership with neurological services.
Hughes et al 2005 UK (151)
Qualitative Semi structured interviews
Aim: to understand peoples experiences and to generate ideas for developing practice and policy in MND health social and palliative care
9 people with MND and 5 carers 15 professionals
Home The findings are presented within three substantive groups: (1) the impact of MND on people's lives (the physical impacts of the illness, including increasing disability; social issues, including restrictions on social activities; and adjustments to people's lives, including methods of coping with the illness); (2) experiences of services (accessing service entitlements, information sources, professionals’ attitudes and approaches, and professionals’ knowledge and understanding of MND); and (3) suggestions for service change (better information and communication, including information on service entitlements; improved knowledge amongst professionals about MND; and some suggestions for service restructuring)
This study brings a fresh approach to understanding the impacts of MND and the ways in which services can be improved to meet people's needs
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 295
Nolan et al 2008 UK (152)
Mixed methods A descriptive correlational design with 16 patient–family member dyads was used. Quantitative findings and in-depth interviews with a subset of five family members following the patient's death. ( (Telephone interviews)
This study compared the preferences of patients with ALS for involving family in health care decisions at the end of life with the actual involvement reported by the family after death.
5 carers Not stated Eighty-seven percent of patients had issued an advance directive. Patients who would opt to make health care decisions independently (i.e., according to the patient's preferences alone) were most likely to have their families report that decisions were made in the style that the patient preferred. Those who preferred shared decision making with family or decision making that relied upon the family were more likely to have their families report that decisions were made in a style that was more independent than preferred. When interviewed in depth, some family members described shared decision making although they had reported on the survey that the patient made independent decisions.
The structure of advance directives may suggest to families that independent decision making is the ideal, causing them to avoid or underreport shared decision making. Fear of family recriminations may also cause family members to avoid or underreport shared decision making. Findings from this study might be used to guide clinicians in their discussions of treatments and health care decision making with persons with ALS and their families
Parker et al 1999 Australia (155)
Semi structured interviews Descriptive analysis
This study examines the potential role for palliative care services in the care of individuals with muscular dystrophy and spinal muscular atrophy, and the support of their families
9 bereaved 4 current family members of people with advanced muscular dystropy
Home Significant issues identified included: (i) a lack of coordination of care and access to skilled, competent carers; (ii) a lack of support for siblings; (iii) inadequate bereavement care; and (iv) limited discussion of options of ventilatory support and advance directives.
The terminal care for individuals with muscular dystrophy and spinal muscular atrophy and their families requires improvement. Although many individuals with these conditions will die following an acute event, palliative care services may be appropriate for those who require a period of terminal care at home.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 296
Table 30. Data Extraction Table End Stage Kidney Disease
Author/ Year/ country
Theoretical perspective/ methods
Study aim Participants Setting Main concepts/themes Recommendations
Berzoff et al. 2007 USA (156)
Qualitative 6 focus group discussions Grounded theory
To explore experiences and views of participants thoughts about the development of Renal Supportive Care Team (RSCT)—a multidisciplinary team designed to provide education and support to severely ill dialysis patients and their families.
Patients, carers, bereaved carers and staff. 36 participants in total.
Dialysis clinics and medical facilities
Education: patients and families that both wanted more education from health care providers than they thought they had received. Support: More support was needed from staff, or peer led, and telephonic possibilities were seen as useful. Communication:There needs to be one key person for ongoing communication. This needs to be culturally competent. Continuity: the team should act as a buddy system. Advance care planning: patients and carers don’t always want to hear about EoL issues at the same time. This info needs to be repeated and written documents are not sufficient.
It is crucial that doctors, nurses, social workers, and chaplains begin to attend systematically to the care of patients and families with renal disease, as this is a population that has not received the attention that they deserve.
Namiki et al. 2009 Australia (157)
Qualitative Interviews. Thematic analysis
To understand daily life activities and challenges of older people living with home haemodialysis.
4 older patients,(3 men, 1 woman) with end stage kidney disease.
Home 3 themes: Accommodating haemodialysis at home Partnership (spousal and professional) Sense of self: In the present – living with ups and downs, the future – hopes and uncertainty
A positive outlook on life in the present enabled participants to look to their future with purpose and hope. Outstanding systems of partnership significantly facilitated this view.
Polaschek 2003 New Zealand (158)
Qualitative. Critical interpretative methodology
To delineate the concerns of one particular group of Caucasian men living on home haemodialysis.
6 Caucasian men on self-care dialysis
Home Patients concerns: Suffering from continuing symptoms Limitations resulting from negotiating dialysis into their lifestyle. Ongoingness and uncertainty of life on dialysis Aletered relationship between autonomy and dependency.
Experience of other ethnicities and women living on dialysis may be significantly different. The nursing role should include attention to the patients experience of living on dialysis, alongside their technical tasks.
Polaschek 2007 New Zealand (159)
Qualitative. Critical interpretative methodology
To describe characteristic attitudes towards their treatment regimen among a group living on home dialysis.
20 people using haemodialysis or peritoneal dialysis.
Home Renal patients managing their condition at home is always a process of negotiation. Optimising the prescription may affect their usual activities in ways that reduce their QoL. Modifying some aspects of the treatment regimen within the limits they know, are possible, may enable them to maintain their normal lifestyle.
Nurses can support clients as they manage their treatment themselves.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 297
Polaschek 2007 New Zealand (160)
Qualitative. Critical interpretative methodology
To describe characteristic attitudes towards their treatment regimen among a group living on home dialysis.
20 people using haemodialysis or peritoneal dialysis.
Home During the initial period of adjustment to treatment many participants learned their need for treatment by experimenting with the therapeutic prescription. They then used their knowledge of the therapy to alter their treatment regime to maintain their normal lifestyle. Having modified their therapeutic prescription, participants’ motivation to continue meeting the demands of treatment was influenced by their individual life situation, including relationships, work and personal attitudes towards life.
Understanding client attitudes towards therapy enables nurses to support people living on dialysis better. Through enhancing their relationships with clients, nurses can assume a key role in service to people living with chronic conditions.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 298
Table 31. Data Extraction Table Older People, including stroke and dementia
Author/ Year/ country
Theoretical perspective/ methods
Study aim Participants Setting Main concepts/themes Recommendations
PLACE OF CARE
Brown et al. 1997 Canada (159)
Qualitative 7 focus groups
This study aimed to identify barriers and facilitators to independence as experienced by seniors with chronic health problems as they interacted with medical, home-based, and public health services.
43 seniors with chronic health problems, carers (most spouses), health professionals
3 primary care areas: medical, home-based, public health
Four main themes characterized the barriers and facilitators to seniors' independence: attitudes and attributes, service accessibility, communication and coordination, and continuity of care.
Health providers need to communicate better, to foster more positive attitudes toward aging, and to participate in refining service access, coordination of services, and continuity of care.
Evans et al. 2006 USA (164)
Qualitative Semi-structured interviews
(1) the reasons for hospitalization, (2) preferences for site of care of patients who are hospitalized, and (3) patients’ and caregivers’ experiences in facilities upon transfer.
Caregivers of deceased hospice patients who transferred to an acute care hospital, a freestanding inpatient hospice facility, or a nursing home while enrolled in hospice and died
Home hospice to inpatient hospice, hospital or nursing home
Patients transferred because of an acute medical event, an uncontrolled symptom, imminent death, or inability to provide needed care safely at home. Although all caregivers expressed a strong preference for care at home, other concerns such as pain and symptom control, safety, and quality and quantity of life became more important with time. We found significant variation in specific preferences regarding care and site of death. Satisfaction with care at the transfer facilities was determined by clarifying goals of care, following treatment preferences, providing personalized care, and the patient’s environment.
Experiences can be improved by treating patients and carers as individuals, exploring and respecting treatment preferences and creating a pleasant physical environment.
Frogatt & Payne 2006 UK (173)
Postal survey of care home managers in one English county. Content analysis of open questions on managers’ understanding of EoL care, priorities for future
To describe the provision of EoL care for older people residing in care homes from the perspective of care home managers.
Managers of 261 care homes
Care homes Managers held diverse understandings regarding the meaning of EoL care. The features of the residents’ conditions and the dying that they experience requires a different way to conceptualise EoL care.
A longer term perspective is required. Also ways to involve residents, relatives, and staff in the development of care at EoL. There is the need for advance care planning, even before entry.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 299
developments
Gessert et al. 2006 USA (166)
Qualitative 8 focus groups
To explore the goals, beliefs, and values used by family members in making decisions on behalf of cognitively impaired, institutionalized elders near the end of life and to identify commonalties and differences between the values invoked by families in rural and urban counties in Minnesota.
38 family members of nursing home residents with severe cognitive impairment
Nursing home Most rural participants: acceptance of death and few conditions on death, beyond their hope that it would be quick and peaceful. Urban respondents: a wider range of attitudes toward death, from unambiguous acceptance of immediate death to evident discomfort with welcoming death under any circumstances. These rural-urban differences had practical implications. Rural respondents were much less likely to endorse interventions that would impede death. Rural respondents tended to express confidence in natural forces; death was seen as neutral or beneficient. Urban respondents: resistance to the approach of death. Some insisted on aggressive medical care in advanced dementia. Uncertainty in decision making
Rural views might be useful in our broad cultural efforts to improve care at the end of life. These may be the result of shared life experiences.
Holley et al. 2009 USA (167)
Mixed-methods Chart review of 74 patients, telephone (22 carers) and face to face (13 carers) interviews, Quantitative analyses between patients who died in 1st year and those who did not., between patients with dementia and without, who died in the 1st year. Qualitative: content analysis.
To assess carers´ expectations of and satisfaction with an urban, home-based geriatrics PC programme (PATCH)
Carers of patients >65, enrolled in Medicare Part B, homebound, limited life expectancy.
Home-based geriatrics PC programme
Patients average age: 85. 86% female, 82% African American. The majority had between 2 and 4 secondary diagnoses. 55% of caregivers were daughters. Carers providing care for patient with dementia was longer than for patients without (80 vs 18 mths), were less likely to work outside the home (35%vs 36 %). More than 2/3 died at home or in in-patient hospice. Amount of care needed and satisfaction was high. Themes: Preferences about care setting, access to practitioner with expertise in PC, challenges with multiple transitions in care, to EoL care and anticipating death.
Early referral to home care programmes so that transitions are made smooth and continuity of care is improved. Continuation of relationship between between PATCH and patients admitted to long term care. Ensure programme sustainability.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 300
McCarty & Volicer 2009 USA (168)
Mixed methods: Questionnaire hospice characteristics, criteria for admission, Services provided, sources of referrals, and outreach used. Open interviews Content analysis
To investigate what promotes and prevents involvement of individuals with dementia in a hospice program.
Convenience sample of hospice organizations 3nurses, 6 social workers, 4 hospice directors
Hospice Facilitators for utilization of hospice for Alzheimers patients: In-patient or residential hospice facility. Services for patients not eligible by Medicare criteria (Bridge, Transitions, PC programs) Barriers: Physician prognosis, lack of recognition of dementia as terminal illness, finances, communication difficulties, institutional barriers, eg when patients are transferred to hospitals, nursing staff may be resistant. Due to misunderstanding of hospice purpose.
Adverse effect of Medicare regulations may be decreased and hospice access improved by: different eligibility criteria. Alternative programs for patients not eligible for Medicare, more intensive education.
Payne et al.2010 UK (175)
Qualitative Semi-structured interviews
To identify patients’ and family members’ experiences of acute stroke and their preferences for end-of-life care, and identify barriers and opportunities to improve care.
28 patients with stroke
2 general hospitals
Communication between patients, family and health professionals was central to a positive experience of stroke care. Honesty and clarity of information was required, even where prognoses were bleak or uncertain. Patients and family attached as much importance to the style of communication as to the substance. Where the focus had shifted from active to passive support, both wished to be included in dialogue with professionals. Where patients were thought to be dying, family was keen to ensure death was peaceful and dignified. Families reported few opportunities for engagement in any form of choice over place or style of EoL care. No family reported being offered the possibility of the patient dying at home.
The findings demonstrate the importance of improving communication between patient, family and health professionals for patients with stroke in UK hospitals.
Treloar et al. 2010 UK (176)
Exploratory retrospective study. Mixed methods: Free interview and semi-structured questionnaire. Analysis: descriptive statistics and identification of themes.
To identify the major factors which make care for dementia patients at home feasible.
Carers of 14 patients interviewed post death
Home Factors for success: Right equipment Expertise about medication, food, social care needs. Understanding and support for funding care commissioning and informal care.
A specialist team is required. Skilled support in a wide range of support methods to be coordinated by someone who understands the issues. Hospital death could be averted by specialist out-of-hours advice.
PREFERENCES FOR CARE
Black et al. 2009
Qualitative Semistructured
1. To determine how surrogate decision
34 surrogate decision makers for
Nursing homes.
Most surrogates reported that patients had previously completed an advance directive (59%), discussed preferences for EoL care (56%), or done
Health care providers may be able to assist patients and
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 301
USA (161)
interviews. Content analysis.
makers for nursing home patients with advanced dementia developed understanding of patients’ treatment preferences, what those preferences were, and how confidently the surrogates held their beliefs about patients’ wishes for EOL care. 2. To identify surrogate-reported factors that either impeded or motivated individuals to express their treatment preferences.
hospice-eligible nursing home patients with dementia.
both (38%). Catalysts for and barriers to completing an advance directive or having EoL care discussions included factors that were both intrinsic and extrinsic to the patient. The most commonly reported wish for EoL care was to not be kept alive by “machines” or “extraordinary measures.”
families by normalizing discussions of dying, encouraging advance care planning, helping them identify goals for EOL care, and providing information to support treatment decisions consistent with patients’ wishes. Finding opportune moments to raise EoL issues early on. Future efforts to improve expression of EoL preferences for EoL care might first examine how patients would like to communicate them (AD,/discussion). Educational interventions can increase discussions. Those tailored to individual is more successful than one general approach. Respect for those who do not wish to discuss preferences.
Fried et al. 1998 USA (148)
Qualitative In-depth interviews Constant comparative analysis
To explore how older persons form preferences for site of medical care by investigating their perceptions of home and hospital care
29 persons who had been hospitalized with CHF, COPD and pneumonia and receiving home care services
Home and hospital
Respondents who initially thought of home care as low-intensity and frequency care were skeptical of expanded home care to treat acute illness. Regardless of their opinions of home or and hospital, all respondents preferred the site associated with greatest survival. If the sites provided equal survival, preferred the home because of freedom from constraints of the hospital and the comfort of home. Those preferring hos pital found home a lonely and frightening place to be sick. Views were shaped by social support, self-reliance, religion, past illness experiences.
Because survival appears to be the most important determinant of preference, home treatment of acute illness is a viable alternative only if it provides outcomes equivalent to those of hospitalisation. Perceptions that home care only can be a low-intensity service may limit preferences for home treatment. When expected outcomes at the 2 sites are similar, the challenge to the health care system will be incorporating pt preference into decisions about the
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 302
appropriate site of care.
Fried et al. 1999 USA (76)
Mixed methods Cross-sectional quantitative and qualitative interviews
To describe older persons’ preferences for home or hospital as the site of terminal care and to explore potential reasons for their preferences.
Community-dwelling persons, 65 years or older, recently hospitalized with CHF, COPD, or pneumonia and not selected according to life expectancy; 246 patients participated in quantitative and 29 in qualitative interviews.
Home, hospital, Nusing home
In quantitative interviews, 118 patients (48%) preferred terminal care in the hospital, 106 (43%) preferred home, and 22 (9%) did not know. One third changed their preference when asked about their preference in the event of a non-terminal illness. Reasons for preference identified during qualitative interviews included the desire to be with family members and concerns about burden to family and their ability to provide necessary care. Concern about long-term care needs resulted in preference for a nursing home when choice was not constrained to home and hospital.
The current debate about home versus hospital as the ideal site for end-of-life care may ignore an important issue to older persons—namely, the care of disabilities that precede death.
Hattori et al. 2005 Japan (177)
Qualitative Descriptive study. Semi-structured interviews
To understand the wishes of the elderly in Japan concerning their EoL care: Wishes for care Preferences for information about their illness Meaning of death
Elderly patients 17 elderly patients hospitalized in a university hospital, and 13 from a university affiliated outpatient facility .
Wishes for care were influenced by a variety of factors: Family, health condition, personal experience, relationship with physician, concept of life and death. Wishes varied during the interview. The wish to die in comfort remained stable.
PC providers need to understand: Patients wishes for EoL care may be dependent on their ability to make decisions, that wishes change and family considerations are a strong influence.
Vig et al. 2002 USA (170)
Descriptive qualitative study. Semi-structured interviews
To explore the attitudes of older adults with medical illness about the end of life, and to investigate whether current values could be extended to end-of-life preferences.
16 older men and women with non terminal heart disease and cancer.
Patients (all men) attending two university-affiliated geriatric clinics
Patients with heart disease and cancer provided similar responses. Participants’ views about good deaths, bad deaths, and EoL scenarios were heterogeneous. Each participant voiced a unique combination of themes in describing good and bad deaths. Participants provided differing explanations for why given themes contributed to good deaths. Currently valued aspects of life were not easily translated into end-of-life preferences.
Communication is important about EoL care preferences. Not only questions about general values need to be asked, also about EoL choices and the reasons for these.
TRANSITIONS
Cassarett et al. 2004 USA (162)
Qualitative Semi-structured interviews
To describe hospice enrolment from the perspective of bereaved family members and to identify information
100 family members of 100 patients who died in hospice
Hospice Almost all family members (n592) and patients (n571) knew about hospice before the patient’s illness. Almost half the patients (n544) were not involved at all in the hospice enrolment decision. The patient’s physician (n551) or the patient or family (n534) initiated most hospice discussions, but patients and families usually obtained
By providing more information about hospice earlier in the illness course, clinicians may be able to facilitate more-informed and more-timely decisions about
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 303
about hospice that would encourage patients and families to enroll sooner.
information about hospice from a hospice representative (n575) rather than from the patient’s physician (n522). Family members identified several kinds of information about hospice that were particularly helpful in deciding whether to enrol and described several aspects of hospice that they wished they had known about sooner. Patients were not involved in decisions due to cognitive impairment.
hospice enrolment.
Chung et al. 2009 (163) USA
Qualitative study Semi-structured interviews: 1 with patient 10 with family 9 with patient and family
(1) To assess the level of hospice knowledge on the part of patients and/or family members; (2) investigating the decision process of hospice enrollment, range of services used, and transitions in the place of hospice care by patients’/family members’ level of knowledge.
20 minority elderly hospice patients
Urban non-profit Medicare-certified hospice.
Half the sample had no/little knowledge of hospice at the time of hospice enrollment decision. The remaining half of the sample had some knowledge or insider knowledge. Those with no prior knowledge fit the profile of the traditional hospice population: cancer and kidney-failure and likely to forgo treatment. Those with some prior knowledge: did not fit the profile: conditions with less clear prognoses and time until death. The highest level of knowledge (through carers’ health care occupations) influences hospice care after enrollment.
To focus future research on how to achieve an optimal knowledge-building strategy about hospice care.
Forbes et al. 2000 USA (165)
Descriptive qualitative study Four focus group discussions. Content analysis.
To describe families’ decision-making processes, both cognitive and affective, regarding EoL treatments for nursing home residents with moderately severe to very severe dementia.
28 family members of residents with moderately severe to severe dementia
4 nursing homes in a Midwestern American city selected on the basis of their racial and economic diversity of residents
Emotional effect, insult-to-life story, two faces of death, values and goals regarding end-of-life treatments,and the unrecognized trajectory of dying. Family members made decisions in an emotional climate of overwhelming burden and guilt, because their loved one’s life had been robbed of personhood, the changes associated with decline from a dementia-related illness were unrecognized as part of a trajectory of dying, death was both a tragedy and a blessing.
Family members need assistance in processing difficult and painful emotions, understanding the trajectory of disease, what decisions might impede a natural death, and comfort or palliative care options.
Lee et al. 1997 UK (174)
Pre-relocation questionnaires to seek the views of carers and staff. Questionnaires sent out 3 months after re-location. Meetings with carers and staff.
To minimize the negative effects in the process of relocation and, to review care procedures, identifying and implementing good practice in the new accommodation.
20 relatives of severe dementia patients, 17 staff.
Relocation of a group of confused elderly people from a long stay psychiatric ward and a similar client group attending a day
Relocation of services affects not only patients and staff, it also has implications for the education of staff and managers. Without appropriate preparation for relocation, the potential negative effects for older patients could be life threatening. The policy of a mixed economy of welfare also implies working in a different culture with different values to which nursing staff would need to adapt.
Planners involved with hospital closure programmes must involve professionals. Re-provision of services should mean an opportunity to evaluate practice and to make changes to improve the service. Staff professionalism needs to be supported by an in-service training agenda for those nursing staff who are
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 304
hospital, to a purpose built nursing home
going to be working in a very different culture of health care. Inclusion of budgetary and health care funding elements in nurse education. There needs to be greater opportunities for trainee nurses to undertake practice placements in the independent sectors of health care. A mandatory policy to provide `preparatory programme' for relocation prior to relocation plans is needed.
Liu & Tinker 2003 Taiwan (182)
Questionnaires on issues about institutional placement. Open questions descriptively presented
To explore the decision-making process of elderly people who entered nursing homes and that of their carers/key families in Taiwan. It traces the admission process, factors influencing perceptions of alternatives and extent of involvement of the carers/key families and the elderly people. • What were the patients’ and carers’/key families’ role in the decision making process? Who was the
235 elderly people in nursing homes 265 primary carers/ key families providing assistance to older relatives
Transition from home to long term care nursing homes
Most elderly people in Taiwan are cared for in their own homes by their families but, in some circumstances, entry to a nursing home seemed inevitable. The decisions were taken mainly within a family context. The adult children of the elderly people, carers/key families’ preferences and the availability of carers influenced the decision. Apart from the important need factors of elders, families’ views about alternatives to nursing homes were significantly influenced by their preferences. There was anxiety about the care provided and QoL in nursing homes. Some people were told their permanent admission was only temporary.
The family setting needs to be a target for intervention (eg carer allowance). Governments should pay more attention to community care. There should be development of new care environments encouraging independence.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 305
most influential person? • What were the factors which influenced their perceptions of alternatives to nursing home entry?
McLennon et al. 2010 USA (169)
Mixed methods Semi-structured interviews with open-ended questions and content analysis. Descriptive statistics
1. To identify common themes from interviews with caregivers who participated in an intervention study to assist caregivers of relatives with Alzheimers and Parkinsons who withdrew because they decided to institutionalize their relative and 2. to describe the characteristics of the sample that withdrew, including any acute event data about crisis events.
11 carers of 9 patients with Alzheimers and 2 with Parkinsons disease
Transition from home to institution in 10 cases, from home to other carer at home in 1 case.
3-4 months before institutionalization carers knew that they were not able to continue caring for their relatives at home. The most frequent reason was serious health events. There were more institutionalizations for the Alzheimers group, indicating that caring for Alzheimers is more difficult than for Parkinsons Disease.
When carers express the need for change in place of care, it may be a signal for immediate assessment and referral to appropriate resources for assistance.
Stadjuhar et al. 2005 Canada (180)
Ethnographic study Participant observation, interviews, focus group discussions Analysis: constant comparative
To describe, from the perspectives of family caregivers and health care providers, the variations in and factors influencing family members' decisions for palliative home care.
13 family members providing care, 47 bereaved family members, 25 health professionals
Home-based palliative care
Decision making process was unique for each family member, but commonalities: Some caregivers made uninformed decisions, giving little consideration to the implications of their decisions. Others made indifferent decisions, reluctantly agreeing to provide care at home, and still others negotiated decisions for home care with the dying person. Decisions were influenced by: 1. fulfilling a promise to the patient to be cared for at home, 2.desiring to maintain a 'normal family life', 3.previous negative encounters with institutional care.
Better preparation of caregivers for their role, enhance their choice in the decision-making process, improve care for the dying in hospital, and consider the development of alternate options for care.
Wakunami et al. 2009 Japan (178)
Qualitative Focus group discussions. Semi-structured interviews with
To develop an understanding of the process by which families accept the elderly with severe
23 family members interested in caring for frail relatives in two sets of male and female focus
Focus groups in community centers. Interviews in homes.
Categories: (i) family affection with wishes for continued survival; (ii) vacillation of desire for death with dignity; (iii) family members’ hierarchy; (iv) awareness that others may make different decisions; (v) family members’ discussion overcoming discordance; (vi) satisfaction with physicians’ explanations; (vii) impressions of life-sustaining measures; (viii) entrusting
Collaborative surrogate decision making by families and physicians.
© Queen’s Printer and Controller of HMSO 2012. This work was produced by Murtagh et al. under the terms of a commissioning contract
issued by the Secretary of State for Health
Project 08/1813/257 306
family members who had made important medical decisions.
brain damage as near death
groups 10participants (5 men and 5 women) who made important medical decisions were interviewed separately. 3 were bereaved.
important decisions to hospital physicians; (ix) significance of family members’ previous experiences; (x) patient’s age; and (xi) duration of medical treatment.
Waldrop & Kirkendall 2010 USA (171)
Qualitative Descriptive study In-depth interviews with and focus groups
To explore the interrelationship between location (rural or urban) and the type of agency (hospice or CHHA-hospice) and to describe how it influences the experiences of older persons and their carers at the EoL.
the directors of CCHA and hospice agencies, key staff constituents (nurses, social workers)
Home-based care
Themes: Rural-Urban Differences (geographic challenges, market forces, and programming issues) and EoL Care Issues (macrosocial, mezzosocial, and microsocial factors). Implications for social work practice.
Importance of: policy-guided practice environmental factors interdisciplinary and inter-agency collaboration Advance Care Planning
Waldrop & Rinfrette 2009 USA (172)
Mixed methods Ethnography of team meetings informed the development of questions for focus groups and written follow-up surveys
Hospice professionals views on the appropriate timing for and communication about hospice.
53 hospice professionals.
Hospice Timely admissions are situation-specific and involve a terminal prognosis, terminal decline, and a mutual understanding.
Hospice needs to be available earlier in the illness trajectory, this will ease transition. Additional education for health professionals about how and when to address EoL issues. Training for health professionals is also important in family dynamics.
© Queen’s Printer and Controller of HMSO 2011. This work was produced by Murtagh et
al. under the terms of a commissioning contract issued by the Secretary of State for
Health
Project 08/1813/257 299