Understanding place of death for patients with non ... · Murtagh FEM, Bausewein C, Petkova H, Sleeman KE, Dodd RH, Gysels M, Johnston B, Murray S, Banerjee S, Shipman C, Hansford

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



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.


Evaluation, Trials and Studies Coordinating Centre

University of Southampton

Alpha House, Enterprise Road

Southampton SO16 7NS

mailto:[email protected]



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.



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



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


6.2 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



Health

Project 08/1813/257 6


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



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



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



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


reported by professionals .......................................................................... 71


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



Health

Project 08/1813/257 10


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



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



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



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



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


Figure 104. Association of functional status with nursing home death in the


Figure 105. Association of hospitalisation with nursing home death in the


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


Figure 108. Association of number of nursing home beds with nursing home


Figure 109. Association of nursing home transfer policy with nursing home


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



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



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.



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).



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.



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



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


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



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.



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.



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.



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



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/



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.



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.



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.



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



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.



© 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



Project 08/1813/257 32



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.



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.



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)



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



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.



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



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



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-



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.



Health

Project 08/1813/257 42

Figure 4. Quality scale used for quantitative evidence (adapted from

Edwards (68, 69))



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)



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.



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



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.



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.



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.



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



Health

Project 08/1813/257 50

6 Results: preferences for place of care


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).



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%


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)



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.



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



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


Participants

Target age group


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


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





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



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



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).





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 ▲



Project 08/1813/257 59

* using Figure 6: Algorithm to assess strength of evidence (15)





Health

Project 08/1813/257 55

Table 7. Evidence on factors associated with preference for hospice enrolment as

reported by patients



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


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


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


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



Health

Project 08/1813/257 62

7 Results: preferences for place of death


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.



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.





Project 08/1813/257 58

Table 8. Included papers which report preferences for place of death in advanced non malignant conditions reported by

patients



Project 08/1813/257 65

Paper

Country

Design

Data collection


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.



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%


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





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%.





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%.





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


Participants

Target age group


Disease group

N


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


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



Project 08/1813/257 70

Paper

Country

Design

Data collection


Participants

Target age group


Disease group

N


Quality score

Comments

Ramon 2006

Spain


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



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


Participants

Target age group


Disease group

N


Quality score

Comments

Abarshi 2009

NL


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



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%



Project 08/1813/257 72

Paper

Country

Design

Data collection


Participants

Target age group


Disease group

N


Quality score

Comments

Van den Block 2007

Belgium



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





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


Participants

Age group


Disease group

N


Quality score

Comments

Groth-Juncker 1983

USA


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


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



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



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



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



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.



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.



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



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


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


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


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


Financial status Average Above average

Low

(1/1)

▲ GP awareness for preferred place of death ▲ GP awareness for preferred place of death



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.



Project 08/1813/257 75

Table 14. Included papers reporting multivariate analyses on factors affecting place of death

Paper

Country

Design

Data collection


Participants

Target age group


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


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


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


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%


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%


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%


71% high

Fried 1999



long term home care program

patients in long term home care program

65+ 80 (8.3) all diagnoses: proportions not given


71% high



Project 08/1813/257 83

Gruneir 2007 USA analysis of routinely collected data


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


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


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%


57% med

Houttekier 2010

Belgium England Nether-lands Scotland Wales

analysis of routinely collected 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%


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%


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%


79% high

Levy 2004


Medicare claims data

US nursing homes


65+ approx 81 all diagnoses: proportions not given


71% high



Project 08/1813/257 84

Liu 2007

China prospective cohort study

hospital records

acute hospital COPD patients requiring mechanical ventilation


64% med

McGregor 2007

Canada retrospective cohort study

administrative data plus nursing home records

publically funded nursing homes


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


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%


n

86% high

Miller 2003


hospice program providers databases

hospice programs (inpatient 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%


n

79% high

Mitchell 2005

USA analysis of routinely collected data



whole national population


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


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%


n

71% high

Muramatsu 2008


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




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



479 14 logistic reg

n

57% med



Project 08/1813/257 85

Ramon 2006

Spain retrospective cohort study


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


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




ecological study

18 multiple linear reg

n

36% med

Shega 2008



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



n

71% high

Silveira 2006


death and birth registration data



all ages 75 (14) all diagnoses: cancer 27% circulatory disease 35% cerebrovascular disease 9% COPD 6% renal disease 1%


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 *


death registration data plus proxy data from family members interviewed after the death

nationally representative sample of all deaths

sample of whole population


all diagnoses: cancer 26.9% chronic cardiovascular 25.7%

12,771** 4 logistic reg

n

79% high



Project 08/1813/257 86

* these papers report different components of the same study, hence **participants only counted once

Weitzen 2003 *


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




all ages overall age not provided


ecological study

3 logistic reg

n

71% high

Yasamura 2000

Japan retrospective cohort study


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



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).



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



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.



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.



Project 08/1813/257 84

Table 15. Evidence on factors associated with home death in the general advanced non-cancer population




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


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


Living alone No informal carer

Moderate Low

66% (2/3)

- (1/1)

Living alone - ▼home death No significant effect


Hospital bed availability

High

100% (4/4)

Increased hospital bed availability - ▼home death



Project 08/1813/257 92


Household or area income

Moderate

100% (2/2)

Increased income - ▲home death


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



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.



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.



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



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.



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.



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



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.



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.



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.



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



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.



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



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.



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.


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.



Project 08/1813/257 100

Table 17. Evidence on factors associated with hospital death in the general non cancer population




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


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


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



Project 08/1813/257 108


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


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


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



Health

Project 08/1813/257 102


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



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.



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.



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).



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



Health

Project 08/1813/257 114


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.



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



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).



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



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



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.


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.



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).



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.


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).



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).



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).



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).



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.



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).



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.



Project 08/1813/257 121

Table 18. Evidence on factors associated with hospital death in the COPD population




Age

Low

- (1/1)

Increasing age -▲hospital death


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



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



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



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



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.



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).



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.



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).



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-



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



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).



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.


We identified 18 factors associated with nursing home death for the general

population with advanced non cancer (see Table 19).



Project 08/1813/257 133

Table 19. Evidence on factors associated with nursing home death in the general non-cancer population




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


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


Home care situation: Had caregiver

Low

- (1/1)

Presence of caregiver –▼nursing home death


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


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



Health

Project 08/1813/257 141


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).



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.



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



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).


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



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.



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).



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).



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).


Figure 91. Association of PC provision with nursing home death



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.


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.



Health

Project 08/1813/257 150

12.8 Factors specific to dementia populations


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


Age Gender

Low Low

- (1/1)

- (1/1)

Increasing age - ▲nursing home death Female gender - ▲nursing home death


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



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).



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



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


Type of factor Factor Strength of evidence Consistency Direction


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


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



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).



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).



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.



Health

Project 08/1813/257 157

Figure 103. Association of cognitive status with nursing home death in the


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


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).



Health

Project 08/1813/257 158

Figure 105. Association of hospitalisation with nursing home death in the


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).



Health

Project 08/1813/257 159

Figure 107. Association of healthcare provision with nursing home death in the


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


The number of nursing home beds was not shown to have a significant

association with nursing home deaths (95, 123).



Health

Project 08/1813/257 160

Figure 109. Association of nursing home transfer policy with nursing home


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).



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



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



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



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



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


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



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



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


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.



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



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).



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



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,



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



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.



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).



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



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).



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”



Project 08/1813/257 171

Table 23. Included papers on transitions towards end of life in non malignant conditions

Paper

Country

Design

Data collection


Participants

Target age group


Disease group

N

No of factors

Type of analysis

Quality score

Ahmed. 2003


Medicare beneficiaries medical records

Regional CHF patients discharged from hospital

65+ 79 (7.5) CHF 908 4 logistic regression

50% medium

Albert. 1999:


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%


36% medium

Anderson. 1993


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


not specified 77.3 (12.7) all diagnoses: Cancer 14% Dementia 16% Depression 24% Diabetes 26% End stage disease 3%


71% high



Project 08/1813/257 179

Paper

Country

Design

Data collection


Participants

Target age group


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%


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%


71%high

Becker. 2009


Medicaid data of nursing home residents

Regional Medicaid nursing home provision and hospitalisations


all ages 77.3 (13.8) all diagnoses: Alzheimers disease 15.1% Other dementias 26.4% Other 23.6%


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



Project 08/1813/257 180

Paper

Country

Design

Data collection


Participants

Target age group


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


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


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


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


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



Project 08/1813/257 181

Paper

Country

Design

Data collection


Participants

Target age group


Disease group

N

No of factors

Type of analysis

Quality score

Dai YT. 2002 Taiwan prospective cohort study


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%


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


71.3(SD 8.2) Parkinson’s Disease 100% Dementia 56%


57%medium

Fillenbaum. 2001


Secondary analysis of research data from CERAD(Consortium to establish a Registry for Alzheimer's disease)


Alzheimer's disease patients living at home or in institutions


<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


88.8 (range 72-101)

Dementia 17% 316 2 logistic regression

71% high



Project 08/1813/257 182

Paper

Country

Design

Data collection


Participants

Target age group


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%


71%high

Gessert. 2006


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%


79% high

Gielen B. 2010

Belgium cross-sectional national survey

Administrative data of one sickness fund


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)


64%medium

Givens. 2010 USA retrospective cohort study

Medicare claims data


CHF medicare beneficiaries with no previous hospice admissions


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


73 (56-89) Alzheimer's disease 66 7 linear regression

71%high



Project 08/1813/257 183

Paper

Country

Design

Data collection


Participants

Target age group


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%


50% medium

Hebert. 2001

Canada prospective cohort study

Longitudinal proxy data from caregivers interviewed

Country Dementia patients living in the community


Not given Dementia 326 15 logistic regression

71% high

Krumholz. 1997


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


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%


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%


79% high



Project 08/1813/257 184

Paper

Country

Design

Data collection


Participants

Target age group


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


<60 (10%), 61-74 (22%) >75(68%)

all diagnoses: CVA 46% other 54%


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%


79% high

Mehr. 1997 USA Prospective cross-sectional survey

national nursing home provider data

National group of nursing homes


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%


79% high



Project 08/1813/257 185

Paper

Country

Design

Data collection


Participants

Target age group


Disease group

N

No of factors

Type of analysis

Quality score

Menec 2009


administrative data plus nursing home and hospital databases


all long term care residents from participating facilities

≤ 74 9.2% 75-84 9.5% 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%


86% high

Miller. 2001 USA retrospective cohort study regional

Medicare claims data and minimum data set assessments


nursing home residents, two groups one enrolled in the hospice(9202) and (27500) not.


83.7(9.3) all diagnoses: Cancer 59%, COPD 18%, CHF 36%, Dementia 16% other 25%


79% high

Mitchell. 2004


Minimum data sets for long term care facilities and community dwelling patients


AD patients from nursing homes and the community (terminal care patients)


83.2 Dementia 3020 5 logistic regression

79% high



Project 08/1813/257 186

Paper

Country

Design

Data collection


Participants

Target age group


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


ESRD- Dialysis patients that died between 2001-2


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


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


(48-88)(median73)

Parkinson’s Disease 50%

178 28 logistic regression * RR+CI alone

71% high

Qartararo. 1991

Australia Randomised control trial


regional Nursing home applicants in a regional sample


81.6 (7.6) all diagnoses: Stroke 22.3% Dementia 34%, Cardiorespiratory disease 31% other 27.4%


70% high



Project 08/1813/257 187

Paper

Country

Design

Data collection


Participants

Target age group


Disease group

N

No of factors

Type of analysis

Quality score

Rich. 1995 USA Randomised control trial


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


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%


86% high

Rosenberg. 2005



Regional Residents from assisted living facilities


85.7 (8.3) all diagnoses: Hip fracture 9.9%, Chronic pain 1.6%


86% high

Rosenwax. 2008

Australia retrospective cohort study



deaths from Alzheimer's disease compared with other causes

75+ 75-84 35.2% >85 64.8%

all diagnoses: Alzheimer’s disease 3.7%


71% high

Rundek. 2000


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%


86% high



Project 08/1813/257 188

Paper

Country

Design

Data collection


Participants

Target age group


Disease group

N

No of factors

Type of analysis

Quality score

Salive. 1993 USA prospective cohort study

longitudinal participant data derived from epidemiologic studies


Community dwelling older adults

65+ Not given all diagnoses: proportions not given


79% high

Smith and Stevens. 2009


Medical and billing records

Regional Patients discharged from hospital (2006)

60+ Not given all diagnoses: proportions not given


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%,


64% medium

Tsuchihashi 2001

Japan prospective cohort study

Medical records Regional hospitals

CHF patients discharged from hospital


69 (14) all diagnoses: CHF 100%, Diabetes 25% Stroke 14% Renal failure 11% Hypertension 41%


79% high

Wheelock. 2003


Huntington Disease Patient database

International - Largest population 68.8% from USA

Huntington Disease patients living in nursing facilities and in communities


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


AD and LBD patients assessed for mortality and time to Nursing home


Not given all diagnoses: Alzheimer’s disease 80%, Lewy Body Dementia 20%

315 2 logistic regression **HR and CI alone

79% high



Project 08/1813/257 189

Paper

Country

Design

Data collection


Participants

Target age group


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


Dementia patients living in the community


78.9(7.8) Dementia 5788 5 logistic regression

85% high

TOTAL PARTICIPANTS:



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



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



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).



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).



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.



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).



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.



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.



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).



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).



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.



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.



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



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.



Project 08/1813/257 197

Figure 120. Model of the factors affecting place of death in non malignant conditions (high strength evidence only)


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



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


Specific symptoms. illness burden and trajectories: Degree of impaired mobility/disability (severe functional impairment ▲ home death)

Macro-social factors:



Project 08/1813/257 198

Figure 121. Model of the factors affecting place of death in non malignant conditions (high & moderate strength evidence)


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



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


Specific symptoms. illness burden and trajectories: Degree of impaired mobility/disability severe functional impairment ▲

home death functional status - no effect on hospital death




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.



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



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.



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.



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



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



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



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.



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.



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


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.



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.



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




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



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.



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.



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.



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.



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.



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.



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.



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.



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.



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



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.



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.



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.



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.



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.



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.



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.



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



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.



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


setting)) OR (admission adj3 (care or place or


setting)) OR (admission adj3 (care or place or



Health

Project 08/1813/257 239


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



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



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.



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.



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.



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.



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.



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.



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.



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.



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.



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.



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



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.



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.



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



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.



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.



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.



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.



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.



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.



Project 08/1813/257 253



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


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


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


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


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



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


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


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



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





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


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



Carey HB, Chorney W, Pherson K, Finkelstein FO, Kliger AS

2001 Continuous peritoneal dialysis and the extended care facility

American Journal of Kidney Diseases


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


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?



Casarett DJ, Hirschman KB, Henry MR 2001 Does hospice have a role in nursing home care at the end of life?



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


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


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


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


Christakis NA 1994 Timing of referral of terminally ill patients to an outpatient hospice

Journal of General Internal Medicine


Christakis NA, Escarce JJ 1996 Survival of Medicare patients after enrollment in hospice programs

New England Journal of Medicine


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


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



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


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



Collins C, Ogle K 1994 Patterns of predeath service use by dementia patients with a family caregiver



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


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


Ellershaw J 2007 Care of the dying: What a differene an LCP makes!


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



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



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


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



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


Galanos AN, Hays JC, Moore JD, Poppe A

2004 Where do continuing care retirement community residents die? [1]


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


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


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



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


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


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


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


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



Project 08/1813/257 268

Hicks F, Corcoran G 1993 Should hospices offer respite admissions to patients with motor neurone disease?


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


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


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


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



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?



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


Kapo J, Harrold J, Carroll JT, Rickerson E, Casarett D

2005 Are we referring patients to hospice too late? Patients' and families' opinions


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


Kite S, Jones K, Tookman A 1999 Specialist palliative care and patients with noncancer diagnoses: The experience of a service


Kovach CR 1998 Nursing home dementia care units. Providing a continuum of care rather than aging in place

Journal of Gerontological Nursing


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


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




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


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


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


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


London MR, McSkimming S, Drew N, Quinn C, Carney B

2005 Evaluation of a comprehensive, adaptable, life-affirming, longitudinal (CALL) palliative care project


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?



Luchins DJ, Hanrahan P, Murphy K 1997 Criteria for enrolling dementia patients in hospice



Ludke RL, Smucker DR 2007 Racial differences in the willingness to use hospice services


Lunney JR, Lynn J, Hogan C 2002 Profiles of older medicare decedents Journal of the American Geriatrics Society




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


MacGregor MS, Agar JWM, Blagg CR 2006 Home haemodialysis - International trends and variation



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


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


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


McCarthy M, ddington-Hall J, Altmann D

1997 The experience of dying with dementia: A retrospective study



McCluskey L, Houseman G 2004 Medicare Hospice Referral Criteria for Patients with Amyotrophic Lateral Sclerosis: A Need for Improvement


McDonald GJ 1981 A home care program for patients with chronic lung disease

Nursing Clinics of North America


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


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



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


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?



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



Miller SC, Weitzen S, Kinzbrunner B 2003 Factors Associated with the High Prevalence of Short Hospice Stays


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


Mitchell SL, Teno JM, Intrator O, Feng Z, Mor V

2007 Decisions to forgo hospitalization in advanced dementia: A nationwide study



Mitty EL, 2004 Assisted living: Aging in place and palliative care Geriatric Nursing 25 3 149 156+163


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


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


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




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


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


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


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



Reddy NC, Korbet SM, Wozniak JA, Floramo SL, Lewis EJ

2007 Staff-assisted nursing home haemodialysis: patient characteristics and outcomes



Reilly RB, Teasdale TA, McCullough LB 1994 Projecting patients' preferences from living wills: An invalid strategy for management of dementia with life-threatening illness



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



Rhodes RL, Teno JM, Welch LC 2006 Access to hospice for African Americans: Are they informed about the option of hospice?




Project 08/1813/257 274

Rich A, Ellershaw J, Ahmad R 2001 Palliative care involvement in patients stopping haemodialysis


Riesenberg D 2000 Hospital care of patients with dementia Journal of the American Medical Association


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


Robertson C, Warrington J, Eagles JM 1993 Relocation mortality in dementia: The effects of a new hospital



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


Sampson WI 1977 Dying at home Journal of the American Medical Association


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


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


Scocco P, Rapattonoi M, Fantoni G 2006 Nursing home institutionalization: A source of eustress or distress for the elderly?



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


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



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




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


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


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



Smith GE, Kokmen E, O'Brien PC 2000 Risk factors for nursing home placement in a population-based dementia cohort



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


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


Steinbach U 1992 Social networks, institutionalization, and mortality among elderly people in the United States


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



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



Teno JM, Weitzen S, Fennell ML, Mor V 2001 Dying trajectory in the last year of life: does cancer trajectory fit other diseases?


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


Tsuji I, Whalen S, Finucane TE 1995 Predictors of nursing home placement in community-based long-term care



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


Wellard SJ, Street AF 1999 Family issues in home-based care International journal of nursing practice




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


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


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


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


284 19 2489 2494 Exclude not original research, overview article



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



Project 08/1813/257 279


Method and data



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



Project 08/1813/257 280


Method and data



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



Project 08/1813/257 281


Method and data



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.



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.



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



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



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.



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.



Project 08/1813/257 287

Table 28. Data Extraction Table COPD




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



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.



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.


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.



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



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.



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.



Project 08/1813/257 293

Table 29. Data Extraction Table Long Term Neurological Conditions




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.



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



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.



Project 08/1813/257 296

Table 30. Data Extraction Table End Stage Kidney Disease




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.



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.



Project 08/1813/257 297



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.



Project 08/1813/257 298

Table 31. Data Extraction Table Older People, including stroke and dementia




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.



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.



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)


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



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.


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



Project 08/1813/257 302

appropriate site of care.


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)


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



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



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.



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.



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.



Health

Project 08/1813/257 299

Documents

Understanding place of death for patients with non ... · Murtagh FEM, Bausewein C, Petkova H, Sleeman KE, Dodd RH, Gysels M, Johnston B, Murray S, Banerjee S, Shipman C, Hansford