Preference for Place of Death in Germany

Abstract

Background:

Dying in the preferred place is considered a key requirement for a “good death.” The aims of our study were to explore preferred places of death of deceased people and their bereaved relatives in Rhineland-Palatinate (Germany). We further wanted to assess the congruence between preferred and actual place of death.

Methods:

The cross-sectional study was based on a random sample of 5000 inhabitants of Rhineland-Palatinate (Germany) who died between May 25 and August 24, 2008. Relatives of these deceased persons were interviewed by a written survey.

Results:

After removing duplicates, 4967 questionnaires were sent out, 3832 delivered, and 1378 completed, yielding a response rate of 36.0%. Regarding the deceased, 93.8% wanted to die at home, 0.7% in a hospital, 2.8% in palliative care, 2.4% in a nursing home, and 0.3% elsewhere. The figures for the relatives were 80.7%, 4.3%, 7.5%, 7.1%, and 0.5%, respectively. Of the deceased 58.9% and of the relatives 59.1% had their wish fulfilled. Logistic regression analysis revealed that living in a rural municipality (adjusted odds ratio [aOR]: 1.88; 95% confidence interval [CI]: 1.02–3.43), rural town (aOR: 2.30; 95% CI: 1.17–4.49) or small town (aOR: 1.95; 95% CI: 1.04–3.68), having a nonworking relative (aOR: 1.79; 95% CI: 1.16–2.76), and living together with a relative (aOR: 2.28; 95% CI:1.57–3.32) increases the probability to die in the preferred place.

Discussion:

Because the availability of a relative was the most important factor to die in the preferred place, relatives of dying people should be supported in providing informal care. The introduction of palliative home care teams should allow more people to die in their preferred place by easing the burden of informal carers.

Introduction

To die in the place of preference is considered a key requirement for a “good death.”¹ A recent methodological review on preferred place of death showed that home was most preferred in 11 of 12 studies.² Reasons to prefer a death at home include being surrounded by loved ones, the maintenance of a sense of normality, physical and emotional support, autonomy and privacy, and a greater chance to control the environment.^3–5 However, it is not always possible or feasible to die at home. Severe symptoms prior to death, a lack of medication and professional caregivers (especially for the old), and the burden of care for families and friends might be obstacles to a death at home.^3,6,7 Some of these reasons might also explain why the proportion of people preferring a home death decreases as death draws nearer. In a UK study 59 cancer patients provided information on their preferred place of death at two different points in time. Fifty-eight percent chose home as the preferred place of death, 20% hospital, 20% hospice, and 2% elsewhere, when asked initially. These figures changed to 49%, 24%, and 25% for home, hospital and hospice, respectively, when asked a second time.⁸ A similar result was found in another study from the UK where 612 patients were asked about their preferred place of care. Eight weeks prior to death more than 75% wanted to die at home. One week prior to death this number decreased to about 50%, shifting mostly toward hospice death.⁹

Investigations on the proportion of people who were able to die in the place of preference ranged from 30% to 90% in the above-mentioned review.² In Germany, the most recent study on preferred place of death dates back to the year 2003 and showed that 75% of 272 cancer patients wanted to die at home, 3% in another private residence, 15% in a hospital, and 2% in a nursing home. In the same study, 72 relatives stated afterwards that 58% of the patients died in the place of their preference, 28% did not, and 14% did not know.¹⁰ Another German study from 2001, interviewing 644 nondiseased people living in the German federal state of Thuringia, revealed that 77% chose home as the most preferable place of death, followed by hospital with 9%, nursing home with 2%, and another private residence with 2%.¹¹ However, actual data on preferred and actual places of death in Germany, comprising a large representative sample, are missing. The aims of our study were to explore the preferred place of death of deceased people in Rhineland-Palatinate (Germany) and of their bereaved relatives. We further wanted to assess the congruence between preferred and actual place of death.

Methods

Data collection

The process of data collection is described elsewhere in detail.¹² Briefly, the EPACS (Establishment of Hospice and Palliative Care Services in Germany) study was carried out in the form of a cross-sectional, written survey between September 2008 and January 2009. The base of the survey was a random sample of 5000 addresses of deceased inhabitants with principal residence in the German federal state of Rhineland-Palatinate (4 million inhabitants), who had died between May 25 and August 24, 2008. A standardized and anonymized questionnaire was sent out to the former address of the deceased. The bereaved, who were not identified beforehand, were kindly asked in a cover letter to complete the questionnaire. Our study was approved by the ethical committee of the medical association of Rhineland-Palatinate and the data protection officer of Rhineland-Palatinate.

Questionnaire

Our questionnaire gathered information on the preferred place of death of the deceased and the bereaved, the actual place of death, sociodemographic characteristics, and the quality of inpatient and outpatient care.

Information on the actual and preferred place of death was surveyed as follows:

• Where did your relative die? (at home, at the intensive care unit of a hospital, at the palliative care unit of a hospital, at a standard hospital ward, in a nursing home, in a hospice, and elsewhere [free-text])

• Did your deceased relative ever tell you where he or she wanted to die? (Yes, No, I don't know)

• If yes, what was his or her preferred place? (at home, in a nursing home, in a hospital, in a hospice, in a palliative care unit of a hospital, and elsewhere [free-text])

• What was your personal wish regarding the place of death of your deceased relative? (at home, in a nursing home, in a hospital, in a hospice, in a palliative care unit of a hospital, elsewhere [free-text], no wish)

Statistical Analysis

All statistical analyses were done with STATA/IC statistical software version 10.1 (StataCorp., College Station, TX). Absolute and relative frequencies were calculated for the general characteristics of the deceased and the respondents. The κ statistic was used as a measure of congruence between the actual and preferred place of death. Univariable and multivariable logistic regression models were applied to estimate influencing factors on the congruence between preferred and actual place of death. The dependent variable was defined as “having died in the place of preference” (1) versus “not having died in the place of preference” (0). We applied backward selection method for the multivariable model, assuming a p value of <0.2 as exclusion criterion.

Results

After removing duplicates from our random sample of 5000 deceased people, 4967 questionnaires were mailed. Questionnaires actually delivered totalled 3832; 1134 were returned as undeliverable. Altogether 1378 questionnaires were completed, leading to a response rate of 36.0% (considering only those questionnaires that actually were delivered).

Table 1 depicts basic sociodemographic characteristics of the deceased and the respondents.

Table 1.

Sociodemographic Characteristics of the Deceased and the Respondents (n=1378)

Characteristic	n	%
(a) Deceased
Age
0–49	51	3.7
50–59	84	6.1
60–69	185	13.4
70–79	322	23.4
80–89	526	38.2
90+	210	15.2
Sex
Male	612	44.4
Female	766	55.6
Nationality
German	1364	99.0
Other	14	1.0
Marital status
Married	654	47.5
Widowed	594	43.1
Divorced	47	3.4
Single	83	6.0
Education^a
High	148	10.7
Average	167	12.1
Low	977	70.9
Missing	86	6.2
Financial status^b
Good	494	35.9
Average	604	43.8
Bad	203	14.7
Missing	77	5.6
Place of residence^c
Rural municipality	426	30.9
Rural town	218	15.8
Small town	329	23.9
Medium-sized town	249	18.1
Large town	156	11.3
Underlying disease^d
Cancer	334 (537)	24.2 (39.0)
Dementia	122 (310)	8.9 (22.5)
Cardiovascular disease	115 (424)	8.4 (30.8)
Other	166 (554)	12.0 (40.2)
Multimorbidity	504	36.6
Missing/I don't know	137	9.9
Progressive incurable disease^e
Yes	998	72.4
No	351	25.5
Missing	29	2.1
(b) Respondents
The deceased was my …
Spouse	491	35.6
Mother/Father	653	47.4
Mother-in-law/Father-in-law	72	5.2
Daughter/Son	14	1.0
Sister/Brother	38	2.8
Aunt/Uncle	47	3.4
Grandmother/Grandfather	20	1.5
Other	42	3.1
Missing	1	0.1
Age
0–39	87	6.3
40–49	221	16.0
50–59	422	30.6
60–69	360	26.1
70–79	185	13.4
80+	85	6.2
Missing	18	1.3
Sex
Male	491	35.6
Female	873	63.4
Missing	14	1.0
Education^a
High	383	27.8
Average	330	24.0
Low	609	44.2
Missing	56	4.1
Respondent lived in the same household as the deceased
Yes	759	55.1
No	604	43.8
Missing	15	1.1

The German educational system does not have exact counterparts in most other countries. After 4 years of primary school pupils are assigned to one of three types of schools (in descending order of duration): the Gymnasium (grammar schools classes A-level), the Realschule (intermediate secondary school), or the Hauptschule (secondary general school). According to that classification we set up a categorical variable containing high (having finished the Gymnasium), average (having finished the Realschule) and low (having finished the Hauptschule/no degree at all) levels of education.

The financial situation of the deceased 4 weeks prior to death was surveyed by asking the respondents to choose between the categories “very bad, rather bad, average, rather good, very good.” We summarized these answers into “good,” “average,” and “bad,”

Rural municipality: <2000 inhabitants, rural town: 2000 to <5000 inhabitants, small town: 5000 to <20,000 inhabitants, medium-sized town: 20,000 to <100,000 inhabitants, large town: ≥100,000 inhabitants.

Respondents were asked to state the underlying diseases of their deceased relative during the last 4 weeks prior to death. Multiple answers were permitted. The figures outside the brackets show the number/percentage of people having suffered only from the aforementioned particular disease. Persons having suffered from more than one disease prior to death were assigned to the category “multimorbidity.” In contrast, the figures in brackets show the number/percentage of people having suffered from the aforementioned disease and potentially from one or more additional diseases. In this regard, the percentage values in brackets always refer to the total number of people (i.e., 1378) and are not to be interpreted as column percentages. The category “cardiovascular disease” includes stroke and chronic, severe heart diseases (including heart failure).

Respondents were asked to state if the death of their deceased relative was related to a progressive incurable disease, leading to death (e.g., cancer, dementia).

The mean age of the deceased was 77.6 years (median: 80; standard deviation [SD]: 13.2). Our sample comprised more deceased women (55.6%) than men (44.4%). The deceased were mostly married or widowed, had a low educational level, and an average or good financial situation.

About 70% of the deceased lived in regions with less than 20,000 inhabitants. Thirty-nine percent of the people in our sample suffered from cancer prior to death.

The respondents were mostly the spouses (35.6%) or children (47.4%) of the deceased. The respective mean age of the respondents was 58.8 years (median: 58; SD: 12.8). There were more female (63.4%) than male (35.6%) respondents.

The distribution of the actual and preferred places of death is shown in Table 2. Whereas almost the same proportion of people died in a hospital (39.3%) as at home (38.2%), respondents and deceased almost exclusively preferred a death at home, if they had a preference.

Table 2.

Actual and Preferred Place of Death (n=1378)

Variable	n	%
Actual place of death
At home	526	38.2
Hospital	541	39.3
Palliative care facility^a	103	7.5
Nursing home	185	13.4
Elsewhere	22	1.6
Missing	1	0.1
Preferred place of death (Deceased)^b
At home	696	50.5 (93.8)
Hospital	5	0.4 (0.7)
Palliative care facility^a	21	1.5 (2.8)
Nursing home	18	1.3 (2.4)
Elsewhere	2	0.2 (0.3)
No wish mentioned	534	38.8
More than one place mentioned	18	1.3
I don't know	62	4.5
Missing	22	1.6
Preferred place of death (Respondent)^c
At home	822	59.7 (80.7)
Hospital	44	3.2 (4.3)
Palliative care facility^a	76	5.5 (7.5)
Nursing home	72	5.2 (7.1)
Elsewhere	5	0.4 (0.5)
No wish	279	20.3
More than one place mentioned	24	1.7
Missing	56	4.1

Contains “palliative care unit of a hospital” and “hospice.”

Figures in brackets show the proportions of the preferred places of death of the 742 deceased who explicitly stated a preferred place of death prior to their death.

Figures in brackets show the proportions of the preferred places of death of the 1019 bereaved who explicitly stated a preferred place of death.

Table 3 shows a cross tabulation of preferred places of death of the deceased and the actual places of death. Of the 696 people who preferred a death at home, 406 (58.3%) actually died at home and 290 persons (41.7%) died in another place.

Table 3.

Preferred Place of Death of Deceased by Actual Place of Death (n=742)

	Actual place of death
Preferred place of death (Deceased)^a	At home	Hospital	Palliative care facility^b	Nursing home	Elsewhere
At home	58.3 (406)	28.6 (199)	4.2 (29)	8.2 (57)	0.7 (5)
Hospital	20.0 (1)	80.0 (4)	0.0 (0)	0.0 (0)	0.0 (0)
Palliative care facility^b	0.0 (0)	28.6 (6)	71.4 (15)	0.0 (0)	0.0 (0)
Nursing home	0.0 (0)	22.2 (4)	0.0 (0)	66.7 (12)	11.1 (2)
Elsewhere	0.0 (0)	0.0 (0)	100.0 (2)	0.0 (0)	0.0 (0)

Presented numbers are row percentages.

For deceased who explicitly disclosed a preference prior to their death.

Contains “palliative care unit of a hospital” and “hospice.”

κ=0.14; overall congruence: 58.9%.

The overall congruence, defined as all met preferences of deceased for all places of death, divided by all preferences for all places in our study, was 58.9%. Thus, 58.9% of the deceased who disclosed their preference on place of death, died in the place they wanted to and 41.1% did not. The κ statistic (κ=0.14) for chance correction showed a slight agreement. In a similar manner, the overall congruence between the preferred place of death of the respondents and the actual place of death was almost the same as for the deceased (59.1%), yet κ was higher (κ=0.31). Comparing the preferred places of death of the respondents and the deceased yields a percent agreement of 92.5% and a κ of 0.52.

To estimate influencing factors on the congruence of preferred and actual place of death, we carried out several univariable and one multivariable logistic regression analyses using backward selection method (exclusion criterion: p<0.2).

Results of the multivariable model in Table 4 show that the possibility to die in the place of preference is significantly higher when living in a rural municipality (aOR: 1.88; 95% CI: 1.02–3.43), rural town (aOR: 2.30; 95% CI: 1.17–4.49) or small town (aOR: 1.95; 95% CI: 1.04–3.68), as compared with a large town, when having a nonworking relative (aOR: 1.79; 95% CI: 1.16–2.76), and when the respondent and deceased lived together in one common household (aOR: 2.28; 95% CI:1.57–3.32).

Table 4.

Univariable and Multivariable Logistic Regression Models Predicting Congruence of Preferred and Actual Place of Death

	Deceased died in the place of his or her preference (1) vs. Deceased died not in the place of his or her preference (0)
	Univariable model			Multivariable model (n=597)
Variable	OR	95% CI^a	P Value	aOR^b	95% CI^a	P Value
Age of decedent
0–49	0.63	0.22–1.80	0.390	^d	^d	^d
50–59	1.18	0.51–2.74	0.694	^d	^d	^d
60–69	1.11	0.64–1.91	0.716	^d	^d	^d
70–79	1.02	0.65–1.61	0.916	^d	^d	^d
80–89	0.93	0.62–1.39	0.718	^d	^d	^d
90+	Ref.
Sex of decedent
Male	1.19	0.88–1.59	0.260	1.29	0.90–1.85	0.167
Female	Ref.
Marital status of decedent
Married	1.21	0.90–1.62	0.206	^d	^d	^d
Unmarried	Ref.
Education of decedent
Good	0.74	0.45–1.23	0.242	0.68	0.38–1.22	0.193
Average	0.70	0.45–1.10	0.122	0.67	0.40–1.13	0.136
Bad	Ref.
Financial situation of decedent
Good	1.50	0.96–2.33	0.076	^d	^d	^d
Average	1.51	0.99–2.31	0.056	^d	^d	^d
Bad	Ref.
Size of town^c
Rural municipality	2.00	1.20–3.33	<0.01	1.88	1.02–3.43	<0.05
Rural town	1.99	1.12–3.53	<0.05	2.30	1.17–4.49	<0.05
Small town	1.47	0.86–2.50	0.160	1.95	1.04–3.68	<0.05
Medium-sized town	1.68	0.96–2.94	0.068	1.77	0.92–3.41	0.086
Large town	Ref.
Disease
Cancer	1.06	0.78–1.44	0.695	^d	^d	^d
Other	Ref.
Age of respondent
0–39	0.42	0.18–0.96	<0.05	0.86	0.30–2.49	0.785
40–49	0.54	0.27–1.06	0.073	1.02	0.41–2.56	0.963
50–59	0.70	0.37–1.32	0.270	1.40	0.59–3.31	0.446
60–69	0.63	0.33–1.18	0.150	0.72	0.32–1.64	0.436
70–79	0.62	0.31–1.24	0.175	0.53	0.22–1.26	0.149
80+	Ref.
Occupational status of respondent
Not working	1.65	1.22–2.23	<0.01	1.79	1.16–2.76	<0.01
Working (Full/Part-time)	Ref.
Respondent lived in the same household as the deceased
Yes	1.99	1.48–2.67	<0.001	2.28	1.57–3.32	<0.001
No	Ref.

CI: confidence interval.

aOR: adjusted odds ratio. (Odds ratio adjusted for any potential predictor included in the model.)

Rural municipality: <2000 inhabitants, rural town: 2,000 to <5000 inhabitants, small town: 5000 to <20,000 inhabitants, medium-sized town: 20,000 to <100,000 inhabitants, large town: ≥100,000 inhabitants.

The variable was entered in the regression model but found not to be significant, and eliminated by the step-wise procedure.

Discussion

Main findings

Information on the preferred place of death of the deceased and the respondents revealed that the order of the preferred place was similar between these two groups. Dying at home clearly was favored, whereas dying in a hospital was least preferred. About two thirds of the deceased and the respondents had their wish come true. Multivariable logistic regression analysis revealed that the probability to die in the place of preference was significantly higher when the deceased lived in a rural municipality, rural town, or small town, as compared with a large town, when having a nonworking relative, and when living together with a relative in one common household.

Relevance of the findings and implications

To the best of our knowledge this is the first population-based study to deliver data on this topic for Germany. The finding that the vast majority of the dying (93.8%) who disclosed their preference on the site of death wanted to die at home is in accordance with recent international studies. In a US study, 87.2% of the dying preferred a death at home (cancer patients),⁴ so did 74% in a Taiwanese study (cancer patients),¹³ 93.5% in an Italian study (cancer patients),¹⁴ and 82.5% in another US study.¹⁵ A further study from South Australia, interviewing the general population concerning preferred places of death, showed that 70% wanted to die at home if hypothetically dying of a terminal illness.¹⁶ These findings reaffirm the importance of a familiar surrounding and the continuous presence of family and loved ones for the dying. In our study, there was a high congruence between wishes of patients and caregivers (92.5%). Nevertheless, more respondents who provided information on their preferred place of death wanted their relative to die in an institution (18.9%) than deceased themselves (5.9%). Two recent studies by Brazil et al. (13.6% vs. 4.7%)³ and Tang et al. (32.0% vs. 26.2%)¹⁷ came to similar findings. This could be due to the fact that after death took place in a hospital or hospice, relatives adapted their view and admitted that death in an institution was the right place for their loved one. The fact that the relatives would have to carry the whole burden of care if death took place at home could be a further reason for the relatives to prefer death in an institution.

It is further interesting to note that 38.8% of the deceased in our sample did not mention a wish regarding the place of death. This could in part be due to the fact that some people died quite suddenly or did not tell the respondent their wish. Another possibility would be that place of death is not as important as it is usually perceived for a substantial part of dying patients.

Altogether, about 40% (305) of the people who wished for a certain place of death could not die in the place they preferred. Approximately the same proportion has also been found in northern Italy (32.9%),¹⁴ England (44.1%⁸ and 43.8%¹⁸), and Australia (40.6%).⁷

In a recent German study, where bereaved were asked about why their deceased relative did not die in the place of his or her preference, 18 persons provided information about an unwanted hospital death. Medical reasons were adduced by 38.8%, hope for improvement until the very end by 33.3%, an acute deterioration in the medical condition by 22.2%, and deficits in nursing care by 5.6%.¹⁰ In a British study from Leeds, 16 patients were not discharged from the hospital to die at home, their place of preference. Reasons included an unexpected deterioration in health (43.9%), lack of community support (27.8%), being too unwell for transfer (11.1%), and active hospital treatment (5.6%).¹⁹

The result of our multivariable regression analysis showed that the probability to die in the place of preference was significantly higher when the deceased had a nonworking relative or lived together with a relative in one common household. It is very likely that decedents, who shared a common household with a relative prior to death or had a nonworking relative, were able to die in their place of preference (usually at home) because there was a person available to provide informal care. The fact that people are also more likely to die in their preferred place when living in a town with fewer than 20,000 inhabitants could be a sign for a higher potential of informal caregivers due to closer family ties and traditional obligations as compared with inhabitants of large towns. These results suggest that health care policies should intend to support and relieve the relatives from the burden of care as much as possible to fulfill the last wish of dying people for a certain place of death.

Limitations

The results of our study are limited in several ways. First of all, we cannot exclude serious selection bias due to the rather low response rate of 36.0%. The distribution of the underlying diseases does not fit the general epidemiologic situation in Germany. In 2008, in Rhineland-Palatinate 44.4% of deaths were of a cardiovascular disease and 26.2% of cancer,²⁰ whereas the corresponding figures for people suffering from these diseases in the last 4 weeks prior to death in our sample were 30.8% and 39.0%, respectively. It seems as if relatives of patients dying from chronically, progressive diseases were more prone to respond. This suggestion is supported when comparing the number of decedents suffering from a chronic disease in our sample with that of relatives who refused participation, but completed a short nonresponder questionnaire. In the sample of participants, 72.4% of the decedents suffered from a chronic disease, compared with 65.4% in the sample of nonparticipants (p<0.001 for Pearson's χ² test). In addition, it is possible that deceased who did not have any relatives at the time of death (widowed/single persons) are underrepresented in our sample due to a lack of potential responders. Because the probability to be widowed or single increases with increasing age, people who died in nursing homes could be underrepresented in our study.

A further limitation relates to recall bias due to the retrospective design of our study. The questioning of the bereaved was carried out as soon as ethically tolerable after death of the concerning relative. Thereby we tried to minimize possible systematic errors due to memory, without unnecessarily disturbing the bereaved relatives.

A further point of criticism is the fact that information on the preferred place of death of the deceased and preferred place of death of the relatives was provided by the latter. It is possible that the answers given by the respondents deviate from what the deceased themselves would have answered. In addition, it is not assured whether information on the preferred place of death reflects what the deceased and respondents felt prior to death or if answers have been adapted after death took place.

Questions aiming at the point in time when the preferred place of death was communicated to the relative and possible changes in the preference over time were also not considered in our study. This could pose a problem because two studies showed that preferences change toward inpatient death as death approaches. However, because the respondents answered the questionnaire after death took place, we presume that the given place was the last one stated by the dying.

Conclusion

The results of our study showed that one third of patients could not die in their preferred place. Because living together with a relative prior to death was the most important factor to facilitate dying in the preferred place (which was usually at home), relatives of dying people should be supported in their daily routine of providing informal care and support. At the time our survey was carried out, 52 local hospice home care services were available for the 4 million inhabitants of Rhineland-Palatinate.²¹ These services however relied mainly on the work of volunteers and only partly on additional palliative care nurses. The introduction of specialized palliative home care teams with 24-hour on-call duty, as intended by recent German legislature in 2007,²² should allow more people to die in their preferred place of death, by easing the burden of care of the relatives. Our findings thus strongly support the importance and exigency of the new German legislation, which hopefully will allow more people to die in their preferred place of death.

Footnotes

Acknowledgments

This work was supported by Fresenius Kabi Germany GmbH and the German Pharmaceutical Industry Association (BPI).

Author Disclosure Statement

No competing financial interests exist.

References

Patrick

, Curtis

, Engelberg

, Nielsen

, McCown

. Measuring and improving the quality of dying and death. Ann Intern Med, 2003; 139:410–415.

Bell

, Somogyi-Zalud

, Masaki

. Methodological review: Measured and reported congruence between preferred and actual place of death. Palliat Med, 2009; 23:482–490.

Brazil

, Howell

, Bedard

, Krueger

, Heidebrecht

. Preferences for place of care and place of death among informal caregivers of the terminally ill. Palliat Med, 2005; 19:492–499.

Tang

. When death is imminent: Where terminally ill patients with cancer prefer to die and why. Cancer Nurs, 2003; 26:245–251.

Choi

, Chae

, Lee

, Kim

, Lee

, Heo

, Kim

, Lee

, Hong

, Yun

. Factors influencing preferences for place of terminal care, of death among cancer patients, their families in Korea. Support Care Cancer, 2005; Aug;13:565–572.

Tiernan

, O'Connor

, O'Siorain

, Kearney

. A prospective study of preferred versus actual place of death among patients referred to a palliative care home-care service. Ir Med J, 2002; 95:232–5.

Agar

, Currow

, Shelby-James

, Plummer

, Sanderson

, Abernethy

. Preference for place of care and place of death in palliative care: Are these different questions? Palliat Med, 2008; 22:787–795.

Townsend

, Frank

, Fermont

, Dyer

, Karran

, Walgrove

, Piper

. Terminal cancer care and patients' preference for place of death: A prospective study. BMJ, 1990; 301:415–417.

Hinton

. Which patients with terminal cancer are admitted from home care? Palliat Med, 1994; 8:197–210.

10.

van Oorschot

, Hausmann

, Köhler

, Leppert

, Schweitzer

, Steinbach

, Anselm

. Patienten als Partner in der letzten Lebensphase. Erste Ergebnisse und Perspektiven eines Modellvorhabens. Bundesgesundheitsblatt, 2004; 47:992–999.

11.

van Oorschot

, Dreßel

, Erdmann

, Hausmann

, Hildenbrand

. Sterben und Tod in Thüringen. Zeitschrift für Palliativmedizin, 2002; 3:30–33.

12.

Escobar Pinzon

, Munster

, Fischbeck

, Unrath

, Claus

, Martini

et al. End-of-life care in Germany: Study design, methods and first results of the EPACS study (Establishment of Hospice and Palliative Care Services in Germany). BMC Palliat Care. 2010 Jul, 30; 9,1:16.

13.

Hsieh

, Huang

, Lai

, Lin

. Grief reactions in family caregivers of advanced cancer patients in Taiwan: Relationship to place of death. Cancer Nurs, 2007; 30:278–284.

14.

Beccaro

, Costantini

, Giorgi Rossi

, Miccinesi

, Grimaldi

, Bruzzi

. ISDOC Study Group. Actual and preferred place of death of cancer patients. Results from the Italian survey of the dying of cancer (ISDOC) J Epidemiol Community Health, 2006; 60:412–416.

15.

Leff

, Kaffenbarger

, Remsburg

. Prevalence, effectiveness, and predictors of planning the place of death among older persons followed in community-based long term care. J Am Geriatr Soc, 2000; 48:943–948.

16.

Foreman

, Hunt

, Luke

, Roder

. Factors predictive of preferred place of death in the general population of South Australia. Palliat Med, 2006; 20:447–453.

17.

Tang

, Liu

T-W

, Lai

M-S

, McCorkle

. Discrepancy in the preferences of place of death between terminally ill cancer patients and their primary family caregivers in Taiwan. Soc Sci Med, 2005; 61:1560–1566.

18.

Koffman

, Higginson

. Dying to be home? Preferred location of death of first-generation black Caribbean and native-born white patients in the United Kingdom. J Palliat Med, 2004; 7:628–636.

19.

Pearse

, Saxby

, Hicks

, Kite

. Patient choice regarding place of death. Palliat Med, 2005; 19:171–172.

20.

Statistical Office of Rhineland-Palatinate. Online-Datenbank des Statistischen Landesamtes Rheinland-Pfalz. 2011 April 10 www.infothek.statistik.rlp.de/lis/onlinedb/index.asp

21.

Sabatowski

, Radbruch

, Nauck

, Roß

, Zernikow

. Wegweiser Hospiz und Palliativmedizin Deutschland 2008/2009. Wuppertal: der hospiz verlag, 2008.

22.

Radbruch

, Voltz

. Was ist speziell an der spezialisierten Palliativversorgung? [What is special in specialised palliative care?] Schmerz, 2008; 22,1:7–8.