Palliative Medicine 2005; 19: 578
/586

Differences between terminally ill cancer patients who died

after euthanasia had been performed and terminally ill
cancer patients who did not request euthanasia
Jean-Jacques Georges, Bregje D Onwuteaka-Philipsen, Gerrit van der Wal Department of Public and
Occupational Health, Institute for Research in Extramural Medicine, VU Medical Center, Amsterdam, Agnes van
der Heide and Paul J van der Maas Department of Public Health, Erasmus MC, University Medical Center
Rotterdam, Rotterdam

Palliative care, directed at improving the quality of life of terminally ill patients, is generally
not aimed at any form of postponing or hastening death. It is possible that high quality
palliative care could prevent requests for euthanasia. However, empirical evidence on this
issue is scarce. In a national survey of end-of-life medical decisions in The Netherlands the
subject of care at the end of life has been addressed. Data on terminally ill cancer patients
who died after their request was granted and euthanasia had been performed were
compared with those of terminally ill cancer patients who did not request euthanasia. The
results show that the prevalence and severity of symptoms e.g., pain, feeling unwell,
nausea, was higher in patients who died after their request was granted and euthanasia had
been performed. No differences concerning the treatment of symptoms or the care provided
were found between the two groups. The results suggest that the practice of euthanasia is
mainly related to the patient’s suffering. Palliative Medicine 2005; 19: 578
/586

Key words: cancer; empirical evidence; euthanasia; palliative care; suffering; terminal illness

Introduction the available reports are mainly based on case descrip-

tions.10
Palliative care, directed at improving the quality of life of In order to obtain reliable overviews of euthanasia,
patients suffering from a terminal illness, essentially physician-assisted suicide and other medical decisions at
through prevention and relief of pain and other distres- the end-of-life, quantitative studies of the prevalence and
sing symptoms, is neither intended to hasten nor to major characteristics of these practices have been carried
out in the Netherlands since 1990.7,11,12
postpone death.1,2 The relationship between euthanasia
In the most recent study, specific attention was paid to
and palliative care is subject to debate. One frequently
the care that was provided at the end-of-life and its
expressed opinion is that with good palliative care,
relationship with EAS.12
requests for euthanasia and assisted suicide (EAS) can
This paper focuses on possible determinants of EAS,
be prevented, and therefore it should never be necessary presenting data on symptoms and their treatment, and
to perform EAS.3
5 However, the wish to die and end-of-life care in two groups of patients, i.e., terminally
requests for EAS do occur in terminally ill patients,6,7 ill cancer patients whose requests had been granted and
and are also reported in terminally ill patients who are euthanasia had been performed, and terminally ill cancer
receiving palliative care.8,9 Multiple factors contribute to patients who did not request euthanasia.
the development of a patient’s request for euthanasia.
Many questions are asked about whether and how
euthanasia contributes to the quality of the end-of-life,
and also about ways in which requests for euthanasia
Methods
could be prevented or altered by the provision of high Design
quality care. However, empirical evidence is scarce, and In the third study on end-of-life decision-making prac-
tices and attitudes in the Netherlands, random samples of
physicians, stratified according to specialty, were inter-
Address for correspondence: Jean-Jacques Georges, VU Uni- viewed about end-of-life decision-making. The interview
versity Medical Centre, Department of Public and Occupa- was the starting point for a prospective study on end-of-
tional Health, Institute for Research in Extramural Medicine,
Van der Boechorststraat 7, 1081 BT Amsterdam, The Nether- life care and its relationship with end-of-life decisions
lands. concerning euthanasia. During the interviews incurable
E-mail: j.georges@vumc.nl cancer patients with a life expectancy of less than three
# 2005 Edward Arnold (Publishers) Ltd 10.1191/0269216305pm1069oa
 Patients who died after euthanasia 579

months and more than one week were included, and including euthanasia. Euthanasia was defined as the
subsequently monitored by means of monthly question- administration of drugs with the explicit intention of
naires filled in by the physician until the patient’s death. ending the patient’s life at his or her explicit request.
The final questionnaire was completed shortly after the Physicians who were able to include a patient in the
patient’s death. prospective study were interviewed about the patient’s
The study design ensured absolute anonymity for all current situation and were asked, at the end of the
participants. Further details about the methods used in interview, to participate in the study by completing a
the study have been described elsewhere.12,13 short questionnaire each month until the patient’s death.
The questionnaire and the interview schedule were
Population developed specifically for this part of the study, and
The specialists included in this part of the study were reviewed by three physicians with expertise in research on
general practitioners, nursing home physicians, surgeons, terminally ill patients. The questionnaire included similar
internists, pulmonologists and neurologists. Samples were questions to those that were asked during the interview,
randomly taken per stratum from the database of the and was designed in such a way that it would not take
professional registries. Of the 434 physicians in the more than 15 min to complete. The questions mainly
sample, 87% agreed to be interviewed (n /376: 174 concerned the patient’s medical situation, the patient’s
medical specialists, 125 general practitioners, 77 nursing symptoms at the time of the physician’s last visit, the aim
home physicians). Of the 376 physicians who participated of the treatment, (para-) medical disciplines involved in
in the study, at the moment of the interview, 120 were the treatment and possible wishes of the patient concern-
treating a patient who met the criteria for inclusion in the ing end-of-life decisions. The final questionnaire had to
prospective study. These criteria were: a diagnosis of be completed shortly after the death of the patient. This
cancer, treatment no longer directed at cure, life expec- questionnaire also contained questions about the physi-
tancy probably no longer than 3 months but longer than cian’s opinion of the (in)adequacy of care in several
1 week, and probably treated by the respondent until domains.
death. Thirty-five physicians were unwilling to partici- Data concerning the first group, e.g., terminally ill
pate or dropped out during the follow-up period, for 85 cancer patients who did not request euthanasia before
of these 120 patients the follow-up was completed until they died, refer to the situation of the patients on the last
their death. Only 7 of these patients had requested occasion the physician had visited them before they died.
euthanasia. This limited number made it impossible to This period had a median length of 1 day (0
/47). The
compare patients who did and did not request euthanasia data concerning the second group, e.g., terminally ill
within the prospective study. For this purpose use was cancer patients who died after euthanasia had been
made of the retrospective descriptions that the inter- performed, refer to the moment when the physician
viewed physicians gave of their most recent case of decided to grant the patient’s request (number of days
euthanasia (if any). Of the interviewed physicians, 119 before death unknown).
were able to describe a case of euthanasia. In order to
make the group as comparable as possible to the group of Analysis
78 patients in the prospective study who had not Descriptive statistical analysis was performed on the
requested euthanasia, we selected only the most recent characteristics of the physicians, the patients and the
cases of euthanasia that had been performed for a patient patients’ symptoms. Symptoms were assessed on a five-
with cancer. We also omitted the cases described by point scale (1 /absence of symptoms to 5/presence of
nursing home physicians in the prospective study because symptoms with maximum severity). The respondents
they had not been asked to describe their most recent were instructed to assess the severity of the symptoms,
case of euthanasia. This resulted in a comparison despite possible treatment of these symptoms. Scores of 2
between 106 euthanasia cases (that occurred between and 3 were defined as moderate and scores of 4 and 5 as
1996 and 2002; median 2001) and 64 cases in which the severe. Independent sample t-tests were performed in
patient did not request euthanasia (study period: 2001
/ order to assess the differences in the number of symptoms
2003). Figure 1 shows this process of selection. between the two groups. Wilcoxon signed ranks tests were
used to asses the differences in age of the physicians and
Measurement instruments the differences in age of the patients, Mann
/Whitney
Experienced physicians, who had been extensively trained tests to assess differences in characteristics of physicians
to use the structured questionnaires, carried out the and differences in the diagnosis of patients, and
interviews. The interviews focussed on experiences and Kruskal
/Wallis tests to assess the differences in the
attitudes with regard to end-of-life decision-making. The prevalence of symptoms. Because physicians could have
physicians were asked to describe their most recent cases described only a case of euthanasia or only a case in
in which various end-of-life decisions had been made, which the patient had not requested euthanasia, or both
580 J-J Georges et al.

Physicians participating in medical end-of-life

decisions study
n=376

Physicians asked about most Physicians with a patient who met the
recent case of euthanasia, if any inclusion criteria for the prospective study
(n=299) (n=120)
(GP n=125, CS n=174) (GP n=47, NHP n=21, CS n=52)

Dropped out (n=35)

Cases of euthanasia performed
(n=119)
Physicians who participated in
(GP n= 68, CS n=51)
follow-up until patient’s death
(n=85)
(GP n=34, NHP n=14, CS n=37)
Exclusion of
patients with other
terminal illness than Patients who had requested
cancer (n=13) euthanasia (n=7)

Terminally ill cancer patients

who died after euthanasia was Terminally ill cancer patients
performed who did not request euthanasia
(n=78)
(n=106)
(GP n=60, CS n=46)
Exclusion of
patients treated by a
NHP (n=14)

Terminally ill cancer patients

who did not request euthanasia
(n=64)
(GP n=29, CS n=35)

Figure 1 Selection of cases. GP, general practitioner; NHP, nursing home physician; CS, clinical specialist.

situations, these three groups were compared to identify Physicians who described the situation of a patient who
possible differences in physician characteristics. did not request euthanasia were more often clinical
specialists and were younger. Furthermore, they appeared
to believe more in the probable influence of palliative
care on euthanasia (requests).
Results
Physician characteristics Patient characteristics
Table 1 describes the characteristics of the physicians who Table 2 describes the characteristics of the patients who
participated in the study. The various cases included in died after euthanasia had been performed and of the
this study were described by a total of 144 physicians, 70 patients who did not request euthanasia before they died.
of whom only described a case of euthanasia, 28 only There were no differences between patients of the two
described a case in which the patient had not requested groups with regard to age or gender. The prevalence of
euthanasia and 36 described both situations. the various primary cancer sites was very similar between
 Patients who died after euthanasia 581

Table 1 Characteristics of physicians who described only a case of euthanasia, of physicians who only described the case of
a patient who did not request euthanasia, and of physicians who described both situations

Physicians of patients Physicians of patients Physicians who

who died after who did not request described both
euthanasia (n /70) euthanasia (n /28) situations (n /36)

n (%) n (%) n (%)

Specialty* General practitioner 41 (59) 10 (36) 19 (53)

Clinical specialist 29 (41) 18 (64) 17 (47)
Age 25
/40 years 4 (6) 6 (21) 2 (6)
41
/50 years 33 (49) 11 (39) 20 (57)
51
/60 years 26 (39) 10 (36) 13 (37)
61
/70 years 4 (6) 1 (4) 0
Gender Male 63 (90) 21 (75) 27 (75)
Female 7 (10) 7 (25) 9 (25)
Have a religious belief 23 (35) 11 (39) 11 (31)
Experience with euthanasia
Have ever had explicit requests for 61 (87) 23 (82) 30 (83)
euthanasia or physician-assisted suicide (PAS)
Have ever performed euthanasia or PAS 43 (70) 16 (70) 22 (73)
Opinion about end-of-life care. (Totally) agree with
the following statements:
In the Netherlands much attention has been 23 (33) 14 (50) 14 (39)
paid to the regulation of euthanasia and too little
to other possibilities to reduce suffering at the
end-of-life
By adequate treatment of pain and adequate 13 (14) 13 (46) 5 (14)
end-of-life care euthanasia is made unnecessary*
The quality of palliative care in the Netherlands 23 (33) 11 (39) 9 (25)
is suboptimal
Development in attitudes during the past five years
Became more restrictive 16 (23) 7 (25) 8 (22)
Became less restrictive 8 (12) 2 (7) 1 (3)
No change in attitude 45 (65) 19 (69) 27 (75)

*PB/0.05 (differences between physicians of patients who died after euthanasia and physicians of patients who did not
request euthanasia).

Table 2 Characteristics of patients who died after euthanasia and of patients who did not request euthanasia

Patients who died after euthanasia Patients who did not request euthanasia
(n /106) (n /64)

n (%) n (%)

Age B/45 years 18 (17) 4 (6)

45
/60 years 26 (25) 16 (25)
61
/75 years 45 (43) 32 (50)
/75 years 17 (16) 12 (19)
Gender Male 58 (55) 37 (58)
Female 48 (45) 27 (42)
Cancer sitea Respiratory 25 (24) 19 (30)
Gastrointestinal 33 (31) 21 (33)
Breast 11 (10) 5 (8)
Gynaecological 9 (9) 3 (5)
Genitourinary 11 (10) 5 (8)
Haematological 6 (6) 7 (11)
Head
/neck 3 (3) 2 (3)
Primary unknown* 0 (0) 3 (5)
Other 6 (6) 1 (1)

a
Two patients with two primary sites.
*PB/0.05.
582 J-J Georges et al.

both groups, with the exception of unknown primary especially depressive feelings, were clearly less often
tumour site, which was more prevalent in patients who treated than physical symptoms in both groups. With
had requested euthanasia. regard to the treatment of confusion, it is difficult to
make a reliable comparison between the two groups
Symptoms because of the low prevalence rates.
Table 3 shows the prevalence and mean number of severe Table 5 presents data on the adequacy of the care
symptoms in the two study groups. provided. In order to assess whether the care provided
The prevalence of some symptoms, such as fatigue, was adequate, the physicians were asked if there was any
feeling unwell and loss of appetite, was noticeably high in shortage or deficiency in several areas of the care that
both groups. was needed by the patients. Table 5 shows that in only a
Patients who died after euthanasia had been performed few cases in both groups the physicians estimated that
more often felt unwell, and suffered from severe pain, the care provided was inadequate. Differences between
nausea, vomiting and coughing. They had fewer pro- the two groups were negligible. When in a certain area the
blems with consciousness and suffered less often from care was described as inadequate, the physicians were
confusion than patients who did not request euthanasia. asked whether this could have encouraged the patient in
The prevalence of anxiety and depressive feelings was low some way to make a request for euthanasia. All
in both groups. physicians stated that this was never the case.
The average number of severe physical symptoms was
higher in patients who died after euthanasia had been
performed, but the average number of psychosocial Discussion
symptoms was higher in patients who did not request
euthanasia before they died. Patients who died after euthanasia had been performed
more often suffered from severe symptoms such as pain,
Treatment of symptoms and adequacy of care vomiting, nausea and coughing, and more often felt
Table 4 describes the treatment of certain severe symp- severely unwell. Patients who did not request for eu-
toms, e.g., pain, nausea, dyspnoea, depressive feelings thanasia before they died were more often confused and
and confusion, in the two groups. In both groups the more often had problems with consciousness. Patients
prevalence of treatment of these symptoms appeared to who died after euthanasia had been performed also had a
be similar. Physical symptoms, especially pain, were higher average number of severe physical symptoms than
usually treated in both groups. Psychological symptoms, patients who did not request euthanasia. However,

Table 3 Prevalence and mean number of severe symptoms in patients who died after euthanasia and patients who did not
request euthanasia

Patients who died after euthanasia Patients who did not request euthanasia
(n /106) (n /64)

n (%) n (%)

Physical symptoms
Feeling unwell* 92 (87) 46 (71)
Fatigue 86 (81) 52 (81)
Loss of appetite 80 (75) 51 (80)
Pain** 58 (55) 13 (20)
Nausea** 44 (42) 7 (11)
Dyspnoea 30 (28) 17 (27)
Vomiting* 27 (25) 4 (6)
Coughing* 24 (23) 5 (7)
Constipation 21 (20) 11 (17)
Pressure ulcer 3 (3) 1 (2)
Number of severe physical symptoms, 4.4 (1.72) [0
/8] 3.2 (1.56) [0
/8]
mean (SD) [range]**

Psychosocial symptoms
Anxiety 17 (16) 12 (19)
Depressive feelings 11 (10) 5 (8)
Confusion* 2 (2) 9 (14)
Unclear consciousness** 0 (0) 21 (33)
Number of severe psychosocial symptoms, 0.28 (0.62) [0
/3] 0.73 (0.82) [0
/3]
mean (SD) [range]**

*PB/0.05; **PB/0.001.
 Patients who died after euthanasia 583

Table 4 Treatment of severe symptoms in patients who died after euthanasia and patients who did not request euthanasia

Patients who died after euthanasia Patients who did not request euthanasia
(n /106) (n /64)

Prevalence Treatment Prevalence Treatment

na n (%) na n (%)

Pain 58 58 (100) 13 13 (100)

Nausea 44 36 (80) 7 7 (100)
Dyspnoea 30 19 (63) 17 14 (82)
Depressive feelings 11 3 (27) 5 1 (20)
Confusion 2 0 (0) 9 6 (85)

a
Number of patients who had the symptom in a severe form.
The percentage is the ratio of patients with the symptom in a severe form who were treated for this symptom.

patients in the latter group had more severe psychosocial by the responding physicians. This change in approach
symptoms. With regard to the treatment of symptoms, in might have introduced bias. The retrospective design in
both groups the symptoms were treated with the same the euthanasia group could have caused recall bias.
pattern of frequency, and in both groups physical However, a request for EAS is an exceptional occurrence,
symptoms were more often treated than psychosocial and will therefore be relatively easy to recall.18 The large
symptoms. With regard to the care that was provided, in majority of respondents described a case of no more than
the majority of cases the physicians estimated that it was 2 years ago. Moreover, the questions asked in the two
adequate. groups probably refer to somewhat different points of
Research on palliative care is ethically and methodo- time in the illness trajectory. The data on patients who
logically challenging, and this partly explains the paucity died after euthanasia had been performed refer to the
of studies that focus on explaining how euthanasia period during which the physician was deciding about
requests could be prevented by providing high quality the request, which is generally earlier in the course of the
care.14
18 The present study is one of the few studies that illness than the time of the physician’s last visit before
address issues related to end-of-life care and euthanasia. the death of a patient who did not request euthanasia. It
However, the study design has several limitations that is therefore probable that the symptoms would have been
might hamper the possibilities of drawing very firm more severe in patients who died of euthanasia if in this
conclusions. It is important to mention that, although group it had also been possible to use data referring to
the original design was intended to monitor a group of the time of the physician’s last visit before the death of
terminally ill cancer patients and compare within this the patient.19,20 Finally, a limitation of this study,
group the patients who did and did not request eutha- regardless of the change in approach, is the fact that
nasia, this was not possible because only seven patients the data were provided by the physicians, making it
requested euthanasia. Therefore, we decided to compare impossible to take the experience of the patients into
the patients who did not request euthanasia with the account. It is possible that the physicians’ responses are
patients in cases of euthanasia described retrospectively influenced by the decision they took. It might have been

Table 5 Care provided for patients who died after euthanasia and patients who did not request euthanasia

Patients who died after euthanasia Patients who did not request euthanasia
(n /106) (n /30)a

n (%) n (%)

Adequate care 101 (95) 27 (90)

Inadequate care 5 (5) 3 (10)

Areas of inadequate care n n

Medical 0 1
Nursing 2 1
Psychosocial 1 0
Spiritual 1 0
Domestic, household 3 0
Other care facilities 0 1

a
When patients, apart from not requesting euthanasia, had made no request concerning other end-of-life decisions (e.g.,
forgoing treatment, terminal sedation) no questions were asked about the adequacy of care.
584 J-J Georges et al.

difficult for physicians in both groups to acknowledge The burden of pain and physical discomfort has been
inadequacies in care, especially in the domains of care for described in other populations of terminally ill patients
which they bear the greatest responsibility. as reasons for a request to hasten death,15,27 and requests
Some interesting results emerged from this study. for euthanasia are not only made because of physical
Compared to physicians who reported the case of a symptoms. A wide range of reasons seems to contribute
patient who died after euthanasia, physicians who only to the development of a request for euthanasia. Among
reported the case of a patient who did not request these, relationships with others, (concerns about) loss of
euthanasia appeared to be more sceptical about the autonomy, loss of control of bodily functions, concerns
quality of palliative care in the Netherlands and more about future distress, becoming a burden on others,
often believed that euthanasia could be prevented by feeling of meaninglessness, dependency, wish to control
providing high quality of care. However, physicians’ the time of death, or loss of dignity are the most
experiences with requests for and the performance of frequently described reasons.8,22,28
euthanasia are very similar. Therefore it is difficult to The results show that in patients who did not request
really understand in which way the differences in euthanasia the number of psychosocial symptoms was
physician characteristics influence the development of a higher than in patients who died after euthanasia had
patient’s request for euthanasia. It is possible that been performed. Considering the fact that psychosocial
physicians who only reported the case of a patient who symptoms e.g., anxiety, feeling of hopelessness, have been
did not request euthanasia provide high quality care, so described as factors contributing to the development of
that their patients request euthanasia less frequently or at requests for euthanasia, we may be surprised by these
a later time in the illness trajectory, or that their patients results.29
31 However, the high prevalence of problems
are hesitant to request euthanasia because they are aware with consciousness and confusion in patients who did not
of the physicians reluctance. In any case, we did not find request euthanasia could explain why, in this group, the
that any one group of physicians reported that their number of psychosocial symptoms is more important, as
patients never made a request for euthanasia. these two symptoms make it less likely, if not impossible,
The fact that inadequate care or lack of attention to to meet at least one of the requirements for prudent
the treatment of some symptoms, e.g., pain, nausea, euthanasia practice that exist for euthanasia, namely that
dyspnoea, depressive feelings, confusion, were not more the request should be well considered. Furthermore, the
prevalent in either of the two groups suggests that lower prevalence of psychosocial symptoms found in
inadequate care did not contribute to the development patients who died after euthanasia had been performed is
of a request for euthanasia in either of the groups. in line with the results of studies on physicians’ decisions
Because of their involvement in the situation, physicians’ about requests for euthanasia, in which it was found that
objectivity could be at stake when assessing the adequacy patients whose requests were not granted had more
of the care provided. When judging the adequacy of care, mental health problems, and especially depression, than
physicians perhaps only refer to the situation of the patients who died after euthanasia had been per-
patient and did not take in to account whether other formed.27,32
kinds of care could have been offered to the patient e.g., The results indicate that physical symptoms were more
specialist palliative care. However, these findings are often treated than psychosocial symptoms, which has
supported by earlier data,21 and are in line with results also been observed in other populations of terminally ill
of other studies describing lack of care as a minor reason patients.31 It is worth noting that not only the prevalence
for patients to request for euthanasia and occurring only of depressive feelings, but also the frequency of treatment
in a minority of patients.8,9,22,23 Conversely, more atten- for depressive feelings was similar in both groups.
tion has to be paid to this issue, because unmet care Depressive feelings did not seem to be an important
needs have been described as contributing to the like- discriminating factor between the two groups, and there
lihood of a patient to consider euthanasia.15 Only a few was no clear influence of depressive feelings on the
studies have focussed on patient needs at the end of life development of requests for euthanasia.
from the perspective of the patients.24
26 In conclusion, it appears that the number of symptoms
It was found that there was a higher prevalence of and the severity of certain symptoms, such as pain,
symptoms, such as pain and nausea, and a higher average nausea, coughing and vomiting, clarify why the patients
of severe symptoms in patients who died after euthanasia in the study population requested euthanasia. These
had been performed than in patients who did not request symptoms seem to make an important contribution to
euthanasia. This could indicate that in the study popula- the unbearable suffering that is a requirement for
tion (a request for) euthanasia is mainly a result of the granting a request for euthanasia. No evidence has
patient’s suffering. However when interpreting these been found to support the assumption that a lack of
results we have to keep in mind that unbearable suffering care and inadequate treatment of symptoms contribute
is a prerequisite for a euthanasia request to be granted. to the development of requests for euthanasia. Until now,
 Patients who died after euthanasia 585

only a few studies have addressed the association between lized inpatient palliative care. J Pain Symptom Manage
end-of-life care and requests for euthanasia.9,15,28 2004; 27: 44
/52.
Further studies are needed to investigate the contribution 9 Seale C, Addington-Hall J. Euthanasia: the role of good
of palliative care in terminally ill patients who wish to care. Soc Sci Med 1995; 40: 581
/87.
hasten death and develop a request for euthanasia. 10 Clark D, ten Have HAMJ, Janssens RJPA. Conceptual
tensions in European palliative care. In ten Have HAMJ,
Taking into account the limitations of this study, future
Clark D eds. The ethics of palliative care. European
efforts should concentrate on describing the illness
perspectives, first edition. Buckingham, Philadelphia:
experience of terminally ill patients in order to gain Open University Press, 2002: 52
/65.
insight into the meaning of their request for euthanasia. 11 van der Maas PJ, van Delden JJM, Pijnenborg L,
Furthermore, more specific attention should be paid to Looman CWN. Euthanasia and other medical decisions
comparing terminally ill patients who request euthanasia concerning the end of life. Lancet 1991; 338: 669
/74.
with terminally ill patients who do not. A prospective 12 Onwuteaka-Philpsen BD, van der Heide A, Koper D,
study design seems to be the most appropriate for this et al . Euthanasia and other end-of-life decisions in the
purpose, among others because attention can also be Netherlands in 1990, 1995, and 2001. Lancet 2003; 362:
paid to requests for euthanasia that do not result in 395
/99.
euthanasia.16,33
35 13 Georges JJ, Onwuteaka-Philipsen BD, van der Heide A,
van der Wal G, van der Maas PJ. Symptoms, treatment
and ‘‘dying peacefully’’ in terminally ill cancer patients: a
prospective study. Support Care Cancer 2005; 13: 160
/
Acknowledgements 68.
We thank all the physicians who participated in the study. 14 Bruera E. Need for increased evidence in palliative care.
Eur J Palliat Care 2004; 7: 160.
Statement about the originality of the data and about the 15 Emanuel EJ, Fairclough DL, Emanuel LL. Attitudes and
absence of conflicts of interest: desires related to euthanasia and physician-assisted
suicide among terminally ill patients and their caregivers.
The results presented in this manuscript are based on JAMA 2000; 284: 2460
/68.
original study data. The authors of this manuscript had 16 Chochinov HM, Tataryn D, Chinch JJ, Dudgeon D. Will
no financial relationships with any commercial compa- to live in the terminally ill. Lancet 1999; 354: 816
/19.
17 Agrawal M, Danis M. End-of-life care for terminally ill
nies of products subject of investigation in the manu-
participants in clinical research. J Palliat Med 2002; 5:
script. There is no potential conflict of interest. 729
/37.
18 van der Vaart W. Inquiring into the past: data quality of
responses to retrospective questions. Veenendaal: Univer-
References sity Press, 1996.
19 Kaasa S, Loge JH. Quality-of-life assessment in pallia-
1 Sepúlveda C, Marlin A, Yoshida T, Ullrich A. Palliative tive care. Lancet Oncol 2002; 3: 175
/82.
care: the World Health Organization’s global perspective. 20 Hwang SS, Chang VT, Fairclough DL, Cogswell J,
J Pain Symptom Manage 2002; 24: 91
/96. Kasimis B. Longitudinal quality of life in advanced
2 World Health Organization. WHO definition of pallia- cancer patients: pilots study results from a VA medical
tive care. 2004. Retrieved 18 September 2004, from: cancer center. J Pain Symptom Manage 2003; 25: 225
/
http://www.who.int/cancer/palliative/definition/en/ 35.
3 Gordijn B, Janssens RJPA. The prevention of euthanasia 21 van der Wal G, van der Heide A, Onwuteaka-Philipsen
through palliative care: new developments in the Nether- BD, van der Maas PJ. Medical decisions at the end of life.
lands. Patient Educ Couns 2000; 41: 35
/46. Practice of national review committees [Medische be-
4 Janssens RJPA, ten Have HAMJ, Zylicz Z. Hospice and sluitvorming aan het einde van het leven. De praktijk en de
euthanasia in the Netherlands: an ethical point of view. J toetsingsprocedure ]. Utrecht: de Tijdstroom, 2003.
Med Ethics 1999; 25: 408
/12. 22 Back AL, Wallace JI, Starks HE, Pearlman RA.
5 ten Have HAMJ. Euthanasia: moral paradoxes. Palliat Physician-assisted suicide and euthanasia in Washington
Med 2001; 15: 505
/11. State. Patient requests and physician responses. JAMA
6 Chochinov HM, Wilson KG, Enns M, et al . Desire for 1996; 275: 919
/25.
death in the terminally ill. Am J Psychiatry 1995; 152: 23 Seale C, Addington-Hall J. Euthanasia: why people want
1185
/91. to die earlier. Soc Sci Med 1994; 39: 647
/54.
7 van der Maas PJ, van der Wal G, Haverkate I, et al . 24 Curtis JR, Wenrich MD, Carline JD, Shannon SE,
Euthanasia, physician-assisted suicide, and other medi- Ambrozy DM, Ramsey PG. Patients’ perspectives on
cal practices involving the end of life in the Netherlands, physician skill in end-of-life care: differences between
1990
/1995. N Engl J Med 1996; 335: 1699
/705. patients with COPD, cancer, and AIDS. Chest 2002; 122:
8 Morita T, Sakaguchi Y, Hirai K, Tsuneto S, Shima Y. 356
/62.
Desire for death and requests to hasten death of 25 Singer PA, Martin DK, Kelner M. Quality end-of-life
Japanese terminally ill cancer patients receiving specia- care: patients’ perspectives. JAMA 1999; 281: 163
/68.
586 J-J Georges et al.

26 Steinhauser KE, Christakis NA, Clipp EC, McNelly M, 31 Blocks SD, Billings JA. Patient requests for euthanasia
McIntyre L, Tulsky JA. Factors considered important at and assisted suicide in terminal illness. The role of the
the end of life by patients, family, physicians, and other psychiatrist. Psychosomatics 1995; 36: 445
/57.
care providers. JAMA 2000; 284: 2476
/82. 32 Haverkate I, Onwuteaka-Philipsen BD, van der Heide A,
27 Meier DE, Emmons CA, Litke A, Wallenstein S, Kostense PJ, van der Wal G, van der Maas PJ. Refused
Morrison RS. Characteristics of patients requesting and granted requests for euthanasia and assisted suicide
and receiving physician-assisted death. Arch Intern in the Netherlands: interview study with structured
Med 2003; 163: 1537
/42. questionnaire. BMJ 2000; 321: 865
/66.
28 Chin AE, Hedberg K, Higginson GK, Fleming DW. 33 Mak YYW, Elwyn G. Use of hermeneutic research in
Legalized physician-assisted suicide in Oregon
/ the first understanding the meaning of desire for euthanasia.
year’s experience. N Engl J Med 1999; 340: 577
/83. Palliat Med 2003; 17: 395
/402.
29 Breitbart W, Rosenfeld B, Pessin H, et al . Depression, 34 Mak YYW, Elwyn G, Finlay IG. Patients’ voices are
hopelessness and desire for hastened death in terminally needed in debates on euthanasia. BMJ 2003; 327: 213
/
patients with cancer. JAMA 2000; 284: 2907
/11. 15.
30 Breitbart W, Rosenfeld B. Physician-assisted suicide: the 35 Willems DL, van der Wal G. Wetenschappelijk onder-
influence of psychosocial issues. Cancer Control 1999; 6: zoek in de palliatieve zorg. Tijdschr Gezondheidswe-
146
/61. tenschappen 1999; 77: 466
/71.