Download Suicidality in terminally ill Japanese patients with cancer

183
Suicidality in Terminally Ill Japanese Patients
with Cancer
Prevalence, Patient Perceptions, Contributing Factors, and Longitudinal Changes
Tatsuo Akechi, M.D., Ph.D.1,2
Toru Okuyama, M.D., Ph.D.1
Yuriko Sugawara, M.D.1
Tomohito Nakano, M.D.3
Yasuo Shima, M.D.4
Yosuke Uchitomi, M.D., Ph.D.1,2
1
Division of Psycho-Oncology, National Cancer
Center Research Institute East, Kashiwa, Japan.
2
Division of Psychiatry, National Cancer Center
Hospital East, Kashiwa, Japan.
3
Division of Psychiatry, National Cancer Center
Hospital, Tokyo, Japan.
4
Palliative Care Unit, National Cancer Center Hospital East, Kashiwa, Japan.
Supported in part by a Grant-in-Aid for Cancer
Research (11-2) from the Japanese Ministry of
Labor, Health and Welfare and by a Grant-in-Aid
for Young Scientists (B) from the Japanese Ministry of Education, Culture, Sports, Science and
Technology.
The Authors thank the patients involved in the
current study, who gave their time so willingly.
They also thank Dr. Tatsuya Morita of the Seirei
Mikatabara Hospital (Hamamatsu, Japan) for his
helpful comments and Yurie Sugihara and Ryoko
Katayama for their research assistance.
Address for reprints: Yosuke Uchitomi, M.D., Ph.D.,
Division of Psycho-Oncology, National Cancer Center Research Institute East, 6-5-1 Kashiwanoha,
Kashiwa 277-8577, Japan; Fax: (011) 81 471 34
7026; E-mail: [email protected]
Received June 9, 2003; revision received September 4, 2003; accepted September 15, 2003.
© 2003 American Cancer Society
DOI 10.1002/cncr.11890
BACKGROUND. The risk of suicide is higher in patients with cancer than in the
general population, making end-of-life care of suicidal terminal patients with
cancer critical. To identify factors and longitudinal changes associated with suicidality among terminally ill Japanese patients with cancer, a prospective cohort
study was performed.
METHODS. Consecutive outpatients with cancer who registered with a palliative
care unit participated. Structured interviews (e.g., Structured Clinical Interview for
DSM-III-R [SCID]) were conducted to assess patient suicidal ideation (Ideation)
and interest in requesting euthanasia (Interest) as main outcome measures of
suicidality. Possible correlated factors also were investigated. The authors analyzed
the data from 140 terminally ill patients with cancer at initial study participation
(baseline) whose subsequent survival time was ⬍ 6 months. Of these 140 patients,
57 (40.7%) completed the follow-up assessment after admission to the unit.
RESULTS. At baseline, 8.6% of the patients had Ideation and 5.0% had Interest.
Self-reported anxiety and depression was significantly associated with Ideation
(P⫽ 0.003). Changes in Ideation and Interest occurred in 38.6% and 15.8% of the
patients, respectively. Ideation was more likely to change than Interest (P ⫽ 0.006).
The current study did not identify factors that predict changes and occurrences of
suicidal ideation and interest in requesting euthanasia.
CONCLUSIONS. Suicidality can change even in terminally ill patients. End-of-life
care that focuses on the psychologic distress of dying individuals may be a way of
preventing suicide. Cancer 2004;100:183–91. © 2003 American Cancer Society.
KEYWORDS: suicide, terminally ill, terminal care, suffering, culture.
P
revious epidemiologic studies indicated that the risk of suicide is
higher among patients with cancer than in the general population
and that one of the important risk factors is advanced disease.1 In
addition, a previous study conducted in the United Kingdom found
that in a 5-year period, patient suicide and suicide attempts occurred
in greater than one-third and two-thirds of hospices, respectively.2
Furthermore, many people, including health professionals, consider
suicide a rational and justifiable choice in the face of the severe
physical and emotional pain that can afflict patients with cancer. This
opinion is reinforced by the general perception of cancer as a disease
characterized by severe suffering, whose inevitable outcome is death,
not cure.3 In contrast, only two psychologic autopsy studies have
actually investigated suicide victims among patients with cancer.4,5
These studies suggested that several physical, psychologic, and existential distress factors (e.g., pain, impairment of physical functioning,
depression, loss of independence, and loss of autonomy) are associ-
184
CANCER January 1, 2004 / Volume 100 / Number 1
ated with suicide in patients with cancer. It recently
was reported that ⬎ 30,000 people in Japan commit
suicide annually, one of the highest suicide rates in the
world. Health problems were cited as the most common risk factor.6 Although suicidality in terminally ill
patients with cancer is critically important in clinical
oncology practice, very few studies have addressed
these issues.
It is not rare for physicians in clinical practice in
Western countries and in Japan to receive requests for
early death from patients.7,8 However, these requests
confront the medical staff with a dilemma. Several
recent studies have demonstrated that the prevalence
of the “desire for death” in terminally ill patients with
cancer is not particularly low (range, 8.5–22.2%).9 –14
These studies demonstrated that pain, weakness, concerns with physical symptoms, especially pain, loss of
control, psychologic distress, such as depression,
hopelessness, demoralization, and perception of being a burden to others, and social support factors are
significantly associated with the desire for death. Cognitive dysfunction and delirium also have been suggested as factors that may increase the likelihood of
suicide.15,16 In regard to cultural differences in attitudes to euthanasia/physician-assisted suicide (PAS),
Western surveys have indicated that ⬎ 50% of the
general population support legalization of euthanasia
or PAS for incurable patients with cancer,15 whereas a
Japanese survey found that only 13% of the general
population endorse euthanasia.17 In Japan, 2.0 –5.0%
of oncologists and/or palliative care physicians have
reported actual experience with euthanasia for terminally ill patients with cancer despite the illegality of
such medical intervention.8,18
Although adequate care for suicidal terminal patients with cancer is critical, very few studies have
addressed the form that the care should take. Identification of patients at increased risk of suicide and
recognition of the underlying factors in patients with
cancer with suicidality are important steps in suicide
prevention and allow appropriate intervention for suicidal patients with cancer. However, to our knowledge, no study has addressed longitudinal changes in,
or predictors of, suicidality, such as suicidal ideation
and interest in requesting euthanasia, in terminally ill
patients with cancer.
The goals of the current study were to replicate
previous findings obtained in Western studies9 –12,14
and to answer the following six questions concerning
suicidality among terminally ill Japanese patients with
cancer with the aim of establishing a useful strategy
for suicide prevention: 1) How prevalent are suicidal
ideation and interest in requesting euthanasia? 2)
What are patient perceptions regarding the reasons for
suicidal ideations? 3) What factors are associated with
suicidal ideation? 4) What factors contribute to interest in requesting euthanasia among patients with suicidal ideation? 5) Are there longitudinal changes in
suicidal ideation and interest in requesting euthanasia
and do any factors predict these changes? 6) What
factors predict occurrences of suicidal ideation and
interest in requesting euthanasia?
MATERIALS AND METHODS
Patients
Consecutive patients who registered with the Palliative Care Unit (PCU), National Cancer Center Hospital
East (Kashiwa, Japan) between October 1997 and November 1999 and met the inclusion criteria were recruited (Table 1). The PCU has an outpatient service
and a 25-bed inpatient service. In principle, patients
register with the outpatient service and are admitted
to the inpatient service mainly for symptom management, terminal care, and respite care for the family.
The most common reasons for patient registration
and admission are physical symptoms. Eligibility criteria for enrollment in the current study included the
following: age ⱖ 18 years, new registration with the
PCU, no current curative cancer treatment, awareness
of the cancer diagnosis, absence of illness that would
prevent the patient from completing the questionnaires and participating in an interview, no apparent
cognitive impairment (Mini Mental State [MMS] score
ⱖ 24),19,20 and no difficulty with verbal communication. Because patients registered with the PCU have a
broad range of life expectancies and because information from terminally ill patients with cancer was considered essential, we followed up all participants.
However, we only analyzed the data from patients
whose death was confirmed to have occurred within 6
months of participation of the initial assessment.
Suicidality (Suicidal Ideation, Patient Perceptions, and
Interest in Requesting Euthanasia)
We assessed patients’ suicidal ideation as suicidality
because it is a well-known indicator of patients’ suicidal behavior.21 We also assessed patients’ perceptions regarding their suicidal ideation to understand
why they wanted to die. First, before interviewing
patients regarding suicidal ideation, trained psychiatrists asked the subjects about the presence or absence
of such feelings as wanting to escape from their disease and going away as a means of avoiding psychologic harm. Second, suicidal ideation was assessed at
the time of the patient’s first visit to the outpatient
clinic after registering with the PCU (baseline) and
again at admission (follow-up) and was classified as
absent, subthreshold, or present by using the suicidal
Suicidality in Terminally Ill Patients/Akechi et al.
TABLE 1
Patient Characteristics (n ⴝ 140)
Characteristic
Age
Mean ⫾ SD
Range
No. of males
Education
ⱖ 12 yrs
Marital status
Married
Never married
Divorced
Separated
Widowed
Unknown
Household size
Living alone
Cancer site
Lung
Colon
Pancreas
Stomach
Head and neck
Other
Time since cancer diagnosis
Mean ⫾ SD
Median
Range
Past cancer treatment
Surgery
Chemotherapy
Radiotherapy
Current medication
Opioids
Antidepressants
Anxiolytics
Religion
Buddhist
Catholic/Protestant
Other
None
Survivala
Median
Mean ⫾ SD
Range
No. of patients (%)
61 ⫾ 10 yrs
38–89 yrs
92 (66)
92 (66)
121 (86)
5 (4)
3 (2)
1 (1)
9 (6)
1 (1)
7 (5)
57 (41)
17 (12)
13 (9)
11 (8)
11 (8)
31 (22)
22 ⫾ 38 mos
13 mos
0–282 mos
61 (44)
81 (58)
51 (36)
62 (44)
4 (3)
19 (14)
3 (2)
2 (1)
8 (6)
127 (91)
61 days
70 ⫾ 43 days
2–176 days
SD: standard deviation.
a
Survival (days) since the date of the initial assessment.
ideation item (recurrent thoughts of death, recurrent
suicidal ideation without a specific plan, or a suicide
attempt or a specific plan for committing suicide) of
the major depressive episode module of the Structured Clinical Interview for DSM-III-R (SCID).22 If suicidal ideation was judged to be present at baseline, the
patient was asked an open-style question, such as,
“May I ask you why you think you have such
thoughts?” The patient’s responses were recorded in a
descriptive manner.
185
As another indicator of suicidality, patients were
asked at baseline and at follow-up about their interest
in requesting euthanasia (“Would you ask a member
of the medical staff to end your life now if the physician agreed with such a procedure?”). Their replies
were scored as 1 (no), 2 (maybe), 3 (probably), or 4
(yes). The wording of this question was based on our
clinical experience. The patients also were asked
about actually requesting euthanasia.
Biomedical and Psychosocial Factors
Sociodemographic, biomedical, and psychosocial factors were investigated at baseline to explore factors
associated with and predicting suicidal ideation, as
well as factors contributing to and predicting interest
in requesting euthanasia. A structured interview was
conducted to identify demographic factors, Karnofsky
performance status, and three common and distressing physical symptoms (pain, dyspnea, and fatigue on
a scale of 1 [none] to 5 [severe]). Current and past
major depression was evaluated on the basis of the
SCID.22 However, because suicidal ideation is one of
the diagnostic criteria, we adopted a modified method
in which we diagnosed current major depression
when patients fulfilled more than five of the eight
original DSM-III-R criteria, excluding the suicidal ideation item.22 The Hospital Anxiety and Depression
Scale (HADS) was used to evaluate psychologic distress.23,24 We asked about several different types of
concerns, including financial burden, future health
status, pain, dependence on others, being a burden on
others, and loss of dignity (scored from 1 [not at all] to
5 [very much]).25 Patients’ use of confidants (presence
and satisfaction) was assessed by a structured interview as an indicator of social support26 and their score
on the MMS was used as a measure of cognitive function.
Consent and Study Procedures
The study was approved by the institutional review
board and the ethics committee of the National Cancer Center of Japan. Written consent was obtained
after each patient had been fully informed of the purpose and the approved protocol of the study.
Patients who registered with the PCU were invited
to participate in the study. After obtaining patients’
consent, trained psychiatrists administered the MMS,
and eligible patients participated in the interview and
completed the questionnaires in the outpatient clinic
(baseline). The baseline investigation was conducted
on the first visit after PCU registration. Of the patients
admitted to the PCU, those whom investigators, physicians, and nurses considered capable of participating in the follow-up study were again administered the
186
CANCER January 1, 2004 / Volume 100 / Number 1
TABLE 2
Factors Associated with Suicidal Ideation in Terminally III Patients with Cancer: Univariate Analysis
Suicidal ideation
Characteristic
Age (yrs)
HADSc
Education (yrs)
Performance statusd
Paine
Dyspneae
Fatiguee
MMSf
Concernse
Financial
Pain
Future physical condition
Dependence on others
Burden on others
Loss of dignity
Satisfaction with confidantsg
Male gender
Marital status (married)
Living alone
Major depression
History of major depression
Absence of confidants
Absence of religious belief
Present (n ⴝ 12)
Mean (SD)/No. (%)a
Analysis
Absent (n ⴝ 128)
Mean (SD)/No. (%)a
Test statisticb
df
P
54 (12)
19.6 (5.2)
13 (3)
62 (12)
2.4 (1.1)
2.1 (0.9)
2.9 (0.9)
27.9 (2.1)
61 (10)
11.8 (6.4)
12 (3)
70 (14)
1.9 (0.9)
2.1 (1.0)
2.2 (1.1)
27.3 (1.9)
2.26
⫺3.95
⫺0.53
0.93
⫺1.68
⫺0.27
⫺2.29
⫺1.01
138
137
—
—
—
—
—
—
0.03
0.0001
0.60
0.05
0.09
0.79
0.02
0.29
3.2 (1.5)
3.8 (1.4)
3.8 (1.4)
2.7 (1.6)
3.4 (1.4)
2.1 (1.6)
4.9 (1.5)
6 (50)
11 (92)
0 (0)
4 (33)
1 (9)
0 (0)
11 (92)
2.0 (1.2)
2.7 (1.4)
3.5 (1.2)
2.0 (1.2)
2.1 (1.2)
1.4 (0.9)
5.5 (1.3)
86 (73)
110 (86)
7 (6)
5 (4)
18 (14)
19 (15)
116 (91)
⫺2.81
⫺2.33
⫺0.94
⫺1.64
⫺3.15
⫺1.49
⫺1.45
1.44
—
—
—
—
—
—
—
—
—
—
—
—
—
1
—
—
—
—
—
—
0.005
0.02
0.35
0.10
0.002
0.14
0.15
0.23
1.00
1.00
0.003
1.00
0.37
1.00
SD: standard deviation; HADS: Hospital Anxiety and Depression Scale; MMS: Mini-Mental State examination.
a
Mean (SD) for all rows above Male gender. No. (%) for all other rows, including Male gender.
b
Test statistic is t for Age and HADS; z for Education, Performance status, Pain, Dyspnea, Fatigue, MMS, Concerns, and Satisfaction with confidants; and ␹2 for Male gender, Marital status (married), Living alone,
Major depression, History of major depression, Absence of confidants, and Absence of religious belief.
c
Total score on the Hospital Anxiety and Depression Scale.
d
Defined by Karnofsky criteria.
e
Coded as 1: absent; 2: mild; 3: somewhat; 4: considerable; 5: severe.
f
Total score on the Mini-Mental State Examination.
g
Coded as 1: very dissatisfied; 2: fairly dissatisfied; 3: slightly dissatisfied; 4: neither satisfied nor dissatisfied; 5: somewhat satisfied; 6: fairly satisfied; 7: very satisfied.
MMS by a trained psychiatrist. Patients who received
an MMS score ⱖ 24 participated in the follow-up
investigation (follow-up).
Statistical Analysis
Suicidal ideation was dichotomized by the interview
ratings (absent/subthreshold vs. present). To explore factors associated with suicidal ideation, the
variables investigated were included in a preliminary univariate analysis. To control type I error
rates, the level of significance was determined by
dividing the significance level (0.05) by the number
of tests performed (Bonferonni correction). As a
result, the significance level was set as 0.0023, because 22 tests were conducted (Table 2). Final associated factors were decided upon by entering vari-
ables with significant P values into a logistic
regression model. Interest in requesting euthanasia
was dichotomized by the interview ratings into
strong interest (yes) versus others and an appropriate univariate analysis was conducted to explore
contributing factors among patients with suicidal
ideation. To investigate baseline predicting factors
of changes of suicidal ideation and interest in requesting euthanasia between baseline and followup, each change was classified into three groups
(increase, no change, and decrease) and an appropriate analysis (e.g., analysis of variance) was conducted with a Bonferroni correction. To explore factors predicting occurrences of suicidal ideation and
interest in requesting euthanasia, a univariate analysis was conducted with a Bonferroni correction at
Suicidality in Terminally Ill Patients/Akechi et al.
baseline among patients with no suicidal ideation
and no interest in requesting euthanasia, respectively. All P values reported are two tailed. All statistical procedures were conducted with SPSS Version 10.0J software (SPSS, San Diego, CA).
Descriptive data expressed by patients regarding
their reasons for suicidal ideation were analyzed by
content analysis.27 First, two investigators (T.A., Y.
Su.) independently categorized the statements with
regard to the similarities and differences between
their reasons in the context. Discordances were resolved through discussions. The validity of the categorization was confirmed by the consensus of the
authors. The reliability coefficients of the interviews
by 2 independent trained psychiatrists were 0.80 for
suicidal ideation, 0.91 for interest in requesting euthanasia, and 1.00 for major depression.
RESULTS
In the current study, 764 terminally ill patients with
cancer were registered with the PCU during the study
entry period. Of these, 507 patients were ineligible for
study entry (too ill, n ⫽ 443; cognitive impairment, n
⫽ 50; age ⬍ 18, n ⫽ 1; not informed of the cancer
diagnosis, n ⫽ 3; difficulty with verbal communication, n ⫽ 10). Of the remaining 257 eligible patients, 28
refused to participate and 20 could not be contacted
(e.g., because of emergency admission to another hospital). Therefore, 209 patients participated in the baseline assessment. Of these, 69 patients were excluded
(survived ⬎ 6 months, n ⫽ 56; unknown survival, n
⫽ 13). Ultimately, the data for the remaining 140 patients whose survival time was confirmed to have been
⬍ 6 months were analyzed (Table 1). The median
survival time after participation in the initial assessment was approximately 2 months. Demographic
data, such as age, gender, and disease site, were similar to the corresponding data for patients receiving
palliative care in Japan.28 The social status (based on
education, marital status, and household size) of patients in the current study also was similar to that of
most Japanese patients with cancer.26 After the baseline investigation, 25 patients were never admitted to
the PCU (e.g., because of emergency admission to
another hospital), whereas the other 115 were admitted. However, 3 refused to cooperate further and 55
were judged to be ineligible (too ill, n ⫽ 39; cognitive
impairment, n ⫽ 16). Therefore, 57 of the subjects
completed the follow-up assessment. The mean interval between the baseline and follow-up investigations
was 45 ⫾ 36 days (median, 35 days), and the mean
⫾ standard deviation and median survival times between the follow-up investigation and death were 39
⫾ 26 and 37 days, respectively.
187
TABLE 3
Factors Associated with Suicidal Ideation in Terminally Ill Patients
with cancer: Multivariate Logistic Regression Analysis (n ⴝ 140)
Characteristic
Beta
SE
Odds
ratio
95% CI
P
HADSa
Concern about being a burden
on othersb
0.17
0.06
1.19
1.06–1.34
0.003
0.49
0.27
1.64
0.97–2.75
0.06
SE: standard error; CI: confidence interval; HADS: Hospital Anxiety and Depression Scale.
a
Total score on the Hospital Anxiety and Depression Scale.
b
Coded as 1: absent; 2: mild; 3: somewhat; 4: considerable; 5: severe.
At the baseline assessment, suicidal ideation was
subthreshold in 37 patients (26.4%) and present in 12
(8.6%). No patients reported an actual suicidal attempt. Interest in requesting euthanasia was strong
(‘yes’) for 7 patients (5.0%), moderate (‘probably’) for
4 patients (2.9%), and weak (‘maybe’) for 7 patients
(5.0%). One patient (0.7%) gave no answer. No patients reported actually requesting euthanasia.
Four major reasons for suicidal ideation were elicited (multiple answers): physical distress (n ⫽ 3), anxiety/fear regarding the future course of the illness (n
⫽ 2), hopelessness (n ⫽ 4), and burden on family (n
⫽ 3). The causes of the physical distress were pain (n
⫽ 3) and dyspnea (n ⫽ 1). Anxiety/fear regarding the
future course of the illness consisted of the fear of
future pain (n ⫽ 1) and uncertainty of the future (n
⫽ 1).
The HADS total score and concerns about being a
burden on others were significant factors associated
with suicidal ideation (Table 2). The logistic regression
model identified only HADS as a final significant factor (Table 3). As a result of this finding, we investigated
whether there was an interaction effect between
HADS and religious belief in regard to suicidal ideation because a previous study had found that spirituality and faith might buffer the effect of depression
on desire for hastened death.29 However, there was no
significant interaction (beta ⫽ 0.014, P ⫽ 0.85).
Among the 12 patients with present suicidal ideation, 6 (50.0%) expressed strong interest (‘yes’) in
requesting euthanasia, 1 (8.3%) moderate interest
(‘probably’), and 2 (16.7%) weak interest (‘maybe’).
One patient (8.3%) gave no answer, and 2 patients
(16.7%) expressed no interest. Among the variables
investigated, none were significant factors contributing to interest in requesting euthanasia (data not
shown).
A change in suicidal ideation between PCU registration (baseline) and admission (follow-up) was documented in 22 (38.6%) of the 57 patients who were
188
CANCER January 1, 2004 / Volume 100 / Number 1
followed up. Eleven patients each experienced an increase and decrease in suicidal ideation (19.3% and
19.3%, respectively). Nine patients (15.8%) had a
change in interest in requesting euthanasia. Four patients (7.0%) had increased interest, and 5 (8.8%) had
decreased interest. The proportion of patients with
observed longitudinal change was significantly higher
for suicidal ideation (38.6%) than for interest in requesting euthanasia (15.8%; chi-square, 7.49, df ⫽ 1, P
⫽ 0.006). Among the variables investigated, none were
significant factors for predicting changes in suicidal
ideation and interest in requesting euthanasia (data
not shown).
Suicidal ideation occurred in 10 of the 40 patients
without suicidal ideation at baseline. None of the
baseline variables investigated significantly predicted
its occurrence (data not shown). Among the 51 patients with no interest in requesting euthanasia at
baseline, interest developed in 4 patients. None of the
baseline factors significantly predicted its occurrence
(data not shown).
DISCUSSION
The prevalences of suicidal ideation and interest in
requesting euthanasia in the current study were essentially consistent with those reported in previous
studies,9 –14 suggesting that neither is uncommon
among terminal patients with cancer. These findings
suggest that the proportion of terminally ill patients
with cancer who are suicidal is similar, irrespective of
cultural and social differences, and that the issues
should be addressed even in routine care.
Our findings regarding perceptions of the reasons for suicidal ideation suggest that individual
patients perceive different factors, including physical, psychologic, and existential distress, as the
causes of their suffering. It is noteworthy that these
findings are almost the same as the results of an
Italian psychologic autopsy study that investigated
the characteristics of terminally ill patients with
cancer who committed suicide5 and a qualitative
Australian study that investigated factors associated
with the wish to hasten death in terminally ill patients with cancer.14 These issues should be addressed in patients with suicidal ideation. The findings also suggest that comprehensive palliative care
may be the best means of preventing suicide among
terminally ill patients with cancer. Another Italian
study suggested that continuous care provided by a
palliative care team may be able to reduce suicide
even among terminally ill patients with cancer.30
Because previous studies investigating the associations between the desire for death and major
depression have consistently indicated the existence
of an association,9,11,12 one noteworthy finding in
the current study may be that major depression is
not independently associated with suicidal ideation.
Because self-reported anxiety and depression were
significant factors associated with suicidal ideation
in the current study, psychologic distress, rather
than a clinical diagnosis of major depression, is a
more important indicator of suicidal ideation in
terminally ill Japanese patients with cancer than in
their non-Japanese counterparts. From this standpoint, the demoralization syndrome proposed by
Kissane et al.31 as a new psychiatric diagnosis for
palliative care may be worth investigating in future
studies because it has been suggested to be associated with desire to die or with committing suicide.
For example, it is characterized by helplessness, loss
of meaning, and existential distress. There may be
other explanations for the finding that major depression is not independently associated with suicidal ideation. For example, several differences between the current and previous studies9,11,12 might
have given rise to such different results, such as
outcome measures (suicidal ideation vs. desire for
death), study setting (PCU registration vs. admission), diagnosis of major depression (modified vs.
standard criteria), and the statistical significance
level finally adopted (conservative vs. standard).
Furthermore, there is a generally lower prevalence
rate of major depression in Asian countries, both
among patients with cancer and among the general
population, compared with Western countries, possibly because of crosscultural differences (e.g., social stigma, cultural reluctance to acknowledge
mental symptoms, and low divorce rate). This epidemiologic difference may be responsible for the
inconsistent findings.32 Nonetheless, our results
confirm that psychologic burden is an important
factor in understanding suicidality among terminally ill
patients with cancer. To enable patients to deal with
psychologic distress, education and training of physicians engaged in end-of-life care are critical.
Physical distress, especially pain, which the general public and patients with cancer often give as a
reason for supporting legalization of euthanasia/
PAS,15 was not associated with suicidal ideation. Our
findings support the study by Emanuel,33 who reported that although pain may be regarded as the major
reason for supporting euthanasia/PAS, psychologic distress actually is more likely than pain to increase patient
suicidality. In any event, because individuals with psychologic distress are more likely to have suicidal ideation, psychologic care focused on terminally ill patients
is essential for reducing patient suicidality.
A previous study suggested a discrepancy between
Suicidality in Terminally Ill Patients/Akechi et al.
suicidality and interest in euthanasia/PAS. Ganzini et
al.34 demonstrated that only 14% of patients with
amyotrophic lateral sclerosis who were willing to consider PAS had thought about committing suicide in
the previous 2 weeks. The results of the current study
did not identify factors that contribute to interest in
requesting euthanasia by patients with suicidal ideation, possibly because of the small sample size. Large
studies are needed to clarify such factors.
Our findings showed changes in suicidal ideation and interest in requesting euthanasia to be
highly variable. Emanuel et al.35 reported that onehalf of terminally ill patients (52% of the subjects
had cancer) who had considered euthanasia or PAS
changed their minds, whereas an almost equal number had begun to consider them. Although there are
several differences between that study and the current one (e.g., subjects, setting, outcome measures,
mean follow-up time), similar results were obtained.
Our findings corroborate the instability of suicidality in terminally ill patients. Another noteworthy
finding is the difference in stability in regard to
suicidal ideation and interest in requesting euthanasia. The results of the current study demonstrate
that suicidal ideation is more changeable than interest in requesting euthanasia. This finding may
reflect the different natures of suicidal ideation and
interest in requesting euthanasia.
The current study did not identify factors that
predict changes and occurrences of suicidal ideation
and interest in requesting euthanasia. This may be
because of the relatively small sample size or lack of
several potentially important factors. Several recent
studies have demonstrated the importance of spiritual
and/or existential factors (e.g., demoralization, hopelessness, dignity, and spiritual well-being) and family
function as associated with suicidality in terminally ill
patients with cancer.14,29,36,37 These may be promising
factors to investigate. Further investigation is needed
to clarify predictive factors for suicidality to prevent
suicide and suffering among terminally ill patients
with cancer.
The current study yielded several findings regarding cross-cultural aspects of suicidality and attitudes toward euthanasia that differ between Western and Japanese terminally ill patients with cancer.
Western individuals generally are more likely to
support euthanasia among terminally ill patients
and more reluctant to accept suicide than Japanese
individuals.15,17,38,39 Japanese public awareness, legal status, and the social movement for euthanasia/
PAS are quite different from those in Western countries.40 For example, Japan is more conservative in
its view of the individual’s right to die and no leg-
189
islation has addressed the question of euthanasia
and/or PAS in Japan.41 Religious differences also
should be considered. There is modest empiric evidence that religion protects against suicide in the
general population in Western countries. However,
in Japanese society, where the Buddhist influence
prevails, suicide is not considered to be sinful, in
contrast to Christianity.41,42 A recent American
study demonstrated that spirituality and faith may
buffer the effect of depression on desire for hastened death.29 Our findings suggest that active religious practice is variable as well as infrequent in
Japanese terminally ill patients with cancer and that
the role and influence of religion on suicidality in
these patients may be different from its role and
influence in Western countries. However, because
the religion of the patients was heterogenous and
the number of the patients was small in the current
study, we could not confirm the role and influence
of religion on Japanese patients with cancer. Additional studies are needed to clarify this issue in
Japan. In contrast, the prevalences of suicidal ideation and interest in euthanasia are similar, as is the
importance of psychologic burden rather than physical symptoms as a factor associated with patient
suicidality, and the stability of suicidality also is
similar. Therefore, there appear to be many similarities in suicidality between terminally ill patients
with cancer in Western countries and Japan, despite
the many cultural differences. These comparative
findings suggest that the suffering experienced by
terminally ill patients with cancer may be universal,
irrespective of individual culture.
The current study has several limitations. First,
our sample size was not very large and the sample
size in the qualitative study was especially small.
Because only 18.3% (140 of 764) and 7.5% (57 of 764)
of patients at baseline and follow-up, respectively,
were included in the analysis, generalization of the
findings obtained may be problematic. Second, institutional bias may have occurred. For example, the
study was conducted in one institution, and the
background of the subjects was somewhat different
from that of individuals in the general population in
Japan who die of cancer. For example, the age of the
participants in the study was younger, and the proportion of patients with stomach malignancies was
lower. Third, potentially important factors, especially validated measurements of hopelessness, personality, and existential distress, were lacking and
may have distorted the results. Fourth, the timing of
the follow-up assessment (admission) may also have
produced distortions because the reasons for admission may have differed. Fifth, because the sui-
190
CANCER January 1, 2004 / Volume 100 / Number 1
cidal ideation and interest in requesting euthanasia
are not the same as completed suicide, the findings
obtained may not be directly applicable to the establishment of a suicide prevention strategy. Finally,
the participants received standard palliative care (at
follow-up, 64% of the patients were satisfied with
the care they received). Therefore, the findings may
not be applicable to terminally ill patients with cancer in other settings.
In conclusion, the results of the current study
indicate that suicidal ideation can change even in
terminally ill patients and that end-of-life care for
dying individuals that is focused on their psychologic
distress may be a way of preventing suicide. The many
similarities regarding suicidality between terminally ill
patients with cancer in Western countries and Japan
suggest that the suffering experienced by these patients may be universal, irrespective of individual cultures.
14.
15.
16.
17.
18.
19.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Harris EC, Barraclough BM. Suicide as an outcome for
medical disorders. Medicine (Baltimore). 1994;73:281–
296.
Grzybowska P. The incidence of suicide in palliative care
patients. Palliat Med. 1997;11:313–316.
Filiberti A, Ripamonti C. Suicide and suicidal thoughts in
cancer patients. Tumori. 2002;88:193–199.
Henriksson MM, Isometsa ET, Hietanen P, Aro HM, Lonnqvist JK. Mental disorders in cancer suicides. J Affect Disord. 1995;36:11–20.
Filiberti A, Ripamonti C, Totis A, et al. Characteristics of
terminal cancer patients who committed suicide during a
home palliative care program. J Pain Symptom Manage.
2001;22:544 –553.
Takahashi Y. [Epidemiology of suicide]. Clin Neurosci.
2002;20:544 –546.
Back AL, Wallace JI, Starks HE, Pearlman RA. Physicianassisted suicide and euthanasia in Washington State—patient requests and physician responses. JAMA. 1996;275:
919 –925.
Asai A, Ohnishi M, Nagata SK, Tanida N, Yamazaki Y.
Doctors’ and nurses’ attitudes towards and experiences of
voluntary euthanasia: survey of members of the Japanese
Association of Palliative Medicine. J Med Ethics. 2001;27:
324 –330.
Chochinov HM, Wilson KG, Enns M, et al. Desire for
death in the terminally ill. Am J Psychiatry. 1995;152:
1185–1191.
Emanuel EJ, Fairclough DL, Daniels ER, Clarridge BR. Euthanasia and physician-assisted suicide: attitudes and experiences of oncology patients, oncologists, and the public.
Lancet. 1996;347:1805–1810.
Breitbart W, Rosenfeld B, Pessin H, et al. Depression, hopelessness, and desire for hastened death in terminally ill
patients with cancer. JAMA. 2000;284:2907–2911.
Wilson KG, Scott JF, Graham ID, et al. Attitudes of terminally ill patients toward euthanasia and physician-assisted
suicide. Arch Intern Med. 2000;160:2454 –2460.
Akechi T, Okamura H, Nishiwaki Y, Uchitomi Y. Predictive
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
factors for suicidal ideation in patients with unresectable
lung carcinoma: a 6-month follow-up study. Cancer. 2002;
85:1085–1093.
Kelly B, Burnett P, Pelusi D, Badger S, Varghese F, Robertson M. Terminally ill cancer patients’ wish to hasten death.
Palliat Med. 2002;16:339 –345.
Rosenfeld B, Krivo S, Breitbart W, Chochinov HM. Suicide,
assisted suicide, and euthanasia in the terminally ill. In:
Chochinov HM, Breitbart W, editors. Handbook of psychiatry in palliative medicine. New York: Oxford University
Press, 2000:51– 62.
Pessin H, Rosenfeld B, Burton L, Breitbart W. The role of
cognitive impairment in desire for hastened death: a study
of patients with advanced AIDS. Gen Hosp Psychiatry. 2003;
25:194 –199.
Miyashita M, Hashimoto S, Kawa M, Kojima M. [Attitudes
towards terminal care among the general population and
medical practitioners in Japan]. Nippon Koshu Eisei Zasshi.
1999;46:391– 401.
Morita T, Akechi T, Sugawara Y, Chihara S, Uchitomi Y.
Practices and attitudes of Japanese oncologists and palliative care physicians concerning terminal sedation: a nationwide survey. J Clin Oncol. 2002;20:758 –764.
Folstein MF, Folstein SE, McHugh PR. Mini-Mental State: a
practical method for grading the cognitive state of patients
for the clinician. J Psychiatr Res. 1975;12:189 –198.
Mori E, Mitani Y, Yamadori A. [Usefulness of a Japanese
version of the Mini-Mental State test in neurological patients]. Shinkei Shinnigaku. 1985;1:82–90.
Mann JJ. A current perspective of suicide and attempted
suicide. Ann Intern Med. 2002;136:302–311.
Spitzer RL, Williams JB, Gibbon M, First MB. Structured
clinical interview for DSM III-R. Washington, DC: American Psychiatric Press, 1990.
Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67:361–370.
Kugaya A, Akechi T, Okuyama T, Okamura H, Uchitomi Y.
Screening for psychological distress in Japanese cancer
patients. Jpn J Clin Oncol. 1998;28:333–338.
Akechi T, Okuyama T, Nakano T, Uchitomi Y. [Desire for
death in terminally ill cancer patients—prevalence and its
associated physical and psychosocial factors]. Sougou Byoin Seishiu Igaku. 2001;13:153–158.
Akechi T, Okamura H, Yamawaki S, Uchitomi Y. Predictors of patients’ mental adjustment to cancer: patient
characteristics and social support. Br J Cancer. 1998;77:
2381–2385.
Pope C, Mays N. Qualitative research in health care (2nd
edition). London: BMJ Publishing Group, 1999.
Kashiwagi T. Palliative care in Japan. Palliat Med. 1991;5:
165–170.
McClain CS, Rosenfeld B, Breitbart W. Effect of spiritual
well-being on end-of-life despair in terminally-ill cancer
patients. Lancet. 2003;361:1603–1607.
Ripamonti C, Filiberti A, Totis A, De Conno F, Tamburini
M. Suicide among patients with cancer cared for at home
by palliative-care teams. Lancet. 1999;354:1877–1878.
Kissane DW, Clarke DM, Street AF. Demoralization syndrome—a relevant psychiatric diagnosis for palliative care.
J Palliat Care. 2001;17:12–21.
Weissman MM, Bland RC, Canino GJ, et al. Cross-national
epidemiology of major depression and bipolar disorder.
JAMA. 1996;276:293–299.
Suicidality in Terminally Ill Patients/Akechi et al.
33.
34.
35.
36.
37.
Emanuel EJ. Euthanasia and physician-assisted suicide. A
review of the empirical data from the United States. Arch
Intern Med. 2002;162:142–152.
Ganzini L, Johnston WS, McFarland BH, Tolle SW, Lee MA.
Attitudes of patients with amyotrophic lateral sclerosis and
their care givers toward assisted suicide. N Engl J Med.
1998;339:967–973.
Emanuel EJ, Fairclough DL, Emanuel LL. Attitudes and
desires related to euthanasia and physician-assisted suicide among terminally ill patients and their caregivers.
JAMA. 2000;284:2460 –2468.
Chochinov HM, Hack T, Hassard T, Kristjanson LJ, McClement S, Harlos M. Dignity in the terminally ill: a crosssectional, cohort study. Lancet. 2002;360:2026 –2030.
Kelly B, Burnett P, Pelusi D, Badger S, Varghese F, Robertson
38.
39.
40.
41.
42.
191
M. Factors associated with the wish to hasten death: a study
of patients with terminal illness. Psychol Med. 2003;33:75– 81.
Nakagawa Y. Death with dignity in the Japanese culture.
Psychiatry Clin Neurosci. 1995;49 Suppl 1:S161–S163.
Cohen LM. Suicide, hastening death, and psychiatry. Arch
Intern Med. 1998;158:1973–1975.
Foley K, Hendin H. The case against assisted suicide. Baltimore: The Johns Hopkins University Press, 2002.
Long SO. Ancestors, computers, and other mixed messages: ambiguity and euthanasia in Japan. Camb Q Healthc
Ethics. 2001;10:62–71.
Maris RW, Berman AL, Silverman MM. Ethical, religious,
and philosophical issues in suicide. In: Maris RW, Berman
AL, Silverman MM. Comprehensive textbook of suicidology. New York: The Guilford Press, 2000:456 – 479.