FULL PAPER (日本語ー抄録のみ)
A COMPARISON OF JAPANESE AND AMERICAN PSYCHIATRISTS' ATTITUDES TOWARDS PATIENTS WISHING TO DIE IN THE GENERAL HOSPITAL
Psychotherapy and Psychosomatics, 66:319-328;1997.
Short title for running head: Psychiatrist's attitudes on suicide in medical patients: Japan vs. United States.
Key words: Suicide, medical patients, Euthanasia, Japan
(1) Department of Psychosomatic Medicine, Tokyo University Branch Hospital, Tokyo Japan. (2) Department of Psychiatry, Albert Einstein College of Medicine, Bronx New York. (3) Tokyo Institute of Psychiatry, Tokyo Japan. (4) Department of Psychiatry and Behavioral Sciences, Tokai University School of Medicine, Kanagawa Japan.
ABSTRACT
Backround: The attitudes of the psychiatrists in Japan and the U.S. were compared in order to investigate their ideas on whether patients in general medical hospitals who have a desire to die should be allowed to, or assisted in this regard, and whether they require psychiatric evaluation and intervention, and the cultural influences on these attitudes. Methods: Japanese and American general hospital psychiatrists' attitudes towards the reasonability of suicide, physician assisted suicide, and removal of life supports under various medical and psychosocial situations were compared. Seventy-two American and sixty-two Japanese psychiatrist's data was collected using the Suicidal Attitudes Inventory (SAI). Results: The majority of both American and Japanese psychiatrists agreed that there may be times when suicidal ideation, or completed suicide in med-surg patients could be reasonable. Significantly more Japanese psychiatrists responded with some agreement to the reasonability of suicide when one is unable to fulfill social role expectations, and had more concern about causing suicidal ideation by informing terminal patients of their diagnosis. Conclusions: The results indicate that psychiatrists' attitudes towards the relationship of psychopathology with suicidal ideation, the effect of depression and other cultural factors on the desire to die in the medically ill, are issues that need better clarification among both the medical profession as well as within society. Looking at how other societies handle these matters may help to understand one's own approach to them.
(Note: Tables are not available on line.)
INTRODUCTION
The current background in Japan on the issue of medical patients who desire death began in 1961 when the son of a hospitalized man suffering from severe pain after a stroke mixed a cup of milk with insecticide and arranged for his unsuspecting mother to give this to his father. After the father requested that his son assist him in dying, the son killed him in order to show his love [1]. Because the procedure was not done by a physician, and because the method used could not be considered ethical, the court ruled this as a case of "Requested Murder". Since the victim acted on his own free will, however, the defendent was sentenced to a one year prison term and three years probation.
The most recent case in Japan of a hospital-based killing took place in 1991 at the Tokai University Medical Center in Kanagawa [2,3,4]. Although the doctor testified that the patient's family had urged him to perform euthanasia (via potassium chloride injection) to ease the patient's suffering, the police investigation reported that the family had not meant for the patient to be killed when they requested intervention to ease his suffering. The court ruled that the physicians actions were unlawful because euthanasia was not the expressed wish of the patient, but because of the family's suspected pressure to end the patient's suffering and the lack of a system to manage this kind of situation in the terminally ill, the doctor was given a two-year suspended sentence.
Although brain death is felt to be equal to clinical death by the vast majority of Japanese physicains, neither the legal system nor the general public seems ready to accept lack of brain activity as legal death in Japan. In 1994 however, the Japan Science Council approved a report advocating passive euthanasia (ie. letting a patient die while providing palliative treatment for pain with morphine) for patients in a deep, irreversible coma, provided the patient has previously stated opposition to life-prolonging measures [4,5]. Because suicide is culturally more acceptable in Japan than in western countries [6] there is the concern that a loose interpretation of this report might put too much decision making power into the hands of physicians. There are also concerns that if dignified death is overvalued in Japan, rescue efforts that can save some patients may be overlooked [7].
In the U.S., considerable action toward legalization of these acts has begun in fits and starts. The U.S. Congress passed the Patient Self-Determination Act in 1990 requiring hospitals to inform patients and their families of their legal right to refuse life-sustaining procedures through advanced directives (living wills and proxy documents) [8]. Unfortunately only 10-20% of Americans have signed an advanced directive, and conflict over how to handle the wishes of terminal patients still arises in the courts. These ongoing problems may have helped to boost support for physician-assisted suicide reflected by two recent decisions. In March of 1996, a jury ruled that retired physician Jack Kevorkian who has admitted to assisting 27 patients suicide had not violated Michigan State law. In the same week a federal court of appeals ruled a Washington State law prohibiting euthanasia invalid [8].
The Death With Dignity Act passed by 51% of the residents of Oregon in 1994, however, was ruled in violation of the Equal Protection clause of the Constitution by a U.S. District Court in 1995 [9]. The court stated that the law "withholds from terminally ill citizens the same protections from suicide the majority enjoys." One example of inadequate protection noted was that non-psychiatric attending physicians who would evaluate the reasonability of request to die could overlook a psychaitric disorder. Some U.S. psychiatrists against the law have voiced that even psychiatrists are not adept at diagnosing depression in the terminally ill, and point out the incentive to encourage euthanasia based on cost savings (60% of healthcare costs occur in the last three months of life). Appeal is now underway [9].
The Japanese Neuropsychiatry Association has no official comment on the issue of euthanasia, while the Japanese Medical Society is officially against active euthanasia and supports passive euthanasia. The American Psychiatric Association and the American Medical Association and both officially opposed to euthanasia. This is thought to be due to the desire to maintain the image of the physician as protector of life and that there is the danger of abuse in legalizing these procedures [10].
Studies of suicide victims in both the west and in Japan have consistently found high rates of pre-existing mental disorders and psychiatrists will need to determine when depression in the terminally ill may affect one's decision to forgo life-sustaining medical treatment [11-15]. Ganzini et al. [13 ] found that remission of major depression with hopelessness in psychiatric patients increased the preferences of these patients for life-saving treatment, while Chochinov et al. [15] and Berger [16-17] found that desire for death in med-surg or terminal patients is associated with clinical depression, that this desire for death is often transient, and that improvement in family support and treatment of pain can diminish desire for death.
A recent four-year, 28 million-dollar multi-site study of more than 9,000 patients in the U.S., the "Study to Understand Prognoses and Preferences for Outcome and Risks of Treatment (SUPPORT)" confirmed substantial shortcommings in the care for seriously ill hospitalized adults [18]. This study found that less than half of the doctor's knew their patient's wishes regarding resuscitation (46% of do-not-resuscitate orders were written within two days of death), 50% of the patients that died and were conscious were in moderate to severe pain in the last three days of life, and 38% spent 10 or more days in an ICU. An intervention using specially-trained nurses to facilitate communication between patients, families, and doctors did not improve the quality of care. A large study of physicians in Washington State found that 48% felt that euthanasia was never ethically justified, while 42% felt the opposite [21]. A slight majority favored legalizing euthanasia in at least some situations, but most would be unwiling to participate. Physicians themselves may be compulsively and perfectionistically dedicated to the treatment of disease [10].
While medically ill patients' refusal of lifesaving treatment is more and more viewed as "allowing to die", in psychiatric settings, a patient's desire to die is generally considered evidence of a mental disorder and disturbed decision making capacity [20]. The burden of proof on the desire to die in psychiatry is on the patient who desires to die; in internal medicine, the burden is to prove that the patient lacks competence to refuse life-sustaining treatment on the physician who wishes to override the refusal. Consultation/Liaison (C/L) psychiatry in Japan is not a well established subspecialty as it is in the United States and psychiatric problems encountered on medical units are often handled by the primary physician, with little experience in this area [21].
It is clear from the above that a more unified and educated approach to this issue is needed within the medical profession in both countries. Issues of whether patients in general medical hospitals who have a desire to die should be allowed to or assisted in this regard, and whether they require psychiatric evaluation and intervention are important areas for both Japanese and American psychiatrists to take the lead in guiding their medical colleagues. The present study compared the attitudes of the psychiatrists in these two countries in order to investigate the status of psychiatrists ideas on these issues and the cultural influences on these attitudes.
METHOD/SUBJECTS
In the U.S., 200 randomly selected psychiatrists in the Academy of Psychosomatic Medicine (a Consultation/Liaison psychiatry organization), and 161 psychiatrists in the American Psychosomatic Society who indicated Consultation/Liaison as a specialty were sent the SAI for a total of 361 mailed questionnaires. The total number of returned questionnaires was 80 (22%), 8 of which stated that because they were not seeing general hospital patients they could not participate, leaving a total of 72 (20%) completed SAI databases for analysis. Repeat mailing or inquiry to those who did not respond was not done. 84% of the U.S. respondents were male and 16% female. The average age was 49.3+14.9 (range 33-79 years). The average number of years of experience of the U.S. psychiatrists was 18.9+11.5 years (range 2-52 years) .
In Japan, five data base sets with return-address stamped envelopes were sent to eighteen psychiatrists working at eighteen different university hospitals chosen from a list of members of the Japanese Society of General Hospital Psychiatry. These psychiatrists were then requested to have up to five members of their departments who do C/L work fill out the questionnaires. These psychiatrists were known by one of the authors (T.H.) to be active in seeing patients on medical services but otherwise there was no knowledge of their attitudes toward medical patients with suicidal ideation, physician assisted suicide or euthanasia.
Fifty-nine data bases were returned from fifteen of the hospitals within a month and on re-mailing after two months one more hospital responded for a total of sixty-two data bases (89.3% male and 10.7% female, average age of 36.3+.9 years, range 25 to 55 years). The mean number of years of exprience of the Japanese psychiatrists was 10.2+6.52 years (range 1 to 29 years). Number of years of experience was found to correlate with age (r=.92, t=18.3, p=.0001).
Average years of experience as well as average age significantly differed between the U.S. and Japanese psychiatrists (for years of experience t=5.9, p<.001, for age t=5.0, p<.001). Because age was not found to correlate with any other variable on the SAI in the U.S. group, it was felt that there was no need to age-match the groups. For years of experience, although there was a low but significant relationship between years of experience and increased agreement on questions 3d (R=.25, p=.045) and 3f (R=.26, p=.034), because this correlation would have actually weakened any difference between the Japanese and U.S. groups, this data was included in the analysis. Breakdown by sex or hospital was not examined due to the small number of respondents from each hospital and the low ratios of female respondents in both groups.
For this study, we developed the Suicidal Attitudes Inventory (SAI) data base [22]. The questions were designed to tap the issues discussed in the introduction and were based on prior surveys and literature on these topics [13-19-20]. The SAI is as follows:
SUICIDAL ATTITUDES INVENTORY
This questionnaire is aimed at understanding the attitudes of psychiatrists working in general hospitals toward medical/surgical inpatients who exhibit suicidal ideation or behaviors while in the hospital for a medical/surgical illness. Your answers should be directed towards patients with medical or surgical illnesses who exhibit suicidal ideation or behaviors to staff or family. Do not include those patients brought to or admitted for the medical consequences of a suicidal attempt. Please circle the choice or fill in the circle that best fits your answer, or fill in the blank where appropriate.
Your name:________________________ Age________ Sex________
Hospital name:__________ Number of years experience as a psychiatrist_______
Suicidal Attitudes Inventory (SAI)
1.) Do you feel there may be times when suicidal ideation is logical or reasonable?
yes no If yes, what percent of the time do you think this occurs at your hospital?
<10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100
2.) Do you feel there may be times when commiting suicide is logical or reasonable?
yes no If yes, what percent of the time do you think this occurs at your hospital?
<10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100
3.) In the list below, in what cases do you feel that commiting suicide is logical or reasonable? Put an "x" next to those cases that apply. Use one x for "feel slightly" two x's for "feel, moderately", and three x's for "feel strongly".
4.) In the list below, in what cases do you feel that a physician assisting the suicide (assuming legal restrictions and safeguards are in place) is logical or reasonable? Put an "○" next to those cases that apply. Use one ○ for "feel slightly" two ○'s for "feel, moderately", and three ○'s for "feel strongly".
(Assuming the patient is competent and knows their diagnosis and prognosis.)
________Terminal illness with poor quality of life no major depression.
________Terminal illness with poor quality of life has major depression.
________Terminal illness with fair to good quality of life, no major depression .
________Terminal illness with fair to good quality of life, has major depression.
________Patient is not terminal but will be a burden on family for care.
________Patient is not terminal but has been unable to fulfill responsibilities at work.
________Patient is not terminal but has been unable to fulfill responsibilities with
family.
________Patient is not terminal but feels they should take responsibility for a superiors' failure.
________Patient is not terminal but they cannot care for their children.
________Patient is not terminal but they cannot care for their children, and it would also be logical or reasonable to commit suicide along with their children ("Shinju" suicide).
5.) A competent non-depressed patient with an incurable though non-terminal illness (ie.-quadreplegia due to an accident) who requires life support (ie.-ventilator or tube feeding) requests termination of this support because of poor quality of life. You would describe termination of the medical care by the physician as (circle one) A: suicide,
B: killing, C: allowing the patient to die naturally.
6.) Suicidal ideas among the terminally ill are linked to psychiatric disorder.
rarely sometimes usually always
7.) Medically ill patients who desire death have distorted thinking.
rarely sometimes usually always
8.) Before acceding to a medical patients wish to refuse life saving treatment it is important to give optimal medical and psychiatric treatment (ie. treat pain and depression)
rarely sometimes usually always
9.) It is valuable to accept a medical patients decision to die even if they are depressed.
rarely sometimes usually always
10.) Psychiatric patients wishes to die are rooted in psychopathology.
rarely sometimes usually always
11.) Medical patients wishes to die are rooted in psychopathology.
rarely sometimes usually always
12.) Withholding a patients' diagnoses can cause suicidal
ideation.
rarely sometimes usually always
13.) Informing patients of a diagnosis of cancer or other
terminal illness can cause suicidal ideation.
rarely sometimes usually always
The SAI given to the Japanese psychiatrists was in Japanese, and was reviewed for consistency with the English translation by three bilingual psychiatrists (D.B, T.H, and Y.T.). To avoid bias, we did not use the term "euthanasia" in the data base. Anonymity was optional.
Data was analyzed as described in the results section. P values of less than 0.05 were considered as statistically significant. N's in the results section that do not add up to the total N of respondents indicates no response on that item.
RESULTS
Summary of responses on the SAI for both the Japanese and American psychiatrists are presented in Table I, and the Chi-squared statistics for those items showing statistically significant differences are presented in Table II.
While there was remarkable similarity on questions 1 and 2 which inquired about whether there might be times when suicidal ideation or commiting suicide could be reasonable, significant differences appeared when the reasonability of commiting suicide was examined under various social, medical, and psychiatric conditions (question 3, b through j), with more Japanese psychiatrists responding with some agreement to the reasonability of commiting suicide in these cases.
Attitudes towards physician assisted suicide (question 4) did not differ. Only on item 4a, terminal illness with poor QOL (quality of life) and no major depression, was there any considerable agreement for both groups on this being logical or reasonable (42% of Japanese and 57% of U.S. psychiatrists responding with some agreement). While not significantly different (p=.09), it was noted that 6.5% of Japanese and 0% of U.S. psychiatrists responded with some agreement to item 4j, that physician assisted suicide that included suicide with one's children could be reasonable in non-terminal cases unable to care for the children.
There was a trend on question 5 for more U.S. psychiatrists to respond that termination of life supports was "allowing the patient to die naturally" and for more Japanese psychiatrists to respond that this was a "killing", although this did not reach statistical significance (p=.09).
On question 8, 40% of Japanese psychiatrists compared to18% of U.S. psychiatrists responded that it is "sometimes" or "usually" important to give optimal medical and psychaitric treatment before acceding to a medical patient's wish to refuse medical life saving treatment. The U.S. responses weighed higher on the "always" response, 83% vs 60% of Japanese.
On question 10, 84% of Japanese psychiatrists felt that psychiatric patients' wishes to die were "usually" rooted on psychopathology, while the U.S. psychiatrists split this between "sometimes" (42%) and "usually" (46%), and this difference in response pattern reached statistical significance.
The responses to question 13 were also significantly different. Although 60% and 66% respectively of U.S. and Japanese psychiatrists responded "sometimes" to the statement that informing patients of a diagnosis of cancer or other terminal illness can cause suicidal ideation, 5% of Japanese psychiatrists compared with 39% of U.S. psychiatrists responded "rarely", and 29% of Japanese, and only 2% of U.S. psychiatrists responded "usually" or "always" to this item.
Significant differences in the patterns of response were not found for items 6,7,9,11, and 12.
Comparison testing among the questions on the SAI for both groups was done in order to study those factors that had significant relationships. A description of those items with significant findings on this analysis is presented below.
On question 5, the Japanese psychiatrists who responded that termination of life support in an incurable non-terminal patient with poor QOL who requests termination is a "suicide"had greater degrees of agreement on question 3g (that commiting suicide is logical or reasonable in cases where a non-terminal patient cannot fulfill family duties) compared with those who labeled termination of life support as a "killing" or "letting the patient die naturally" . (ANOVA p=.03). For the Americans, however, a response of "suicide" question 5 was significantly correlated with greater degrees of agreement on question 4b, 4c, and 4d (that physician assisted suicide could be reasonable in, 4b: terminally ill with low QOL and major depression [ANOVA p=.0001], 4c: terminally ill with good QOL and no major depression [ANOVA p=.0009], 4d: terminally ill with good QOL and major depression [ANOVA p=.003]).
For question 8, Japanese psychiatrists responding that it is "sometimes" important to give optimal medical and psychiatric treatment before acceding to a medical patients wish to refuse life saving treatment had significantly greater degrees of agreement on question 4 items f,g,h,i (that physician assisted suicide is reasonable in situations related to failure in duty to others) than those subjects who responded "usually" or "always" on this item (ANOVA p=.03). For the Americans, those responding "sometimes" on question 8 had significantly greater degrees of agreement on question 3a (commiting suicide is reasonable in terminal illness with low QOL and no major depression [ANOVA p=.01]) and question 4a (that physician assisted suicide is reasonable in terminal illness with low QOL and no major depression [ANOVA p=.04]) compared to those subjects who responded "usually" on this item.
On question 9 (it is valuable to accept a medical patient's decision to die even if depressed), the American psychiatrists who responded "sometimes" to this item had increased agreement on the following items compared to those who responded "rarely": 3a (reasonable suicide in terminal illness with low QOL and major depression [ANOVA p=.007]), 4b (terminally ill with low QOL and major depression [ANOVA p=.008]), 4c (terminally ill with good QOL and no major depression [ANOVA p=.001]), 4d (terminally ill with good QOL and major depression [ANOVA p=.014]). There was no significant correlation found on question 9 in the Japanese group.
On question 11, the Japanese psychiatrists who responded that medical patients wishes to die are "rarely" rooted in psychopathology had significantly greater degrees of agreement on question 4e (that physician assisted suicide is reasonable in cases where a non-terminal patient is a burden on family for care) than those subjects who responded "sometimes" or "usually" on this item (ANOVA p=.047). For the Americans, increased degrees of agreement on question 11 was correlated with increased degrees of agreement on question 10 (psychiatric patients wishes to die are rooted in psychopathology [ANOVA p=.0001]).
Years of experience had a number of correlations for both groups. The Japanese who responded on question 5 that termination of life support in an incurable non-terminal patient who requests termination is a "suicide" had a significantly greater number of years experience as a psychiatrist (20+4.8) compared with those who labeled termination of life support as a "killing"
(11.7+5.1), or those labeling this as "letting the patient die naturally"(8.1+5.7) (ANOVA p=.0001.). Also on question 11, Japanese responding that medical patients wishes to die are "rarely" or "sometimes" rooted in psychopathology had a significantly lower number of years of experience (8.7+5.8) than those subjects who responded "usually" on this item (years experience 12.8+6.8), [t=2.54, p=.014].
For the Americans, years of experience correlated positively with degree of agreement to questions 3d (that suicide could be reasonable in terminal illness with good QOL and major depression [r65=.25, p=.045], 3f (that suicide could be reasonable in a nonterminal patient unable to fulfill work duties [r65=.26, p=.03], and 4b (that physician assisted suicide could be reasonable in terminal illness with poor QOL and major depression [r65=.33, p=.007].
