|
|
| |||||||||||||||||||||||||||||||||
Background Since the Oregon Death with Dignity Act was passed in November 1994, physicians in Oregon have faced the prospect of legalized physician-assisted suicide. We studied the attitudes and current practices of Oregon physicians in relation to assisted suicide.
Methods From March to June 1995, we conducted a cross-sectional mailed survey of all physicians who might be eligible to prescribe a lethal dose of medication if the Oregon law is upheld. Physicians were asked to complete and return a confidential 56-item questionnaire.
Results Of the 3944 eligible physicians who received the questionnaire, 2761 (70 percent) responded. Sixty percent of the respondents thought physician-assisted suicide should be legal in some cases, and nearly half (46 percent) might be willing to prescribe a lethal dose of medication if it were legal to do so; 31 percent of the respondents would be unwilling to do so on moral grounds. Twenty-one percent of the respondents have previously received requests for assisted suicide, and 7 percent have complied. Half the respondents were not sure what to prescribe for this purpose, and 83 percent cited financial pressure as a possible reason for such requests. The respondents also expressed concern about complications of suicide attempts and doubts about their ability to predict survival at six months accurately.
Conclusions Oregon physicians have a more favorable attitude toward legalized physician-assisted suicide, are more willing to participate, and are currently participating in greater numbers than other surveyed groups of physicians in the United States. A sizable minority of physicians in Oregon objects to legalization and participation on moral grounds. Regardless of their attitudes, physicians had a number of reservations about the practical applications of the act.
If enacted, the Oregon law will allow an attending physician to prescribe a lethal dose of medication requested by a terminally ill patient for self-administration.1 Only competent adult residents of Oregon who are expected to live less than six months are eligible. The law specifies that the patient must make two oral requests and one written request during a 15-day period. A second physician must confirm the diagnosis, the patient's decision-making capacity, and the voluntary nature of the request. Referral to a mental health professional is required if the patient's judgment appears to be influenced by depression or some other mental disorder. The physician must ask the patient to disclose the decision to family members, but the patient may refuse to do so. Physicians must report their participation in assisted suicide to the state health division, but they are protected from professional and legal liability.
Although there is growing public acceptance of assisted suicide,3 many professional organizations remain opposed to it.4-6 Surveys reveal, however, that practicing physicians are divided on the issue. In U.S. studies, 31 to 54 percent of physicians polled have expressed neutral or positive attitudes toward legalizing physician-assisted suicide or euthanasia.7-10
Since the 1994 election, physicians in Oregon have faced the strong possibility of practicing in the first state where physician-assisted suicide is legal. We conducted a survey of physicians who would be eligible to prescribe a lethal dose of medication under the new law. Our objectives were to describe the personal and professional characteristics associated with particular attitudes toward physician-assisted suicide; current practices in the care of terminally ill patients, including responses to requests for assisted suicide and interaction with other health care professionals; and practical issues of concern in implementing the Oregon Death with Dignity Act.
Methods
The study was designed as a cross-sectional mailed survey. We purchased a list of all licensed physicians practicing in Oregon from the Oregon State Board of Medical Examiners (BME) in November 1994. The Oregon Death with Dignity Act defines an attending physician as "the physician who has primary responsibility for the care of the patient and treatment of the patient's terminal disease." For the purpose of the study, we defined "attending physicians" as all physicians in Oregon licensed in the following specialties: internal medicine, family practice, general practice, neurology, gynecology, therapeutic radiology, and surgery.
We developed a questionnaire after reviewing earlier surveys7-16 and issues raised during debates about the Oregon initiative and plans for its implementation. The questionnaire was reviewed by a national panel of experts in bioethics and survey research, then tested in a pilot study with a sample of 65 Oregon physicians randomly selected from the BME list of physicians in the specialties noted above.
The questionnaire contained 56 questions. The characteristics of the respondents were determined by 28 closed-ended questions. Attitudes and practices were determined by the degree of agreement with 27 statements, based on a five-point scale. Information about the medications that respondents would prescribe was elicited with an open-ended question. To be consistent with the language of the Oregon Death with Dignity Act, we defined participation in physician-assisted suicide as writing a prescription for a lethal dose of medication ("lethal prescription") that a terminally ill patient intends to use to end his or her life.
After the study had been approved by the Oregon Health Sciences University Human Subjects Committee, the survey was mailed to the 4046 eligible physicians on March 31, 1995. A reminder was mailed two weeks later, followed by a second questionnaire mailed to nonresponders one month after the initial mailing. All responses received before July 1, 1995, were included in the analysis. The physicians were assured that their responses would be confidential and anonymous. The questionnaire contained no numbers or other information that could link the responses to specific persons. Nonresponders were tracked by the use of code numbers preprinted on return envelopes. When each questionnaire was returned, a research assistant immediately removed it from the envelope, logged in the code number on the envelope, and then destroyed the envelope. The questionnaires were then assigned consecutive identification numbers for the purpose of analysis.
Statistical Analysis
Responses were entered into the computer with an Optical Mark Reading scanner (OpScan 5 with Scantools software; National Computer Systems, Edina, Minn.). Data entry was verified manually on a randomly selected set of 5 percent of the questionnaires, and no scanning errors were detected.
Statistical analyses were performed with version 6.09 of the Statistical Analysis System software and JMP 3.1 Statistical Software (SAS Institute, Cary, N.C.). After assessing the frequencies, we collapsed the Likert-scale ratings into three categories: strongly disagree or disagree, neutral, and agree or strongly agree. On the basis of discussions with religious leaders in our state, religious denominations were categorized as Catholic, Protestant, other Christian, Jewish, and other non-Christian. Differences between subgroups of physicians defined on the basis of characteristics, attitudes, and practices were assessed with the use of chi-square tests and analyses of variance as appropriate. We used univariate and multiple-logistic-regression modeling to analyze associations between characteristics of the physicians and their attitudes toward assisted suicide.
Results
Characteristics of the Physicians
Of the 4046 physicians sent questionnaires, 102 were ineligible for the survey: 52 had moved out of state, 36 had stopped practicing in the designated specialty and had begun to practice in an excluded specialty, 8 had left no forwarding address, and 6 had died. Of the remaining 3944 physicians, 2761 (70 percent) returned the questionnaire. The demographic and professional characteristics of the responding physicians are shown in Table 1. Their mean age was 47 years, and most (82 percent) were men. Ninety percent of the respondents had completed training and were in practice at the time of the survey, and in the preceding year, 81 percent had cared for at least one patient with a life expectancy of less than six months. On the basis of a comparison with 1993 Oregon BME data, the sample was representative of physicians in Oregon in terms of age, sex, years since graduation from medical school, and specialty distribution.
|
The statement that competent, terminally ill patients have a right to commit suicide was endorsed by 73 percent of the respondents (Figure 1). Sixty-six percent said that physician-assisted suicide would be ethical in some cases, and 60 percent said it should be legal in some cases. Approximately a third of the respondents agreed with the statements that physician-assisted suicide would be immoral (33 percent), would violate professional ethics (34 percent), and would violate personal religious beliefs (30 percent). Most respondents thought that some patients might request physician-assisted suicide because of concern about being a burden to others (93 percent) or financial pressure (83 percent). A minority of the respondents (29 percent) thought that legalizing physician-assisted suicide could result in lethal overdoses being given to patients without their consent.
|
|
The characteristics of the physicians who might be willing to participate in physician-assisted suicide, if it were legal to do so, were compared with the characteristics of those who would not be willing to participate (Table 3). These characteristics were then added to a multiple-logistic-regression model that included the following variables: religious affiliation (reference category, Protestant), specialty (reference category, internal medicine), size of practice locale (reference category, large city or suburb), age, and sex. Variables significantly associated with a willingness to participate in physician-assisted suicide included Jewish affiliation (odds ratio, 2.94; 95 percent confidence interval, 2.10 to 4.11), no religious affiliation (odds ratio, 2.87; 95 percent confidence interval, 2.34 to 3.53), other non-Christian affiliation (odds ratio, 1.92; 95 percent confidence interval, 1.29 to 2.85), and age (a 13 percent increase in willingness to participate for each 10-year increase in age). An unwillingness to participate in physician-assisted suicide was associated with a Catholic affiliation (odds ratio for willingness, 0.43; 95 percent confidence interval, 0.32 to 0.57) or other Christian affiliation (odds ratio for willingness, 0.32; 95 percent confidence interval, 0.18 to 0.56) and a small-town or rural locale (odds ratio for willingness, 0.67; 95 percent confidence interval, 0.53 to 0.84). Sex, specialty, and whether or not the respondent cared for terminally ill patients were not significant predictors of the willingness to participate in physician-assisted suicide.
|
The 1874 physicians who were not morally opposed to physician-assisted suicide were asked about practical issues of concern regarding writing a prescription for a lethal dose of medication (Figure 2). Fifty-three percent of the respondents were concerned that the patient's family might sue, and 51 percent were concerned about the possible harm if an attempt failed or a complication developed. Some respondents were concerned that someone other than the patient might use the prescription (33 percent) or that writing it might violate federal laws governing the prescription of drugs (25 percent), jeopardize their license to practice in another state (23 percent), or lead to sanctions by hospitals (24 percent) or ostracism by colleagues (25 percent).
|
Current Practices
A total of 570 respondents (21 percent) said that they had been asked for a prescription for a lethal dose of medication within the preceding year (Table 4), and 187 (7 percent) stated that they had written such a prescription before the passage of the Oregon measure. Most of these physicians (124) stated that the patients for whom they had written the prescriptions had taken the medication.
|
A total of 759 respondents (27 percent) had discussed a patient's request for a lethal prescription with another physician, a nurse, a social worker, or a member of the clergy. A majority (1681, or 61 percent) believed other providers should be informed only at the patient's request, although 874 (32 percent) thought other providers have the right to know, whether the patient agrees or not. A total of 149 physicians had been approached by other professionals in the preceding year to discuss a patient's request for a lethal prescription.
Discussion
The high response rate (70 percent) in this survey attests to the importance of this topic to physicians in Oregon, who have practical issues about legalizing physician-assisted suicide. The results reveal a greater acceptance of physician-assisted suicide among Oregon physicians than among other groups of physicians in previous U.S. studies. Sixty percent of doctors in Oregon believe that physician-assisted suicide is ethical and should be legal in some cases, and nearly half might be willing to write a prescription for a lethal dose of medication if it were legal to do so. this support for the concept of assisted suicide, however, is counterbalanced by concern about its practical application. Strikingly, 50 percent of the respondents were not sure what they would prescribe if they decided to comply with a patient's request for a lethal dose of medication. This uncertainty raises grave questions about the potential for incomplete suicides in the absence of reliable prescribing information. Currently, there are no published data on the effectiveness of drugs and doses when used orally as the sole means of committing suicide. In Oregon, only the Hemlock Society has stepped forward as a potential resource for pharmaceutical information. The Oregon Death with Dignity Act would not require a physician to be present when the patient takes the medication and would forbid the administration of a lethal injection if ingestion of the medication does not result in death. In contrast, both in the Netherlands and in Australia's Northern Territory, physicians are expected to be present at the time the patient ingests the medication and are permitted to administer a lethal injection if the oral dose does not result in death.17,18 The results of our survey suggest that inadequate consideration has been given to these issues to date.
Most of the respondents believe that at times patients are motivated to request assisted suicide because of financial pressure or fear of becoming a burden to loved ones. It has been argued that in the United States, unlike Australia and the Netherlands, which have national health insurance, patients who are unable to pay for their medical care may be motivated by financial constraints to choose physician-assisted suicide over palliative care.19-22 There are 418,000 medically uninsured citizens in Oregon.23 At present, physicians in Oregon have no guidelines for responding to a patient's request for assisted suicide if they think the request is prompted by financial pressure.
Half the respondents in our study are not confident they could predict that a patient had less than six months to live. This uncertainty probably reflects the lack of clear criteria for estimating life expectancy,24-26 but also points to the gap between the measure's requirements and physicians' assessment of their skills. In addition, one third of the respondents are not confident they could recognize depression in a patient asking for a lethal dose of medication. Previous research has shown that primary care physicians overlook depression about 50 percent of the time.27,28 If laws allowing assisted suicide require physicians to confirm a short life expectancy or to identify depression, these gaps between expectations and skills must be bridged.
The finding that a substantial minority of physicians (32 percent) believe that other health care providers have a right to be informed of a patient's wish to commit suicide, whether or not the patient agrees, raises an additional question about assisted suicide. The Oregon Death with Dignity Act allows the attending physician to refuse to comply with a request for assistance but is not specific about the right of other professionals caring for the patient either to know the patient's plan or, if they do know, to withdraw from the case. If physician-assisted suicide is legalized, both the patient's right to confidentiality and the health care provider's right to refuse to participate in an act he or she finds morally unacceptable must be protected.29,30
We could find only one previous report of the prevalence of physician-assisted suicide in the United States.9 In that study, as in ours, one fifth of the physicians surveyed had previously been asked by patients for a prescription for a lethal dose of medication. Two percent of the respondents in that study had written such a prescription for at least one patient, as compared with 7 percent of the respondents in our sample.
The results of our study show that religious beliefs strongly influence the willingness to participate in physician-assisted suicide. Similar associations have been demonstrated previously.10 We also found that physicians practicing in small towns or rural communities were less likely to be willing to participate in physician-assisted suicide than those practicing in cities. In Oregon, 62 percent of the populace resides in rural communities and towns with populations under 25,000, and close to a third of the population resides in the Portland metropolitan area.31 The finding that the location of physicians' practices predicts their willingness to participate in physician-assisted suicide suggests that threats to confidentiality, lack of anonymity, and social disapproval may make such participation riskier for physicians, as well as for patients and their families, in smaller communities. Finally, the association between the increasing age of physicians and their willingness to participate in physician-assisted suicide is consistent with the findings reported by one group of investigators8; others, however, have found no such correlation.10
The main limitation of our study is the lack of information about nonresponders. The large sample, the high response rate, and the representative distribution of the responders in terms of sex, age, and specialty lend credibility to the findings, yet we do not know whether the 30 percent of physicians who did not respond to the questionnaire held views or had experiences similar to those of the respondents. In this survey, unlike previous studies, we limited the questions to physician-assisted suicide and excluded questions about euthanasia (lethal injection administered by someone other than the patient). Finally, it may not be appropriate to generalize these findings to areas of the United States where the population is more concentrated in urban areas than it is in Oregon, where demographic characteristics are more diverse, or where physician-assisted suicide has received less attention in the media.
The findings of our survey raise several issues in the debate about legalizing physician-assisted suicide in this country. Reliable information on oral medications and doses is not readily available to physicians.19 Many of the respondents in our survey were uncertain about their ability to predict accurately whether a patient had less than six months to live and their ability to recognize depression — two requirements of the Oregon act. Because comprehensive care for dying patients is not universally available, many of the respondents fear its absence may create financial incentives for choosing physician-assisted suicide.30 The respondents expressed divergent views about how and when to protect both a patient's right to privacy and a provider's right to refuse to participate in a practice he or she finds morally objectionable. Although the ultimate fate of the Oregon Death with Dignity Act is unknown, its approval by voters has raised practical issues for policy makers to consider.
Supported by grants from the Meyer Memorial Trust and Denison Family Fund–Oregon Community Foundation. The views expressed in this article are those of the authors and do not necessarily reflect the views of the sponsors.
We are indebted to the Oregon Medical Association for endorsing this survey; to Annette Matthews for assistance with the design and layout of the questionnaire and with data management and analysis; to Jonathan Fields, M.S., for statistical assistance with the questionnaire, study design, and data analysis; and to Carol Stocking, Ph.D., for reviewing the questionnaire.
Source Information
From the Departments of Medicine (M.A.L., H.D.N., S.W.T.), Psychiatry (L.G.), and Emergency Medicine (T.A.S.), School of Medicine; the Office of Research, School of Nursing (V.P.T.); the Center for Ethics in Health Care (M.A.L., V.P.T., L.G., T.A.S., S.W.T.); and the Biomedical Information Communication Center (H.D.N.), Oregon Health Sciences University, Portland.
Address reprint requests to Dr. Lee at the Center for Ethics in Health Care (L-101), Oregon Health Sciences University, Portland, OR 97201.
References
| |||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
