Attitudes of Tennessee physicians toward euthanasia and assisted death. (Original Article).

Background: Although many studies of euthanasia and physician-assisted death (PAD) have been performed in the United States, none have specifically addressed attitudes among physicians practicing in Tennessee.

Methods: In January 2001, we mailed a 30-item survey instrument to a stratified random sample of 1,117 physicians drawn from the Tennessee Licensing Bureau.

Results: Tennessee physicians are highly polarized over the issues of euthanasia and assisted death. A slight majority (47%) did not favor euthanasia or PAD and would oppose the legalization of such procedures. Of the physicians supporting euthanasia or PAD (43%), only 25% would administer a lethal overdose and less than a third would counsel/prescribe medication for an overdose. Attitudes were influenced by three primary factors: ethics, religion, and the role of the physician to relieve pain and suffering.

Conclusion: Regardless of their overall position, the majority of physicians agreed on basic restrictions and safeguards to prevent abuses and to protect vulnerable patients.

**********

Key Points

* Attitudes toward euthanasia and PAD were influenced by three primary factors: ethics, religion, and the role of the physician to relieve pain and suffering.

* The majority of physicians agreed on basic restrictions and safeguards to prevent abuses and to protect vulnerable patients.

* A slight majority (47%) did not favor euthanasia or PAD and would oppose legalization.

* Of the physicians supporting euthanasia and PAD (43%), only 25% would administer a lethal overdose and less than a third would counsel/prescribe medication for an overdose.

The question whether a physician should assist patients with an intolerable terminal condition to end their life has been debated since the practice of medicine began with Hippocrates. The debate surrounding physician-assisted death (PAD) and voluntary euthanasia is still alive. (1)

In the United States, patients now generally have a legal right to refuse treatment, and physicians can honor advance directives, living wills, and surrogate decisions. Physicians currently withhold or withdraw treatment such as respiratory support, cardiopulmonary resuscitation, dialysis, and sometimes nutrition and hydration, thereby allowing patients to die. Studies indicate that, as of 1993, approximately half the deaths in nonemergency hospital settings involved some form of withholding or withdrawing treatment. (2) Patients with living wills and end-of-life directives have had these instruments made with the expectation that their wishes will be acted on by the physician. Persons with such a document are concerned about losing control, being a burden, being dependent, and losing dignity, in addition to an existence in an irreversible vegetative state of unconsciousness.

By many indications, American society is ready for the legalization of PAD and euthanasia. A 1991 General Social Survey conducted by the National Opinion Research Center asked the following question: When a person has an illness or disease that cannot be cured or is in a permanent vegetative state, do you think doctors should be allowed by law to end the patient's life by some painless means if the patient and his or her family request such measures? More than 70% of those queried responded "yes." A Gallup poll in 1994 indicated that 65% of the U.S. public is in favor of allowing doctors to help the terminally ill end their suffering. (3) In a Harris poll conducted in 1999, 73% of respondents agreed that physicians should, without penalty, be allowed to assist the terminally ill with their death. The fact is that many patients now want to trust that their physician will stay with them and not abandon them when the only way out of their suffering is to help them die as they choose, with dignity. (2)

The Declaration of Lisbon, adopted in 1981, details some of the ways in which human rights can be translated into the patient-physician interaction. Patients have the right to choose their physicians, to receive adequate information and to accept or refuse treatment, to have care based on judgments free from outside interference, to have confidences kept, to receive or decline spiritual and moral comfort, and to die with dignity.

Referring to the Pythagorean priesthood, on which the Hippocratic Oath is based, one finds clear injunctions against abortion and euthanasia (poisons), and for the keeping of confidences and the father-son relationship of the teacher and student. The Hippocratic Oath states, "I will neither give a deadly drug to anybody if asked for it, nor will I make a suggestion to this effect. Similarly, I will not give to a woman an abortive remedy. I will apply dietetic measures for the benefit of the sick according to my ability and judgment; I will keep them from harm and injustice." (4)

Herein the conflict arises: the autonomy of the individual, not giving a deadly drug or abortive remedy, yet keeping them from harm and injustice. The ruling in Roe v. Wade sustaining a woman's right to an abortion (ie, the "right to choose") is noted here because it is specifically prohibited in virtually all medical oaths. Yet, physicians whose attitude or belief could be that they are keeping patients from harm or injustice are involved in abortion regularly by performing the procedure. The principles used to justify abortion could be used by the physician to justify other forms of mercy killing such as euthanasia and PAD. Arguments that the patient is medically compromised, brain damaged or in a comatose permanent vegetative state, or has no quality of life can be applied to the taking of life in utero and ex utero.

The U.S. Court of Appeals for the Ninth Circuit pointed out that abortion and assisted suicide share a common rationale. That rationale is found in the Due Process Clause of the Fourteenth Amendment, which states, in part, "No State shall deprive any person of life, liberty, or property without due process of law." Citing abundant U.S. Supreme Court precedent, the lower court pointed out that liberty is an evolving concept whose content cannot be limited by historical understanding or customary usage. Although the specific content of one's liberty at any given time may be difficult to assess, the choices central to personal autonomy are also central to liberty under the Fourteenth Amendment. A right of autonomy broad enough to cover a woman's right to abort, declared the Ninth Circuit, is broad enough to cover a terminally ill person's fight to determine the time and manner of death. Judge Reinhardt of the Ninth Circuit issued a 109-page opinion in 1996 on assisted suicide and abortion. (5) In his opinion, he stated, "Our conclusion is strongly influenced by, but not limited to, the plight of the mentally competent, terminally ill patient. We are influenced as well by the plight of others, such as those whose existence is reduced to a vegetative state or a permanent and irreversible state of unconsciousness." Judge Reinhardt acknowledged that he was "marvelously struck by the compelling similarities between abortion and assisted suicide: both involve matters of life and death, both arouse similar moral and religious passions, and in both, the strength of the State's interest may vary with the circumstances and both raise the fundamental questions about an individual's right of choice." Judge Reinhardt notes one other similarity: "As with abortion before legalization, assisted suicide is widely although secretly practiced."

When facing the ethical and moral decision to assist a patient with death or to administer a lethal overdose at the request of the patient, how will the physician react? Will the physician support the patient's wishes in end-of-life decisions or be guided by his or her own ethical and moral values?

Studies performed in Hawaii, (6) Massachusetts, (7) Michigan, (8) Oregon, (9) South Carolina, (10) and Washington state (11) indicate that the majority of physicians are agreeable to forms of assisted death and euthanasia. However, none of these studies have been directed at physicians practicing medicine in Tennessee, indicating that research was needed to determine the attitudes of Tennessee physicians on euthanasia and PAD.

Methods

Study Population

As of January 2000, Tennessee had 12,396 licensed physicians actively practicing medicine in a variety of specialties. A computer disk with lists of names and addresses of all licensed physicians in five selected specialties (internal medicine, family practice, general surgery, general practice, and oncology) was obtained from the Tennessee Licensing Bureau. These areas were selected because previous surveys have used similar populations. A stratified random sample was selected from the list of licensed physicians in those areas for a total population of 1,117. Every nth physician on the list was selected to obtain the required sample size in each stratum. The Small Sample Method (12) was used to determine the required sample sizes. At the 95% confidence level (ie, [+ or -]5% sampling error) with a 50-50 split, the required sample sizes were 324 internists, 304 family practitioners, 240 general surgeons, 169 general practitioners, and 80 oncologists.

Instrument

The instrument used in this survey was a modified version of the questionnaire developed by Dickinson et al (10) and used with their permission. The instrument used to gather data for this survey had 30 items divided among four sections. The first section consisted of demographic information, which included age, sex, area of medical specialty, and number of terminally ill patients (if any) seen during the preceding month. Section 2 consisted of six items regarding overall position about helping critically ill patients to end their lives. The third section--reason for position--comprised eight items. The final section used 12 items to determine restrictions, stipulations, and safeguards. Sections were responded to using a 5-point Likert scale with the available responses being SA, strongly agree; A, agree; N, neutral; D, disagree; and SD, strongly disagree.

Administration

The survey was conducted in phases during January 2001. A computer was used to generate a systematic, stratified random sample and produce the mailing list and labels. An identifying letter was sent to the physicians informing them of the survey, and the survey form with cover letter and return stamped envelope were mailed 5 days later. Thirty days after mailing the survey, it was concluded that all who chose to participate had done so by returning a completed instrument. To ensure anonymity, there was no tracking information on any of the correspondence.

Analysis of Data

Both descriptive and inferential statistics were used to analyze the data. Data were transferred from each questionnaire and entered into a database using SPSS software (SPSS, Inc., Chicago, IL). Data were initially broken down into categories to determine the percentage of total responses received, the percentage of each group responding, and group-specific and overall percentages of male and female respondents. The [chi square] test was used to determine whether significant associations existed between sex, age, and/or religious influence, and physicians' willingness to assist in a patient's death. Analysis of variance was used to test the significance of the differences between the mean scores of specialties of physicians on items pertaining to their willingness to support and/or engage in euthanasia or physician-assisted suicide. The level of significance for this analysis was P = 0.05.

Results

The results of the study are divided into four sections. The first section addresses demographic data such as overall response rate by specialty, age, sex, and number of terminally ill patients seen during the preceding month. Section two addresses physician responses toward euthanasia and assisted death. The next section summarizes the relationships between age and religion and the administration of a lethal overdose and/or prescribing medication and counseling to end the life of the terminally ill patient. In the last section, the self-reported responses to the legalization of euthanasia and assisted death are presented, along with an analysis of restrictions, guidelines, and stipulations considered important by the responding physicians.

Demographic Information

Of the 1,117 Tennessee physician-respondents, 42 were found to be ineligible by having moved out of the state or leaving no forwarding address. Of the remaining 1,075 instruments, 414 were returned, and 49 of these were ruled ineligible because of retirement (n = 4), failure to complete the instrument (n = 27), or failure to return the survey within the established time frame (n = 18), leaving a total of 365 (34%) usable data collection instruments. Table l presents a summary of instruments mailed with the corresponding return rate.

The ages of the participants ranged from 28 to 87 years, with a mean age of 48.7 years (Table 2). Of the eligible respondents, 84.7% were male and 15.3% were female. Table 3 presents the breakdown of each specialty by sex. A comparison of respondent characteristics from research by Dickinson et al (10) in South Carolina and Cohen et al (11) in Washington state with those of the Tennessee physicians is shown to demonstrate similarities.

The number of terminally ill patients seen by the physicians over the preceding month ranged from 0 to 30, with two oncologists reporting having seen more than 50 terminally ill patients within that time frame. The mean number was 5.9 terminally ill patients seen during the previous month. Table 4 illustrates the number of terminally ill patients seen by the physicians, both overall and by medical specialty.

Overall Attitudes toward Euthanasia and Assisted Death

As a group, more than half of the physicians (53%) responded that it was not ethical to administer a lethal overdose (euthanasia) or to counsel the patient to use a lethal overdose (assisted death), and 36% did not think it unethical to do either. Fifty-seven percent (n = 209) agreed that the deliberate administration of an overdose of medication to an ill patient on request with the primary intent to end life (euthanasia) was never ethically justified, whereas 34% (n = 124) did not consider it unethical to administer an intentional overdose on request (Table 5). On the issue of counseling a patient in the use of a fatal overdose, 52% (n = 188) agreed that it was never ethically justified and would not provide counseling, whereas 38% (n = 137) would offer counseling (Table 6).

Physicians were almost equally divided over whether a situation could arise in which the deliberate administration of a lethal overdose or counseling a patient in self-administering a lethal overdose should be legal. As a group, 44% agreed that a situation could arise in which administration of an overdose or counseling a patient should be legal, and 46% disagreed with situational legalization. Forty-four percent (n = 160) agreed that in some situations, administering an overdose should be legal, whereas 47% (n = 172) thought intentional overdose should not become legalized (Table 7). Physicians were evenly split at 45% (164 for and 164 against) on the issue of legalization of situational counseling (Table 8).

Responding to items concerning participation in euthanasia or assisted death was optional; however, the majority opted to respond. Overall, the large majority (61%) stated they personally would neither administer an overdose nor counsel a patient on taking an overdose, whereas fewer than one third (29%) stated they would do either or both. Sixty-six percent (n = 234) would not personally administer an overdose, and 56% (n = 199) said they would not personally prescribe medication for or counsel a patient to assist a death. Twenty-nine percent (n = 90) would personally administer an overdose, and 32% would prescribe medication and/or offer counseling (Tables 9 and 10).

As a group, female physicians were slightly more likely to approve of situational euthanasia than male physicians. Male physicians were more likely to think that counseling a patient to use an overdose was never justified. Male and female physicians responded almost equally that situational counseling for use of an overdose should be legal. Male and female physicians in equal numbers responded that they would personally administer a lethal overdose (Table 11).

Reasons for Attitudes Reported

When physicians agreed that to deliberately administer an overdose or to counsel for or prescribe medication for an overdose was unethical, they were less likely to personally do either. If it were legal to administer an overdose or to counsel or prescribe medication for an overdose on a situational basis, however, the physician would be more likely to comply with the request.

Age was not a determining factor of the individual physician's willingness to participate in assisted death or to counsel/prescribe medication for an overdose. However, the younger the physician, the more receptive and likely they would be to administer an overdose or to counsel or prescribe medication for an overdose. Younger physicians were also more likely to approve of legalization for situational deliberate administration of an overdose and counseling/prescribing medication for the purpose of the patient overdosing.

Respondents who opposed euthanasia considered the practice to be inconsistent with the physician's role to relieve pain and suffering. The stronger the disagreement with euthanasia, the less concern for the autonomy and right of self-determination for the patient. Physicians who opposed euthanasia also strongly objected to legalization of assisted death and euthanasia, citing the potential for abuse.

Physicians agreed that in some situations euthanasia should be legal if requested by the patient. The two appropriate situations cited were poor quality of life and pressure from external factors. Physicians who opposed euthanasia as being unethical cited that guidelines would not be helpful in preventing or minimizing the potential for abuse (Table 12).

Effects of Legalization

Physicians were asked whether they thought supporting the legalization of euthanasia and assisted death would affect the social process of dying and grieving, reduce futile medical care, or prevent deterioration to an unacceptable quality of life for the patient. Almost half (182 of 365) of the respondents disagreed that such legalization would undermine the social process of dying and grieving. Fewer than one-third thought that legalization would have a social impact. Thirty-four percent of physicians agreed that legalization would reduce the delivery of futile medical care, and 43% responded that legalization may prevent deterioration to an unacceptable quality of life for the patient. Fifty-seven percent thought that legalization would not have an effect. Fifty-seven percent of physicians disagreed that legalization of euthanasia and PAD would reduce the delivery of futile medical care, and 49% disagreed that it would prevent deterioration to an unacceptable quality of life for the patient.

Restrictions, Stipulations, and Safeguards

Regardless of the physicians' overall position on euthanasia and assisted death, they were asked to rate the importance of guidelines a physician should consider when using an overdose (euthanasia or PAD) to end the life of a patient requesting such assistance. Forty-four percent agreed and 34% disagreed that hastening death should be restricted to adult patients. Eighty percent agreed that the patient should not be unduly distressed, and 82% agreed the patient should be mentally competent. Seventy-two percent agreed that pain should be beyond medical control and that life expectancy of the patient should be less than 6 months (Table 12).

The more the physicians agreed that legalization would prevent abuses, the more they agreed that adequate pain control but poor quality of life or not wanting to be a burden may be situations in which an overdose could be appropriate. Eighty-two percent of physicians supporting legalization also agreed that the administering physician should have an established relationship with the patient. Sixty-eight percent thought the patients' immediate families should agree with the patient's decision to die. Seventy-six percent of respondents thought that two physicians should agree with the decision of the patient to terminate life. Eighty-one percent of the physicians agreed there should be a specific waiting period between the patient's request for a drug overdose and granting such a request. The physicians were more evenly divided about whether a drug overdose should be recorded as the cause of death. Forty-one percent of the respondents agreed, 34% disagreed, and 25% were neutral. However, those supporting legal ization did not think that a drug overdose should be recorded as the cause of death (Table 13).

Findings

Physicians in Tennessee have sharply polarized attitudes toward euthanasia and assisted death. These attitudes were reflected by the responses to the ethics of euthanasia and PAD and the willingness of the physician to engage in such activity. The results are limited to physicians in the selected areas of medical practice and the honesty of their responses. Conversely, to the extent that respondents may have been reluctant to report that they might engage in euthanasia or assisted death in some situations, the actual number of physicians may be underreported.

Fifty-four percent of the physicians who responded that relieving pain and suffering was consistent with the role of the physician simultaneously responded that euthanasia and assisted death were unethical. The more the physician thought that euthanasia and assisted death were unethical, the more they opposed situational legalization (P < 0.001) for either euthanasia or assisting death. However, if it were legal to counsel or prescribe medication for an overdose (assisted death), physicians who agreed with assisted death would be likely to give counseling or prescribe medication for an overdose (P < 0.001). Of the physicians who responded that euthanasia or assisted death was ethical, only a minority expressed a willingness to participate in euthanasia (25%) and in PAD (32%).

Physicians who supported the legalization of euthanasia and assisted death stated that ending the life of a terminally ill patient was consistent with the role of the physician to relieve pain and suffering (P < 0.001). Of those responding that relieving pain and suffering were consistent with the role of the physician, the majority responded that they supported a physician's helping a patient hasten death because it may prevent deterioration to an unacceptable quality of life for the patient (59%) and because autonomy/self-determination should be respected (56%). However, physicians who responded that euthanasia was never ethically justified did not think that euthanasia was consistent with the role of the physician to relieve pain and suffering (P < 0.001), nor did they think that patient autonomy/self-determination should be respected (P < 0.001).

Physicians who opposed euthanasia as unethical would not support the legalization of euthanasia/assisted death, because they thought that guidelines would not prevent abuses (P < 0.001), and they thought that vulnerable patients might be forced into requesting euthanasia (P < 0.001). Physicians who agreed that for some patients euthanasia should be legal cited two situations in which euthanasia would be appropriate (P < 0.001): adequate pain control, but poor quality of life; and external factors (burden to family, depleting family resources).

Physicians supporting legalization did not think that vulnerable patients would be forced to participate (P < 0.001) or that legalization would undermine the social process of death and dying (P < 0.001), but agreed that legalization would reduce futile medical care and prevent deterioration to an unacceptable quality of life, and agreed that autonomy/self-determination should be respected (P < 0.001).

Physicians indicated that their positions on euthanasia and assisted death were influenced by their religion. Physicians who supported the legalization of euthanasia and assisted death stated their position was not influenced by religion (P < 0.001). Of the physicians who think that euthanasia and assisted death are never ethically justified, the majority (50%) stated that their overall position was influenced by their religion.

The legalization of assisted death and euthanasia were not favored; however, this was by a slim margin of less than 3%. Forty-seven percent were against and 44% favored legalization, and less than 10% were neutral.

Discussion

End-of-life dilemmas are common in all areas of medicine, and the majority of physicians can anticipate a request to assist with hastening the death of a patient during their career. Despite the numerous legal, ethical, and moral constraints involved, many physicians are assisting patients in the death process. The situations epitomized by Dr. Jack Kevorkian (13) and the story of "Debbie's Dying" (14) are not isolated instances of euthanasia or assisted death. As one physician anonymously stated, "Physician assisted death is a common but private occurrence."

The Tennessee study, performed in the "buckle of the Bible Belt," is a replication of the Washington study in the Northwest and the South Carolina study in the South, thus representing three geographic regions of the country. When physicians were asked whether a situation might arise in which they themselves would administer a lethal overdose, only 25% of Tennessee physicians responded that they would participate in euthanasia (58% in Washington and 29% in South Carolina). If the polarized attitudes of physicians in these states are near exemplary of those among physicians in other states, it will be difficult at best to formulate and implement laws and policies concerning assisted death and euthanasia.

These issues are currently a "close call" among physicians and the general public as noted by passage of initiative 119 in Oregon and in Michigan where 44% of physicians and 53% of the public supported some form of situational voluntary euthanasia. However, as one anonymous Tennessee physician noted on a survey response:

A much greater problem regarding terminally-ill patients are those whose family demand prolonging life (and agony) at any cost due to the family's own feelings of guilt, denial, inadequacy, etc. Euthanasia will never help these patients or their families. Our society has a deep seated fear of death and a total lack of sincere religious beliefs and until these two problems are addressed, euthanasia has little to offer most Americans. At least another 100 years will be required for the attitudes of Americans to change and accept euthanasia.

The results of this study will have important implications for patients, physicians, and policy makers. Whether euthanasia and assisted death will have a future role in the care of terminally ill patients remains an issue for continued debate. Patients and physicians will be caught up by changing technology and demographics. Patients will look to their physicians for assistance in ending pain and suffering with an assurance and finality that only death can provide. At present, the medical profession does not seem prepared to answer the questions created by their plight.

The current debate surrounding euthanasia and assisted death will continue in the literature, in the state houses, and at the patient's bedside. Information crucial for resolving this debate is the physician's attitude toward euthanasia and assisted death.

Conclusion

The majority of Tennessee physicians do not favor allowing euthanasia or assisted death and would oppose legalization. Even physicians who responded in support of allowing euthanasia and assisted death had major reservations about active participation in either type of procedure. Physician attitudes toward euthanasia and assisted death were influenced by three primary factors: ethics, religion, and the role' of the physician to relieve pain and suffering of the terminally ill patient. Although closely divided, the majority of physicians were against euthanasia or assisting death to prevent deterioration to an unacceptable quality of life for the patient, even more so when the patient had adequate pain control, but the majority of physicians did not think that external factors were justification for euthanasia or assisted death. The variables of age and sex were not factors, but the greater the influence of religion, the less likely the physician was to be supportive of or to participate in either euthanasia o r assisted death. Physicians are acknowledging growing public acceptance, if not demand, for euthanasia and assisted death and may be tactfully deciding they can deal with these issues without rather than with settled public policy; therefore, less than half (44%) supported legalization of euthanasia and assisted death. Regardless of their overall position, the majority of physicians agreed on the basic restrictions, stipulations, and safeguards to prevent abuses and protect vulnerable patients.

Recommendations

Many physicians have received or will receive a request for assisted death or euthanasia. Medical schools should allocate more time training students to relate to the very sick or terminally ill patient, including interpersonal skills and interpretation of body language. Medical education needs to assist future practitioners who may not value or understand the role of the patient in making medical decisions (autonomy) or who may be unwilling to relinquish control of the decision-making process (self-determination). Educational efforts are needed to prepare physicians for this eventuality and enable them to assess the patient's mental state, the underlying cause for such a request, and the adequacy of palliative care before responding.

More effective pain management and control of other symptoms, as well as further development of the hospice and similar programs, should be included in efforts to improve care for terminally ill and extremely ill patients. Some physicians view death as a failure and thus make a concerted effort to avoid actions contributing to death. Future research may support the thought that physicians who work closely with terminally ill patients desire to prolong life and attempt to relieve suffering by means other than death. Additional future studies should explore professionally based values, implications of withholding or withdrawing treatment, the principle of "double effect," and the recognition of and support for end-of-life directives in addition to the issue of children's rights.

Comparison of Results: Tennessee, South Carolina, and Washington

The demographic characteristics of the participants from Tennessee in this study were similar to those in the South Carolina and Washington studies. Responses to items in the instrument were also similar to those obtained by Dickinson et al (10) and Cohen et al (11) (Tables 14 and 15). Of the Tennessee physicians surveyed, 57% were in agreement that the deliberate administration of a lethal overdose was never ethically justified, compared with 48% in Washington and 44% in South Carolina. In Tennessee, 52% of participating physicians responded that it was unethical to counsel a patient in self-termination of life, whereas 50% of those in Washington and 44% of those in South Carolina considered such counseling ethical. As for legalizing the administration of a lethal overdose in certain situations, 54% of the physicians in Washington, 52% of those in South Carolina, and 44% of those in Tennessee were in favor of legalization. In Washington, 58% of physicians surveyed responded that they personally would adminis ter a lethal overdose in some situations, compared with 29% of physicians in South Carolina and 25% in Tennessee. In Washington, 53% of physicians responding agreed that in some situations it should be legal to counsel a patient in terminating life, compared with 48% in South Carolina and 45% in Tennessee.

Table 1

Summary of instrument responses by medical specialty

                   Initial   No. of
Medical            sample   surveys   Usable   Validity
specialty           size    returned  surveys    (%)

Family practice      304      121      103      28.2%
Internal medicine    324      104       97      26.6%
Surgery              240       81       70      20.3%
Other (a)                      41       38      10.4%
General practice     169       36       30       8.2%
Oncology              80       31       23       6.3%

Total              1,117      414      365      33.9%

* Change in area of specialization since list was obtained from
Tennessee Licensing Bureau.

Table 2

Ages of participants, by medical specialty

Medical               No. of     Minimum  Maximum          Mean
specialty          participants    age      age            age

General practice        30         29       74     56.10 [+ or -] 12.64
Surgery                 74         29       86     50.95 [+ or -] 11.98
Family practice        103         29       87     49.05 [+ or -] 14.23
Oncology                23         36       71     47.57 [+ or -]  8.13
Internal medicine       97         28       82     46.74 [+ or -] 12.78
Other                   38         29       69     43.00 [+ or -] 11.04

Table 3

Sex of participants, by medical specialty

                           Men                  Women

Medical
specialty          No.         (%)     No.         (%)

Internal medicine  76        (78.4%)    21       (21.6%)
Family practice    86        (83.5%)    17       (16.5%)
General practice   27        (90.0%)     3       (10.0%)
Surgery            67        (90.5%)     7        (9.5%)
Oncology           22        (95.7%)     1        (4.3%)
Other              31        (81.6%)     7       (18.4%)

Table 4

Terminally ill patients seen during previous month

Medical
specialty          Minimum  Maximum         Mean

Oncology (a)          3       30     16.44 [+ or -] 8.89
Internal medicine     0       30      7.92 [+ or -] 7.92
Other                 0       25      6.66 [+ or -] 6.78
Surgery               0       30      4.47 [+ or -] 4.29
Family practice       0       15      3.95 [+ or -] 3.65
General practice      0       15      2.87 [+ or -] 3.33

(a)Two oncologists reported treating more than 50 terminally ill
patients.

Table 5

Deliberate administration of an overdose is never ethically justified
(a)

Response  Frequency  Validity (%)

SD/D         124         34.0%
Neutral       32          8.8%
SA/A         209         57.2%

Total        365        100.0%

(a)Mean score, 3.49 (SD, 1.42). Items were recorded on the basis of
5-point Likert scale, with 5 indicating strongly agree, 4 indicating
agree, 3 indicating neutral, 2 indicating disagree, and 1 indicating
strongly disagree.

Table 6

Counseling patient in taking an overdose is never ethically justified
(a)

Response  Frequency  Validity (%)

SD/D         137         37.5%
Neutral       40         11.0%
SA/A         188         51.5%

Total        365        100.0%

(a)Mean score, 3.32 [+ or -] 1.42.

Table 7

Situational administration of overdose should be legal (a)

Response  Frequency  Validity (%)

SD/D         172         47.2%
Neutral       33          9.0%
SA/A         160         43.9%

Total        365        100.0%

(a)Mean score, 2.79 [+ or -] 1.44.

Table 8

Situational counseling for overdose should be legal (a)

Response  Frequency  Validity (%)

SD/D         164         44.9%
Neutral       37         10.1%
SA/A         164         44.9%

Total        365        100.0%

(a)Mean score, 2.87 [+ or -] 1.39.

Table 9

Would personally administer overdose (ie, perform euthanasia) (a)

Response  Frequency  Validity (%)

SD/D         234         65.5%
Neutral       33          9.2%
SA/A          90         25.2%

Total        357        100.0%

(a)Mean score, 2.29 [+ or -] 1.36.

Table 10

Would personally counsel or prescribe medication for overdose (a)

Response  Frequency  Validity (%)

SD/D         199         56.0%
Neutral       41         11.5%
SA/A         115         32.4%

Total        355        100.0%

(a)Mean score. 2.51 [+ or -] 1.42.

Table 11

Group statistics

                                     No. of
Statement                      Sex  responses       Mean (a)

Situational administration of   M      309     2.79 [+ or -] 1.44
 overdose should be legal       F       56     2.80 [+ or -] 1.42
Never justified to counsel      M      309     3.33 [+ or -] 1.43
 for overdose                   F       56     3.29 [+ or -] 1.36
Situational counseling for      M      309     2.88 [+ or -] 1.40
 overdose should be legal       F       56     2.80 [+ or -] 1.31
Personally would administer     M      309     2.25 [+ or -] 1.37
 overdose                       F       56     2.27 [+ or -] 1.33

(a)Strongly agree, 5; neutral, 3; strongly disagree, 1.

Table 12

Guidelines (a)

                                    SA/A                 Neutral

Restriction                 No.        (%)      No.         (%)

Adults only                 171       (44%)     88         (22%)
Not unduly distressed       290       (79%)     43         (12%)
Mentally competent          300       (83%)     28          (8%)
Less than 6 months to live  263       (72%)     54         (15%)
Pain beyond control         263       (72%)     55         (15%)

                                    SD/D

Restriction                 No.        (%)

Adults only                 123       (34%)
Not unduly distressed        32        (9%)
Mentally competent            7        (9%)
Less than 6 months to live   48       (13%)
Pain beyond control          47       (13%)

(a)SA/A, strongly agree/agree; SD/D, strongly disagree/disagree.

Table 13

Stipulations (a)

                                SA/A              Neutral

Stipulation               No.         (%)     No.         (%)

Established relationship  300       (82%)      37       (10%)
 with the patient
Patient's family should   247       (68%)      66       (18%)
 agree with decision
Two physicians should     276       (76%)      51       (14%)
 agree with decision
Specific waiting period   296       (81%)      43       (12%)
Drug overdose as cause    149       (41%)      90       (25%)
 of death

                                SD/D

Stipulation               No.         (%)

Established relationship   28        (8%)
 with the patient
Patient's family should    52       (14%)
 agree with decision
Two physicians should      38       (10%)
 agree with decision
Specific waiting period    26        (7%)
Drug overdose as cause    126       (34%)
 of death

(a)SA/A, strongly agree/agree; SD/D, strongly disagree/disagree.

Table 14

Characteristics of respondents in Tennessee, South Carolina, and
Washington State (a)

                                                   South
Characteristic                   Tennessee        Carolina

Mean age (yr)                  49 [+ or -] 13  47 [+ or -] 11
Male-to-female ratio, M/F (%)      85/15           86/14
Medical specialty (%)
 Family practice                     28              21
 Internal medicine                   27              23
 General surgery                     20              22
 General practice                     9               0
 Oncology                             6              12
 Psychiatry                           0              22
 Other                               10               0
Mean no. of terminally ill      6 [+ or -] 6   11 [+ or -] 23
 patients seen in last month


Characteristic                   Washington

Mean age (yr)                  46 [+ or -] 10
Male-to-female ratio, M/F (%)       81/19
Medical specialty (%)
 Family practice                     20
 Internal medicine                   10
 General surgery                     14
 General practice                     0
 Oncology                            12
 Psychiatry                          18
 Other                               26
Mean no. of terminally ill      7 [+ or -] 16
 patients seen in last month

(a)Plus-minus values are means to whole number [+ or -] SD (Washington
State and South Carolina data are from Dickinson at al(10)).

Table 15

Comparison of responses (a)

                                           State (%)

Statement                      TN            SC       WA

Deliberate administration
 of overdose (ie, euthanasia)
 is never justified

 SA/A                          57%          44%       48%
 N                              9%          10%        9%
 SD/D                          34%          45%       42%

Never justified to counsel
 for overdose

 SA/A                          52%          42%       39%
 N                             11%          13%       11%
 SD/D                          38%          44%       50%

Situational administration
 of overdose should be legal

 SA/A                          44%          52%       54%
 N                              9%           7%        8%
 SD/D                          47%          41%       38%

There are some situations
 in which I would
 administer overdose

 SA/A                          25%          29%       58%
 N                              9%          11%        8%
 SD/D                          64%          60%       58%

Situational counseling for
 overdose should be legal

 SA/A                          45%          48%       53%
 N                             10%          11%       10%
 SD/D                          45%          41%       37%

Personally would prescribe
 medication for overdose

 SA/A                          32%          33%       40%
 N                             11%          13%       10%
 SD/D                          55%          53%       49%

(a)Comparison of the responses from Tennessee, South Carolina, and
Washington State. The values shown are percentages of respondents
strongly agreeing or agreeing (SA/A), strongly disagreeing/disagreeing
(SD/D), or indicating neutrality (N). Aggregate percentages may not
equal 100% because of rounding. Statements are paraphrases of survey
items.

Accepted November 26, 2001.

References

(1.) Sade RM, Marshall MF. Legistrothanatry: A new specialty for assisting in death. Perspect Biol Med 1996;39:547-549.

(2.) Hall R. Final act: Death with dignity. Humanist 1994;54:l0-15.

(3.) Carr W. A right to die. Saturday Evening Post 1995;267:50-54.

(4.) Beauchamp TL, Childress JF. Principles of Biomedical Ethics. New York, Oxford University Press, 1994, ed 4.

(5.) Angell M. The Supreme Court and physician-assisted suicide: The ultimate right. N Engl J Med 1997;336:50-53 (editorial).

(6.) Siaw LK, Tan SY. How Hawaii's doctors feel about physician-assisted suicide and euthanasia: An overview. Hawaii Med J 1996;55:296-298.

(7.) Huang FY, Emanuel LL. Physician aid in dying and the relief of patients' suffering: Physicians' attitudes regarding patients' suffering and end-of-life decisions. J Clin Ethics 1995;6:62-67.

(8.) Bachman JG, Aleser KH, Doukas DJ, Lichtenstein RL, Corning AD, Brady H. Attitudes of Michigan physicians and the public toward legalizing physician-assisted suicide and voluntary euthanasia. N Engl J Med l996; 334:303-309.

(9.) Ganzini L, Fenn DS, Lee MA, Heintz RT, Bloom JD. Attitudes of Oregon psychiatrists toward physician-assisted suicide. Am J Psychiatry 1996; 153:1469-1475.

(10.) Dickinson GE, Lancaster CJ, Sumner ED, Cohen JS. Attitudes toward assisted suicide and euthanasia among physicians in South Carolina and Washington. Omega (Westport) 1997-1998;36:201-218.

(11.) Cohen JS, Fihn SD, Boyko EJ, Jonsen AR, Wood RW. Attitudes toward assisted suicide and euthanasia among physicians in Washington State. N Engl J Med 1994;331:89-94.

(12.) Salant P. Dillman D. How to Conduct Your Own Survey. New York, John Wiley & Sons, 1994.

(13.) Gallagher H. New morality. New Mobility 1996;14:70.

(14.) Vaux KL. Debbie's dying: Mercy killing and the good death. JAMA 1988;259:2140-2141

This article is adapted from a dissertation presented at the University of Tennessee, Knoxville, Knoxville TN.

Reprint requests to Douglas Essinger, PhD, 1455 Lonesome Pine Trail, Greeneville, TN 37745. Email: [e-mail]

Copyright [c] 2003 by The Southern Medical Association 0038-4348/03/9605-0427