Physician-Assisted Suicide

Pro: Autonomy Meets Non-maleficence
by Jon F. Merz

Con: A Better Prescription
by Felicia G. Cohn

(Issue 7 ?&nbspposted May 2, 1997; archived May 30, 1997)

What is "physician-assisted suicide"? If people buy guns from Joe's gun shop and take their own lives, do we refer to this as "Joe-assisted suicide"? Joe, of course, does not stand in the same shoes respecting his customers as does a physician to her patient. Physicians somehow are more responsible for their patients, and this paternalism is a firmly embedded ethos of medical practice. Unlike the physician, Joe is not bound by ethical principles that help define the bounds of his profession, and it is those principles that impute a duty of physicians to act in their patients' best interests. Nonetheless, society well might hold Joe accountable (for Joe-assisted suicide) for sale of a gun if he knows (or reasonably should have known) that his customer is going to take his own life in short order. So, too, does the physician who controls the drugs by which a patient might take his own life feel responsible for that act, and society - at least on its face - appears willing to hold physicians responsible as well.

But on what dimensions does physician-assisted suicide (PAS) - prescribing a drug and giving information regarding self-administration for the purpose of allowing a patient to end his life - differ from other acts performed by physicians that result in patient death? The table below presents some of the important dimensions, including: patient consent to die; the specific intent of the actor to cause death; whether the actor is the physician or the patient; and whether the primary cause of death is an action or a failure to act. Inaction is that which permits the natural consequences of a chain of events placed in motion by nature or other factors beyond the control of the (potential) actor. These factors yield a 2 x 2 x 2 x 2 table, the individual boxes of which have been conveniently numbered to facilitate discussion.

Looking at the table, we see that accidents may be attributed when there is neither the intent or consent to die (numbers 1-4). This table also presumes that, for patient acts, it is incongruous to discuss intent inconsistent with consent (5-8). Further, the table shows that consensual deaths due to external causes are fortuitous (10 and 13).

Now looking at physician involvement in patient death, we see that patients have the "right" to choose to die from some act other than one performed by their physician with the primary intent of causing death. That is, physicians must withhold or withdraw medical care on the refusal of the patient (or family), knowing that the lack of intervention will lead to death from the patient's underlying disease or condition, because they are constrained from intervening by the patient's right to be free of unwanted bodily invasions (15). Further, physicians may increase doses of morphine to patients in intractable pain for the purpose of relieving pain, despite the knowledge (and perhaps the desire) that such an act will cause the death of the patient (the principle of double effect) (9). Physicians may not, of course, act with the specific intent to kill with (16) or without (8) a patient's consent. Withdrawing or withholding care without the patient's consent (7) is far more fuzzy because of issues of appropriateness, futility of care, and surrogate decision-making authority.

Some acts that cause death (9) and the (highly frequent!) failure to act to prevent death (15) are acceptable. Note that the main difference between suicide (14) and the right to die (15) is that suicide is caused by the purposive act of the patient instead of by some cause that can be ascribed to "nature." The main difference between suicide and indirect euthanasia (9) is lack of primary intent to cause death. Both double effect and the right to die offer physicians plausible deniability of intent. On the other hand, physician assistance with a patient's suicide - by writing the needed prescription with knowledge of its intended use - implies complicity in the act, and the intent of the physician to participate in what many view as an irrational (and for some, immoral) act may not be denied.

Part of the debate about physician-assisted suicide turns on the reasonableness of suicide in some instances. Those who suffer from painful, fatal diseases - most often end-stage metastatic cancers - face the lamentable choice between pain control that leaves them unconscious (or nearly so) and fully dependent upon caregivers, or living with the pain in a more sentient and perhaps less dependent state. For some persons, suicide is rational because it permits them to retain some dignity, relieve intolerable pain, and control their lives through orchestration of their own deaths. Often, having the ability to control their deaths affords peace of mind, and the drugs go unused. The principle of autonomy suggests these judgments should be respected.

Nonetheless, the autonomy of these persons runs smack into the ethical principles of beneficence and non-maleficence. Physicians, simply, should do no harm, and should act to preserve life and health (physical, mental, and social). As the AMA states in its Code of Medical Ethics :

The social commitment of the physician is to sustain life and relieve suffering. Where the performance of one duty conflicts with the other, the preferences of the patient should prevail.

The concerns about PAS fall into three areas. First, it is asserted that participation in suicide is fundamentally at odds with physicians' roles as healers, and could lead to a loss of trust by patients. But others assert that permitting physician help in dying will increase, not decrease, trust. As summarized by Christine Cassel of Mount Sinai Medical School, New York City, "The public appears to be losing faith in doctors, at least partly because of our paternalistic and sometimes cruel insistence on life at any cost." This is a basic empirical question that cannot be answered by simplistic proclamations.

Second, it has been asserted that PAS would be difficult to control. This concern is about specification errors. To whom would the law apply, and how would regulations be implemented that would prevent PAS from being used too liberally? Of course, physicians currently participate in patients' deaths, with an estimated 80% of Americans dying in hospitals and nursing homes, and 70% of these hospital deaths involve some level of withholdingor withdrawing care. The SUPPORT study suggests that many of these cases are without express consent, as well. Physicians also rely on the principle of double effect in titrating morphine to dying patients (and they routinely do not intubate patients to prevent the foreseeable consequence of suppressed ability to breathe). Studies in the United States and in the Netherlands suggest that few patients - on the order of only 7% - request aid in dying, and that 75% of the time these requests are not honored. It is also asserted that suicide may become the standard of care for old, infirm, and particularly poor or underinsured persons. Nonetheless, it isn't clear why regulations ensuring the competency and voluntariness of decisions would permit broadening of the inclusion criteria or how persons could be coerced to take their own lives.

Third, it is asserted that allowing PAS will start us on a slippery slope toward active euthanasia, involuntary euthanasia, and, before long, mercy killing of the weak and infirm. This complaint voices the worry that future policy makers will not be able to draw a line between impermissible acts and those currently under consideration. Of course, we already permit withdrawal and withholding of care and indirect euthanasia. As highlighted in the figure above, PAS differs in important ways from these activities, and differs markedly from other forms of euthanasia as well. The fact that public and medical discourse about PAS is as thorough and exhaustive as it is ensures that this activity can be controlled and does not risk sliding us inexorably toward eugenic infanticide.

What do we know about PAS? It is generally recognized that there is some small cohort of patients for whom palliative care is inadequate. Many people are fearful of a painful death, and a majority of the public support PAS in limited circumstances. Further, a large percentage of physicians support PAS, and many doctors report patient requests for lethal prescriptions and information about how to effectively use them. Finally, some physicians report complying with these requests.

Physicians who are willing to help patients in this ultimate and desperate act should be able to do so. Fatally ill patients who are in intractable pain are often quite vulnerable and at risk of becoming prisoners of a technological imperative. They likely do not have the means to relieve themselves from their burdensome existence, and physicians have not only the expertise to help patients, but they have a state-granted monopolistic power over the controlled substances which offer - for some small number of patients - the means for relief. Joe the gun salesman is an unacceptable option.

Permitting PAS has several benefits beyond the benefit to those patients who most desire to be helped in dying. Instead of being performed in secret, open, legalized PAS permits greater medical and societal oversight, reducing the chances of abuse. Greater discussion about PAS may also help providers understand the methods of administration, reducing the risk of harm to people. Further, patients who attempt suicide with firearms, poisons, or by other self-inflicted violence can seriously harm themselves in the attempt, leaving themselves worse off.

The fundamental empirical concerns about PAS can be measured only if PAS is permitted and its use and outcomes assessed. Questions that must be addressed include :For what patients would PAS be "appropriate" and why? Where are the shortcomings in palliative, end-of-life care, and from where will improvements come (particularly if PAS is legalized)? Will PAS undermine patient trust in their health care providers? And will PAS lead inexorably toward devaluing human life? Physicians already play a role in their patients' deaths, and some physicians are willing to help patients take their own lives. Better than the status quo is to permit PAS under well regulated and reported conditions.

Jon F. Merz is a Research Assistant Professor of Bioethics in the Department of Molecular and Cellular Engineering and Faculty Associate in the Center for Bioethics at the University of Pennsylvania.

The controversy over physician-assisted suicide (PAS) has become a focal point of American law and policy since the United States Supreme Court accepted two PAS cases in October 1996 :Dennis C. Vacco et al. v. Timothy E. Quill, M.D., et al., 80 F.3d 716 (2d Cir. 1996), and State of Washington v. Harold Glucksberg, et al. (formerly Compassion in Dying v. Washington) 79 F.3d 790 (9th Cir. 1996). The current debate focuses on individual rights :whether an individual has a right to physician assistance in taking his or her own life. Advocates claim that the ability to control the time and circumstances of one's death is a matter of autonomous decision making, and that physician assistance is required to facilitate and assure the successful implementation of those decisions. A number of arguments regarding public opinion, personal choice, suffering, physician expertise, end-of-life care decisions, and the need for regulation appear to present compelling support for PAS. However, upon examination, they prove to be misleading or unpersuasive. Construing the debate in terms of an individual's "right to die" masks other important considerations about health care at the end of life.

"The public wants PAS." One opinion poll after another has indicated that a majority of the American public favors PAS. However, a recent survey by the American Medical Association demonstrates that, with education, individuals often change their minds about PAS. Support for PAS appears to be based on a lack of information about its prevalence and abuses in the Netherlands and knowledge of other options for care at the end of life. In the Netherlands, PAS is illegal but tolerated when performed in accord with a certain set of conditions. Calculations based on the rate of PAS in the Netherlands applied to the American population indicate that PAS could account for over 61,000 deaths in the United States annually. The Dutch experience also suggests a significant number of unreported cases and instances in which PAS was provided without patient request or consent. [1] When presented with the Dutch data, one-half of those who supported PAS reversed their positions. Once provided with descriptions of hospice and palliative care options, only 14% said they would still opt for PAS. [2]

"Physicians are obligated to respect a patient's autonomous choice for assistance in suicide. Distrust will result if physicians refuse to honor patients' wishes." The debate has been framed in terms of a "right to die," a slogan that suggests that PAS is a matter of individual choice between either suffering or death. Individual choices are now constrained by clinical reality; currently, pain relief is inadequate, palliative and supportive care uncommon and underfunded, and hospice eligibility limited. Despite public assumptions to the contrary, physicians are not trained to provide adequate end-of-life care, and instead fight death with aggressive interventions that are costly and often inappropriate for persons nearing the end of their lives. End-of-life care does not have to be this way. Improvement is possible, but the availability of PAS may serve as a disincentive to improve clinician training and practice and change health-care funding mechanisms. Thus, PAS could become a coercive offer, an option that one feels he or she must exercise to avoid suffering, indignity, and even impoverishment. Given this context, distrust seems more likely to result from uncertainty regarding a physician's role - caretaker or killer - than from physician refusal to comply with a patient's request for PAS.

"PAS is a compassionate response to situations of dreadful suffering, in which a miserable patient requests and immediately receives assistance in bringing about a desired death. " A number of misperceptions plague this image. Suffering need not be a part of dying and PAS is not the only remedy for suffering. Rates of intractable pain, even among cancer patients, are generally not higher than 3%, and even this can be relieved if the patient is willing to accept sedation. It is already legal for physicians to provide pain medications, even if they might have the side effect of hastening death. The real effect of large doses of pain medications on life span is not well-studied. Experience suggests that adequate pain alleviation might even extend life. Other symptoms, such as nausea, fatigue, and shortness of breath, can also be managed.

If all efforts at comfort care fail, patients may seek to bring about their deaths in other ways, e.g., committing suicide unassisted, refusing life-sustaining treatment, or forgoing nutrition and fluids. Dehydration, for example, requires no assistance and is a relatively comfortable way to die, especially for a very ill person who may have little interest in food and water and despite the negative view of what healthy people might call starvation. Furthermore, according to virtually every proposed PAS statute, a physician could only act on a request for PAS following a substantial waiting period, often longer than two weeks. This is hardly the immediate response envisioned and may take longer than death by dehydration.

"Physicians possess the unique expertise and singular ability to bring about an easy suicide, and so are necessarily involved. " The PAS debate is more about making suicide easy than a right to die. Yet physician involvement does not assure an easy suicide. Physicians often are not taught effective symptom management, and certainly do not learn how to take life. Additionally, physician assistance would be limited to providing a lethal prescription, something for which physicians receive no training. Due to either incorrect dosing or patient condition, many patients will be unable to swallow or keep the pills down. This raises questions as to the need for further assistance and may even suggest that active euthanasia, or lethal injection, would be more humane. The specter lurks of a slippery slope to abuse. Furthermore, the question remains of how to deal with a failed attempt, particularly if that act has rendered the patient worse off or unable to request another attempt. What is needed is more end-of-life care education, not PAS. It is interesting to note that even if PAS were legal, it might be difficult to find a physician willing to assist. Among those who favor PAS, only a minority are actually willing to participate. [3]

"Physician-assisted suicide is a logical extension of the right to forgo life-sustaining treatment or receive potentially death-hastening pain medications. " Despite claims that PAS is no different from either forgoing life-sustaining treatment or using drugs that may shorten life, such as those reasons offered by the Second Circuit Court in the decision mentioned above, there are important differences in intent, legal justification, and clinical description. The explicit intent in PAS is to end life; this is not the case in either other action where the intent is to avoid medical battery or alleviate pain. For example, if a patient does not die when a ventilator is withdrawn, the physician does not proceed to smother the patient to death. Forgoing life-sustaining treatment is based on a negative right to be left alone, and the use of medications is based on practice standards for effective pain management. PAS would create a positive legal right to assistance. The physician who assists a patient's suicide may feel as if he or she is responsible for that patient's death, whereas death in the other scenarios is the result of the underlying disease process. Both the use of pain medications and refusal of life-sustaining treatment are options available to all patients, while proposals for PAS restrict its application. This in itself raises questions as to the fears and dangers associated with allowing PAS.

"PAS is occurring clandestinely. With legalization, it might be regulated appropriately. " The studies that demonstrate that PAS is regularly occurring are generally empirically unreliable. They contain no standard definition of PAS and so may be counting a variety of activities inappropriately. Furthermore, the fact that it may be occurring hardly provides a reason for making it legal. Although many in the United States now use illicit drugs and engage in underage drinking and agitate for the legalization of these activities, our society has decided that there are important reasons for maintaining their illegality. The same is true of PAS. Unless our society achieves a consensus on the value of life, works out the practical difficulties, and resolves the role conflict for physicians, any policy allowing PAS will continue to be troublesome.

Even regulation of PAS does not assure that its practice will be limited according to the boundaries we create. Most proposed legislation requires that PAS be confined to terminally ill, suffering, competent patients who voluntarily and repeatedly request assistance from their physicians. Physicians are limited to providing a lethal prescription that must be self-administered by the patient. Each of these conditions is problematic, whether undefinable or unsustainable, and will lead to a number of contested cases.

Regardless of the Supreme Court's decision this summer regarding whether physician-assisted suicide is a constitutional right, the debate, far from being over, will move to the states. The public focus on a "right to die" has seriously miscast the debate, obscuring the real issues in end-of-life care. The calls for PAS arise in a social system that is inattentive to the complex physical, emotional, and spiritual needs of patients as they near the end of life. Additionally, there is a genuine fear of abuse among those who have experience with this often frail, disabled, and economically disadvantaged patient population. Resolving a patient's suffering should not rely upon assisting that patient's suicide. Rather, providing comfort care, especially when cure is no longer possible, is an important task for physicians as well as an important societal goal. Making good palliative care a real option involves developing a new research agenda, enhancing medical education, and changing priorities in health-care delivery and funding. Physician assistance in dying should come to mean supportive and comfort care rather than lethal prescription. The most powerful calls for PAS come from individuals who experience suffering we would all prefer to avoid. As we develop policy, we must remember that PAS is about more than individual rights and tragic situations. As Oliver Wendell Holmes reminded us, "hard cases make bad law."

Felicia Cohn is a senior scientist with the Center to Improve Care of the Dying and an assistant research professor in the Department of Health Care Sciences at The George Washington University, Washington, DC.

Andrzej Krauze is an illustrator, poster maker, cartoonist and painter who illustrates regularly for HMS Beagle, The Guardian, The Sunday Telegraph, Bookseller and New Statesman.

Tell us what you think about physician assisted suicide.

Endlinks

Information on the Internet about end-of-life care and physician assisted suicide is abundant. The following sites discussing physician assisted suicide were recommended by J. Merz:

Compassion in Dying
Choice in Dying
Euthanasia Research & Guidance Organization
The Hemlock Society
DeathNET

The following sites relating to end-of life care were suggested by F. Cohn:

American Foundation for Suicide Prevention
American Hospice Foundation
American Medical Association
Center to Improve Care of the Dying
Choice in Dying
Hospice Foundation of America
Last Acts
National Council on Aging
National Decisions for Health Network
National Family Caregivers Association
National Hospice Organization
PBS: "Before I Die: Medical Care and Personal Choices" (or alternate address)
Project on Death in America
The Robert Wood Johnson Foundation

Readers may also be interested in the following news articles on legislation of voluntary euthanasia, as well as Internet discussion groups on the topic:

	Associated Press. 1997. House votes to bar federal funding for assisted suicide lopsided vote belies partisan division on bill. Washington Post, April 11, p. A20. Article describing recent federal legislation.
	Parsons, B. and Middleton, K. 1997. Euthanasia ban no barrier to care: doctors. The Age, April 15. New article on study of PAS in Australia.
	Hess, D. 1997. House vote reiterates ban on U.S. aid for assisted suicide. Philadelphia Enquirer, April 11. Article describing recent federal legislation.
	&nbspZinn, C. Australian voluntary euthanasia law is overturned. Br. Med. J. 314(7086):7086

In Slate, a round of "debate memos" has begun on the topic. In addition to the memos, there are many links therein of interest. And be sure to check out what readers say in The Fray, Slate's interactive discussion forum.

Previous Op-Ed Articles:

Money Isn't Everything

by Yoji Arata (Issue 6; posted April 18, 1997)

Reflections on Grant Peer Review

by Harry Brodie (Issue 5; posted April 4, 1997)

Mammalian Cloning: The Science Of The Lambs

by Alan P. Wolffe (Issue 4; posted March 21, 1997)

Non-ethics of cloning

by John Maddox (Issue 3; posted March 5, 1997)

Introns 'R' Us: Why the Great Debate Deadlocked

by Andrew D. Ellington (Issue 2; posted February 20, 1997)

Is Pasteur's Day Past?

by Andrew Murray (Issue 1; posted February 1, 1997)