New York State Task Force on Life and the Law

When Death is Sought

Assisted Suicide and Euthanasia in the Medical Context

Supplement to Report
April 1997

In our 1994 report, When Death is Sought: Assisted Suicide and Euthanasia in the Medical Context,[1] we unanimously recommended that New York retain its prohibition of assisted suicide and euthanasia. After extensive study, we concluded that legalizing these practices would be profoundly dangerous for large segments of the population, especially in light of the widespread failure of American medicine to treat pain adequately or to diagnose and treat clinical depression in many cases. Some of us concluded that assisted suicide is inherently unethical in all circumstances; others found that the practice can be ethically appropriate in extraordinary cases, but that legalizing it would pose serious and insurmountable risks of mistake and abuse that would greatly outweigh any benefit that might be achieved. Those risks center on the likelihood that many individuals would request suicide assistance because of improper medical care, unrecognized lack of decision-making capacity, or coercion, not because of a voluntary, settled commitment to die. Our recommendations on assisted suicide and euthanasia were grounded in over ten years of efforts to promote patients’ right to control their medical care, including the right to refuse unwanted treatment.

Earlier this year, the United States Supreme Court heard arguments in two cases challenging the constitutionality of laws that make assisted suicide illegal in all cases. The first case, Washington v. Glucksberg, was an appeal from a ruling of the United States Court of Appeals for the Ninth Circuit, which found that Washington State’s prohibition of assisted suicide violates the constitutional guarantee of due process, because it denies competent, terminally ill patients the right to “hasten death.” [2]The second case, Vacco v. Quill, was an appeal from a decision of the United States Court of Appeals for the Second Circuit, which rejected the Ninth Circuit’s due process analysis, but found that New York’s prohibition of assisted suicide is unconstitutional because it denies certain terminally ill individuals the equal protection of the laws. Specifically, the court found that, because New York allows competent, terminally ill patients who require life-sustaining treatment to “hasten death” by refusing such treatment, it must also allow physicians to help competent, terminally ill patients who do not require life-sustaining treatment to “hasten death” by prescribing lethal drugs for patients to self-administer. [3]

The Supreme Court arguments generated substantial public interest in the care and treatment of patients at the end of life. We are concerned, however, that the public’s focus on the narrow result of the two decisions under review by the Court has obscured the potential impact of the reasoning that led the Ninth and Second Circuits to reach that result. In particular, we are deeply troubled by the manner in which the decisions blur long-standing distinctions between physician-assisted suicide and the refusal of unwanted medical treatment, and between physician-assisted suicide and the use of high doses of opioids for the relief of pain. Whatever one thinks about the legalization of physician-assisted suicide, the implications of abandoning these distinctions for patients’ control over their medical care should present significant cause for concern.

The Decisions

In both Washington v. Glucksberg and Vacco v. Quill, a group of physicians and dying patients challenged state prohibitions on assisted suicide as applied to physicians who “facilitate the exercise of the decision of competent, terminally ill adults to hasten inevitable death by prescribing suitable medications for the patient to self-administer for that purpose.”[4] The plaintiffs asserted two legal theories in support of their claim, one based on due process and the other based on equal protection.

Washington v. Glucksberg. In Glucksberg, the Ninth Circuit agreed with the plaintiffs that, under the Constitution’s due process clause, competent, terminally ill individuals have a “liberty interest in choosing the time and manner of [their] death.” Reviewing cases affording constitutional protection to personal decisions about marriage, procreation, family relationships, child rearing, and heterosexual activity, Judge Reinhardt concluded that the “common thread” of the cases “is that they involve decisions that are highly personal and intimate, as well as of great importance to the individual.” [5] According to Judge Reinhardt, “few decisions are more personal, intimate or important than the decision to end one’s life, especially when the reason for doing so is to avoid excessive and protracted pain.”[6] The court also found that the Supreme Court’s decision in Cruzan v. Director, Missouri Department of Health,[7] “by recognizing a liberty interest that includes the refusal of artificial provision of life-sustaining food and water, necessarily recognizes a liberty interest in hastening one’s own death.” [8]

The court next examined six interests asserted by the state in support of its prohibition of assisted suicide: (1) the preservation of life; (2) the prevention of suicide; (3) preventing the influence of third parties; (4) the interests of third parties; (5) protecting the integrity of the medical profession; and (6) concern about adverse consequences (or the “slippery slope”). None of these interests, the court found, is sufficient to override a competent, terminally ill individual’s liberty interest in committing suicide with a physician’s aid. Most of these interests, the court concluded, apply equally to the refusal of life-sustaining medical treatment, which the court saw as legally and ethically indistinguishable from the self-administration of drugs to “hasten inevitable death.”[9] In addition, the court argued that assisted suicide could not be distinguished from the lawful administration of high doses of opioids to relieve pain, which, according to the court, has the “known effect” of “hasten[ing] the end of the patient’s life.” [10]

Vacco v. Quill. Unlike the Ninth Circuit in Glucksberg, the Second Circuit in Quill rejected the contention that decisions about assisted suicide implicate a constitutionally-protected liberty interest. Citing Bowers v. Hardwick,[11] the decision upholding Georgia’s sodomy statute, Judge Miner emphasized the Supreme Court’s reluctance to extend the list of fundamental rights and liberty interests beyond those recognized in the text of the Constitution itself or in long-standing Supreme Court precedents. To qualify for heightened due process protection under existing Supreme Court doctrine, Judge Miner wrote, a right must be “‘deeply rooted in this Nation’s history and tradition,’”[12] or “so ‘implicit in the concept of ordered liberty’ that ‘neither liberty nor justice would exist if [it] were sacrificed.’”[13] According to the Second Circuit, assisted suicide does not satisfy either of these tests, particularly in light of the long-standing societal opposition to suicide and assisted suicide.

Despite this finding, the court accepted the plaintiffs’ contention that New York’s prohibition of assisted suicide violates the constitutional guarantee of equal protection. First, the court found that the patient-plaintiffs -- competent, terminally ill adults in the final stages of their illness but not in need of life-sustaining treatment -- were “similarly situated” to terminally ill patients “whose treatment includes life support.” Second, the court found that the State treats these two groups of patients differently: “those in the final stages of terminal illness who are on life-support systems are allowed to hasten their deaths by directing the removal of such systems, but those who are similarly situated, except for the previous attachment of life-sustaining equipment, are not allowed to hasten death by self-administering prescribed drugs.”[14] Rejecting the State’s effort to distinguish the refusal of treatment from assisted suicide, the court emphasized that, in both situations, “a patient hastens his death by means that are not natural in any sense.” The court also concluded that the distinction is not rationally related to any legitimate state interest. According to the court, “[a]t oral argument and in its brief, the state’s contention has been that its principal interest is in preserving the life of all its citizens at all times and under all circumstances.”[15] This interest, the court concluded, is simply not legitimate when applied to competent, terminally ill patients who wish to end their lives.

The Risks of Legalization

The Ninth and Second Circuits both dismissed the risks associated with legalizing physician-assisted suicide as insubstantial, and claimed that, to the extent risks exist, they can effectively be eliminated through state law or regulation. Our concerns about the risks of legalizing assisted suicide are set forth in detail in When Death is Sought, and will not be restated in depth here. We take this opportunity, however, to outline briefly the primary risks associated with legalization: We continue to believe that these profound dangers associated with legalizing physician-assisted suicide outweigh any benefits such a change in law might achieve in isolated cases. Yet, one need not accept this conclusion to be troubled by the analysis of the Ninth and Second Circuits. The fundamental premise underlying the reasoning of both courts -- that assisted suicide is no different from the refusal of life-sustaining medical treatment (and, for the Ninth Circuit, the use of high doses of opioids to relieve pain) -- undermines a variety of legal and clinical tenets critical to the care of patients at the end of life. The question of legalizing assisted suicide should be addressed on its own merits, not on the simplistic assumption that all practices resulting in death must be treated the same. The following discussion outlines the primary factors distinguishing assisted suicide from the refusal of treatment and the use of opioids for the relief of pain. These distinctions include both inherent logical differences and practical differences in the balancing of the benefits and risks at stake.

The Distinction Between Refusing Medical Treatment and Suicide

The distinction between the refusal of life-sustaining medical treatment and suicide has been a critical component of all of our recommendations on end-of-life care. In our report on do-not-resuscitate orders, we grounded our legal analysis on the premise that suicide relates only to self-inflicted deaths and “not to a decision to refuse life-sustaining treatment.”[16] We reaffirmed this position in our report on the health care proxy, which found that “as a matter of public policy the taking of human life must not be granted legal sanction.”[17] Based on that conclusion, the health care proxy law, as enacted by the New York State Legislature, provides that “[t]his article is not intended to permit or promote suicide, assisted suicide, or euthanasia.” Our proposed legislation on surrogate decision-making for incapacitated patients without advance directives contains a similar statement; the report accompanying the proposed legislation expressly states that surrogate decision-making is “not intended either as a step on the road to assisted suicide or as a vehicle to extend the authority of family members beyond the traditional boundaries established by consent to provide treatment or not to treat.”[18] Most recently, our report on assisted suicide and euthanasia proposed “a clear line for public policies and medical practice between forgoing medical interventions and assistance to commit suicide or euthanasia,”[19] and outlined the legal, ethical, and policy considerations distinguishing the two practices.[20]

We recognize that “the moral distinction between assisting to die and withdrawing treatment is hard to discern in certain cases.”[22] The alleged distinction between “acts” and “omissions,” for example, is “particularly nebulous,”[22] given that physicians who comply with requests to refuse treatment are often required to undertake affirmative acts, such as disconnecting respirators or feeding tubes. Resting the distinction on a difference in intent is also not always persuasive, because “[i]n the act of disconnecting a life-sustaining ventilator ... some physicians actually intend, not just to rid the patient of unwelcome technology, but to help the patient end her suffering by dying sooner.”[23] Finally, as an empirical matter, it is undeniable that withdrawing or withholding life-sustaining treatment, at least in some cases, can play a causal role in any death that ensues. “When a doctor detaches a feeding tube from a patient who could have lived for an additional decade, albeit in a profoundly diminished state, she is certainly ‘the cause’ of death insofar as she determines when and how the patient dies.” [24]

Nonetheless, the fact that the refusal of life-sustaining treatment and assisted suicide are similar in certain respects does not mean that the practices implicate identical legal, clinical, ethical, and public policy concerns. The following factors, taken together, present compelling reasons to distinguish between the refusal of life-sustaining treatment and assisted suicide for law and public policy, despite the similarities that might exist in individual cases. Although these distinctions may not, in themselves, compel the conclusion that assisted suicide should remain illegal, they undermine the claim that the legal recognition of a broad right to refuse treatment requires recognition of a right to assisted suicide as a matter of constitutional law.

The right to refuse medical treatment is based on the long-standing right to resist unwanted physical invasions, not on a right to “hasten death.”

Critics of the distinction between the refusal of life-sustaining treatment and assisted suicide contend that both practices are based on the proposition that dying patients have a right to “hasten death.” For example, in Glucksberg, Judge Reinhardt characterized the judicial recognition of the right to refuse life-sustaining treatment as a “drastic change regarding acceptable medical practices,” reflecting the courts’ belief “that terminally ill persons are entitled ... to hasten their deaths, and that ... physicians may assist in the process.”[25] This description of the development of the right to refuse treatment simply cannot be reconciled with the cases originally recognizing that right.

Courts that affirmed the right to refuse treatment, including life-sustaining measures, consistently grounded that right in the long-standing doctrine of informed consent, which forbids physicians from performing invasive medical procedures without the patient’s knowing and voluntary agreement[26]. That doctrine is based on the common-law concept of battery, under which any nonconsensual “touching” is a “tort” -- a legal wrong -- providing grounds for the victim to sue.[27] While patients who refuse treatment may become sicker, and sometimes will die, that result has always been regarded as an unavoidable consequence of applying the doctrine of informed consent consistently and without exception, not as a reason to recognize individuals’ right to refuse treatment capable of prolonging life. Contrary to Judge Reinhardt’s assertion, the fact that courts did not explicitly recognize the right to refuse life-sustaining treatment until relatively recently does not mean that the right represented a “drastic change.” Rather, the courts’ recognition of the right to refuse life-sustaining treatment was simply an application of the long-standing prohibition of battery to “the advance of medical technology capable of sustaining life well past the point where natural forces would have brought certain death in earlier times.”[28] Until the widespread use of devices such as respirators, dialysis machines, and feeding tubes, there was simply no occasion for courts to consider the right to refuse life-sustaining treatment, outside the narrow context of “patients who refused medical treatment forbidden by their religious beliefs.” [29]

The fact that courts grounded the right to refuse treatment in the long-standing right to resist unwanted physical invasions, rather than in a broader “right” to “hasten death,” strongly undermines the claim that the refusal of treatment and assisted suicide are legally and ethically the same. Prohibiting individuals from refusing medical treatment would represent “a violation of personal autonomy and physical integrity totally incompatible with the deepest meaning of our traditional respect for liberty.”[30] In some cases, such prohibitions would violate sincerely-held religious beliefs opposing certain medical interventions, such as the belief among Jehovah’s Witnesses against receiving transfusions of blood.[31] Decisions about assisted suicide do not implicate these interests. For this reason alone, assisted suicide is fundamentally different from the right to refuse treatment; moreover, the difference is not simply in degree but in kind.

Characterizing the refusal of medical treatment as “the cause” of any deaths that result would undermine society’s commitment to respecting patients’ decisions about medical care.

In case after case, courts have concluded that deaths following the refusal of treatment are caused primarily by the patient’s underlying disease, not the patient’s decision or act. As the New Jersey Supreme Court has held, “a patient does not die because of the withdrawal of a kidney dialysis machine, but because his underlying disease has destroyed the proper functioning of his kidney.” Likewise, a patient does not die “from the withdrawal of a nasogastric tube, but because of her underlying medical problem, i.e., an inability to swallow.”[32] If these statements were meant to suggest that the refusal of treatment plays absolutely no causal role in the patient’s death, they would obviously be untrue. For example, when a physician withdraws a respirator or disconnects a feeding tube from a seriously ill patient, it is undeniable that these actions causally contribute to the patient’s death: but for the withdrawal of treatment, the patient would probably have continued to live. This empirical question of “but-for” causation, however, is clearly not what the courts had in mind. Instead, the law’s traditional analysis of the cause of deaths following the refusal of life-sustaining treatment reflects important judgments about the nature and goals of medicine, which society should be extremely hesitant to revise, particularly in the guise of constitutional interpretation.

The law has long distinguished between the determination of causation as a factual matter and the determination of causation for the purpose of assessing legal and ethical accountability. When a variety of factual causes are necessary, but not individually sufficient, to bring about a particular result, the determination of which among them are properly cited as causative for legal purposes becomes a policy judgment, reflecting underlying assumptions about rights, duties, and moral blame.[33] This is precisely the case when patients die following the refusal of life-sustaining treatment. In contrast to patients who take lethal drugs, patients who refuse life-sustaining treatment will not die unless they are suffering from a condition that makes it impossible to live without invasive medical support (such as an inability to breath, or an inability to swallow or assimilate food taken orally). As Daniel Callahan has put it, “there must be an underlying fatal pathology if allowing to die is even possible.”[34] In light of the multiple causes of death following the refusal of life-sustaining treatment, the determination of legal causation cannot be based on simple empirical observation, but requires a deliberate judgment about legal and ethical accountability. The traditional view that the disease, not the refusal of treatment, is the primary cause of death affirms widely-shared beliefs about the nature of medical care – in particular, that consent to medical treatment is not obligatory, but a matter of individual choice.[35] Because the technology is optional, patients who refuse it are not considered to be accountable for causing their own deaths.[36]

By claiming that patients who refuse life-sustaining treatment are the primary cause of any deaths that result, the Second and Ninth Circuits unfairly stigmatize patients who choose not to submit to every available technology capable of prolonging life. The implication of assigning causal accountability to the patient, rather than the underlying injury or disease, is that consent to life-sustaining treatment is expected, and that those who refuse treatment are therefore responsible for bringing about their own deaths. In Quill, Judge Miner actually states that deaths following the refusal of life-sustaining treatment are unnatural: “[b]y ordering the discontinuance of ... artificial life-sustaining processes or refusing to accept them in the first place, a patient hastens his death by means that are not natural in any sense.”[37] This statement reflects a curious – and, we believe, dangerous -- view of the relationship between nature and technology. Until the development of respirators and feeding tubes, patients who lost the ability to breathe or swallow would inevitably die, and no one would think to argue that such deaths were self-inflicted or anything but natural consequences of injury or disease. If, as Judge Miner claims, it is now “unnatural” to die from an inability to breathe or swallow, it is only because technologies have been developed that can forestall many of these deaths. The invention of new technology, however, does not make the choice to allow events to proceed without the technology “unnatural.” To claim otherwise is to establish a “technological imperative,” in which the very existence of technology becomes a mandate for its use.[38] Such reasoning is actually more consistent with the claim that the use of life-sustaining treatment should be obligatory, and should be disturbing to those who support patients’ right to control their own medical care.

Indeed, if patients who refuse life-sustaining treatment are responsible for “causing” the deaths that result, few deaths could be attributed to natural causes. Death often follows a decision to forgo an available medical treatment that could potentially prolong the patient’s life. For example, death is typically preceded by the cessation of breathing and heartbeat, and only sometimes are efforts made to resuscitate dying patients by performing CPR. If deaths resulting from the failure to provide medical treatment are “nothing more nor less than assisted suicide,” the consensual failure to perform CPR is a form of assisted suicide, and the doctor who does nothing when a patient’s heart has stopped has “caused” the patient’s death.[39] Likewise, patients who refuse chemotherapy, because they are unwilling to endure the painful and debilitating side effects, would no longer be victims of cancer, but of their own “unnatural” acts. While some might distinguish these examples of withholding life-sustaining treatment from more active instances of turning off respirators or disconnecting feeding tubes, it is widely accepted that “withdrawing” and “withholding” medical treatment implicate identical legal and ethical concerns.[40] Thus, if turning off a respirator is suicide, so too is refusing to be connected to the respirator in the first place. And if refusing a respirator is suicide, the same must be true for refusing an organ transplant, or refusing any other grueling procedure with an uncertain outcome. Such characterizations defy common sense. There is an obvious difference between refusing invasive technologies that have the potential to prolong life long after the body is able to survive on its own, and deciding to commit suicide by causing the body to stop functioning before death would otherwise occur.

Finally, as a practical matter, telling physicians that they are the primary cause of death when patients refuse medical treatment is likely to backfire, by leading physicians, especially those opposed to assisted suicide, to question their participation in the withdrawal and withholding of medical care. Despite patients’ clear legal right to refuse life-sustaining treatment, many doctors -- particularly those who are opposed to any participation in efforts to “hasten death” -- must still be persuaded to comply with patients’ requests to forgo aggressive measures.[41] By assuring doctors that they are not the legal cause of death when patients refuse treatment -- in other words, by affirming that withholding and withdrawing treatment are not assisted suicide -- existing legal and ethical standards allow physicians to honor patients’ wishes about treatment without having to feel responsible for causing the patient’s death. If physicians are told that deaths following the refusal of treatment are “unnatural,” and that the refusal of treatment is the primary legal and ethical “cause of death,” many physicians are likely to rethink their participation in the withholding and withdrawal of treatment. The result would be a disastrous setback for patient autonomy.

Equating the refusal of treatment with suicide would make it impossible to limit physician-assisted suicide to competent, terminally ill patients, or to legalize physician-assisted suicide without also legalizing euthanasia.

Equating the refusal of treatment with assisted suicide is also inconsistent with the claim that assisted suicide could be limited to competent, terminally ill patients, or that assisted suicide could be legalized while physician-administered lethal injections remain illegal. These practical consequences of abandoning the distinction between the refusal of treatment and suicide should not be ignored. In contrast to the broad right to refuse medical treatment, few proponents of legalizing assisted suicide argue that the practice should be available to anyone on demand. Instead, advocates of legalization have argued that assisted suicide should be treated as a “nonstandard medical practice reserved for extraordinary circumstances,”[42] or as a “response to medical failure,” for those “few patients” who “will face a bad death despite all medical efforts.[43] In fact, most advocates of legalization acknowledge that laws prohibiting assisted suicide serve valuable societal interests, especially when applied to healthy individuals suffering from reversible physical or psychological problems. As Judge Reinhardt observed in Glucksberg, “the state has a clear interest in preventing anyone, no matter what age, from taking his own life in a fit of desperation, depression, or loneliness or as a result of any other problem, physical or psychological, which can be significantly ameliorated.” In such cases, he wrote, “the heartache of suicide is the senseless loss of a life ended prematurely,” and the state can legitimately take steps to prevent these suicides from taking place.[44]

Recognizing the need for limits, the plaintiffs in Glucksberg and Quill propose a right to physician-assisted suicide only for competent patients who are terminally ill. In addition, they argue that recognizing a right to a prescription for lethal drugs does not mean that patients should be permitted to direct their physicians to administer a lethal injection, even if they are unable to commit suicide by any other means. Legislative proposals to legalize assisted suicide now pending in many states contain similar limitations.[45] In fact, these distinctions were critical to the success of the referendum to legalize physician-assisted suicide in Oregon.[46]

Characterizing the refusal of life-sustaining treatment as a form of assisted suicide, however, would make it impossible (and probably unconstitutional) to limit assisted suicide to these narrow categories of cases. First, the claim that assisted suicide could be limited to terminally ill patients ignores the fact that the right to refuse treatment has not been limited to patients who are terminally ill. For example, in Bouvia v. Superior Court,[47] the California Court of Appeals authorized the removal of a feeding tube from a young woman afflicted with severe cerebral palsy, who had years of life ahead of her, rejecting efforts to limit the right to refuse treatment to patients who are terminally ill. As the Bouvia court observed, “if [the] right to choose may not be exercised because there remains to [the patient], in the opinion of a court, a physician or some committee, a certain arbitrary number of years, months, or days, [the] right will have lost its value and meaning.” Other state courts have also rejected terminal illness as a constitutional benchmark,[48] and the opinions of the Supreme Court justices in Cruzan -- a case involving a patient who was not terminally ill -- suggest that the United States Supreme Court would do the same thing if it were directly confronted with the question.[49] If terminal illness is not an appropriate prerequisite for the refusal of life-sustaining treatment, and the refusal of treatment “is nothing more nor less than assisted suicide,” how can other forms of assisted suicide[50] be limited to patients who are terminally ill?

The same is true for the claim that assisted suicide could be limited to competent patients who make a contemporaneous request for physician-assisted death. As the New Jersey Supreme Court observed in the case of Karen Ann Quinlan, if a patient’s decision to forgo life-sustaining treatment is deserving of legal recognition, “it should not be discarded solely on the basis that her condition prevents her conscious exercise of the choice.”[51] To protect individuals’ right to refuse unwanted life-sustaining treatment after a loss of capacity, the law has created mechanisms like living wills, health care proxies, and surrogate decision-making, all of which rely on a good-faith assessment of the incapacitated patient’s wishes and/or best interests by health care professionals, family members, and close friends. If the refusal of treatment is a form of suicide, and it is permitted for patients without decision-making capacity, other forms of suicide would have to be permitted for incapacitated patients as well. At a minimum, it would be impossible to deny the right to incapacitated patients who have specifically requested assisted suicide as part of an advance directive, or who have given a relative or friend explicit decision-making authority over treatment decisions by signing a health care proxy. In fact, a footnote in Judge Reinhardt’s opinion in Glucksberg directly opens the door to such practices, by stating that “a decision of a duly appointed surrogate decision maker is for all legal purposes the decision of the patient himself.”[52] What this means is that, even if the law is never changed to legalize euthanasia for the incapacitated, surrogate decision-makers could authorize the provision of lethal drugs to incapacitated patients by consenting to assisted suicide on the patient’s behalf.

Moreover, characterizing the refusal of life-sustaining treatment as a form of suicide is inconsistent with the claim that physicians could be allowed to help patients commit suicide by prescribing lethal drugs, but not by providing lethal injections at a patient’s request. If it is unfair to distinguish between “hastening death” by refusing life-sustaining treatment and “hastening death” by self-administering lethal drugs, how can it be acceptable to distinguish between self-administering lethal drugs and instructing a physician to administer those same drugs directly?[53] Allowing physicians to prescribe lethal drugs but not to provide lethal injections would discriminate against patients who want to commit suicide but are physically unable to pick up or swallow a pill. These patients may in fact be suffering more than their able-bodied counterparts, and their claims for assistance may therefore appear more deserving of societal respect.

Finally, even what are characterized as “procedural” limitations on the right to physician-assisted suicide, approved by the Second and Ninth Circuits and endorsed by virtually all supporters of legalization, would be difficult to defend if the refusal of life-sustaining treatment is seen as a form of assisted suicide as a matter of law. If, as Judge Miner claims, any distinctions between the refusal of treatment and other forms of “hastening death” are arbitrary and unconstitutional, there would be no basis for requiring candidates for assisted suicide to submit to waiting periods, second opinions, and committee review. Such requirements are not imposed on individuals who seek to refuse life-sustaining treatment, and any effort to introduce them would undoubtedly be seen as burdensome and intrusive. “[I]f there is really no moral or legal difference between ‘allowing to die’ and ‘assisting suicide’ -- if, as Judge Miner opines, adding [physician-assisted suicide] to our repertoire of choices would not add one iota of additional risk to individuals or society over and above those we already countenance -- then it would seem that encumbering the choice for [physician-assisted suicide] with all sorts of extra protective devices would lack constitutional validity.”[54]

The balance between the benefits and risks likely to result from the legalization of physician-assisted suicide is extremely different from a similar balancing in the context of decisions to refuse medical treatment.

In When Death is Sought, we concluded that the legalization of assisted suicide would create insurmountable risks of mistake and abuse. To the extent the Second and Ninth Circuit recognized these risks, they dismissed them as irrelevant, on the theory that similar risks apply when patients refuse life-sustaining medical treatment. The fact that similar risks exist in both situations, however, does not mean that the risks have the same implications for law and clinical practice. The critical question is whether the risks can be mitigated through careful regulation, and, if not, whether they outweigh the reasons advanced for changing the law. On both of these grounds, the risks associated with legalizing assisted suicide are fundamentally different from those involved in respecting patients’ refusals of life-sustaining medical treatment.

First, the risks associated with legalizing assisted suicide would be far more difficult to regulate than the risks involved in refusing life-sustaining treatment. Many decisions to refuse life-sustaining treatment -- particularly decisions to withdraw respirators and feeding tubes -- take place in hospitals and nursing homes. By contrast, decisions about assisted suicide are likely to take place at home or in a physician’s office. It is comparatively easy to require second opinions, committee oversight, and retrospective monitoring in institutional settings. Outside of hospitals and nursing homes, “effective oversight to minimize error or abuse would be more difficult, if not unrealizable.” [55]

Second, with the refusal of treatment, the balance between the risks and the underlying individual right at stake yields different results from a similar balancing in the context of assisted suicide. On the risk side of the ledger, any harms that might result from the inappropriate refusal of treatment extend only to individuals who are suffering from an underlying condition that makes it impossible to live without invasive medical support. The size of this group, although not negligible, is inherently limited, and “just about everyone in this category must be very bad off indeed.” [56] With assisted suicide, by contrast, the risk of mistake and abuse is considerably larger, because anyone who takes lethal drugs will die, regardless of any underlying pathology. As Seth Kreimer has argued, “[t]he quantitative distinction between some and all can be a legitimate predicate for the qualitative distinction between permission and prohibition.” [57]

At the same time, the individual and societal need for a broad right to refuse treatment is far greater than the need for changing the law to allow physicians to help patients commit suicide with lethal drugs. If the law did not permit patients to refuse life-sustaining treatment, dying patients would be forced to submit to any procedure that might potentially extend their lives, no matter how burdensome. The result -- strapping patients down, pumping them with drugs, sticking tubes into them, and cutting them open to perform surgery -- would be a brutal assault on individual rights and, in many cases, sincerely-held religious beliefs.[58] By contrast, the legal prohibition of assisted suicide prevents patients from obtaining a physician’s assistance in escaping a situation imposed by nature, but does not impose any additional harm not caused by the patient’s own injury or disease.[59] Moreover, “to the extent that laws prohibiting assisted suicide and euthanasia impose a burden, they do so only for individuals who make an informed, competent choice to have their lives artificially shortened, and who cannot do so without another person’s aid. As studies have confirmed, very few individuals fall into this group, particularly if appropriate pain relief and supportive care are provided.” [60] The refusal of treatment, by contrast, is an integral part of everyday medical practice.[61] Prohibiting such decisions would therefore constitute a burden to individual autonomy in a significantly larger number of cases.

The Distinction Between Administering High Doses of Opioids to Relieve Pain and “Physician-assisted Death”

Some proponents of legalizing assisted suicide argue that the practice is indistinguishable from another, widely-accepted, aspect of medical care: the use of morphine and other opioids to relieve pain. In a 1994 article in The New York Times, Thomas Preston, a cardiologist, stated that the use of morphine drips “is undeniably euthanasia, hidden by the cosmetics of professional tradition and language.” According to Dr. Preston, the continuous injection of morphine into a patient’s vein will inevitably lead to the patient’s death by “curtailing her breathing.” Acceptance of the practice, he wrote, is “society’s wink to euthanasia,” and demonstrates that, despite existing legal prohibitions, “euthanasia is widespread now.”[62]

Judge Reinhardt relied heavily on this argument in his opinion in Glucksberg. “As part of the tradition of administering comfort care,” he wrote, “doctors have been supplying the causal agent of patients’ deaths for decades.” When physicians administer morphine drips for the relief of pain, “the actual cause of the patient’s death is the drug administered by the physician or by a person acting under his supervision or direction.” Because physicians are already causing patients’ deaths by administering morphine drips, Judge Reinhardt concluded, the State cannot assert an interest in preventing physicians from causing death by prescribing lethal drugs for patients to self-administer.[63] The court dismissed the State’s reliance on differences in intention, because “one of the known effects in each case is to hasten the end of the patient’s life.”[64]

The effort to characterize morphine drips as a form of covert euthanasia is extremely misguided. First, as a factual matter, the causal relationship between morphine drips and patients’ deaths is far less clear than Dr. Preston or Judge Reinhardt contend. While high doses of morphine can depress respiration when administered to patients who have not developed tolerance to the drug, physicians who treat patients with morphine for the relief of pain increase the doses gradually, so that tolerance can develop. Dr. Kathleen Foley, chief of the pain service at Memorial Sloan-Kettering Cancer Center, has concluded that “[t]he rapid development of tolerance to the respiratory depressant effects allows for escalation of the opioid dose in some patients to very high doses,” and that “[t]here appears to be no limit to tolerance” when the drug is administered properly. [65] The claim that the use of morphine at properly titrated levels “hastens” patients’ deaths, based on the effects of high doses of morphine on patients who have not developed tolerance, is entirely unfounded. It represents one of many myths about the consequences of using narcotics in the clinical setting, which have themselves contributed to the undermedication of patients experiencing treatable pain.

Second, and more importantly, the fact that morphine drips may accelerate patients’ deaths in some cases does not make their use equivalent to assisted suicide or euthanasia. “Just as a surgeon might undertake risky heart surgery knowing that the patient may die on the table, so the conscientious physician can risk suppressing the patient’s respiratory drive and thus hastening death so long as she is pursuing a valid medical objective and there are no better (less risky) options at hand.”[66] As the President’s Commission observed, “the moral issue is whether or not the decisionmakers have considered the full range of foreseeable effects, have knowingly accepted whatever risk of death is entailed, and have found the risk to be justified in light of the paucity and undesirability of other options.”[67] These observations are consistent with the legal concept of recklessness, which is defined as the conscious disregard of a substantial and “unjustifiable” risk. “This definition necessarily excludes situations where the benefit of taking action outweighs the likelihood that the action will cause harm.” [68] Thus, physicians are not permitted to prescribe morphine for minor headaches, when ordinary aspirin would work as well, but they can (and, indeed, should) for the pain associated with terminal illness, assuming that no other less risky options exist. This does not mean that the physician can administer opioids indiscriminately: the doctrine of recklessness requires the physician to undertake a good-faith balancing of the benefits and risks.[69] Instead, it recognizes that medical treatment sometimes requires significant trade-offs, and that acceptance of negative consequences for legitimate medical purposes is not equivalent to causing those consequences for their own sake.

Just as conflating the refusal of treatment with assisted suicide is likely to undermine patients’ ability to control their medical treatment, telling physicians that an unintended death resulting from the provision of necessary palliative treatment is a form of covert euthanasia is likely to result in many more patients experiencing unrelieved pain. As John Arras has pointed out, “many physicians would sooner give up their allegiance to adequate pain control than their opposition to assisted suicide and euthanasia.” [70] Characterizing the provision of pain relief as a form of euthanasia may well lead to an increase in needless suffering at the end of life. Advocates of legalizing assisted suicide should think carefully about the consequences of this argument for compassionate end-of-life care.


For all of the reasons set forth in When Death is Sought, many of which have been highlighted here, we continue to believe that legalized physician-assisted suicide would be profoundly dangerous for large segments of the population. Even those who support the legalization of physician-assisted suicide, however, should be concerned about the premises on which arguments for legalization are based. Assisted suicide for relatively rare cases of unrelievable suffering should not be justified by arguments that undermine the right to refuse medical treatment, which affects virtually every individual who ever seeks out medical care. The legalization of assisted suicide should also not jeopardize physicians’ willingness to administer effective medication for the treatment of severe pain, by claiming that death is an inevitable consequence of high doses of opioids, or by implying that physicians are legally and ethically accountable for the unintended harmful consequences of legitimate medical care. Maintaining the distinctions between assisted suicide, the refusal of treatment, and the use of high doses of opioids for the relief of pain, is essential to a coherent policy of end-of-life medical care. Conflating these issues may be rhetorically powerful for those who wish to legalize assisted suicide, but it will ultimately weaken the autonomy of patients at the end of life.

The widespread public interest in physician-assisted suicide represents a symptom of a much larger problem: our collective failure to respond adequately to the suffering that patients often experience at the end of life. Improving palliative care, and attending to the psychological, spiritual, and social needs of dying patients, must be a critical national priority. Whether or not assisted suicide is ultimately legalized, we hope that those on all sides of the debate over legalization will join forces to help achieve this important goal.


[1] The New York State Task Force on Life and the Law, When Death is Sought: Assisted Suicide and Euthanasia in the Medical Context (New York: The New York State Task Force on Life and the Law, 1994).

[2] Compassion in Dying v. Washington, 79 F.3d 790 (9th Cir.) (en banc), cert. granted sub nom. Washington v. Glucksberg, 65 U.S.L.W. 3254 (1996).

[3] 80 F.3d 716 (2d Cir.), cert. granted, 65 U.S.L.W. 3254 (1996).

[4] 80 F.3d at 719; see also 79 F.3d at 797 n.7.

[5] 79 F.3d at 813. Although we will not address the point at length here, we believe that Judge Reinhardt’s analogy to the right to choose an abortion is seriously flawed. As argued eloquently by the United States Solicitor General, that right “implicates a constellation of liberty and equality rights” not applicable to decisions about “hastening death.” Brief for the United States as Amicus Curiae Supporting Petitioners, Washington v. Glucksberg, No. 96-110, November 1996, at 15.

[6] 79 F.3d at 813.

[7] 497 U.S. 261 (1990).

[8] 79 F.3d at 816.

[9] Id. at 824.

[10] Id.

[11]478 U.S. 186 (1986).

[12] 80 F.3d at 723 (quoting Moore v. City of East Cleveland, 431 U.S. 494, 503 (1977)).

[13] Id. (quoting Palko v. Connecticut, 302 U.S. 319, 325-26 (1937)).

[14] Id. at 729.

[15] Id.

[16] The New York State Task Force on Life and the Law, Do-Not-Resuscitate Orders: The Proposed Legislation and Report of the New York State Task Force on Life and the Law (New York: The New York State Task Force on Life and the Law, 2d ed. 1988): 14.

[17] The New York State Task Force on Life and the Law, Life-Sustaining Treatment: Making Decisions and Appointing a Health Care Agent (New York: The New York State Task Force on Life and the Law, 1987): 41.

[18] The New York State Task Force on Life and the Law, When Others Must Choose: Deciding for Patients Without Capacity (New York: The New York State Task Force on Life and the Law, 1992): 222.

[19] When Death is Sought, at vii.

[20] Id. at 146-48.

[21] The New York State Task Force on Life and the Law, Life-Sustaining Treatment, at 40.

[22] In re Conroy, 98 N.J. 321, 486 A.2d 1209 (1985).

[23] J. Arras, “Physician Assisted Suicide and Euthanasia: Three Approaches to Social Policy,” Journal of Contemporary Health Law and Policy (forthcoming, 1997).

[24] Id.

[25] 79 F.3d at 821-22.

[26] See, e.g., In re Conroy, 98 N.J. 321, 347 (“The patient’s ability to control his bodily integrity through informed consent is significant only when one recognizes that this right also encompasses a right to informed refusal. ... Thus, a competent adult person generally has the right to decline to have any medical treatment initiated or continued.”).

[27] See, e.g., Bartling v. Superior Court, 163 Cal.App.3d 186, 194 (1984) (“[W]here a doctor performs treatment in the absence of an informed consent, there is an actionable battery. The obvious corollary to this principle is that a competent adult patient has the legal right to refuse medical treatment.”).

[28] Cruzan v. Director, Missouri Department of Health, 497 U.S. 261, 270 (1990).

[29] Id.

[30] Arras, “Physician Assisted Suicide and Euthanasia.”

[31] See, e.g., Munn v. Algee, 924 F.2d 568 (5th Cir. 1991).

[32] In re Peter, 108 N.J. 365, 529 A.2d 419 (1987).

[33] This point is addressed at length in H.L.A. Hart & T. Honore, Causation and the Law, 2d ed. (Oxford: Oxford University Press, 1985).

[34] D. Callahan, The Troubled Dream of Life: In Search of a Peaceful Death (New York: Simon & Schuster, 1993): 77.

[35] Although there are exceptions to this principle, such as when an individual’s refusal of treatment will endanger the welfare of other persons, “in order to override the patient’s right to control his care and treatment, the State’s interest must be compelling.” Rivers v. Katz, 67 N.Y.2d 485, 504 N.Y.S.2d 74, 80 (1986) (describing limited circumstances in which the forcible medication of patients with decision-making capacity would be justified as an exercise of the State’s police power).

[36] By contrast, when the withdrawal of life-sustaining treatment is not based on the patient’s choice, it is appropriate to consider it the cause of death as a legal and ethical matter. For example, the withdrawal of a respirator by a “greedy and hostile son” who thinks “his inheritance will be dissipated by a long and expensive hospitalization” is considered the cause of his mother’s death, and any attempt to claim that the underlying disease was the cause of death “would rightly be dismissed as transparent sophistry.” D. Brock, “Voluntary Active Euthanasia,” Hastings Center Report 22, no. 2 (1992): 10-22, at 13. Characterizing the disease as the cause of death would deny the blameworthiness of the son’s wrongful conduct, and is therefore unthinkable, despite the fact that it is at least partially true.

[37] 80 F.3d at 729.

[38] See, e.g., E.J. Cassell, “The Sorcerer’s Broom: Medicine’s Rampant Technology,” Hastings Center Report 23, no. 6 (1993): 32-39 (“Technologies come into being to serve the purposes of their users, but ultimately their users redefine their own goals in terms of the technology.”).

[39] G.J. Annas, “The Promised End – Constitutional Aspects of Physician-Assisted Suicide,” N. Eng. J. Med. 335 (1996): 683-87 (“Since the failure to perform cardiopulmonary resuscitation always ‘hastens death,” … patients who refuse cardiopulmonary resuscitation would always be committing suicide (and doctors who write do-not-resuscitate orders would always be assisting suicide).”).

[40] See, e.g., T.L. Beauchamp & J.F. Childress, Principles of Biomedical Ethics, 3d ed. (New York: Oxford University Press, 1989): 148; President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Deciding to Forego Life-Sustaining Treatment (Washington: U.S. Government Printing Office, 1983): 73-77.

[41] See generally The SUPPORT Principal Investigators, “A Controlled Trial to Improve Care for Seriously Ill Hospitalized Patients,” JAMA 274 (1995): 1591-98 (reporting widespread use of aggressive treatment in dying patients, regardless of existence of advance directive).

[42] F.G. Miller, T.E. Quill, H. Brody, J.C. Fletcher, L.O. Gostin, and D.E. Meier, “Regulating Physician-Assisted Death,” N. Eng. J. Med. 331 (1994): 119-23, at 119.

[43] H. Brody, “Assisted Death -- A Compassionate Response to a Medical Failure,” N. Eng. J. Med. 327 (1992): 1384-88, at 1385.

[44] 79 F.3d at 820.

[45] See generally D. Callahan & M. White, “The Legalization of Physician-Assisted Suicide: Creating a Regulatory Potemkin Village,” University of Richmond Law Review 30 (1996): 1-83.

[46] E.J. Emmanuel & E. Daniels, “Oregon’s Physician-Assisted Suicide Law,” Archives of Internal Medicine 156 (1996): 825-29.

[47] 225 Cal. Rptr. 297 (Cal. Ct. App. 1986).

[48] See, e.g., Fosmire v. Nicoleau, 75 N.Y.2d 218, 551 N.Y.S.2d 876 (1990); In re Bergstedt, 106 Nev. 808 (1990).

[49] G.J. Annas, “The Promised End,” at 686 (Cruzan stands for the proposition “that an adult need not be terminally ill to refuse treatment”).

[50] Quill, 80 F.3d at 729.

[51] 70 N.J. at 41, 335 A.2d at 664.

[52] 79 F.3d at 832, n.120.

[53] As Erich Lowey has argued, “[k]illing others who are terminally ill at their own request when they are incapacitated and unable to implement their own wishes is a form of assisted suicide in circumstances where nonassisted suicide is no longer possible.” E.H. Lowey, “Healing and Killing, Harming and Not Harming: Physician Participation in Euthanasia and Capital Punishment,” Journal of Clinical Ethics 3 (1992): 29-34.

[54] Arras, “Physician Assisted Suicide and Euthanasia.”

[56] C.H. Coleman & T.E. Miller, “Stemming the Tide: Assisted Suicide and the Constitution,” Journal of Law, Medicine & Ethics 23 (1995): 389-97, at 394.

[57] Arras, “Physician Assisted Suicide and Euthanasia.” S. Kreimer, “Does Pro-Choice Mean Pro-Kevorkian? An Essay on Roe, Casey, and the Right to Die,” American University Law Review 44 (1995): 803, 841.

[58] See, e.g., In re A.C., 573 A.2d 1235, 1244 n.8 (D.C. App. 1990) (en banc) (“Enforcement could be accomplished only through physical force or its equivalent. [The patient] would have to be fastened with restraints to the operating table, or perhaps involuntarily rendered unconscious by forcibly injecting her with an anesthetic, and then subjected to unwanted major surgery. Such actions would surely give one pause in a civilized society, especially when [the patient] had done no wrong.”).

[59] See J. Rubenfeld, “The Right of Privacy,” Harvard Law Review 102 (1989): 737, 795.

[60] When Death is Sought, at 72.

[61] See G.R. Scofield, “Exposing Some Myths About Physician-Assisted Suicide,” Seattle University Law Review 473 (1995): 473, 481 (“[T]he only way we can offer patients and doctors the chance to prolong life -- use life-sustaining treatment -- is by also allowing them to decide when to cease such efforts.”).

[62] T.A. Preston, “Killing Pain, Ending Life,” The New York Times, Nov. 1, 1994, at A27.

[63] 79 F.3d at 823.

[64] Id. at 824 (emphasis added).

[65] K.M. Foley, “Controversies in Cancer Pain: Medical Perspectives,” Cancer 63 (1989): 2257-65, at 2261-62; see also W.C. Wilson, N.G. Smedira, & C. Fink, “Ordering and Administering of Sedatives and Analgesics During the Withholding and Withdrawal of Life Support From Critically Ill Patients,” JAMA 267 (1992): 949-53 (finding “no evidence that death actually was hastened by the administration of drugs,” and that, “if anything,” the data “suggest that death occurred earlier in the patients who did not receive drugs”).

[66] Arras, “Physician Assisted Suicide and Euthanasia.”

[67] President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Deciding to Forgo Life-Sustaining Treatment (Washington: U.S. Government Printing Office, 1983), 82.

[68] When Death is Sought, at 62; see also W.R. LaFave and A.W. Scott, Jr., Substantive Criminal Law, vol. 1 (St. Paul: West Publishing Co., 1986), § 3.7, at 327 (“The test for reasonableness in creating risk is ... said to be determined by weighing the magnitude of the risk of harm against the utility of the actor’s conduct.”).

[69] See N.J. Cantor & G.C. Thomas, III, “Pain Relief, Acceleration of Death, and the Criminal Law,” Kennedy Institute of Ethics Journal 6 (1996): 107-27.

[70]Arras, “Physician Assisted Suicide and Euthanasia.”