Clinical Ethics in Anesthesiology. A Case-Based Textbook

2. End-of-life issues

22. Physician aid-in-dying and euthanasia

Alex Mauron and Samia Hurst

The Case

Mr. B., a successful 79-year-old scientist, has a very aggressive malignancy, is no longer able to work or function independently and is weakening rapidly. Oral medication helps somewhat with his bone pain but does not alleviate his severe dyspnea. He asks his physician for medication to “put him out of his misery”, but is told that such treatment is “illegal.” The day before his planned admission to a nursing facility for hospice care, Mr. B., with a final surge of energy, goes to the shed behind his house and shoots himself fatally. His written farewell note to his family ends with “it is a shame that a man has to do this for himself.”

This case occurred 50 years ago; what would have happened today?

Should physicians end a patient’s life at his or her request (as in euthanasia), or otherwise collaborate with the ending of life at a patient’s initiative (as in assisted suicide)? This question unavoidably raises two separate ethical issues. One regards the general prohibition of killing, its justifications, and possible exceptions; the other concerns the specific obligations of physicians and healthcare personnel in this respect. This question also demands a consideration of historical context, since the phenomenology of dying has changed a great deal over time. For centuries “natural death” was conceived of by physicians and lay people alike as an experience that was at once tamed by familiarity, and yet utterly unavoidable. Dealing with death had much to do with social mores and religious rituals, little with medical interventions, which were in any event quite incapable of forestalling an inevitable demise. Modern medicine, especially emergency medicine and critical care, has changed all that and often made impending death the locus of complex decisions and choices. Research, including the classical ETHICUS study,1 has shown conclusively that for a majority of patients dying in a Western intensive care unit, treatment limitations were in force, albeit with significant variations according to culture and religion.2 Foregoing life-sustaining treatment on account of futility and/or patient wishes has become standard procedure and there is a vast literature discussing how to do it in an ethically defensible way.

The fatality of impending death has been replaced by what is aptly called “end-of-life decisions.” Do-not-resuscitate (DNR) orders, advance directives, practice guidelines and the like are tools for choosing and deciding where formerly no choice existed and no decision was required. This shift has implications for the way physicians should classify their death-related actions. The smug distinction between “letting nature take its course” (innocent) and “active killing” (guilty) is no longer tenable once it is clear that acceptable forms of treatment limitation are causally connected to the patient’s death. That does not necessarily mean that the traditional distinction between killing and letting die is moot, but it does imply that the relevant distinctions are more subtle and go beyond a mechanical analysis of causes and effects, actions and omissions.

The intentional ending of a human life at this person’s request thus raises two distinctive ethical issues. It is obvious that the moral and legal permissibility of voluntary death, be it at another’s hand, or with another’s help, or wholly self-administered, has been a topic of philosophical and religious controversy for millennia. The debate still goes on and is reflected in the heterogeneity of legislation concerning euthanasia, assisted suicide (with or without physician involvement), and even suicide itself. At the same time, there has been a parallel debate in medical ethics, which is about the professional duty of physicians: does it necessitate the preservation of life no matter what? If not (as nearly everybody agrees), what are the sorts of end-of-life management strategies that are ethically acceptable? Do they include physician-assisted suicide and voluntary active euthanasia? What about terminal sedation? Both the broader societal discussion and the debate within medical ethics are important. They are related, to be sure, but must not be confused.


“Physician aid-in-dying” is a somewhat ambiguous term, requiring clarification. Although general questions regarding end-of-life care and decision-making are related to our topic, not all of them are part of it. This chapter is not about interrupting life support, DNR orders, or palliative care, but about the wilful termination of a patient’s life by a physician and the willing collaboration of a physician to the patient’s suicide. The term “physician-assisted death” used by some authors is clear and covers our subject matter. Some practices may be seen as borderline, such as terminal sedation, and we will therefore discuss it also since it is a specific concern of anesthesiologists. The issue of the withdrawal of food and hydration from comatose patients, or from dying patients who request it, is often drawn into the legal controversy on physician-assisted death. Examples are the case of Terry Schiavo3 or the very recent case of Eluana Englaro in Italy (a bill currently before the Italian senate would make hydration and alimentation compulsory).4 In fact, these broader end-of-life issues have a way of spilling over into the debate on physician-assisted death, in part because similar arguments have been used in both areas. As noted by Dworkin in 2007:

When these practices (i.e., withdrawal of life-support, accepting a refusal of artificial hydration and alimentation, etc.) were proposed by physicians, and when they were sanctioned by the courts, there were the same kinds of prediction made by opponents as are now made about physician-assisted suicide. Patients would be manipulated or coerced into making, or agreeing, to such practices. Patients who might be cured by new discoveries would mistakenly request to be removed from life-support. The slippery slope from voluntarily requesting DNR to having such orders entered on their chart without consent would be traversed.5

Nevertheless, we will largely limit ourselves to a consideration of voluntary active euthanasia and physician-assisted suicide. As regards euthanasia, this means that both nonvoluntary and involuntary euthanasia are outside our remit (and universally seen as beyond the pale). We will also focus on physician-assisted suicide, even though we have to broaden our perspective occasionally, as in discussing the Swiss legal situation.

Ethical arguments and legislation

Physician-assisted death harks back to a general debate on voluntary death that has been going on for millennia, and some of the arguments given today have been present throughout this history. They closely parallel arguments made in favor of or against euthanasia and assisted suicide.6

Arguments against euthanasia and assisted suicide

(1) That although respect for autonomy is important, it should not extend to euthanasia or assisted suicide. For long periods of history, lives were considered to belong to God, or the State, or both. In monarchies, suicide was a common law felony against the king. More recently, philosophers have argued that suicide cannot be autonomous because it ends the possibility of autonomy. Nor can a request for assisted death be considered autonomous if it is a symptom of a disease, such as depression: if this is always the case, these requests are never autonomous.

(2) That beneficence does not necessarily extend to authorizing assisted death, and should concentrate exclusively on the use of alternatives and the promotion of well-conducted and universally available palliative care. Furthermore, requests for assisted suicides are often cries for help, which should be answered with appropriate understanding and care, rather than assisted death. This is part of suicide prevention, a major public health goal.

(3) That although we allow withdrawal or withholding of life-sustaining interventions, this is significantly different from allowing euthanasia or assisted suicide. The actual acts are different: injecting a lethal medication, or removing an invasive apparatus, are not the same. Moreover, physicians who withdraw treatment do not intend to end the patient’s life, but to put aside a medical intervention which could itself be the cause of further suffering.

(4) That allowing euthanasia or assisted suicide would have severe adverse consequences, such as a trivialization of suicide, or a “slippery slope” leading to cases of euthanasia without patients’ consent. Opponents point to data suggesting that such involuntary euthanasia takes place in the Netherlands. They fear that patients may sometimes choose euthanasia under pressure, or to avoid becoming a burden on loved ones.

Arguments in favor of euthanasia and assisted suicide:

(1) That respect for autonomy does support allowing voluntary death, because autonomy means self-ownership of persons and the right to pursue our own goals in life, as long as we harm no one else. If autonomy means control over how we live, it also entails control over the way we die, which has been considered since Antiquity to be the one undeniable human freedom, and must be authorized as a liberty-right in a liberal polity. Decriminalization of suicide also recognizes that choosing to die can be a reasonable option, and makes assisted suicide a victimless crime.

(2) That even alleviation of physical pain has a failure rate, and even well developed palliative care cannot address all types of suffering with equal success. Sometimes, living with a painful disease may reasonably be seen as worse than death, we should therefore allow euthanasia or assisted suicide in cases where palliation is ineffective. The epidemiology of legal assisted suicide is very different from that of suicide generally, so it is possible to allow the first without jeopardizing public health efforts to prevent the latter. Although some requests for assisted death are cries for help, this is not always the case: attempting to understand all such requests in this way would represent “hermeneutic obstinacy” similar to the “therapeutic obstinacy” involved in persisting with life sustaining treatment over a patient’s refusal.7 The possibility of assisted death also reassures many people who are healthy, but know that they could become terminally ill someday, and that this could involve intractable suffering. The possibility of euthanasia or assisted suicide can alleviate some of their fears.

(3) That euthanasia and assisted suicide are not significantly different from practices such as withdrawal of life sustaining treatments at the end of life, which are broadly accepted as ethically justified. In both types of cases the patient requests to die, the physician intends to end the patient’s life, and the patient indeed dies. The only difference is the “proximate cause”: when life-sustaining treatment is removed, the patient is killed by his disease. In the case of euthanasia, the direct cause of death is the physician’s act, in the case of assisted suicide, it is the patient’s act. Proponents of voluntary euthanasia question the moral significance of this distinction.

(4) That “slippery slope” arguments are not valid: controversial practices can be regulated and controlled. It would be surprising if we were able to abide by a rule forbidding euthanasia, but not by a rule limiting the circumstances where it can take place. Countries where assisted suicide and/or euthanasia are legal have not experienced a slippery slope; since legislation authorizing euthanasia and assisted suicide often comes with stricter requirements for notification, it is difficult to know whether cases of involuntary euthanasia increase or decrease following the legislative change.

The ethical status of physician participation in euthanasia and suicide assistance is a distinct point in this debate. Opponents highlight the tension between assisting death and duties to preserve life, and risks to the physician–patient relationship. Proponents view assisting death as an intrinsic part of a humane response to suffering, and a form of the respect owed to patients’ self-determination. Moreover, the idea that the physician–patient relationship would suffer from allowing euthanasia or physician assisted suicide is speculative, and data suggests that legalizing aid-in-dying would not affect patients’ trust in their physician negatively.8

Currently, assisted suicide is legal in the US states of Oregon, Washington and Montana, and in Switzerland, Luxemburg, Belgium, and the Netherlands. Euthanasia is legal in the last three countries. All but one of these legislations provides a detailed description of the circumstances where assisted death is permissible, require physician involvement, and mandate notification. The exception is Switzerland, where nonphysicians can assist suicide and where the only legal requirements are that the patient be capable of decision-making and perform the lethal act himself, and that the helper be motivated only by altruistic motives. Typically, requirements for legal euthanasia or assisted suicide include that patients need to be capable of decision-making and persistently requesting death, and that there be otherwise intractable suffering.

Attitudes and practices

Public attitudes

Public attitudes toward assisted death vary between countries and with the sort of situation considered. In the US two-thirds of surveyed members of the general public favor euthanasia or assisted suicide when presented with the case of a patient in unremitting pain. When the case is changed to a patient suffering from functional debilitation, or who views life as no longer worth living, one-third no longer supports assisted dying, and one third still does.9 In Europe, public attitudes differ between countries, with higher rates of acceptance in countries such as The Netherlands, but also Denmark, Sweden, and France, and much lower rates in countries such as Romania, Malta, or Turkey. Factors associated with higher acceptance include weaker religious beliefs, but also younger age, and higher educational level.10

Professional attitudes

The American Medical Association, as well as physician associations in countries such as Norway, Germany, and the UK, have opposed euthanasia and assisted suicide, and physician participation specifically, maintaining that assisted dying conflicts with the role of doctors. This opposition, however, is not universal. Physician associations in The Netherlands do not maintain that medical assistance in dying conflicts with the physician’s role; the Swiss Academy for Medical Sciences states that assisted suicide is “not a medical act,” but nevertheless described circumstances where it finds it to be professionally acceptable for a physician to assist suicide.11


Data from The Netherlands is the most systematically reported, and shows rates of notification increasing from 18% in 1990 to 80% in 2005.12 In Switzerland, most suicide assistance is performed through right-to-die organizations rather than individual physicians. From 1990 to the mid 2000s, the number of suicide assistance cases has remained stable, with possibly a slight increase in the number of patients suffering from chronic rather than fatal disease.13

Serious requests for assisted death are made by only a minority of seriously ill patients. Intractable pain motivates requests infrequently. Psychological, social, or existential suffering, such as loss of community, autonomy, or meaning, seem to motivate euthanasia requests more often.14 Patients with symptoms of depression have been shown to be four times more likely to request euthanasia than those without such symptoms.15 Although these results should be taken seriously, they do not signify that all requests for assisted death are associated with depression. Fears that assisted dying would disproportionately affect vulnerable groups have also been voiced, but current data show otherwise.16

The link between assisted dying legislation and practices is complex. Up to 3.7% of US physicians reported having – illegally – practiced assisted suicide and up to 9.4% reported having practiced euthanasia.17 Intensive care unit physicians in France, where euthanasia is illegal, are more likely to report having practiced “deliberate administration of medication to speed death in patients with no chance of recovering a meaningful life” compared with their colleagues from 11 European countries, including The Netherlands.18 The Dutch Euthanasia Act seems to have been followed by a small decrease in the number of euthanasia and physician-assisted suicide cases.19 As Belgian doctors have pointed out, more transparent reporting is likely to increase the effectiveness of social control of these practices.20

Responding to requests for aid-in-dying

A request for euthanasia or assisted suicide is an emotionally taxing situation, where a thoughtful and respectful response is particularly important. Several elements are crucial. First, some requests are expressions of suffering rather than direct concrete requests for assisted death, but not all of them are. Second, adequate symptom management can significantly decrease the number of such requests. Third, the patient’s decision-making capacity is a necessary condition of euthanasia and assisted suicide also where these practices are legal. Fourth, addressing the issue of aid-in-dying explicitly and non judgmentally is preferable to the alternatives. Fifth, even in countries where euthanasia or assisted suicide is legal, there is no entitlement right to obtain it from a physician who disagrees.

Several approaches have been proposed. Two stages in responding to requests for euthanasia or

Figure 22.1

Figure 22.1 Responding to requests for euthanasia or assisted suicide

assisted suicide are outlined in Fig. 22.1. First, these requests should be carefully listened to. Many requests for assisted death are expressions of suffering, and are withdrawn by patients after appropriate symptom management. However, this is not always the case. Although the controversies and emotions surrounding this topic can make these discussions difficult, remaining non-judgmental is important to patient management. The patient’s request should be discussed and clarified. If she is truly making a request to die, and is competent to make decisions, a structured discussion about advance care planning should take place, and the possibility of writing advance directives should be offered.

In all circumstances, suffering should be assessed using a holistic “total suffering” palliative care approach. Depression should be screened for and treated if present. Where available, specialized palliative care expertise should be offered. Whenever possible, symptom management should be handled by a team of healthcare providers with complementary expertise. When they are likely to be called upon for specific interventions, such as palliative sedation, anaesthesiologists should be included in discussions regarding their indication in the patient’s specific circumstances. Importantly, palliative sedation does not constitute assisted dying in the sense discussed in this chapter. In some cases, it can be a part of symptom management at this stage.

Many requests will be dropped following these steps. However, some do persist. In such cases, where a competent patient persists in asking to die despite appropriate management of suffering, several situations exist.

One is the situation where a patient is under life-sustaining therapy. Controversies surrounding treatment withdrawal tend to increase as the invasiveness of the intervention decreases: withholding food and water is often more controversial than withholding ventilator support. However, if any life-sustaining intervention is being applied, its refusal by a competent patient is usually considered sufficient grounds to withdraw it.

When no life-sustaining intervention is being used, the next steps will depend on the legal status of euthanasia or assisted suicide, and on the physician’s own convictions. Where assisted death is not legal, or in situations that do not fulfill legally described criteria, palliative sedation can again be considered as an alternative at this stage. Indeed, it is sometimes preferred even in areas where assisted death would be legal. Finally, even in situations where euthanasia or assisted suicide would be legally authorised, there is no duty on the part of physicians to perform either intervention. In countries where assisted death is legal, an objecting physician can refer the patient whose request persists to a colleague. Where it is known in advance that a physician would in any case refuse to assist death, this should be made clear to the patient as early as possible. It is of course important to be clear on this point both in countries where assisted death is legal, and in countries where it is not.

Key points

• Palliation with therapeutic intent, including palliative sedation, is not equivalent to assisted death even in cases where the treatment may hasten death.

• Physician assisted suicide and voluntary euthanasia are legal in some jurisdictions.

• Arguments against physician assisted death usually refer to a violation of the physician’s role and/or the potential for coercion of vulnerable patients.

• Arguments for physician-assisted death usually refer to beneficence and patient autonomy.

• Although many patients who request aid-in- dying are depressed, not all requests are associated with depression, nor does the presence of depression necessarily invalidate a request for aid-in-dying.

• Requests for aid-in-dying should be taken seriously and carefully listened to. Screening for and treatment of problematic symptoms and depression may lead to withdrawal of the request.

• When a request for aid-in-dying persists and no life-sustaining treatments are being employed, the response will depend on the legal status of euthanasia or assisted suicide, and on the physician’s own convictions.


1* Sprung, C.L.S.L. Cohen, P. Sjokvist, al. (2003). End-of-life practices in European intensive care units: the Ethicus Study. JAMA290(6), 790–7.

2* Sprung, C.L.Maia, P.Bulow, al. (2007). The importance of religious affiliation and culture on end-of-life decisions in European intensive care units. Intens Care Med33, 1732–9.

3 Schindler v. Schiavo, 780 So.2d 176 (Fla. 2001).

4 Senate, I. (2009). “Bill no. 1369.” Retrieved 5 May, 2009, from

5* Dworkin, G. (2007). Physician-assisted death: the state of the debate. In The Oxford Handbook of BioethicsB. Steinbock, (ed). Oxford and New York: Oxford University Press, pp. 376–92.

6 Emanual, E.J. (2009). Euthanasia and physician-assisted suicide. UpTo Date online. Accessed June 3, 2010.

7* Mauron, A. (2006). La médecine moderne et l’assistance au suicide en Suisse. Synthèse du point de vue de la CNE. Beihilfe zum Suizid in der Schweiz, Beiträge aus Ethik, Recht, und MedizinC.Rehmann-SutterA. BondolfiJ. Fischer, and M. Leuthold (eds.). Nerm: Peter Lang.

8* Hall, M.Trachtenberg, F., and Dugan, E. (2005). The impact on patient trust of legalising physician aid-in-dying. J Med Ethics31(12), 693–7.

9 Wolfe, J.Fairclough, D.L.Clarridge, al. (1999). Stability of attitudes regarding physician-assisted suicide and euthanasia among oncology patients, physicians, and the general public. J Clin Oncol17(4), 1274.

10* Cohen, J.Marcoux, I.Bilsen, al. (2006). European public acceptance of euthanasia: socio-demographic and cultural factors associated with the acceptance of euthanasia in 33 European countries.Soc Sci Med63(3), 743–56.

11 Académie Suisse des Sciences Médicales. (2004). Directives médico-éthiques pour la prise en charge des patientes et patients en fin de vie.

12 Rurup, M.L.Buiting, H.M.Pasman, al. (2008). The reporting rate of euthanasia and physician-assisted suicide: a study of the trends. Med Care46(12), 1198–202.

13 Fischer, S.Huber, C.A.Imhof, al. (2008). Suicide assisted by two Swiss right-to-die organisations. J Med Ethics34(11), 810–14.

14* Georges, J.J.Onwuteaka-Philipsen, B.D.van der Heide, al. (2006). Requests to forgo potentially life-prolonging treatment and to hasten death in terminally ill cancer patients: a prospective study. J Pain Symptom Manage31(2), 100–10.

15 van der Lee, M.L., van der Bom, J.G.Swarte, al. (2005). Euthanasia and depression: a prospective cohort study among terminally ill cancer patients. J Clin Oncol23(27), 6607–12.

16* Battin, M.P., van der Heide, A.Ganzini, al. (2007). Legal physician-assisted dying in Oregon and the Netherlands: evidence concerning the impact on patients in “vulnerable” groups. J Med Ethics33(10), 591–7.

17 Emanuel, E.J. (2002). Euthanasia and physician-assisted suicide: a review of the empirical data from the United States. Arch Intern Med162(2), 142.

18 Vincent, J.L. (1999). Forgoing life support in western European intensive care units: the results of an ethical questionnaire. Crit Care Med27(8), 1626–33.

19 van der Heide, A.Onwuteaka-Philipsen, B.D.Rurup, al. (2007). End-of-life practices in the Netherlands under the Euthanasia Act. N Engl J Med356(19), 1957–65.

20 Smets, T.Bilsen, J.Cohen, al. (2009). The medical practice of euthanasia in Belgium and The Netherlands: legal notification, control and evaluation procedures. Health Policy90(2–3), 181–7.

Further reading

Ariès, P. (1974). Western Attitudes Toward Death: From the Middle Ages to the Present. Baltimore and London: The Johns Hopkins University Press.

Breitbart, W. (2010). Physician-assisted suicide ruling in Montana: struggling with care of the dying, responsibility, and freedom in big sky country. Palliat Support Care18, 1–6.

Emanuel, L. L. (1998). Facing requests for physician-assisted suicide: toward a practical and principled clinical skill set. JAMA280(7), 643–7.

Hurst, S. A. and Mauron, A. (2006). The ethics of palliative care and euthanasia: exploring common values. Palliat Med20(2), 107–12.

O’ Neill, B. and Fallon, M. (1997). ABC of palliative care. Principles of palliative care and pain control. BMJ315(7111), 801–4.