Death With Dignity or Life With Health Care Rationing

Patricia Illingworth
Northeastern University

Harold Bursztajn
Harvard Medical School

Patricia Illingworth, Department of Philosophy and Religion, Northeastern University; Harold Bursztajn, Department of Psychiatry, Harvard Medical School.
Correspondence concerning this article should be addressed to Patricia Illingworth, 361 Holmes Hall, Northeastern University, Boston, Massachusetts 02115-5000.

Psychology, Public Policy, and Law
2000, Vol. 6, No. 2, 314-321

The observations made of Oregon's Death with Dignity Act are reassuring to the extent that they accurately state that patients who seek relief under the Act view it as a mechanism to further their autonomy. But, given the nature of social data gathering, including the potential for response bias and self-validation, and the degree to which the concept of "autonomy" is open to a broad or narrow interpretation, one has to be careful not to accept this rinding at face value. It is also reassuring to read that contrary to what some people have supposed, the presence of an option to hasten death has not had a chilling effect on the prescription of pain medication. At the same time, however, the authors believe that it is important to ensure that these expressions of autonomy-based reasoning are indeed valid and authentic.

In this commentary, we argue that in view of what is reported in "Observations on the First Year of Oregon's Death With Dignity Act," it is reasonable to conclude that Oregon's Death with Dignity Act has the potential to serve the autonomy interests of the terminally ill by enriching opportunities for choice for those who choose to hasten death and those who decide not to do so. Nonetheless, in order for this potential to be fulfilled in the current health-care context, those who implement the Act need to ensure that the terminally ill understand how their choices are influenced by economic, psychological, and social factors. To this end, we identify three conditions that need to be met in order for true informed decisionmaking to take place and suggest areas that need to be monitored if the choices of the terminally ill are to be truly autonomous.

Before Jumping to Conclusions

The report (Coombs Lee & Werth, 2000) [l] stated that there were many more inquiries about physician-assisted suicide (PAS) than just those from people who opted to hasten their deaths under the Act. Some callers did not qualify for reasons having to do with their illness, and others took different courses of action. Apparently, for some callers a consultation with Compassion in Dying is a component of a deliberation process. Not all people have made up their minds in advance, and Compassion seems to be facilitating dialogue and a thoughtful approach. Nonetheless, because depressed and frightened people often have difficulty changing their minds because of cognitive constriction and rigidity, we are still concerned that the subgroup of callers who opt for PAS may suffer from masked depression where a desperate cry for help and overwhelming fear of death comes to be expressed as a giving up or as a counter phobic cry for assisted suicide. [2]

According to "Observations," all 10 patients who committed suicide took their medications at home, in the company of family, physicians, clergy, and Compassion's volunteers and hospice staff. Coombs Lee and Worth underscored that no patient died alone. It is worth noting, however, that most had lived alone. The ability to provide patients with control over the timing of their death, to enable them to surround themselves with their loved ones, and to ensure that they are with others at the time of death, are surely among the strengths of the Act. Still, the troubling thought arises that perhaps at least for some, the only way to secure the company of others when suffering is to arrange for a presuicide family-and-friends party.

Autonomy and Choice in the Shadow of a Terminal Illness

Concerns about autonomy and self-determination figure importantly among the reasons that patients gave for their decision to hasten death, especially when these terms are construed broadly. Under "elements of suffering," patients are reported to have identified "physical discomfort other than pain," "dependence on others for personal care," "restriction to bed or wheelchair," "loss of control over bodily function," "loss of control," "severe pain," and "loss of autonomy" as their primary reasons for seeking PAS. On a broad understanding of "autonomy" and "self determination," concerns about dependence, independence, and loss of control are really just different ways of talking about autonomy and self-determination. It is both understandable and sad that these autonomy issues figure so importantly in a list of "elements of suffering."

To learn from the report that many of those who seek PAS give autonomybased reasons for their choice confirms much of the moral raison d'Stre behind the policy giving people the option to choose PAS. But the desire to hasten death in order to protect one's autonomy and sense of control is surely ironic. Because desires and preferences are not writ in stone, one has to wonder if it is not possible to help people feel more at peace with their fear of loss of control as well as with actual loss of control.

Nonetheless, unlike White and Callahan, we are not disturbed because only a few people express this particular concern. First, although only a subset of people appear sufficiently concerned with control to make use of the Act, it would seem to be a mistake to stigmatize them for this. The world consists of a diverse group of people with varying needs for control. For many, an excessive concern for control would qualify as a neurotic wish. But as Woody Allen has made clear in movie after movie, and Sigmund Freud before him, neurosis is more the rule than the exception. Surely we do not want to be in a position of choosing among neuroses. In principle, Oregon's Death with Dignity Act simply leaves the options open to those who choose to avail themselves of it.

Moreover, the people who choose to die under the Act are not the only ones to benefit from it. As Batavia underscores in his concluding remarks, "every person dying from a terminal illness in Oregon benefited from the law." [3] The value of choice per se should not be underestimated. One does not have to choose an option in order to benefit from having it. Human beings thrive on choice. [4] John Stuart Mill, philosopher and psychologist, understood this point. Consider the following: [5]

The human faculties of perception, judgment, discriminative feeling, mental activity, and even moral preference are exercised only in making a choice . . . He who lets the world choose his plan of life for him has no need of any other faculty from the ape-like one of imitation. He who chooses his plan for himself employs all his faculties. He must use observation to see, reasoning and judgment to foresee,... and when he has decided, firmness and self-control to hold to his deliberate decision.

Mill's point is relevant to the case at hand. In the face of a choice about whether to die with assistance (by way of the Act) or to die without assistance, the terminally ill must now think about what best suits them as individuals; what, in the parlance of "autonomy," is authentic. Thus, the Act will benefit both those who choose to die with assistance as well as those who choose to die without it because it enables both groups to make a choice. People who choose not to die with physician assistance may feel greater comfort with their decision because they have made a choice. As Victor Frankl, in his "Man's Search for Meaning," [6] reiterates: even in the face of almost certain death, choice is possible, and where there is choice, there can be meaning and hope.

Second, although only a small number of people chose to die under the Act and specifically identified "control" as the reason for doing so, it cannot be presumed that their wish to do so is necessarily excessive. As the decision to control the time of death becomes easier for people to make, the group that makes it may become larger. Whether a desire is viewed as excessive or not is relative to the size of the class of people who make it. We can not really tell what the size of that class will be until the possibility of choosing PAS is widely available, public, and stigma free.

Concerns about autonomy surfaced again with respect to the four patients who indicated that an element of their suffering was the burden they would cause to others during an extended dying process. It is unclear how much weight should be given to patients' desires to protect their loved ones from the burden of caring for them, especially once their loved ones have indicated that they do not consider such care a burden. Arguably, the wish to protect others in this way is often consistent with people's overall conception of themselves and in this way is authentic. Parents, for example, may have always been the caretaker and may find that having their children care for them is inconsistent with their conception of self. But it can also indicate a depressive and even paranoid view of what one's dying will mean to others. Self-centered parents, accustomed to "being boss" and dominating their children, may, when depressed, fear that they will be treated as badly as they have cared for their children.

Physicians who fail to be fully open and honest about their beliefs and values concerning hastened death, and who either blindly refuse to participate or conversely encourage suicide can add to the patient's distress. The authors tell us of a case in which the patient had requested assisted death, had been refused, and had been promised a quiet and peaceful death but had been seriously disappointed. Although we are not given enough information to do much but speculate about what was going through the physician's mind, it is likely that the physician actually believed that he or she could manage the patient's symptoms. The question is, "did he allow his values and moral beliefs to lead the doctor into self-deception about what exactly the probabilities were that the physician could supply an easy death and that there was no alternative?" This is an area that is ripe for self-deception in both directions. Physicians who are overburdened and busy, frightened of running afoul of state or federal drug enforcement agencies, or favor PAS and are overly concerned about scarce-medical-resource use at the end of life, may be more prone to read their preferences into patients' wishes. The Act does have some built-in monitoring mechanisms to ensure that the physician's wish to hasten death is not determinative. This is important because overall we are concerned that nonreversible decisions that affect the patient are not made without the patient's full consent. Of course, physicians who object to PAS may find that they are more likely to be overly optimistic about their ability to provide a comfortable life before death. Because people only die once, there is a very real sense that whatever is done is always irreversible.

The requirement of a mental health evaluation can be a mechanism to ensure autonomy, but if applied mechanically, or in a perfunctory manner, it can also be an invasion or a neglect of individual autonomy, at a time when patients are sensitive to issues of autonomy and loss of control. Any requirement that ultimately makes psychiatrists gatekeepers is problematic because for many people, including some psychiatrists and psychologists, the decision to hasten the end of life can be prima facie evidence of a mental impairment. Moreover, it is questionable whether psychologists and psychiatrists have the requisite knowledge to detect subtle but significant sleep deprivation, medication, or metabolic compromises of mental function. Without a comprehensive capacity evaluation, however, many primary care or attending physicians will be insensitive to such subtle, but significant, impairments of capacity and will never suspect that a forensic consultation is needed or will be too concerned that managed care will not approve a forensic consultation. Rather than face rejection, they simply will not ask for such a consultation.

Given the evidence, it is too early to conclude that vulnerable members of the community—the socioeconomically disadvantaged, racial minorities, the disabled, and women—will not be disproportionately represented. For example, many of the concerns related to the disabled involve long-term effects of internalizing certain attitudes about the disabled. [7] One year is not enough time for any of these social attitudes to have been internalized and, in turn, have a visible effect. Moreover, one wonders about the definition of "disability" and whether it may not be too restricted. [8]

The fact that the overwhelming numbers of those requesting suicide assistance were living alone raises the question as to why they were alone. Certainly social impairment and withdrawal are among the early signs of disabling depression.

What ought to be made of the fact that palliative care improved after a request for assisted death had been made? First of all, one wonders what the evidence for this is. Is it simply patient avowals? Second, although the natural assumption is that once a request for hastened death has been made, caretakers feel freer to use pain medications that put life at risk, there may be other explanations for the change. In the extreme, it is also possible that caretakers, annoyed that patients are not making use of the option for hastened death, are less than fully helpful. A more benign explanation is that the request for assisted suicide drops the veil of denial and ignorance from the often all-too-busy health care providers and leads to more attention and effective response to patients' suffering (i.e., "the squeaky wheel gets the grease"). By the same token, given the climate of physician intimidation by state and federal bureaucracy regarding the prescription of controlled substances, a patient's request for medication to hasten death may embolden otherwise frightened physicians to prescribe needed medication for anxiety and panic relief.

We have a deep wish to share the optimism proffered by proponents of the Oregon assisted-suicide model. However, given what is at stake here, the initial data collected by the proponents is to be viewed with a critical eye. Our concern is not simply that the case numbers are small and pose a potential selection bias. The finding that fear of loss of control is the avowed reason for considering hastened death by terminally ill patients is ambiguous. It raises the question of whether many patients' stated preference for assistance in suicide is actually a cry for help. In other words, to the extent that this request takes the form of help in hastening death, an important corollary is whether such a solution is the result of subtle but significant impairment and restriction due to psychiatric disorders, social conditions, and managed-care influence on clinical practice.

Even in everyday life, relatively healthy individuals may manifest in some contexts an overvalued and fixed belief bordering on delusions such as perfectionism, omnipotence, or invulnerability. [9] With serious medical illness, both depressive and posttraumatic disorders are common and share as hallmark symptoms helplessness and fear of loss of control. [10] Whatever the external and bodily function reality, there are often biopsychosocial solutions to such fears of which patients are unaware, either because they have not been informed or because they are prevented from integrating such information by the restrictions on their own ability to communicate as well as the helplessness and shame they are experiencing. We are concerned that Compassion is unaware of these issues and therefore unable to include such information in the dialogue with patients. In the absence of social support, and with cost-based restrictions on meaningful doctorpatient communication ranging from time crunches to preapproval for forensic consultation, real but treatable helplessness, silent fear, and shame become all too often overlooked and overwhelming.

Moreover, PAS represents a certain outcome. It may be irresistible to people when they are faced with a painful and terminal, yet uncertain, death. The more certain dying patients can be that they will be assisted by their physicians in facing death by being comforted and will not be abandoned by a profit-seeking medical system, the less they may have to rely on the certainty associated with PAS. Thus autonomy and a sense a control can be achieved for some people in other ways—namely, by ensuring certainty through other mechanisms. It may be fruitful to explore the psychological connection between sharing uncertainty and autonomy. Perhaps the need to control death through PAS is greatest where uncertainty is high and the patient is alone to bear it. [11]

The Informed Consent Process in the Era of Managed Care

When confronted with the uncertainty of life's finitude and helplessness, and facing the certainty of death, many patients do not want to be alone. The sharing of life's uncertainty in a mutually respectful and supportive relationship may be an antidote to an otherwise unbearable aloneness. The informed consent process for terminal patients can be of some help in establishing a trusting and supportive frame for exploration. [12] In order for such a process to work, it has to include three elements.

First, there must be a competent decisionmaker. Given the high prevalence of persons living alone among those requesting assisted suicide, together with the difficulty of detecting depression in terminal-care medical settings, there is a likelihood that this subset of patients are covertly depressed. The masking of depressive and panic symptoms, which can subtly, but significantly, impair cognitive faculties (such as the ability to consider alternatives and to weigh the risks and benefits of being able to extend meaningful life) is well-recognized. This masking of depressive symptoms can occur because of the withdrawal and dissociation that often accompanies trauma and depression itself, because of undertreated pain and fear that often accompanies terminal illness, or because of the euphoric or dulling side-effects of pain medications. Such patient denial and dissociation of affect when frightened, alone, and facing death is increasingly likely to be experienced, unrecognized, and untreated when patients are offered relatively short-term primary care doctor-patient relationships with little or no access to psychodynamically informed psychiatric consultation. Moreover, caregiver reactions in terminal care are inherently vulnerable to unrecognized, but significant, perceptual biases such as an overidentification with a patient's occult depressive states extending to overwhelming yet hidden dread and horror. Such overidentification can take the form of glib rationalization in the form of largely rhetorical questions such as "is it not natural to feel depressed if you are alone and dying," [13] and lead to a premature resignation from the responsibility of caring for the patient by providing meaningful biopsychosocial support.

Second, there must be an informed decisionmaker. The above considerations, together with a variety of cost-based restrictions on care and cost and governmental restrictions on prescribing practices; intimidation of health-care providers and institutions by economic credentialing and deselecting of those individuals and institutional providers who engage in burdensome and often futile appeals; as well as governmental prosecution of physicians prescribing anxiety and painrelieving substances, result in physicians developing the habit of abandoning suffering patients. Because consistency and habit guide practice all the more in an increasingly time-squeezed clinical milieu, it is today likely that even when a patient has traditional fee-for-service insurance, the attitudes and habits fostered by managed care will restrict the clinician in what treatment alternatives are communicated to the patient. A deceptive and demoralizing silence, a kind of de facto "gag clause" as to relevant alternatives, is most often unrecognized, or defensively rationalized as "not wanting to disappoint the patient" or "hanging crepe." By the same token, once some physicians feel there is nothing more they can do to ease the patient's suffering, a kind of self-fulfilling prophesy of helplessness together with the difficulty of obtaining managed-care approval for consultation can prevent meaningful psychiatric referral. Rather, it takes a patient wanting to commit suicide to bring up a too little and too late referral.

Third, there must .be a free decisionmaker. Patients who are ill and alone are vulnerable to undue influence and coercion. In the managed-care era, where the patient's interest in extending meaningful life may clash with social and institutional provider interests, the "no exit" ambiance of ignored suffering may lead patients to feel that the only relief is a final exit in compliance with their caregivers unstated but reasonably inferred wishes. Such a preference, a variation of the "Stockholm Syndrome," often seen in de facto captive situations involves desperate patients overidentifying with the preferences of those on whom their lives depend rather than acting on their own authentic values. Under such conditions, combined with the covert coercion by psychically and financially pressured caregivers and families, the decision to hasten death cannot be said to be a voluntary or free choice.

Concluding Questions

Each of the above conditions for meaningful informed consent in providing an assisted-suicide alternative for terminally ill patients is, in principle, realizable. [14] A minimum assessment of patients requesting assistance with suicide needs to include an independent psychodynamically informed assessment of capacity untainted by potential economic conflicts of interest and value bias for or against hastening death. Such an assessment must also be extended to include economicinformed consent of the patient as to material provider and institutional conflicts of interests, and an ethically and legally informed assessment of the caring context to rule out undue managed-care-based provider or institutional influence. In the event that patients are adamantly against being initially seen by a psychologist or a psychiatrist, meeting with a psychodynamically informed attorney or bioethicist may also be helpful. Trained to be impartial and neutral With respect to their own values and with a psychodynamically informed mental health consultant for support, an attorney or bioethicist may also be able to make contact with a frightened, shame sensitive, and defensively depressed or illness-traumatized patient.

Finally, in the name of autonomy, self-determination, and respect for persons, the language associated with hastening death should be free from all spin. [15] To identify PAS as "Death with Dignity" can be interpreted to imply that other kinds of death (slow, painful, and uncertain) are undignified. Surely, what makes death dignified is not the presence or absence of control over death, but the attitudes of the patients themselves in both situations. Some patients, regardless of controlling the time and manner of their death, will be bitter, angry, and resentful that they are dying, whereas other patients, with little external control over time and manner, may experience internal control and end with grace. In view of the potential for economic gain from PAS, it is important that any language associated with it be impartial and neutral, thus helping the patient to make a truly autonomous decision. Such decisions begin not with the conclusion, but with questions such as "What does it mean to make an autonomous decision here and how?"


  1. Coombs Lee, B., & Werth, J. L., Jr. (2000). Observations on the first year of Oregon's Death With Dignity Act. Psychology, Public Policy, & Law, 6, 267-289.
  2. Maltsberger, J. T. (1986). Suicide risk: The formulation of clinical judgment. New York: New York University Press.
  3. Batavia, A. (2000). So far, so good: Observations on the first year of Oregon's Death With Dignity Act. Psychology, Public Policy, & Law, 6, 291-304.
  4. Mill, J. S. (1978). On liberty. New York: Elizabeth Rapaport Hackett Publishing.
  5. Id., page 56.
  6. Frankl, V. E. (1984). Man's search for meaning: An introduction to logotherapy. New York: Simon Schuster.
  7. Silvers, A., Wasserman, D., & Mahowald, M. B. (1998). Disability, difference, discrimination. New York: Rowman & Litflefield.
  8. Illingworth, P., & Parmet, W. (2000). Positively disabled: The relationship between the definition of disability and rights under the ADA. In A. Silvers & L. Francis (Eds.), Americans with disabilities: Exploring implications of the law for individuals and institutions. New York: Routledge.
  9. Dubovsky, S. L. (1997). Mind body. New York: W. W. Norton.
  10. Green, B. L., Epstein, S. A., Krupnick, J. L., & Rowland, J. H. (1997). Trauma and medical illness: Assessing trauma-related disorders in medical settings. In J. P. Wilson & T. M. Keane (Eds.), Assessing psychological trauma and PTSD. New York: Guilford Press.
  11. Bursztajn, H. J., Feinbloom, R. L, Hamm, R. M., & Brodsky, A. (1990). Medical choices, medical chances: How patients, families, and physicians can cope with uncertainty. New York: Routledge, Chapman & Hall.
  12. Id.
  13. Bursztajn, H. J., Gutheil, T. G., & Brodsky, A. (1998). Ethics and the triage model in managed care hospital psychiatry. Psychiatric Times, 15, 33-38.
  14. Grisso, T., & Appelbaum, P. S. (1998). Assessing competence to consent to treatment: A guide for physicians and other health professions. New York: Oxford University Press.
  15. Illingworth, P. (2000). Bluffing, puffing and spinning in managed care. Medicine and Philosophy, 62-76.