/ Death, Medicine, and the Moral Significance of Family Decision Making

Abstract

Policy, practice, and theory concerning the choices medicine presents to us at the end of life suffer from a romantic picture of death's special character and from a cynical view of families. A dash of realism about our deaths and our intimates is presented that challenges the legal, professional, and ethical orthodoxies about death and families.

Key Words: death, families, ethics, proxy decision making

    1. James Lindemann Nelson, Ph.D., is Associate for Ethical Studies, The Hastings Center, Briarcliff Manor, New York, 10510.return to text


     
    It is one of the best known pieces in literature, a staple of undergraduate curricula: The Death of Ivan Ilych. (Tolstoy, 1886/1960). Having access to the title, readers know what's going to happen right from the start: As though to eliminate any possible doubt, we watch the unfolding of Ivan's life, character, and relationships in flashback from his obsequies. Ivan himself, of course, is not so advantageously positioned. A good part of the story's drama consists precisely of his coming to understand that his illness is fatal. This task turns out to be complex and difficult, marked by ambivalence, insight, and denial.

    Ivan's story offers us a powerful and particular image of what is involved in coming to grips with dying. It stresses the importance of the jobs we have to do as our lives come to a close, and the value of the insights we can then gain. It also offers an equally forceful and vivid image of the place of the family at the end of life, one which highlights the falsity that permeates relationships, and the unreliability of those who are closest to us.

    Tolstoy wrote "Ivan Ilych" in 1886. We die differently now, many of us in hospitals, many in the aftermath of some deliberation and choice about using, withholding, or withdrawing therapies. Should a very low birthweight, brain-damaged baby be removed from her ventilator, a step which will end her suffering, but also any chance she has at life? Should an elderly man with "multiple-organ failure" undergo the violence of cardiopulmonary resuscitation if his heart stops, trading a peaceful death for a tiny chance at staying alive long enough to leave the hospital? Contemporary medicine has introduced new complexities into dying, complexities which often force patients and their families into making choices of a sort Ivan did not face. Yet current clinical practice and legal and ethical policy concerning those decisions reflects a very Tolstoyan construction of what's at stake and what's in danger.

    The response of Ivan's family to his dying was not notable for its moral insight. This fact is most marked by the translucent curtain of deceit with which his family veils Ivan's descent to death. Ivan is dying, but his dying is a forbidden subject; Ivan in particular must not acknowledge, or even allude to it. The terrible consequence is that he must suffer his dying without familial recognition.

    What tormented Ivan Ilych most was the deception, the lie, which for some reason they all accepted, that he was not dying but was simply ill, and that he only need keep quiet and undergo a treatment and then something very good would result ... this deception tortured him—their not wishing to admit what they all knew and what he knew, but wanting to lie to him concerning his terrible condition, and wishing and forcing him to participate in that lie. (p. 137)

    In this essay, I pose a counter-image to Tolstoy's, in two parts. My leading idea will be that our most intimate connections—which is what I will take "family" to mean here—will often have very important constructive roles to play in the tasks we face as our lives come to a close. But I will also underscore the fact that families are often deeply involved in those tasks and significantly affected by how they are discharged. Accordingly I will argue that families ought to have some say in how pertinent choices are made. Both these considerations should enrich and help direct our policy concerning end-of-life decision making.

    The Standard Approach: Romanticizing Death, Demonizing Families

    We enjoy a considerable measure of social consensus that treatment too burdensome for the benefits it promises may be withheld or withdrawn—even if rejection of treatment is tantamount to acceptance of death. This consensus was perhaps most clearly flagged by the Supreme Court's decision in Cruzan vs. Missouri Department of Health (1990), which upheld a patient's right to decide against life-sustaining therapy. There has also been wide agreement that such decisions are solely authorized by the principle of patient autonomy; that is, the moral claim that people enjoy a certain kind of sovereignty over what interventions in their bodies are consistent with their values, and which are not (Faden & Beauchamp, 1986; Buchanan & Brock, 1989).

    This consensus has a certain instability packed in it. Economic pressures and a reassertion of the autonomy of health care professionals have led some to think that life-prolonging health care can in principle be withheld despite patient/family desires, if it is expensive enough (Fleck, 1992), or withdrawn over patient/family objection, if the odds of it working are low enough (Tomlinson & Brody, 1990). But the major practical problem with the patient sovereignty view has been that when people are sick enough to require decision making of this kind, they are often too sick to make any decisions at all. For the past few decades, states have been experimenting with different means of extending a person's decision-making authority regarding health care, a movement culminating in the federal Patient Selfdetermination Act of 1990, which mandates that all patients be informed of the procedures approved by their state for directing their health care even if they should become incapacitated.

    Practically speaking, what this boils down to is allowing other people to convey a patient's treatment preferences, if she cannot exercise this authority in her own voice at the time a decision is required. Others might assist in the interpretation of "living wills," or more generally, written treatment directives, which are often both vague and ambiguous. They may simply make a decision as the patient's proxy, trying to judge as she would have judged. But in either case, the interpreter or proxy decision maker enjoys her position by virtue of her relationship to the will of the patient: either because she had been explicitly delegated to fill these roles, or in the absence of an explicit declaration made by the patient, because she is assumed to be able to transmit or reproduce the patient's preferences better than anyone else.

    The natural assumption is that close relatives will typically be in the best position to decide. But there is an equally natural objection: Family members hardly count as disinterested parties. Because of their very closeness, relatives often have a sizable stake in how treatment decisions go, and if their interests influence the decision making, the orthodoxy regards the process as morally contaminated. This "standard approach" to end-of-life decision making shares the suspicion about intimates found in Tolstoy's depiction of Ivan's decidedly nasty family, in which those who have some kind of relationship to the dying man—his wife and adult daughter—don't love him, and those who do—which is to say, his young son—seem to be permitted no relationship to him. Our thinking about end-of-life decision making, particularly concerning patients who cannot make decisions on their own behalf, seems to be haunted by specters closely resembling Ivan's wife and daughter, who saw him largely as a means to fulfilling their own desires. Therefore, judgments about starting or stopping life-sustaining therapy are carefully guarded to prevent such manipulation of vulnerable people.

    Family members, then, have no standing simply as family members, but only as conduits to the preferences that the patient actually had, or would have had. If their interests influence whether medical treatment of various kinds continues or not, then the patient is at great risk of abuse: either suffering the continual burdens of invasive care for an inadequate goal, or forgoing desired care and with it the chance to extend life.

    In fact, the picture for families is even darker. Not only are their motives suspect; it turns out that even their readings of the patient's desires are questionable. Recent studies of proxy decision making have indicated that families are, as it turns out, not very good at guessing the preferences of their relatives when it comes to the end of life (Seckler, Meier, Mulvihill, & Cammer Paris, 1991; Emanuel & Emanuel, 1992). In the standard view, then, their main claim to decision-making authority is undermined, while the main caution against them seems as strong as ever.

    The picture for incompetent patients also appears grimmer than has yet been suggested. What they have at stake is not simply the possibility of undergoing extended discomfort or premature death because their families are either mistaken about their preferences or malignly indifferent to them. Equally significant is the loss of the ability to invest their deaths with the kind of meaning that best comports with their sense of their life overall. Ivan Ilych provides us with a hint of this theme. Recall his painful examination of his life, his insight into how misdirected and trivial he had allowed his life to become, and how his final task is to accept himself, and his suffering, and hence to achieve salvation.

    The idea that there is often a "terminal perspective" on life, from which we can get an especially accurate view of our lives, and the idea that how we end our lives is crucial to the success or failure of those lives overall, strike me as at least loosely linked. Together they make up what might be called a romantic view of death, clearly present in Tolstoy, and not at all foreign to contemporary sensibilities. Consider this passage from Life's Dominion, the most recent book of the influential philosopher and legal scholar, Ronald Dworkin:

    There is no doubt that most people treat the manner of their deaths as of special, symbolic importance: they want their deaths, if possible, to express and in that way vividly to confirm the values they believe most important to their lives. (Dworkin, 1993, p. 211)

    Whether or not Dworkin is right about "most people" on this point, he is, I think, surely right about what most ethical and legal theorists think when they take up the issue of how choices should be made at the end of life. We find here another significant reason why our evolving policy on this matter has, since the 1970's, been directed toward empowering the patient. It isn't simply to defend them from assaults on what Dworkin (1993) would call their "experiential" interests, or how things feel to them: It is also to protect their ability to live and die in accordance with their "critical" interests; that is, with their reflective sense of what is truly significant and characteristic about their lives. How we die is of particular significance to whether or not our critical interest in having lived a good life is achieved, and it is crucial to our achieving such a life that our deaths be as much as possible orchestrated according to our own ideas.

    A Revised Account: Dying in Intimacy

    The contemporary context of decision making in the face of death, then, is in very important respects much the same as it was in the late 19th century. We are cynical about families, romantic about death. What's wrong with this standard, "Tolstoyan" conception of the significance of death and the suspicious character of families? In my view, pretty much everything.

    This is not to say that there are no abusive and otherwise untrustworthy families out there. Nor is it to deny that for some people, the process of dying is transformative, offering new and deep insights. Finally, I am not implacably hostile to the idea that the way we die can be crucial to the success or otherwise of our lives overall. Rather, my attitude toward all these claims is that they are all overstated; they ought not to be taken as the predominating feature of either families or death. Lots of people have families who are not decidedly nasty; lots of people die without gaining deep insights into the nature of things; lots of people can have bad deaths who had quite acceptable lives over all; lots of people have good deaths which are not good because they, personally or through carefully directed proxies, have orchestrated every step. And the worst of the overstatement is what might be regarded as its cumulative implication: We face death alone, most often as "vulnerable adults," whose chief need is protection from rapacious relatives.

    General practice, as opposed to policy and theory, indicates that my misgivings about this tableau are not idiosyncratic. Relatively few people avail themselves of formal advance directives, despite the publicity given to the importance of advance health-care planning; the few who do draw up such directives tend to be disproportionately white, well off, and well-educated. While there are many possible explanations here, one plausible suggestion is that different subcultures within our nation have different views about how important it is to take a direct hand in end-of-life decision making.

    Part of the problem may be that medical practice and legal policy regarding death correctly assume that most people want to die well, but that both practice and policy are confused about what dying well means to many of us. Doing something "well" does not necessarily mean doing it according to our own self-regarding desires; it may mean acting in accord with what strikes us as right, seemly, meet—where these notions guide us in ways that we believe to be good in themselves, and not simply because we happen to accept them. More particularly, many of us may believe that our deaths should cohere with a life lived in important connection with other people. The course of our dying should express concern about their burdens, not because doing so is the crucial task of our lives, nor because death has vouchsafed to us some special moral insight at the end, but because such concern is consistent with long held views about how to live well, views that need not be abandoned when the job at hand is how to die well.

    Not simply speculation, data, drawn largely from the work of High (1988, 1993, 1994) show that many people feel no need to file a formal document because they think of their families as their advance directives. The Harvard-based medical ethicists Linda and Ezekiel Emanuel (1992) have wondered whether High's results don't simply reflect most people's uncritical acceptance of the view that families know best what we ourselves would want, and that this enthusiasm for relatives would not survive the growing evidence to the contrary. But their critique makes two crucial assumptions. First, it assumes that the kind of medical choices which are open to us as we die are typically such that we have considered preferences about them, preferences expressing something that matters to us deeply. It also assumes, perhaps even more significantly, that our choices rule the day, however they might affect the interests of those with whom we have been intimate.

    But both these assumptions seem unwarranted. The Michigan law professor Patricia White, drawing on her experience in the presumably less emotionally charged area of estate planning, has pointed out that "people find it difficult to predict accurately how they would react to some hypothetical future crisis" (1992, p. 850). The idea, then, that the job of a proxy decision maker is to somehow elicit just what the patient would have wanted if she could speak for herself in the present situation assumes that there is some one thing she would have wanted, and this assumption may well be false.

    One of course could simply make determinations about one's future care, rather than predictions. That is, the decision maker would be exerting her autonomy now, reflecting her current preferences, rather than making a guess about what she would want in a future in which she is incapacitated, if, contrary to fact, she could make a considered decision at that time. But if we are to understand advance decision making as a determination rather than as a prediction, then it isn't clear that decision making at the end of life retains the kind of special moral significance the romantic perspective gave it. Dying "romantically"—i.e., in a way that reflects something crucially important about your life—might well require, not a blunt determination now of how a future event be handled, but fine-grained sensitivity to the details of that future time. How much pain or discomfort is at issue? What are the chances that a medical intervention will achieve its end, at what cost to the patient or to others the patient cares about? It is not implausible that making decisions of this kind could, in principle, allow the decision maker an opportunity to express and even develop her moral character. But, if so, what would allow her this opportunity is the ability to fit her decision precisely to the circumstances.

    The upshot is that it is far from clear that all, or even many, of the preferences healthy, self-aware people have about hypothetical future crises really count as considered or authoritative in any event. And we have yet to consider the point that, even if we assume incompetent patients typically have well-considered and well-ordered preferences that others might put into practice, the interests of their families remain morally relevant to decision making, even if those interests run counter to patient preferences.

    As John Hardwig (1990) has powerfully argued, there is no good reason to think that the ill are totally excused from their moral obligations to their intimates simply because of their illness. Nor is it appropriate to think that family members are required to bear any imaginable burden to further any interest of a relative, if that interest happens to be medical. Not a plea to endorse selfishness, the standard approach to decision making at the end of life proceeds as though selfishness were the appropriate standard: The patient's needs must be served, and the only way to assure they are met is to forbid family members to think of anything other than what the patient herself would. But families quite often have a different way of organizing the distribution of caring work that goes on within them. Sometimes that organization may be open to moral criticism—as when women are assigned an unequal share of caring labor simply because they are women—but the very fact that they distribute the family's resources in a way that is sensitive to many needs ought not to be regarded as beyond the moral pale simply on its face. Maintaining that proxy decision making by family members is to be censured is particularly ironic in the present context of health care delivery, in which the medical interests of patients are sometimes subordinated to the needs of the Health Maintenance Organization in which they are enrolled, or the resources their state is willing to make available for Medicaid.

    It is on the basis of considerations of this sort that I think that Ivan Ilych's sort of death ought to be seen as unusual—fit to be the subject of an immortal short story—rather than as a good guide to what challenges and choices will standardly face people as they die. We needn't construct a policy that assumes families are to be carefully controlled, suspected of guilt until proven innocent. We needn't think that putting our own stamp on the precise character of our death is a crucial determinant of the quality of our lives. Therefore, we needn't be so enamored of systems that rely primarily on explicit advance directives, seeing their authority as stemming solely from the patient and, in effect, disadvantaging the many patients and families without advance directives to whom death will come. It seems to me both more realistic, as well as quite defensible morally, to reverse the burden of proof here: We ought to recognize that families have a certain kind of moral authority to serve as proxies, unless perhaps the patient has made an explicit declaration to the contrary, or unless that authority is misused to a point that constitutes abuse.

    But this strategy is only part of what should be an overall rethinking of the contexts in which we die, and the assumptions that are prevalent in those contexts—assumptions which tend to undermine the kind of closeness that very ill patients can have with their families. Health care institutions should be set up to be as transparent as possible to these connections, not now the case. Hospitals, for example, remain places in which certain value commitments are evident and powerful: They are hierarchial, unfamiliar places, which separate you from daily routines and common sources of identity affirmation, running all the way from your own clothes to your most intimate connections. Hospitals have their own, clear agenda, to which patients are strongly invited to subscribe. The notion that patients need to be empowered in such settings is exactly right; the mistake is in thinking this is likely to happen if patients are allowed to be alienated from their own sources of personal affirmation and authority in the name of giving such authority formal protection.

    Conclusion: An Objection and a Reply

    It might be alleged that, the institutional structure of health-care systems apart, the system of deciding currently in place for incapacited people is actually very well suited to accommodate just the values sketched out here. Many people have families in which there are people whom they trust. Many people don't think it essential that their death reflect precisely what their own decisions would have been, had they been able to make them directly. Such people can easily execute advance directives which say, in effect, "My spouse gets to decide any feature of my medical care, if I am not able to do so." For those people who either do not trust their families, or do not wish to burden them with the task of making end-of-life decisions, appointing nonfamily proxies will be possible. For people who think that it is crucial that the circumstances of their deaths as closely as possible fit some overriding conception of the integrity of their lives, more specific treatment directives are possible. What really gets left out of the standard view?

    This very reasonable question has a pragmatic answer, to which I have already alluded, and a rather deeper answer. The pragmatic response is simply that the majority of people will, for the foreseeable future, die without a formal advance directive. At the very least, this fact suggests that we pay more attention to how to make health care decisions for this group of people, and the most reasonable response would seem to be a system of proxies arranged in descending order of priority: spouse, adult children, parents, siblings, and so on. This system would certainly not be without problems—for instance, understanding what "spouse" means in a society where people often live together without formal marriage—but it would at least have the right scope and the right slant. Individuals who felt uncomfortable with the ordering or wanted to leave specific instructions to their proxies would be within their rights to execute specific directives to change it.

    The deeper reason is that the standard approach is not neutral among different views of what a person owes to her family, or more broadly, of the nature of intimate connections. It contains a certain expressive force suggesting that our intimate ties are insignificant unless formalized by an explicit exercise of our own sovereign authority. This view is neither self-evidently true, nor altogether innocuous with regard to its impact on how we think about family ties generally in this society. Rereading Ivan Ilych reminds us that skepticism about the family is not a new phenomenon, but should not distract us from the distinct possibility that new forms of defensiveness about intimate connections can make things worse, as well as better.

    References

    Buchanan, A. & Brock, D. (1989). Deciding for others. Cambridge: Cambridge University Press.

    Cruzan v. Director, Missouri Department of Health,(1990) 110 S. Ct. 2841..

    Dworkin,R. (1993). Life's dominion. New York: Knopf.

    Emanuel, E. & Emanuel, L. (1992). Proxy decision making: An ethical and empirical analysis. JAMA, 267, 2067-2171.

    Faden, R. & Beauchamp, T. (1986). A history and theory of informed consent. New York: Oxford University Press.

    Fleck, L. (1992). Just health care rationing: A democratic decision making approach. University of Pennsylvania Law Review, 140, 1597-1636.

    Hardwig, J. (1990). What about the family? Hastings Center Report, 20, 5-10.

    High, D. (1988). All in the family: Extended autonomy and expectations in surrogate health care decision making. Gerontologist, 28, 46-51.

    High, D. (1993). Why are elderly people not using advance directives? Journal of Aging and Health, 5, 497-515.

    High, D. (1994). Families' roles in advance directives. Special supplement, Hastings Center Report, 24, S16-S18.

    Seckler, A., Meier, D., Mulvihill, M.& Cammer Paris, B. (1991). Substituted judgment: How accurate are proxy predictions? Annals of Internal Medicine, 151, 289-294.

    Tolstoy, L. (1960). The death of Ivan Ilych and other stories (Alymer Maude, Trans.). New York: New American Library (Original work published 1886).

    Tomlinson, T.& Brody, H. (1990). Futility and the ethics of resuscitation. JAMA, 264, 1276-1280.

    White, P. (1992). Appointing a proxy under the best of circumstances. Utah Law Review, 1992, 849-860.