What About the Family? - the Role of Family Interests in Medical Decision Making

by John Hardwig

IN: Hastings Center Report (1990) 20 [March/April], 5-10.

We are beginning to recognize that the prevalent ethic of patient autonomy simply will not do. Since demands for health care are virtually unlimited, giving autonomous patients the health care they want will bankrupt our health care system. We can no longer simply buy our way out of difficult questions of justice by expanding the health care pie until there is enough to satisfy the wants and needs of everyone. The requirements of justice and the needs of other patients must temper the claims of autonomous patients.

But if the legitimate claims of other patients and other (non-medical) interests of society are beginning to be recognized, there is another question that is still largely ignored. That is the question of the extent to which the patient's family can legitimately be asked or required to sacrifice their own interests so that the patient can have the care and treatment he wants.

This question is not only almost universally ignored, it is generally implicitly dismissed, silenced before it can even be raised. This tacit dismissal results from a fundamental assumption of medical ethics: medical treatment ought always to serve the interests of the patient. This, of course, implies that the interests of family members should be irrelevant to medical treatment decisions or at least ought never to take precedence over the interests of the patient. All questions about fairness to the interests of family members are thus precluded, regardless of the merit or importance of the interests that will have to be sacrificed if the patient is to receive optimal treatment.

My task here will be to call this pervasive assumption into question. My main contention will be that because there is a whole range of cases in which important interests of family members are dramatically affected by decisions about the patient's treatment, medical decisions often should be made with those interests in mind. Indeed, there are many cases in which family members have a greater interest than the patient in which treatment option is to be exercised. In such cases, the interests of the family members often ought to override the interests of the patient.

Nor can the problem of family interests be resolved by simply redefining the other members of the family as "patients," thereby redefining the issue as the problem of conflicting interests among patients. No, the other members of the family are not always ill, and even if ill, they still may not be patients. Nor will it do to define the whole family as one patient. Granted, the slogan, "the patient is the family," was coined partly to draw attention to precisely the issues I wish to raise. But the idea that the whole family is one patient is just too monolithic: the conflicts of interests, beliefs, and values among family members are often too real and run too deep. Thus, if I am correct, it is sometimes the moral thing for a physician to sacrifice the interests of her patient to those of non-patients -- specifically to those of the other members of the patient's family.

But what is the "family?" As I will use it here, it will mean roughly "those who are close to the patient." "Family" so defined will often include close friends and companions. It may also exclude some with blood or marriage ties to the patient. "Closeness" does not, however, always mean care and abiding affection. In fact, closeness need not even be a positive experience -- one can hate, resent, fear, or despise a mother or brother with an intensity not often directed toward strangers, acquaintances, or associates. But there are cases where even a hateful or resentful family member's interests ought to be considered.

This use of "family" gives rise to very sensitive ethical -- and also legal -- issues in the case of legal relatives with no emotional ties to the patient. I cannot pursue these issues here. I can only say that I do not mean to suggest that the interests of legal relatives who are not emotionally close to the patient are always to be ignored. They will sometimes have an important financial interest in the treatment even if they are not emotionally close to the patient. But blood and marriage ties can become so thin that they become merely legal relationships. (Consider, for example, "couples" who have long since parted but who have never gotten a divorce, or cases in which the next of kin cannot even be bothered with making proxy decisions.) Obviously, there are many important questions about just whose interests are to be considered in which treatment decisions and to what extent. If the thesis I present here is correct, all of these issues will require extensive discussion.

CONNECTED INTERESTS

There is no way to detach the lives of patients from the lives of those who are close to them. Indeed, the intertwining of lives is part of the very meaning of closeness. Consequently, there will be a broad spectrum of cases in which the treatment options will have dramatic and importantly different impacts on the lives of the patient's family.

I believe there are many, many such cases. To save the life of a newborn with serious defects is often to dramatically affect the rest of the parents' lives and, if they have other children, it may seriously compromise the quality of their lives, as well. . . . The husband of a woman with Alzheimer's disease may well have a life that is totally dominated for 10 years or more by caring for an increasingly foreign and estranged wife. . . . The choice between aggressive and palliative care or, for that matter, the difference between either kind of care and suicide in the case of a father with terminal cancer or AIDs may have a dramatic emotional and financial impact on his wife and children. . . . Less dramatically, the choice between two medications, one of which has the side effect of impotence, may radically alter the life a couple has together. . . . The drug of choice for controlling high blood pressure is too expensive (i.e., requires too many sacrifices) for many families with incomes just above the ceiling for Medicaid. . . . My own grandfather committed suicide after his heart attack as a final gift to his wife -- he had plenty of life insurance but not nearly enough health insurance, and he feared that she would be left homeless and destitute if he lingered on in an incapacitated state.

Because the lives of those who are close are not separable, to be close is to no longer have a life entirely your own to live entirely as you choose. To be part of a family is to be morally required to make decisions on the basis of thinking about what is best for all concerned, not simply what is best for yourself. In healthy families, characterized by genuine care, one wants to make decisions on this basis, and many people do so quite naturally and automatically. But even if one is not so inclined, it is irresponsible and wrong to exclude or to fail to consider the interests of those who are close. It is only when the lives of family members will not be importantly affected that one can rightly make exclusively or even predominantly self-regarding decisions.

Although "what is best for all concerned" sounds utilitarian, my position does not imply that the right course of action results simply from a calculation of what is best for all. No, the seriously ill may have a right to special consideration, and the family of an ill person may have a duty to make sacrifices in order to respond to an illness in the family. But it is one thing to claim that the ill deserve special consideration; it is quite another to maintain that they deserve exclusive or even overriding consideration. For that would mean that other family members have no effective rights vis-a-vis the ill. Surely we must admit that there are limits to the right to special treatment by virtue of illness. Otherwise, everyone would be morally required to sacrifice all other goods to better care for the ill. We must also recognize that patients, too, have moral obligations, obligations to try to protect the lives of their families from destruction resulting from their illnesses.

Thus, unless serious illness excuses one from all moral responsibility -- and I don't see how it could -- it is an oversimplification to say of a patient who is part of a family that "it's his life" or "after all, it's his medical treatment," as if his life and his treatment could be successfully isolated from the lives of the other members of his family. More accurate would be to say "it's their lives" or "after all, they're all going to have to live with his treatment." Then the really serious moral questions are not whether the interests of family members are relevant to decisions about a patient's medical treatment and whether their interests should be included in his deliberations or deliberations about him. Instead, they will often be questions of how far the family and friends can be asked to support and sustain the patient; what sacrifices they can and cannot be morally required to make for his health care; how far they can reasonably be asked to compromise the quality of their lives so that he will receive the care that would improve the quality of his life; how far he can reasonably expect them to put their lives "on hold" in order to preoccupy themselves with his illness to the extent necessary to care for him.

THE ANOMALY OF MEDICAL DECISIONMAKING

The way we analyze medical treatment decisions by or for patients is plainly anomalous to the way we think about other important decisions family members make. I am a husband, a father, and still a son, and no one would argue that I should or even responsibly could decide to take a sabbatical, another job, or even a weekend trip solely on the basis of what I want for myself, without considering the interests of my family. Why should decisions about my medical treatment be different? Obviously, I think they are not different. But why should we have even thought that medical treatment decisions are different?

--Is it because medical decisions, unlike other decisions, involve life and death matters? But most medical decisions are not matters of life and death, and we as a society risk or shorten other peoples' lives -- e.g., through our toxic waste disposal decisions -- quite apart from considerations of whether that is what they want for themselves.

--Have we been misled by a preoccupation with the biophysical model of disease? Perhaps it has tempted us to think of illness -- and hence also of treatment -- as something that takes place within the body of the patient. What happens in my body does not -- barring contagion -- affect my wife's body. But it usually does affect her.

--Have we been operating on a tacit desire to simplify the practice and the ethics of medicine by considering only the medical or health-related consequences of treatment decisions? Perhaps, but it is obvious that we need a broader vision of and sensitivity to all the consequences of action, at least by all those who are not simply technicians following orders from above. Generals need to consider more than military consequences, businessmen more than economic consequences, teachers more than educational consequences, lawyers more than legal consequences.

--Does the weakness and vulnerability that is part of serious illness imply that the ill need our protection and therefore that we should serve the interests of the ill? Those who are sick may need special protection, but this can only mean that we must take special care to see that the interests of the ill are duly considered. It does not follow that the interests of the ill are to be served exclusively or even that their interests must always predominate. Moreover, we must also remember that in terms of the dynamics of the family, the patient is not always the weakest member, the member most in need of protection.

--Does it make historical, if not logical, sense to view the wishes and interests of the patient as always overriding? Historically, illnesses were generally of much shorter duration; patients got better quickly or died quickly. Moreover, the costs of the medical care available were small enough so that rarely was one's future mortgaged to the costs of the care of family members. But, although this was once truer than it is now, there have always been significant exceptions to these generalizations.

Thus, none of these considerations adequately explains why the interests of the patient's family have been thought to be appropriately excluded from consideration. At the very least, those who believe that medical treatment decisions are morally anomalous to other important decisions owe us a better account of how and why this is so.

LIMITS OF PUBLIC POLICY

Now, it might be thought that this whole problem is a problem only because our society does not shelter families from the negative effects of medical decisions. If, for example, we adopted a sufficiently comprehensive and inclusive system of national health insurance and also a system of public insurance to guarantee the incomes of families, then my sons' chances at a college education and the quality of the rest of their lives might not have to be sacrificed if I were to receive optimal medical care.

However, it is worth pointing out that we are still moving primarily in the opposite direction, the Medicare Catastrophic Coverage Act notwithstanding. Instead of designing policies that would increasingly shelter family members from the adverse impact of serious and prolonged illnesses, we are still attempting to shift the burden of care to family members in our efforts to contain medical costs. Thus, a social system that would safeguard families from the impact of serious illness is nowhere in sight in this country. And we must not do medical ethics as if it were.

But it is perhaps even more important to recognize that the lives of family members could not be sheltered from all of the important ramifications of medical treatment decisions by any set of public policies. In any society in which people get close to each other and care deeply for each other, treatment decisions about one will often and irremediably affect more than one. Thus, if a newborn has been saved by aggressive treatment but is severely handicapped, the parents may simply not be emotionally capable of abandoning the child to institutional care. A man whose wife is suffering from multiple sclerosis may simply not be willing or able to go on with his own life until he sees her through to the end. A woman whose husband is being maintained in a vegetative state may not feel free to marry or even to see other men again, regardless of what some revised law might say about her marital status.

Nor could we desire a society in which friends and family would quickly lose their concern as soon as continuing to care began to diminish the quality of their own lives. For we would then have alliances for better but not for worse, in heath but not in sickness, until death appears on the horizon. And we would all be poorer for that. A man who can leave his wife the day after she learns that she has cancer, on the grounds that "I have my own life to live," is to be deplored. The emotional inability or principled refusal to separate ourselves and our lives from the lives of ill or dying members of our families is not an unfortunate fact about the structure of our emotions. It is a desirable feature, not to be changed even if it could be. Not to be changed even if the resulting intertwining of lives debars us from making exclusively self-regarding treatment decisions when we are ill.

Thus, our present individualistic medical ethics is isolating and destructive. For, by supporting those who make self-regarding decisions and by implicitly suggesting that they make "their own" treatment decisions on a self-regarding basis, such an ethics does what it can to encourage each of us to see our lives as simply our own. We may yet turn ourselves into beings who are ultimately alone.

FIDELITY OR FAIRNESS?

Fidelity to the interests of the patient has been a cornerstone of both traditional codes and contemporary theories of medical ethics. The two competing paradigms of medical ethics -- the "benevolence" model and the "patient autonomy" model -- are simply different ways of construing fidelity to the interests of the patient. Both must be rejected or radically modified. The admission that treatment decisions often affect more than just the patient thus forces major changes on both the theoretical and the practical level. Obviously, I can only begin to explore the needed changes here. But I should first say something about the theoretical changes that are required. I will then turn to some of the implications for treatment decisions.

Instead of starting with our usual assumption that physicians are to serve the interests of the patient, we must build our theories on a very different assumption: both medical and non-medical interests of both the patient and other members of the patient's family are to be considered. It is only in the special case of patients without family that we can simply follow the patient's wishes or pursue the patient's interests. In fact, I would argue that we must build our theory of medical ethics on the presumption of equality: the interests of patients and family members are to be weighed equally; medical and non-medical interests of the same magnitude deserve equal consideration. Like any other moral presumption, this one can, perhaps, be defeated in some cases. But we will always need a special argument to defeat the presumption of equality. The burden of proof will always be on those who would advocate special consideration for any family member's interests, including those of the ill.

Even where the presumption of equality is not defeated, life, health, and freedom from pain and handicapping conditions are extremely important goods for virtually everyone. They are thus very important considerations in all treatment decisions. In the majority of cases, the patient's interest in optimal health and longer life may well be strong enough to outweigh the conflicting interests of other members of the family. But even then, some departure from the treatment plan that would maximize the patient's interests may well be justified in order to best harmonize the interests of all concerned or to require significantly smaller sacrifices by other family members. The patient's interests will, no doubt, often outweigh the conflicting interests of others in treatment decisions. But that is no justification for failing to recognize that an attempt to balance or harmonize different, conflicting interests is often morally required. Nor for overlooking the morally crucial cases in which the interests of other members of the family ought to override the interests of the patient.

Reconceptualization of the ethical roles of both physician and patient will be required by this basic change, since our understanding of both has been built on the presumption of patient primacy, rather than fairness to all concerned. Recognition of the moral relevance of the interests of family members thus reveals a dilemma for our understanding of a physician: Should we retain a fiduciary ethic in which the physician is to serve the interests of her patient? Or should the physician attempt to weigh and balance all the interests of all concerned? I do not yet see which way out of this dilemma is preferable. All I can now do is to try to envision the options.

If we retain the traditional ethic of fidelity to the interests of the patient, the physician should excuse herself from making treatment decisions that will affect the lives of the family on grounds of a moral conflict of interest. For she is a one-sided advocate. A lawyer for one of the parties can not also serve as judge in the case. Thus, it would be unfair if a physician conceived as having a fiduciary relationship to her patient were to make treatment decisions that would adversely affect the lives of the patient's family. Indeed, a physician conceived as a patient advocate should not even advise patients or family members about which course of treatment should be chosen. For it is similarly unfair if one party in a conflict has an advocate to speak for his interests while the others have none. All a patient advocate can say is what course of treatment would be best for the patient. As advocate, she must remain silent about what's best for the rest of the family and also about what should be done in light of everyone's interests.

The other way out of the dilemma would, of course, be for physicians to renounce their fiduciary relationship with their patients. On this view, physicians would no longer be agents of their patients and would not strive to be advocates for their patients' interests. Instead, the physician would aspire to be an impartial advisor who would stand knowledgeably but sympathetically outside all of the many conflicting interests of those affected by the treatment options, and who would strive to discern the treatment that would best harmonize or balance the interests of all concerned.

Although this second option contradicts the Hippocratic Oath and most other codes of medical ethics, it is not, perhaps, as foreign as it may at first seem. Traditionally, many family physicians -- especially small-town physicians who knew patients and their families well -- made decisions in precisely this way. They attempted to attend to both medical and nonmedical interests of all concerned. Many contemporary physicians still make decisions in this way. But we do not yet have an ethical theory that explains and justifies what they are doing.

Nevertheless, we may well question the physician's ability to act as an impartial ethical observer. Increasingly, physicians do not know their patients, much less the patients' families. Moreover, we may doubt physicians' abilities to evenhandedly weigh medical and nonmedical interests. Physicians are trained to be especially attuned and responsive to medical interests. We may well want them to remain that way. Physicians also tend to be deeply involved with the interests of their patients, and it may be impossible or undesirable to break this tie in order to enable physicians to be more impartial advisors. Finally, when one person retains the services of a physician, it seems reasonable that she should be able to expect that physician to be her agent, pursuing her interests, not those of her family.

We must also rethink our conception of the patient. The moral relevance of the interests of the patient's family also poses a dilemma for the way we think of patients. On one hand, if we continue to stress patient autonomy, we must recognize that patient autonomy implies that patients have moral responsibilities. But if, on the other hand, we do not want to burden patients with weighty moral responsibilities, we must abandon the ethic of patient autonomy.

If I am morally empowered to make decisions about "my" medical treatment, I am also morally required to shoulder the responsibility of making very difficult moral decisions. In fact, I often have the obligation to make decisions that run contrary to my own interests and to what I want. That's simply ethics: the right course of action for me to take is not always the one that promotes my own interests.

Some patients, motivated by a deep and abiding concern for the well-being of their families, will undoubtedly consider the interests of other family members. For these patients, the interests of their family are part of their interests. But not all patients will feel this way. And the interests of family members are not relevant if and because the patient wants to consider them. They are not relevant because they are part of the patient's interests. They are relevant whether or not the patient is inclined to consider them. Indeed, the ethics of patient decisions is most poignantly highlighted precisely when the patient is inclined to decide without considering the impact of his decision on the lives of the rest of his family.

Confronting patients with tough ethical choices may be part and parcel of treating them with respect, as fully competent adults. We don't, after all, think it's right to stand silently by while other (healthy) adults ignore or shirk their moral responsibilities. But if we believe that most patients, gripped as they often are by the emotional crisis of serious illness, are not up to shouldering the responsibility of such decisions or should not be burdened with it, then I think we must simply give up on the ethic of patient autonomy. Patient autonomy would then be appropriate only when the various treatment options will affect only the patient's life.

Recognizing that moral responsibilities come with patient autonomy will require basic changes in the accepted meanings of both "autonomy" and "advocacy." Because medical ethics has ignored patient responsibilities, we have interpreted "autonomy" in a sense very different from Kant's original use of the term. In medical ethics, patient autonomy has come to mean simply the patient's freedom or right to choose the treatment he believes is best for himself. But as Kant knew well, there are many situations in which people can achieve autonomy and moral well-being only by sacrificing other important dimensions of their well-being, including health, happiness, even life itself. For autonomy is the responsible use of freedom and autonomy is therefore diminished whenever one ignores, evades, or slights one's responsibilities. Human dignity, Kant concluded, consists in our ability to refuse to compromise our autonomy in order to try to achieve the kinds of lives (or treatments) we want for ourselves.

Thus, the responsibilities of patients imply that there is often a conflict between patient autonomy and the patient's interests (even as those interests are defined by the patient). And we will have to rethink our understanding of patient advocacy in light of this conflict: Does the patient advocate try to promote the patient's interests (as defined by the patient)? Or does she promote the patient's autonomy even at the expense of those interests? Responsible patient advocates can hardly encourage patients to shirk their moral responsibilities. But can we really expect health-care providers to promote patient autonomy when that means encouraging their patients to sacrifice health, happiness, sometimes even life itself?

If we could give an affirmative answer to this last question, we would obviously thereby create a third option for reinterpreting the role of the physician: The physician could maintain her traditional role as patient advocate without being morally required to refrain from making treatment decisions whenever interests of the patient's family are also at stake if 1) patient advocacy were understood as promoting patient autonomy and 2) patient autonomy were understood as the responsible use of freedom, not simply the right to choose the treatment one wants.

Much more attention needs to be paid to all of these issues. However, it should be clear that absolutely central features of our theories of medical ethics -- our understanding of physician and patient, and thus of patient autonomy, patient dignity, and patient advocacy -- have been predicated on the assumption that the interests of family members should be irrelevant to medical treatment decisions or should always take a back seat to the interests of the patient. Basic conceptual shifts will be required if we acknowledge that this assumption is not warranted.

AUTONOMY AND ADVOCACY

Such basic conceptual shifts will necessarily have ramifications that will be felt throughout the field of medical ethics. To cite only the most obvious examples, the discussions of privacy and confidentiality, of withholding/withdrawing treatment, and of surrogate decision making will all have to be reconsidered in light of the interests of the family. Many treatment decisions will also be affected, becoming much more complicated than they already are. For a host of new and very different issues are raised by the inclusion of family interests. I can not even begin to discuss all of these issues. I can only offer a few remarks about treatment decisions, organized around the central issue of who should decide.

There are at least five answers to the question of who should make treatment decisions in cases where important interests of other family members are also at stake: 1) the patient, 2) the family, 3) the physician, 4) an ethics committee, and 5) the courts. The issue of the physician's role in treatment decisions has already been discussed. Resort to either the courts or to ethics committees for treatment decisions is too cumbersome and time-consuming for any but the most troubling cases. So I will focus here on the contrast between the patient and the family as appropriate decision makers. It is worth noting, though, that we need not arrive at one, uniform answer to cover all cases. On the contrary, each of the five options will undoubtedly have its place, depending on the details of the case at hand.

Should we still think of a patient as having the right to make decisions about "his" treatment? As we have seen, patient autonomy implies patient responsibilities. What, then, if the patient seems to be ignoring the impact of his treatment on his family? At the very least, responsible physicians must caution such patients against simply opting for the treatments they want. Instead, physicians must speak of responsibilities and obligations, not just wants and desires. They must raise considerations of the quality of many lives, not just that of the patient. They must explain the distinction between making a decision and making it in a self-regarding manner. Thus, it will often be appropriate to make plain to patients the consequences of treatment decisions for their families and to urge them to consider these consequences in reaching a decision. And sometimes, no doubt, it would be appropriate for family members to present their cases to the patient in the hope that his decisions would be shaped by their appeals.

And yet, we sometimes permit people to make bad or irresponsible decisions and excuse their decisions because of various pressures they were under when they made their choices. Serious illness is undoubtedly an excusing condition, and perhaps we should allow some patients to make self-regarding decisions, especially if they insist on doing so and the negative impact of their decisions on others is not too great.

Alternatively, if we doubt that most patients have the ability to make treatment decisions that are really fair to all concerned, or if we are not prepared to accept a policy that would assign them the responsibility of doing so, we may conclude that patients should not be empowered to make treatment decisions in which the lives of their family members will be dramatically affected. Indeed, even if the patient were completely fair in making the decision, the autonomy of other family members would have been systematically undercut by the fact that he alone made it.

Thus, we need to consider the autonomy of all members of the family, not just the patient's autonomy. Considerations of fairness and, paradoxically, also of autonomy therefore indicate that the family should make the treatment decision, with all competent family members whose lives will be affected participating. Such family conferences undoubtedly already often take place. But family conferences would, on this view, often be morally required. And these conferences would not be limited to cases involving incompetent patients; cases involving competent patients would also often require family conferences.

Obviously, it would be completely unworkable for a physician to convene a family conference every time a medical decision might have some ramifications on the lives of family members. But such discussion need not always take place in the presence of the physician. And we can recognize that formal family conferences become more important as the impact of treatment decisions on members of the patient's family grow larger. Family conferences may thus be morally required only when the lives of family members would be dramatically affected by treatment decisions.

Moreover, family discussion is often morally desirable even if not morally required. Desirable, sometimes, even for relatively minor treatment decisions: After the family has moved to a new town, should the parents commit themselves to two-hour drives so that their teenage son can continue to be treated for his acne by the dermatologist he knows and whose results he trusts? Or should he seek treatment from a new dermatologist?

Sometimes family conferences about treatment decisions would be characterized throughout by deep affection, mutual understanding, and abiding concern for the interests of others. But not always. Other conferences might begin in an atmosphere charged with antagonism, suspicion, and hostility but move toward greater understanding, reconciliation, and harmony within the family. Such conferences would be significant goods in themselves, as well as means to ethically better treatment decisions. They would leave all family members better able to go on with their lives.

Still, family conferences cannot be expected to always begin with or move toward affection, mutual understanding, and a concern for all. So if we opt for joint treatment decisions when the lives of several are affected, we need to face the fact that family conferences will sometimes be bitter confrontations in which past hostilities, anger, and resentments will continue to boil to the surface. Sometimes, too, the conflicts of interest between patient and family and also between one family member and another will be deep and irresolvable, forcing families to invoke the harsh perspective of justice, divisive and antagonistic though that perspective may be. Those who favor family decisions when the whole family is affected will have to face the question of whether we really want to put the patient, already frightened and weakened by his illness, through the self-conflict and bitter confrontations that family conferences would sometimes precipitate.

We must also recognize that family members may be unable to unwilling to press or even state their own interests in the face of a member of the family who is ill. Sometimes such refusal is admirable, even heroic: it is sometimes evidence of willingness to go "above and beyond the call of duty," even at great personal cost. But not always. Refusal to press one's own interests can also be a sign of inappropriate guilt, a crushing sense of responsibility for the well-being of others, acceptance of an inferior or dominated role within the family, or lack of a sense of self-worth. And all of these may well be mobilized by an illness in the family. We must also not minimize the power of the medical setting to subordinate nonmedical to medical interests and to emphasize the well-being of the patient at the expense of the well-being of others. Thus, it will often be not just the patient, but also the other family members who will need an advocate if a family conference is to reach the decision that best balances the autonomy and interests of all concerned.

Granted, the existing theory of patient autonomy and also of proxy decisionmaking has been designed partly as a buttress against pressures from family members for both overtreatment and undertreatment of patients. The considerations I have been advancing will enable us to see that sometimes what we've seen as undertreatment and also overtreatment is not really so. Both concepts will have to be redefined. Still, I do not wish to deny or minimize the problem of family members who demand inappropriate treatment. Treatment decisions are extremely difficult when important interests of the other members of the family are also at stake. The temptation of family members simply to demand the treatment that best suits their interests is often very real.

But I do not believe that the best safeguard against pressures from family members for inappropriate treatment is to issue morally inappropriate instructions to family members in the hope that these instructions will prevent abuses. Asking a family member to pretend that her interests are somehow irrelevant often backfires. Rather, I think the best safeguard against pressures for inappropriate treatment would be to candidly admit the moral relevance of the interests of other members of the family and then to support the family through the excruciating process of trying to reach a decision that is fair to all concerned.

WHO SHOULD DECIDE?

Acknowledging the interests of family members in medical treatment decisions thus forces basic changes both on the level of ethical theory and in the moral practice of medicine. Many issues are involved, and none will be easy to resolve. Indeed, the sheer complexity of the issues raised might seem to some a sufficient reason to ignore family interests in favor of the much simpler ethic of absolute fidelity to the patient. But that would be the ostrich approach to the complexities of medical ethics. And I would insist that we must not abandon our patients' families to lives truncated by an over-simplified ethic. That would be an unconscionable toll to exact in order to make our tasks as ethicists and moral physicians easier.

Reconstruction of medical ethics in light of family interests would not be all pain and no gain for ethicists and physicians, however. Acknowledging family members' interests would bring benefits as well as burdens to medical practitioners. For the practice of medicine has rarely been as individualistic as codes and theories of medical ethics have advocated. Indeed, much of what now goes on in intensive care nurseries, pediatricians' offices, intensive care units, and long-term care institutions makes ethical sense only on the assumption that the interests of other members of the family are also to be considered.

Contemporary ethical theory and traditional codes of medical ethics can neither help nor support physicians, patients, and family members struggling to balance the patient's interests and the interests of others in the family. Indeed, our present ethical theory can only condemn as unethical any attempt to weigh in the interests of other family members. If we would acknowledge the moral relevance of the interests of the family, we would be able to understand why many treatment decisions being made today make sense and are not unethical. And perhaps we could then develop an ethical theory that would guide and support physicians, patients, and families in the throes of agonizing moral decisions.

FOOTNOTES

* I wish to thank Mary Read English, Michael Lavin, Gary Smith, Joanne Lynn, Larry Churchill, and the editors of Hastings Center Report for valuable suggestions about this paper.