Justice, Reason, and Luck in Rationing Lifesaving Medical Resources

Abstract

When lifesaving medical resources such as organs or medical technologies are scarce, how should they be distributed? Among the patients needing such resources, who should be selected, and how? James F. Childress addresses these questions in his landmark article, “Who Shall Live When Not All Can Live?,” providing a careful analysis of the ethical dimensions of rationing lifesaving medical resources and defending a method of randomized patient selection. Childress's argument is not only a model of practical reasoning in bioethics, it also influenced the design of public policy regarding rationing, now institutionalized by the United Network of Organ Sharing (UNOS), a nonprofit organization under contract with the federal government to oversee the allocation of scarce organs and tissues.This is not to say that Childress's argument is uncontroversial or that it is not subject to strong challenges and the need for revision or amendment. In this article I want to identify one challenge to Childress's argument and show how it opens up an alternative pathway for bioethical public policy that Childress would do well to note and that he might want to avoid. To that end, I will unpack that challenge, indicate the stakes involved, identify its practical implications, and defend a way to deflect it. Before doing that, however, I need to describe Childress's constructive position.First published in 1970, Childress's argument conceives of rationing lifesaving medical resources as occurring in two stages. At the first stage, prospective patients are to be screened according to the criterion of medical utility to identify a class of people deemed “medically acceptable” for treatment. The aim would not be to identify precise prospects of success at the individual level but to determine a class of potential recipients who meet medical criteria for a reasonable prospect of successful response to treatment (Childress 1997d, 172–73). Medical criteria, applied in this way, would exclude some persons in order to create a pool of prospective patients, but those criteria would not allow physicians to then select specific patients who seem to have relatively better prospects than others. At the second and final stage, patients who are to receive scarce resources would instead be selected randomly—without regard to race, gender, economic background, nationality, creed, or social value. Childress specifies the process of random selection in terms of queuing (first come, first served) or in terms of using a lottery (175). The central rationale, about which I will say more below, turns on the principle of human dignity and equality of opportunity, as opposed to social value or utilitarian considerations. Randomness, Childress argues, is the best means for rationing resources that honors the equal dignity of all persons.Childress refined his ideas in subsequent writings and in the postscript written for the 1997 republication of “Who Shall Live When Not All Can Live?” by distinguishing medical utility from social utility (1997d, 184; see 1997a, 220). One reason for the refinement was the need to think not only about how to ration dialysis machines—the question behind his first foray into rationing—but also about the ethics of allocating scarce organs. Henceforth he allows considerations of medical utility to inform decisions after the first stage of screening, noting that considerations of medical benefit are independent of a patient's social worth and are in accord with his main principle for rationing, namely, respect for equal dignity. On this revised account, utilitarian considerations may intrude on, and modify, the randomness of the second phase of the selection process for reasons that prescind from considerations of the recipient's social worth. This last stage is constrained by the need to responsibly steward scarce resources, such as organs, that cannot be reused. An eligible patient waiting for an organ might undergo health problems that render him or her medically unacceptable after having been initially included in the pool but before an organ becomes available to him or her. Considerations of medical utility require physicians to consider the urgency of patient need and the procedure's probability of success. A physician may thus reallocate a randomly assigned organ away from a recipient with no reasonable prospect for success to another patient who can palpably benefit from it. Childress refined his views after 1970 in light of these ideas, adding that it was impossible to provide an a priori lexical ranking of medical need and hope for success (Childress 1997a). Childress's final position still has two stages, but each stage is constrained by considerations of medical utility: (1) establishing a pool of potential recipients according to the criterion of medical acceptability; (2) randomly selecting recipients from the pool of medically acceptable patients and modifying that process when necessary according to updated judgments regarding a rationed resource's prospect for success.1Of these stages, the second is arguably the more controversial, and it is about this stage that I will discuss the ideas of justice, reason, and luck as they factor into Childress's proposal for using random selection for rationing lifesaving medical resources. I will simplify matters by referring to random selection as a lottery. Random selection can be institutionalized in different ways: choosing straws, queuing, or entering into a drawing, for example. What these examples have in common, and what is captured by the term lottery, is the fact that the outcome is a matter of chance rather than effort or, for that matter, anything specific about the applicant's background or profile (Childress 1997d, 173–75). My use of the term lottery is meant to accentuate the role of chance in Childress's proposal and aims to highlight the fact that, on his line of argument, those who receive and those who are denied scarce lifesaving medical resources are the subjects of good and bad luck, respectively.Using a lottery ignores features about the eligible candidate except for the barest essential fact, namely, membership in the human race. Random selection thereby implies some metaphysical claims insofar as it presupposes the distinction between essential and accidental dimensions of human personhood. That distinction enables us to make sense of the fact that the lottery ignores certain features of medically acceptable applicants—features such as inherited wealth, proven talent, social position, or potential social value—as irrelevant when determining how medical resources should be rationed.I want to explore metaphysical presuppositions associated with using a lottery to ration lifesaving medical resources and what those presuppositions imply about justice, luck, and personhood as those ideas bear upon matters of equitable rationing and patient selection. At stake in ethical debates about “who shall live when not all can live” is the relationship between justice and chance in ethics and metaphysics. Childress's argument advances the counterintuitive idea that the use of chance can be a fair way to allocate scarce lifesaving medical resources. On his line of argument, those who are saved when not all can be saved are winners in a lottery, the beneficiaries of good luck, and those who are not saved are losers in a lottery, victims of bad luck. Although I believe that Childress is correct in the substance of his proposal, I think that the reasons he offers in its defense need to be strengthened.Allowing for a marriage of justice and luck in rationing poses difficult questions for proponents of a lottery who wish to discount putative accidental features of personhood when determining a fair method for allocating scarce lifesaving medical resources. Accepting the results of luck while bracketing other effects of fortune in the same overall account of fairness seems incongruous. Consider a line of thought that would seek to avoid such incongruity by refusing to discount or bracket putative accidental features of personhood. Theorists who believe that the distinction between essential and accidental properties is otiose when thinking about personhood, or that it ought to have no bearing on determining the requirements of justice, might consider specific features of an individual's profile—what might otherwise be classified as contingent aspects of a person—when determining how to ration lifesaving medical resources. That is to say, those who reject the distinction between essential and contingent features of personhood might have reasons to consider certain aspects of luck—family background, potential value, or political or financial power, for example—when establishing what is appropriate in matters of resource rationing.We can sharpen some of these initial concerns about justice and luck from the vantage point of liberal political and moral theory. One version of that theory would drive a wedge, as John Rawls does, between justice-as-fairness, on the one hand, and accepting the results of what Rawls calls the natural and social lottery, on the other, when theorizing about justice. For Rawls, the results of the natural or social lottery are “arbitrary from a moral point of view” (1971, 72). They distribute advantages or disadvantages to persons for reasons entirely beyond their control. Allocating resources without taking measures to correct for the lottery's effects will only replicate distributions and advantages that are already in place. Those allocations would be fair, then, only if the original distribution is fair. From a Rawlsian perspective, finding a marriage between fairness and luck seems contrary to justice or, more precisely, seems to migrate in the direction of libertarian accounts of justice, in which matters of luck are not considered morally arbitrary or necessarily contrary to justice. I will elaborate on these ideas below. For now, note the following: insofar as Childress's argument weds justice and chance, his view may fail to close the door to being co-opted by libertarians and turned in an entirely different direction that would allow for the rationing of lifesaving medical resources on the basis of financial assets.2Rawls's claim that a theory of justice should aim to correct for the morally arbitrary results of the natural or social lottery would seem to prevent Childress from getting his argument off the ground. A libertarian view, on the other hand, would reject Rawls's wedge between justice and fortune and would accept the results of the natural and social lottery as part of the background assumptions for a theory of justice. Libertarian Robert Nozick asks why matters that are classified as “accidental” to human personhood—including natural or economic endowments—should be ignored when thinking about fairness in various forms of commercial and other exchanges (1974, 226). Nozick writes, “Why we, thick with particular traits, should be cheered that (only) the thus purified men within us are not regarded as means [in Kantian theories such as Rawls's] is … unclear” (228). Although Childress draws a metaphysical distinction between essential and contingent aspects of personhood that Nozick would question, his proposal for rationing lifesaving medical resources is nonetheless congenial with accepting lottery results. Consider, then, these questions: If we may accept the results of a randomizing lottery after a pool of medically acceptable recipients has been established, what should prevent us from accepting other aspects of the natural or social lottery as soon as the pool of eligible patients has been determined? Why put up a firewall between essential and accidental features of personhood after potential patients have been identified, but before rationing occurs? Why ignore other effects of a lottery and accept only a randomized version to distribute scarce lifesaving medical resources? Childress argues against a utilitarian approach to rationing lifesaving medical resources, but there is nothing particularly utilitarian about allocating them on the financial market. Childress's anti-utilitarian argument, in other words, rules out too little. As a result, his view is less secure against being marshaled to endorse an alternative, libertarian method for rationing scarce resources, one based on financial assets rather than on individuals' incomparable worth—quite a different procedure from what his interlocutors might have imagined when he first engaged the issue.Below I will address this problem—what I will call the libertarian problem—in Childress's argument and indicate how it might be deflected. Childress seems to agree with libertarian thinking that luck is an acceptable consideration when rationing and seems to disagree only about when luck can be introduced in the rationing process. This idea suggests—but does not substantiate—that Childress has overlooked a feature of libertarian philosophy, namely, the refusal to distinguish between chance and fairness. This feature would allow eligible patients who have secured the necessary economic resources to purchase scarce resources such as organs rather than risk being denied in a randomizing lottery.All this is to say that ideas that resonate with libertarian philosophy seem present in Childress's argument about rationing and patient selection, but their deeper aspects are not given full voice in his 1970 article and subsequent writings on the topic.3 In what follows, I will draw out those ideas, focusing on issues of justice, luck, and ethics. I aim to clarify different ways of distinguishing between justice and chance and then indicate how Childress can defend the idea that random selection is morally preferable to various alternatives for rationing lifesaving medical resources, including those of a libertarian stripe.Arriving at that point will require several steps and distinctions. First, I will sort out how Childress arranges his reasons in defense of using a lottery, making plain his core principle and the reasons that support it. Second, I will turn to the two alternatives to random selection that Childress argues against: letting everyone die, and rationing resources on the basis of social worth. There I will indicate how Childress's argument seeks to avoid nihilism and reductionism—more general orientations of thinking that shadow the alternatives he seeks to rebut. Third, I will turn to matters of metaphysics and ethics, focusing in particular on luck and justice. I will direct our attention to recent discussions of morality and luck to illuminate the seeming oddity of Childress's position and to sort out different versions of luck to indicate how they bear upon morality. To those versions I will add an account of instrumental and corrective luck to show how Childress may endorse a procedure that relies on chance while maintaining an account of the human person that is largely Kantian in spirit, presupposing a distinction between essential and accidental features of personhood. With those arguments in place, I will conclude by showing how Childress's argument can offer us an ethics of luck without a philosophical anthropology of luck and simultaneously avoid the charge of incongruity and deflect efforts to co-opt his position toward permitting money to play a role in the distribution of scarce lifesaving medical resources.Childress's argument in defense of using a lottery appeals to several moral and nonmoral values. These values are all coordinated to defend the idea that using a lottery is morally sound, practically feasible, and preferable to the alternatives that he invites us to consider, namely, allowing everyone to die or selecting patients on the basis of their social worth.Childress's principal justification of random selection defends it “because it preserves a significant degree of personal dignity by providing equality of opportunity” (1997d, 176; emphasis in the original). More precisely, he writes, “the individual's personal and transcendent dignity … can be protected and witnessed to by a recognition of his equal right to be saved. Such a right is best preserved by procedures that establish the equality of opportunity” (176).Childress then appeals to reasons that are not central to his principal claim but that lend auxiliary support to it. Foremost among them is the idea that trust between patients and doctors is better preserved when patients are viewed as ends in themselves rather than as instruments of medical progress or in terms of social usefulness (178–79). Here we see in explicit terms the Kantian undercurrents of Childress's thinking. Moreover, he notes, there are practical advantages to his proposal in that implementing a process of random selection does not require a committee to make the selections. There exist psychological and emotional benefits to using the lottery as well. The stress and strain of being rejected would be exacerbated by the feeling of being judged as relatively unworthy by persons of authority applying social values (178). The experience of rejection can be more easily accepted by an unsuccessful applicant in a situation of equality of opportunity, fairness, and trust—values that are best preserved by random selection (179). Along similar lines, Childress asks us to consider how criteria for distributing scarce lifesaving medical resources would be conceived from behind a veil of ignorance, and he concludes that the rational choice arising from deliberations behind the veil would be random selection, “since this [random selection] alone provides equality of opportunity” (177).With these ideas in mind, Childress targets two alternatives regarding the ethics of rationing medical lifesaving resources: letting everyone die or deciding who receives scarce resources on the basis of social worth.The first alternative, letting everyone die, would disavow rationing altogether and would capitulate to circumstances of bad luck. One might label it a position of negative equality: each person has an equal right to live, creating a situation in which each person's claim to scarce lifesaving resources leads to a wholesale cancelation of rights claims. Because of this wholesale cancelation, all must accept the fate to die. I will call this an egalitarian bad luck position. Stated more dramatically if not more precisely, it is a nihilistic proposal, claiming as it does to preserve no life and no demonstrable value. But if everyone is fated to die, one might reasonably ask whether there is a way to preserve some value. While not exactly the same kind of case as a trolley dilemma, in which a conductor must choose between taking one life or taking several, the case of determining who shall live when not all can live confronts us with a similar problem, namely, whether and how it is justifiable to preserve some value in an otherwise irresolvable life-against-life, or a rights-against-rights, conflict.Childress identifies two ways of considering moral possibilities within the nihilistic proposal, the first of which aims to preserve some value. That option is to hope that some persons will step forward and volunteer to die, thereby resolving the egalitarian bad luck position with a saintly/heroic answer. Nihilism would thus be averted by heroism. Childress puts aside that possibility because it does not accord with the idea of limited altruism, which would paint the heroic picture as unrealistic, given what we may reasonably presume about the general reluctance to perform self-sacrifice.Similarly, Childress rejects the egalitarian bad luck position as “irresponsible” because it ignores the human “inclination to seek [one's] own good” (172). That is to say, the nihilistic alternative ignores another feature of human nature, namely, the inclination of self-preservation and welfare. At a minimum, that idea suggests that anyone asked to consider the problem of rationing should seek an equitable resolution that aims to preserve some life.Two features about human nature are thus cited to discount the egalitarian bad luck alternative: it would be redeemed only by moral heroics, which seems implausible; or it runs contrary to the human instinct of self-preservation. These ideas about human nature's limited altruism and inclination for self-preservation show how demanding the nihilistic alternative is when thinking about the problem of rationing lifesaving medical resources. The effort necessary to redeem a situation of fate, and the demand to accept it, both seem inhuman.There is a problem, however, with the second line of argument that Childress offers against the egalitarian bad luck position: if the inclination to seek one's own good is universal, and a right to self-preservation is premised upon it, then those rights cancel each other out in conflict of life-with-life circumstances. That everyone has an equal right to self-preservation prevents any of them from drawing an asymmetry that justifies differential claims to scarce lifesaving medical resources.4 That fact returns us to the egalitarian bad luck position: everyone is, and should be, equally fated to die. What appeared to be one way of challenging the acceptance of fate—the idea of self-preservation and rights that presumably arise from it—lends ethical support to it.Seen in this light, calling the requirement to accept fate “irresponsible” is to beg the moral question. It fails to indicate what values must be served as a basis for defining a distributive procedure as responsible. Addressing that problem requires us to recast Childress's argument and to introduce consequentialist ideas into his overall account. I will return to this point below.The second alternative to random selection is to decide who is to receive scarce medical resources on the basis of their social worth. This alternative seeks to preserve some value when all value would otherwise be lost. Childress rejects this alternative for familiar reasons associated with utilitarian thinking: admitting considerations of social worth submerges the individual to the social whole; introduces cultural, racial, gender, or class bias into social assessments; presupposes the ability to forecast future social needs; and embroils us in interminable debates about relevant criteria for measuring relative social worth, debates that are unresolvable in a society characterized by a pluralism of value (1997d, 173).Moreover, Childress understands his appeal to equal human dignity to offer a thicker understanding of the human person than that associated with utilitarian approaches to distributing scarce lifesaving medical resources. On this matter of philosophical anthropology, he challenges the impression that deontological views of the self are “thin” and lack an adequate understanding of human historicity and situatedness (cf. Sandel 1998). In this respect Childress supplements his desire to avoid nihilism with the effort to avoid reductionism. Utilitarian appeals to social value, he argues, “would in effect reduce the person to his social roles, relations, and functions. Ultimately it dulls and perhaps even eliminates the sense of the person's transcendence, his dignity as a person, which cannot be reduced to his past or future contribution to society” (Childress 1997d, 174). The appeal to human dignity, in short, exposes limits to notions of personhood often attached to role morality or putatively “thick” accounts of human agency. Viewing the self as radically situated omits features of human personhood that deontological views of agency can capture.With the main lines of Childress's argument now clarified, let us turn to the central ideas in his position regarding the moral bases for using a lottery as a method of rationing lifesaving medical resources to patients deemed medically acceptable. One of my aims in unpacking Childress's argument is to consider the deep foundational challenge I identified in my opening remarks. After putting aside the nihilistic alternative, Childress devotes the bulk of his argument to defending the relative superiority of using a lottery for rationing lifesaving medical resources by reckoning with challenges to his ideas along metaethical lines, focusing on utilitarianism. The gist of his position is that using a lottery is superior to a utilitarian approach for the reasons given above: first, his egalitarian position does not submerge the individual into the social whole and, second, the values for determining social worth lack consensus in a pluralistic society and are subject to various contingencies that render them unusable when applied as moral criteria. But casting his views in opposition to this metaethical alternative distracts him, and perhaps even his critics, from thinking about the distinction between fairness and chance that Rawlsians would draw at a foundational level when theorizing about justice. That potential blind spot and its practical implications require attention.In broaching this discussion of justice and chance, we do well to note Bernard Williams's seminal contribution to the topic in his essay “Moral Luck” and how his ideas about luck and morality were developed by Martha Nussbaum in her discussion of tuchē in The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy and commented upon by Thomas Nagel in an important article on the subject. Williams, Nussbaum, and Nagel cite the fact of luck to challenge modern theories of morality, especially (in the cases of Williams and Nagel) Kantian views of moral agency and theories of justification.Williams and Nussbaum encourage us instead to think about the moral life in terms of Greek literature and philosophy. They believe that deontology and utilitarianism fail to take adequate account of the fact of luck and how luck exposes the fragility, contingency, and vulnerability of our existence. A fuller exposition of their arguments lies beyond the purposes of this article. My point here is to note how Williams and Nussbaum suggest that luck and “moral reason” are, in modern thought at least, at odds with one another. In their view, modern moral philosophy and the picture of the human subject to which it is tied are too cramped to account for the experience of luck, human vulnerability, and dependence in everyday life. Thus Williams and Nussbaum hark to Greek philosophy and tragedy to garner insights about the moral life that are ignored or closed off by deontological and utilitarian theories. Writing on the “still powerfully influential idea that there is one basic form of value, moral value, which is immune to luck,” Williams provides a summary account of the Kantian view that he rejects in this way: Just as, in the realm of character, it is motive that counts, not style, or powers, or endowment, so in action it is not changes actually effected in the world, but intention. With these considerations there is supposed to disappear even that constitutive luck from which the ancient sages were happy to benefit. The capacity for moral agency is supposedly present to any rational agent whatsoever…. The successful moral life, removed from considerations of birth, lucky upbringing, or indeed of the incomprehensible Grace of a non-Pelagian God, is presented as a career open not merely to the talents, but to a talent which all rational beings necessarily possess in the same degree. Such a conception has an ultimate form of justice at its heart, and that is its allure. Kantianism is only superficially repulsive—despite appearances, it offers an inducement, solace to a sense of the world's unfairness. (1981, 20–21) In a similar vein, Nussbaum (1986) observes how Aristotle's account of virtue presupposes being the beneficiary of certain goods (good birth and upbringing being the most notable) as conditions for the good life—aspects of morality that lie outside the agent's control. Nagel's views differ from Williams's and Nussbaum's, but the general thrust of his account moves in the same direction, questioning the elevation of the will for theorizing about morality and, in Nagel's (1979) case, reckoning with the paradox that attends to rendering compatible the idea of volition and responsibility, on the one hand, and the idea that we are subject to external forces to explain our conduct, on the other. With these thoughts in the background we can see, from different angles, the stakes involved in defending luck in light of moral reason, as Childress does in his defense of using a lottery to ration lifesaving medical resources. For Childress, moral reason can justify the place of luck in rationing procedures, whereas for Williams, Nussbaum, and Nagel, the fact of luck renders problematic the modern project of providing moral justification.Clearly Childress's defense of luck does not echo ideas about human contingency, vulnerability, and virtue in ways that resonate with the work of Williams and Nussbaum. His argument instead has a decidedly Kantian and deontological cast, focusing as it does on the egalitarian ideas of human dignity and respect for persons as a basis for understanding the equal claims that potential recipients have for scarce lifesaving medical resources. His defense is precisely of the sort of commitment to equal opportunity that Williams describes in the quotation above.Yet, contrary to inferences that might be drawn from the work of Williams and Nussbaum, a Kantian attracted to Childress's proposal has resources for thinking about tragedies, luck, and loss in moral experience. The tragedy in this biomedical case consists in the fact that some patients will be the victims of bad luck in more than one kind of particular lottery. Here, tragedy is both the context and the result of efforts to redeem a situation in which life conflicts with life. When “some shall live when not all can live,” those who shall live in Childress's account are beneficiari