The debate around age-based healthcare rationing was precipitated by two books in the late 1980s, one by Daniel CallahanFootnote 1 and the other by Norman Daniels.Footnote 2 These books ignited a firestorm of criticism,Footnote 3 best captured in the claim that any form of age-based healthcare rationing was fundamentally ageist, discriminatory in a morally objectionable sense. That is, the elderly had equal moral worth and an equal right to life as the nonelderly. If an elderly and nonelderly person each had essentially the same medical problem requiring the same medical treatment, then they had an equal right to receive that treatment no matter what the cost of that treatment. Alternatively, if cost was an issue because the benefits of the treatment were too marginal, then both the elderly and nonelderly patients requiring that treatment ought to be denied it. If there were something absolutely scarce about the treatment, then some fair process would have to be used to make an allocation decision (and that fair process could not use some age cutoff among the allocation criteria).
The Beginning of the Age-Based Rationing Controversy
Callahan and Daniels may have opened themselves up to these criticisms because their defense of age-based rationing was too broad and unqualified. Callahan, for example, can be fairly read as saying that beyond some age (75–80) individuals ought to be denied very expensive life-prolonging medical care. He appealed to the notion of a “natural life span” as a reference point for justifying this admonition. Once individuals had had the opportunity to achieve what it was reasonable to achieve over a natural life span, they no longer had a strong moral claim to expensive life-prolonging resources, especially if such claims threatened the financial resources needed to afford younger members of society the same opportunities already enjoyed by the now elderly. Callahan's hope was that a cultural shift in social expectations would freely accomplish this result rather than coercive government policies. Callahan did not try to make subtle distinctions among life-threatening circumstances faced by the elderly. Two 80-year-olds with the same life-threatening cardiac condition and surgical need might have radically different prospects for long-term survival (1 year vs. 10 years), but neither would have a stronger claim to those medical resources because both had exceeded a natural life span.
Daniels offered a very different (but very general) argument in defense of age-based rationing, the “prudential life span account.” What he wanted to avoid was a situation in which the healthcare needs of the young were pitted against the healthcare needs of the elderly. He asks us to imagine ourselves (in a Rawlsian-like thought experiment) as prudent young adults who have to allocate limited healthcare resources across our entire life span without knowing what specific health problems we might encounter that would threaten us with premature death unless we had assured access to some expensive lifesaving intervention. What he reasonably assumes is that prudent individuals (in this state of healthcare ignorance) would want to maximize the likelihood of their achieving old age, and, consequently, they would reduce substantially the amount of life-prolonging resources they would allocate to their future possible elderly selves (especially if only very marginal benefits could be achieved regarding length or quality of life) and allocate those resources to younger stages of their life in order to diminish their risks of premature death or disability. This strategy was intended to defuse the criticism that his views were ageist, but it is a very broad general argument.
John Harris has introduced into the literature the “fair innings” argument.Footnote 4 The general idea is that individuals who have achieved a normal life span have had their “fair innings,” and, consequently, they ought not to demand extra innings, especially if making such demands denied younger individuals the opportunity to have their fair innings. This broad argument has intuitive moral legitimacy for many.Footnote 5 However, Harris himself is a critic of this argument (and its ageist implications) except in a couple of very narrowly framed circumstances.Footnote 6
My Key Claim
In this essay I want to defend a “limited” form of age-based rationing. One implication of this claim is that I do not regard age-based rationing as intrinsically morally objectionable (unjust). There are some circumstances in which an age-based rationing decision would be unjust, but other circumstances in which such a decision would be just, or at least “just enough.” In other words, it will sometimes be true that it will be morally permissible to deny an elderly individual a costly life-prolonging intervention that we would be morally obligated to provide a nonelderly individual in those same medical circumstances, and the primary reason for that denial will be the age of that individual as opposed to the probability of success of that intervention, or degree of effectiveness, or comorbidities, or any other morally and medically relevant considerations.
If age-based healthcare rationing has such a morally obnoxious quality about it (but we need to find effective ways to control escalating healthcare costs), why should we not identify other approaches to rationing justly that avoid using any age-based reference points? The short answer I would offer to that question is that we would have to tolerate significant injustices in our healthcare system if we avoided altogether some forms of age-based rationing.
Social Origins of the Age-Based Rationing Problem
In brief, here are the social, political, and economic circumstances that prompted Callahan and Daniels to write their books in the late 1980s. First, the current elderly population in the United States represents about 13% of the overall population, but approximately 35% of all health dollars spent in the United States are devoted to meeting the healthcare needs of the elderly. The statistical fact is that, as things are now, the elderly consume a hugely disproportional share of health resources compared to the nonelderly.
Second, in the United States and other advanced nations in the West, we are just beginning to experience the post-World War II “baby boom” reaching retirement age. In the United States this is a cohort of about 78 million individuals who will raise the fraction of our population over age 65 to about 21% in the year 2030.Footnote 7 If that future generation were to consume healthcare resources at the same level as the current generation of the elderly, we would have extraordinarily difficult moral and political and economic challenges in meeting those demands without sacrificing the just claims (of many sorts) of the nonelderly.
But, third, it is very unlikely that those future demands would be so restrained because of the continued proliferation of increasingly expensive medical technology throughout our healthcare system (as well as the healthcare systems of other advanced nations).Footnote 8 The likelihood is that per capita healthcare spending on the elderly will increase very significantly and make even more difficult the resulting moral and political problems. In 1993 the Medicare program for the elderly (which only covers about half the healthcare expenses of the elderly) spent $150 billion. In 2009 the projections are for $503 billion, and in 2018 the projections are for $932 billion (still 12 years out from having the entire baby boom generation reach age 65).Footnote 9 The conclusion I would draw from these three points is that whatever the data were that prompted Callahan and Daniels to propose age-based rationing in the late 1980s, those data have become substantially worse since then, thereby justifying examining age-based rationing again.
Medicare and Its Unintended Consequences
There were good reasons for creating the Medicare program in the United States in 1965. Relatively few of the elderly then could afford to purchase health insurance from private insurance companies after they had retired (if insurance would even be offered to them) because they were seen as being at high risk for serious health problems. This was properly seen as a justice issue corrected by the creation of the Medicare program. In short, what we created was national health insurance for the elderly. At the time this did not represent a threat to the just healthcare claims of the nonelderly because so little of the costly high-tech medicine we take for granted today had been invented or disseminated. But that dissemination has taken place over the past 40 years. The overall result is that the elderly are living longer with a greater burden of chronic illness that requires constant costly medical assessment and management.Footnote 10
The Medicare and Medicaid programs (the latter covering long-term care for the impoverished elderly) represent public expenditures, which are adding to the current and projected deficit problems of the United States. As I write we are in the middle of wrestling with health reform, the hope being that virtually all the currently uninsured in the United States (47 million) would be provided with a substantial package of health benefits, equal at least to what we provide Medicare beneficiaries. But the total cost of the package to the federal government is now estimated at $1.2 trillion over a 10-year period. It is unthinkable that these costs would simply be added to the federal deficit. That means savings have to be achieved elsewhere (read: Medicare) from other federal healthcare expenditures in order to cover at least part of those costs. Critics will be prompted to ask whether that is fair to the elderly. The obvious retort is to ask whether the status quo (47 million uninsured) is fair to the nonelderly.
To a large extent the elderly have access to whatever the healthcare system has available for meeting their health problems, no matter how costly, no matter how marginally beneficial.Footnote 11 The relatively well-off elderly can demand these extraordinarily expensive cancer drugs that yield only extra weeks or months of life whereas the uninsured nonelderly with early-stage cancers will have no assured access to less expensive, more effective cancer treatments that can either save their lives or prolong them for very long periods of time. Many of these individuals will unnecessarily die prematurely.Footnote 12 A number of recent medical journal articles have been discussing a relatively dramatic increase in the number of individuals in their 80s and 90s who are having complicated surgical procedures for their heart problems.Footnote 13 The logic behind these efforts is nicely captured by Shim:
Together, the hope of medical intervention, the difficulty of prognostication and the actual gains of life that a great many cardiac procedures confer, contribute to the ethical viability and imperative of treating risk. For practitioners and patients alike, understandings of the mandate to treat at ever-older ages contribute, pragmatically, to the elimination of any significant deliberation about whether or not to treat. Instead, standard practice replaces choice.Footnote 14
The point is that the availability of the technology for prolonging life in the very old, together with assured funding for its use in the very old, and the perception of a likely net benefit create a clinical and (unreflective) moral imperative for its use that undermines our society's capacity (and willingness) to meet the just claims of the uninsured nonelderly. These are the clinical and social facts that generate a need for a renewed moral appraisal of the legitimacy of some forms of age-based healthcare rationing. The alternative is the continued toleration of the multiple, diffuse, invisible injustices that are inflicted on the nonelderly regarding their healthcare needs.Footnote 15
Arguments against Age-Based Rationing
Let us briefly rehearse the major moral arguments that speak against the legitimacy of any form of age-based rationing. Baruch Brody contends that allowing any elderly individual to die when we have the capacity to prolong that individual's life (and that individual wants that intervention) represents a failure to fulfill our “duty to rescue.”Footnote 16 He gives special moral weight to the fact that we are talking about identifiable individuals who are patients of physicians (as opposed to statistical lives that might be sacrificed as a result of a social nonfunding decision). We will spend millions of dollars to try to save the life of some hapless hiker in the Arctic wilderness, and, therefore, according to Brody, we have at least as strong an obligation to save older patients who are conveniently at hand in hospital beds. That duty to rescue does not have an age cutoff, which would represent a failure to accord equal respect to older individuals whose lives were in peril.
Someone with a Rawlsian bent might be inclined to argue that the terminally ill elderly are among the “medically least well-off,” and consequently, justice requires that they be provided with more in the way of medical resources, not less (as would be characteristic of age-based rationing decisions).
Again, with reference to the terminally ill elderly someone could invoke a “last chance therapy” argument: It is one thing to insist that someone with a life-threatening medical need accept a less expensive effective intervention for their medical problem rather than a more expensive medical intervention that might be only marginally better. But when someone has only one therapeutic alternative or death, then we are morally obligated to provide that alternative no matter what the cost. The suppressed premise in this argument is that human life is priceless (no matter what the age of a person), and, hence, it is morally wrong to deny individuals with a terminal cancer these cancer drugs that might cost $100,000 for a course of treatment that could yield only a few extra months of life.
Yet another argument invoked in these circumstances is that denying the terminally ill elderly life-prolonging medical resources is just a form of involuntary euthanasia. For those who propound this line of argument the offering of hospice instead of the life-prolonging care these patients want is simply a more socially genteel form of involuntary euthanasia.
Finally, John Harris defends what he calls an “age indifference” principle as a reference point for condemning virtually all forms of age-based rationing. He writes, “An individual's entitlement to the concern, respect, and protection of the community, as expressed in its public health care system does not vary with age or life expectancy. I believe this principle is the only one consistent with justice.” Harris then adds as a supportive argument, “Those who believe in discriminating in favor of the young or against the old must believe that in so far as murder is an injustice, it is less of an injustice to murder the old than the young.” But he adds that our common law tradition does not see things that way. Murder is just as bad, whether an individual is denied 40 additional years of life or a single hour of life.Footnote 17
In Defense of Limited Age-Based Rationing
What we need to notice with regard to the above arguments is that they are perfectly general. They are not aimed at any particular types or circumstances in which age-based rationing might be considered. All age-based rationing is judged to be fundamentally flawed. Here, however, are some examples that ought to precipitate some critical reflection.
We should start by taking note of the fact that subtle age rationing is already part of accepted (and morally defensible) medical practice. We typically limit lots of screening tests by age, such as prostate cancer screening. We stop doing screening mammograms beyond age 70. We focus certain public health programs, such as antismoking campaigns, on the relatively young, believing (correctly) that we are much more likely to achieve substantial positive results there than in someone with a 30-year habit of smoking.Footnote 18 We use 10-year duration hip prostheses in patients over age 80 (as opposed to much more costly longer duration prostheses) when it is unlikely those patients will survive more than 10 years due to assorted comorbidities. This is morally warranted so long as the resources saved are reallocated to meet higher priority healthcare needs of the elderly. Still, my critic might argue that these examples are not persuasive enough because they are not weighty enough; age-based rationing matters most when life itself is at stake. So we shall consider some of those cases.
The DeBakey Case
First, consider the case of Dr. Michael DeBakey, the famed cardiac surgeon who pioneered a number of important surgical procedures. He diagnosed himself with an aortic aneurysm at age 98. He knew that in theory it was surgically repairable, but he feared the brain injury such surgery could occasion. His family and colleagues convinced him to have the surgery. The surgery was successful but Dr. DeBakey spent eight months in the hospital at a cost of more than $1 million (and he survived for a year after that). Our question is: Would Dr. DeBakey have been treated unjustly if he had been denied that surgery, largely because of his advanced age and the likely excess costs associated with a prolonged hospitalization?
Several of the critical arguments advanced above could be invoked to justify providing Dr. DeBakey with the surgery. That is, someone could argue there was a duty to rescue, or that this was a last chance intervention, or that Dr. DeBakey was among the medically least well. However, the very generality of each of these rationales has the morally problematic implication of justifying virtually every costly marginally beneficial life-prolonging intervention for every patient on Medicare, thereby adding to the problem of escalating health costs and diminishing the likelihood of health reform that would provide health insurance for America's uninsured.
Invoking the “duty to rescue” argument makes moral sense when the numbers are small and the circumstances rare, even if the costs per rescue are extraordinarily high. But in the world of life-prolonging medical technology we have today, the opportunities for rescue are ubiquitous among the elderly, and the likelihood is that a majority of Medicare recipients would need to be rescued at great expense several times before disease overmastered them. This is a radically different situation from the “usual” circumstances in which we invoke the “duty to rescue.” The command of public resources by the Medicare elderly, especially under this alleged ethical obligation, is very much a threat to the just claims of the uninsured and marginally insured nonelderly.
A similar analysis will apply if the ethical argument involves an appeal to last chance therapies or to the status of being medically least well-off. There is a range of circumstances in which these moral considerations will be legitimate and determinative of a just outcome. For example, a relatively small number of children are born each year in the United States with either Gaucher's or Hunter's syndrome. These are metabolic disorders that will result in very premature death. Today we have the drugs imiglucerase and idursulfase, each of which costs $300,000 per person per year. But they are effective in preventing very premature death for these young afflicted individuals. These are last chance therapies; there are no alternatives. Further, these children can be regarded as being among the medically least well-off. What makes the critical moral difference in these cases is that these interventions are very effective in sustaining both length and quality of life, and these individuals are very far from having had their fair innings. The same cannot be said for Dr. DeBakey.
My critic might respond that this is not really a matter of age-rationing; rather, what matters is whether a life-sustaining intervention is going to be sufficiently effective in a very old individual. So the proverbial vigorous elderly individual (Hazel Homer [age 99] in note 13) who gets 5 extra years of life from an expensive cardiac intervention has a just claim because that intervention proved to be so effective. However, I would argue that is too quick a moral judgment. What can be argued instead is that there are compelling considerations of compassion that would warrant giving Hazel her bimodal cardiac resynchronization device at $60,000. Compassion must not be confused with justice, nor will compassion trump justice if a compassionate response to the health needs of the very old threatens to fail to meet the just health claims of the relatively young. We can illustrate this last point with our second critical example.
The Case of the Artificial Heart
In the United States we are currently clinically testing totally implantable artificial hearts [TIAHs]. These devices may be regarded as last chance therapies for patients in end-stage heart failure. Each year in the United States 550,000 individuals enter early stage heart failure. It is expected that improved models of this device will yield 5 extra years of life expectancy on average for patients who would otherwise die of heart failure. Computer modeling suggests that we could implant as many as 350,000 of these devices per year at a cost of $300,000 each. That would add $105 billion to the cost of healthcare in the United States, roughly 70% of which would be added to the cost of Medicare. About 100,000 of these devices would be needed by those below age 70. If the societal decision were made that we could afford only 100,000 of these devices, do we want to argue that they ought to be distributed by lottery over the elderly and nonelderly (so long as minimal criteria for likelihood of success are met)? Or would we have good reasons for saying that the nonelderly had the stronger just claims because they had not yet had their fair innings?Footnote 19 Should we be morally comfortable with accepting Harris's age indifference principle as justification for using a lottery to distribute access to TIAHs? If we were, then a 40-year-old in heart failure facing death in 6 months with a prospect of 20-plus years survival with a TIAH would be entitled to no more a chance for that TIAH than a 90-year-old in heart failure with only a 2-year prospective gain in life expectancy. This does not look like a morally plausible result that most of us would be comfortable endorsing.
We must ask ourselves how the 90-year-old managed to achieve that age. He might have gotten there because substantial and costly healthcare resources got him there, the very sorts of resources that the 40-year-old might now be denied. Alternately, he might have simply been the beneficiary of an excellent genetic endowment (whereas our 40-year-old was the unfortunate recipient of a deficient genetic endowment). From the perspective of healthcare justice, what would justify society conferring additional life-years on the fortunate 90-year-old at the expense of life-years lost by the 40-year-old? Neither egalitarian nor utilitarian considerations would warrant such an allocation.
Harris's view is that both such individuals equally desire to experience “the rest of their lives,” and this is what requires equal moral respect. But that expression of desire looks more like plain self-interest than any consideration having substantial moral weight. If we perform a Rawlsian-like mental experiment with two 20-year-olds and tell them that one of them will have the fate of our 90-year-old whereas the other will have the fate of our 40-year-old, but we must decide now which of those two future possible selves will have assured access to the TIAH, I am confident that the impartial answer that would emerge would be that the 40-year-old gets the TIAH.
Harris is correct that there are numerous rationing scenarios in which age indifference would be a moral requirement. But that does not imply that age indifference is required in every rationing scenario. Implantable cardiac defibrillators (ICDs) cost $40,000 each. They are intended to prevent sudden death from an arrhythmia. In the United States we may be implanting 200,000 per year. However, 81% of these devices never fire over a 5-year period, at which point batteries need to be replaced at a cost of $20,000. There are now tests that can determine with 98.7% accuracy which patients will likely not suffer an arrhythmic event over the next 2 years. We could reduce by 70,000 the number of devices implanted and save $3 billion per year, but there would have been 800 (in theory) preventable deaths as a trade-off for saving those dollars. In this scenario age indifference would be a moral requirement. The 60-year-old and the 80-year-old would both be justly denied an ICD if the relevant medical test indicated it was extremely unlikely the ICD would make a difference. In both cases the marginal benefit is too small to justify the cost for either individual. Higher priority health needs of both the elderly and nonelderly could be better served by the achieved savings.
Harris invokes the analogy with murder as a basis for rejecting all age-based rationing that involves loss of potential life-years. However, this is an inapt analogy. The murderer is violating the liberty rights of his victim. His victim would have gone on living happily but for his murderous intervention. His victim does not require any expensive social resources for continued survival. No question of distributive justice is raised in the case of murder. But that is precisely the moral issue at stake with respect to the problem of healthcare rationing. We have only limited resources to meet virtually unlimited healthcare needs. What criteria may we use to determine which needs are fairly met or unmet? If we spend $300,000 to give a child with Gaucher's an extra year of life, are we equally morally obligated to spend that same amount to give an 80-year-old an extra week of life, or an extra day of life, as Harris's use of the murder analogy seems to imply? This is neither just nor reasonable. The loss of that extra week of life is rightly regarded as being unfortunate, not unjust.
There are numerous reasonable and morally relevant considerations that might be invoked for making fair healthcare rationing decisions at the level of social policy. These considerations will often need to be balanced against one another in complex circumstances. We have no reason to believe there exists one “perfectly just” balancing that could be achieved through human reason in all these circumstances.Footnote 20 Numerous options for allocating artificial hearts can be reasonable and “just enough.” But it would be unwise and unfair if such allocation decisions were left to the moral intuitions of tens of thousands of individual physicians, all vulnerable to the normal range of personal biases.
Rationing and Rational Democratic Deliberation
The better option, as I have argued at length in a recent book,Footnote 21 would be a public process of rational democratic deliberation. This would not be a Rawlsian mental experiment; this is a real world option. The vast majority of us in middle age are completely ignorant of our future possible health needs, and in that respect we are capable of the moral impartiality required by these choices. Further, we are mindful of what we regard as reasonable financial expenditures of our money (taxes) for healthcare. That creates the budgets (limits) within which we can collectively set healthcare priorities and determine what rationing protocols we are willing to impose upon our future possible selves that we judge to be “fair enough.”
If we decide that we are willing to pay for only 100,000 artificial hearts and that these should not be available at social expense to those over age 70, we have made an age-based rationing decision. But it is not a decision we are imposing on others we deem unworthy of equal moral respect. And it is a judgment we make with the understanding that this represents a last chance therapy for a future possible version of myself. We recognize that we are telling future caregivers that we have freely given up a right to be rescued with an artificial heart at age 77 because we have chosen, in effect, to allocate limited and expensive life-prolonging healthcare resources to earlier stages in our life when they would do much more good and be judged to be much more valuable to us. Such judgments represent fair terms of cooperation, commendable moral reciprocity. They cannot properly be described as nonvoluntary forms of euthanasia.Footnote 22
If we collectively decided, we could pick age 75 instead of age 70, or we could choose age 80 as an absolute limit for an artificial heart at social expense. We could choose not to use any age at all, relying instead upon some complex mix of clinical criteria related to probability of survival for some number of years. But that would create considerable fuzziness and uncertainty (and consequent risk to fairness) in actual practice. This is what Callahan refers to as the “ragged edge” problem.Footnote 23 The virtue of a definite age limit in this context is that it creates a bright line that might better protect fairness. What a fair deliberative process allows us to say at the very least is that choosing a specific age beyond which individuals would not have a claim to an artificial heart at social expense is not unjust.
Another circumstance in which we may have to invoke some form of age-based rationing would be a serious influenza pandemic, comparable in severity to the 1918 Spanish flu that may have killed 50 million people. If we did not have enough respirators to save the lives of all who would otherwise be virtually certain to die, then it would be morally defensible (not unjust) if we were to deny respirators to those over the age of 70. The substantive moral justification for such a choice would be the fair innings argument. The procedural moral justification would be a rational democratic deliberative process that endorsed such a policy, even if the endorsement was as the least unjust option available. We might wonder how this could be fair to those who are currently in their 60s or 70s. I will mention that I am 65 years old, but I would certainly endorse this policy if any alternative arrangement put my children at risk of being denied a life-saving respirator (say, as a result of a lottery process). I suspect this same commitment would be endorsed by the vast majority of my fellow nearly senior citizens.
Conclusion
What we are defending in this essay is the need for surgically precise age-based rationing protocols that have been fairly and democratically legitimated in a range of rationing circumstances where an age-based criterion would yield a more just outcome. We are not denying that the greater health needs of the elderly justly command more social resources to meet those needs. But we are saying that there must be just limits to meeting those needs. We must also note the morally relevant point that some of the savings that result from age-based rationing protocols redound to the benefit of the elderly in the form of health services that yield more health good for them at a lower cost. Purchasing very expensive marginally beneficial life-prolonging healthcare can adversely affect the just claims of the elderly to needed healthcare as much as the just claims of the nonelderly. This last point applies as much to the U.S. healthcare system as to European healthcare systems.
