AMA Journal of Ethics^®

Death panels do not fit anyone’s conception of health reform. Yet this is what the Obama administration was accused of endorsing when it proposed paying physicians to talk about advance care planning with their older patients. The rhetorical implication of the charge was that physicians were being paid to counsel terminally ill patients to refuse life-prolonging care in order to save Medicare money. The painful clinical reality often enough is a minimally conscious moribund patient whose family demands that “everything be done,” despite the obvious and pointless prolongation of terminal suffering that entails.

Medicine today is suffused with the need to control costs. How can that be done ethically? What does it mean to be a “just” and “caring” society (or physician) when we have only limited resources to meet virtually unlimited health care needs? The “limited resources” are money, taxes, or insurance premiums that we (collectively) are willing to pay to meet our health care needs. Our health needs are “unlimited”; the last 40 years have seen a proliferation of new and expensive medical technologies that effectively create needs, driving health costs skyward [1]. In 2009 the United States spent $2.5 trillion on health care, about 17.6 percent of our GDP [2]. It is predicted that by 2019, this figure will have risen to $4.5 trillion, or about 20 percent of the likely GDP [3]. Medicare expenditures were about $500 billion in 2009; in 2019, they could be as high as $1 trillion [3].

Some researchers believe that cost control can be achieved “painlessly,” without violating moral norms or deep political values [4]. This is supposed to be achieved by getting rid of “waste and inefficiency.” Unfortunately, one person’s waste and inefficiency is often another person’s life-sustaining care. How much life-sustaining are we morally obligated to provide to patients in a persistent vegetative state (PVS)? Think of Terri Schiavo and the moral and political controversy that case generated. There may be 25,000 such patients in the U.S. whose care costs more than $2.5 billion annually. We might be tempted to regard this expenditure as being merely “imprudent,” not unjust. But the Urban Institute estimates that 22,000 Americans die prematurely each year as a result of being uninsured and unable to afford timely primary care that might have addressed a curable cancer before it became incurable [5]. These are patients who clearly could have benefited substantially from timely medical care, unlike PVS patients. This should nudge us out of any moral complacency about “imprudence.”

Renal dialysis was perfected in the late 1960s. Very few Americans in end-stage kidney failure could afford the annual cost ($90,000 in 2010 dollars) of dialysis at that time, which meant they would die prematurely. Advocates were outraged that we would allow these individuals to die when we had an effective technology that could prolong their lives indefinitely, and only money stood in the way of saving them. The rallying cry of some was that human life was priceless. Congress responded by creating the 1972 ESRD amendments to Medicare, which paid the full costs of dialysis or kidney transplants for all patients in end-stage renal failure, no matter what their employment or insurance status might be. The belief at the time was that this was a unique technology and that the cost of the program would top out at $500 million 20 years out. The actual cost of the program in 2009 was about $31 billion, and it was sustaining about 480,000 lives. This was national health insurance for end-stage kidney disease [6].

The obvious moral question today, a question of justice, is why patients who need equally expensive care for their heart disease or cancer or liver disease do not also have national health insurance coverage. What became apparent by the late 1970s was that dialysis was not a morally unique life-prolonging intervention; rather, it was the beginning of a torrent of such technologies.

Today we have dozens of extraordinarily expensive cancer drugs that cost $50,000 to $130,000 for a course of treatment and yield median survival gains measurable in weeks or months [7]. More than 600,000 cancer patients each year are candidates for these drugs. We have dozens of drugs to manage heart disease. None are as expensive as the cancer drugs, but with 5.5 million Americans in various stages of heart failure and about 70 million Americans with some form of heart disease [8], costs add up quickly.

In 2010 we did more than 1.2 million coronary angioplasties at $40,000 each and almost 500,000 coronary bypass surgeries at $65,000 each. We implanted almost 200,000 cardiac defibrillators (ICDs) at $40,000 each with the intention of preventing fatal cardiac arrhythmias. Does it matter that 81 percent of them never fired over a 5-year period, at which time a battery would have to be replaced for $20,000 [9]? Does that represent a wasteful use of health resources? We have a test that can identify with 98.7 percent accuracy who among these potential ICD recipients will not have a fatal arrhythmia over the next 2 years. We could save $2 billion per year by using that test. But getting it wrong 1.3 percent of the time represents 800 lives that would be lost each year. How should we assess that outcome, morally speaking? Does that represent a morally objectionable “pricing of human life”?

We have about 4.5 million Alzheimer patients in the U.S. At least 500,000 of them in any given year will be in the end stages of that disease. If physicians caring for those patients detected a heart irregularity suggestive of a potentially fatal arrhythmic event, would it be unjust if they failed to offer the option of an ICD to those patients? Would this be a morally objectionable “pricing of human life”?

More than one million Americans are HIV-positive. Today the vast majority of those individuals have their lives sustained by three or four drug combinations that cost $35,000 per person per year. Each year in the U.S. 550,000 individuals are in end-stage heart failure. Dick Cheney, our former vice president, is one such individual. But he is having his life prolonged for an extra year or two because he received a left-ventricular assist device (LVAD) at a cost of $200,000. Does every one of those 550,000 individuals have a right to one of these devices for an extra year or two of life? That would represent an extra $100 billion in health care costs. If we failed to provide that option to all those individuals for financial reasons, would that represent a morally objectionable pricing of human life?

If we have only limited resources to meet virtually unlimited health care needs, at what social cost might a just and caring society limit access to LVADs to relatively younger individuals (70 or fewer years of age) in order to provide access to needed and effective health care for the 50 million uninsured in the U.S.? Or to assure sufficient resources for meeting the life-prolonging needs of HIV-positive patients? About 1.3 million Americans have rheumatoid arthritis. Roughly 20 percent of them will have the most serious form of that disease, requiring treatment with infliximab at a cost of $25,000 per person per year. Infliximab does not save or prolong the lives of these patients; it “only” improves the quality of their lives. Does this mean that if we are concerned about “pricing human life” that we are morally obligated to fund LVADs for all patients in end-stage heart failure who want one along with any cancer drug that offers only very small gains in life expectancy before we would consider funding the costs of infliximab for patients with rheumatoid arthritis?

My point is that the rhetoric of “pricing human life” badly distorts our moral judgment when we must set health care priorities and control health care costs. We cannot avoid the need to make rationing decisions; we have only limited resources and the domain of what we call health care “needs” expands with each new medical technology [10].

The fundamental moral challenge is this: should we allow the nonrational and nonmoral forces of the market and the relative power of conflicting political interest groups to determine who is denied access to needed health care (which is how rationing now occurs in the U.S.)? Or should rationing decisions come about through a rational, respectful, inclusive, democratic deliberative process that allows us to decide together what expensive, marginally beneficial, life-prolonging health care we would be willing to deny to our future possible selves [11]? The virtue of such a process is that it is open and transparent, unlike the hidden, unaccountable workings of markets and interest groups that allow us to accept with equanimity and moral indifference the premature deaths annually of 22,000 uninsured Americans. Such a process also has the virtue of sparing physicians primary moral responsibility for making these rationing decisions.

If I am unwilling to pay additional taxes or insurance premiums for an LVAD for an 85-year old stranger in end-stage heart failure because I believe it is “not worth it,” then I am morally obligated to say the same holds true for a future possible version of me in those same circumstances. We have the moral right to judge for our future possible selves that there are many other health care interventions that I might need that yield much more benefit at a much lower cost than that LVAD. Such a public choice for our future possible selves is neither unjust nor uncaring.

The same cannot be said for the life-threatening rationing decisions imposed by the Arizona governor and legislature upon 100 Medicaid recipients needing various transplants [12]. Their claim was that this was not a cost-effective use of taxpayer money. If this was an honest assessment, then such transplants ought to also be excluded from the health plans of these government officials. This is what it means to make rationing decisions that are just and caring and responsible.