In 2003, I and other members of the Medicare Coverage Advisory Commission (MCAC) met to consider the benefits and risks of using an implanted but externally powered pump, the left ventricular assist device (LVAD), for “destination therapy.” Destination therapy is for people with congestive heart failure whose age or other diseases make them ineligible for transplants.
The one randomized trial studying its effects showed it reduced the death rate by 48 percent, but half of recipients were dead after a year and three quarters were dead after two years. Many recipients had infections in the first three months after surgery.
The estimated cost of the device and the open-heart surgery to implant it was over $200,000, not including costs of complications. A Blue Cross/Blue Shield estimate of the cost per quality adjusted life year (QALY) was between $500,000 and $1.4 million, making it a true outlier in terms of cost effectiveness. With 5,000 recipients per year – and up to 100,000 potentially eligible – costs add up to $350 million to $7 billion per year.
What is the “opportunity cost” of destination therapy using LVADs? What other benefits could be purchased with the cost of this procedure?
For example, if Medicare invested some of those costs in outreach programs to provide better access to blood pressure and lipid screening, plus treatment, then one could prevent many congestive heart failure cases – lowering mortality rates much more than with LVADs. In short, much more effective and cost-effective alternatives exist than “destination” coverage for LVADs.
Unfortunately, asking this question about opportunity cost – and backing it up with evidence about cost-effectiveness—is not on the agenda for MCAC or the Centers for Medicare and Medicaid Services (CMS). Although the language in the Medicare act says coverage must be provided for “reasonable and necessary” services, it would take an explicit act of Congress to enable MCAC and CMS to consider opportunity costs and to make cost effectiveness analysis a part of deliberation about coverage. That is because no one – the managers of Medicare and the politicians in Congress – wants to face in a public way the need to consider opportunity costs. No one wants to be accused of rationing health care.
Cost effectiveness analysis is an attempt to measure the health benefit per dollar spent. Health benefit can be measured in various ways, including cases of a disease avoided, lives saved, life years saved, and health-adjusted life years saved (HALYs). Since many disease conditions significantly reduce quality of life but may not be fatal, lives saved and life years saved do not allow us to compare health benefits across as many diseases and interventions for them as we may want to.
A life year discounted for a health decrement provides a way make broader comparisons. In medical contexts, the Quality Adjusted Life Year is the construct most often used, as in the Blue Cross/Blue Shield study of the LVADs cost effectiveness. In public health contexts, Disability Adjusted Life Years (DALYs) are used to estimate the burden of disease and cost effectiveness studies focus on the cost per DALY of reducing that burden.
Public Health Service and Institute of Medicine reports on cost effectiveness recommend that it be an input into a broader deliberation about coverage. It should not be used as a mechanical decision-making procedure. The decision process should be free to consider other ethically relevant factors that cost effectiveness by itself is insensitive to, such as distributive issues. For example, cost effectiveness is intended to help us maximize the health benefit produced per dollar spent, regardless of who gets that benefit or where in a life it goes. Yet most people are not straightforward maximizers of such health effects.
Most people want to give some priority to those who are sickest, for example; cost effectiveness give no such priority. (This point is made famous by the experiment in rationalizing Oregon’s Medicaid benefit. The Oregon Health Services Commission originally thought it could simply rank condition treatment pairs by their cost effectiveness and cover them in descending order of cost effectiveness until funds ran out. But the method ended up ranking some lifesaving treatments lower than some quality of life treatments, and public outrage forced a change in methods.) Most people also want to give people a fair chance at some benefit rather than always favoring those who will benefit the most.
If, however, a decision process about coverage is structured to allow input about cost effectiveness, and to encourage deliberation about distributive or other ethical issues, then the flaws of cost effectiveness can be addressed and the valuable information it gives us can also be acted on.
For example, in thinking about the LVAD case, knowing that LVADs are far less cost effective than preventive programs aimed at reducing the risk of heart failure means that we are holding constant the seriousness of the disease. In listening to information about cost effectiveness, but in the context of treating or preventing the same condition, we are less subject to the criticism that we are ignoring the seriousness of the condition.
Consider another way to entertain ethically relevant considerations while still using information about cost effectiveness. In England and Wales, the National Institute for Health and Clinical Excellence (NICE) has been criticized for using a particular monetary threshold – 30,000 pounds per QALY – as a guide to recommendations about coverage. This threshold has only weak normative support for it, and applying it mechanically overlooks important considerations.
For example, if the only treatment for a condition costs more than that threshold, then many would want to be flexible about paying more. A public method, such as NICE’s Citizen’s Councils, should search for relevant reasons for flexibility.
In short, we need to modify how we understand the language of “reasonable and necessary” in the Medicare law by insisting that opportunity cost be considered. Not imposing great opportunity costs on others would be one way to understand “reasonable.” We could then get relevant information about the opportunity costs from cost effectiveness studies, although we would need to view those studies as but one input into a deliberative process that is ethically informed.
Norman Daniels, Ph.D., is Mary B. Saltonstall Professor and Professor of Ethics and Population Health in the Department of Global Health and Population at Harvard School of Public Health. His most recent books are Just Health: Meeting Health Needs Fairly (Cambridge, 2008), and Setting Limits Fairly: Learning to Share Resources for Health, 2nd Ed, (Oxford, 2008). He works in the U.S. and abroad on health disparities, health reform, and priority setting for health. firstname.lastname@example.org; 617–905-4937.