Just over 1 in 4 dollars spent by the Medicare program last year was spent on someone who was in their last year of their life.  This is nothing new–the basic proportion has not changed since it was first noted in the 1970s.  Other nations that spend much less on health care nevertheless spend a similar proportion of total system spending on persons in their last year of life

Many believe that a health policy solution focused on reducing spending at end of life (EOL) is a practical cost-saving strategy for the United States, but this consistent proportion (across time and spending level) of total costs spent during the last year of life directly challenges the potential success of this approach.

A focus on EOL spending will likely receive increased attention over the coming months and years because the Medicare program is financially unsustainable, a fact that is exacerbated and made acute by the looming retirement of the baby boomers.  In addition, cuts in planned future Medicare spending were used to finance insurance expansions in health care reform, and a means of absorbing such cuts in a manner that does not harm patients is a necessity.  

Medicare, along with Medicaid, is the primary driver of the long-term structural federal deficit that exists and will exist in the foreseeable future unless changes are made. (See page 3, figure 1.1 of CBO’s June 2009 Long-Term Budget Outlook, linked to earlier in the preceding sentence.) This fact makes further cuts in the future growth of Medicare spending needed if the system is to become sustainable, and there are no easy choices.  The demographic effect of the baby boomers on Medicare was inevitable once they had fewer children than their parents; this reality can be lamented, but not altered.  Slowing the rate of growth in program expenditures (payment cuts) and/or an increase in taxes are the only ways to make the program sustainable.

Reducing End-Of-Life Spending: A Painless Solution?

This is where EOL spending comes in.  Reducing wasteful, futile or even harmful EOL expenditures is an obvious place to start, and such reductions may offer the only “painless” cuts that could conceivably be imagined.  The logic goes like this:  Health care costs rise appreciably near death because most people are sick before they die, and medical care is expensive.  According to the Census Bureau, 83 percent of all the deaths occurring in the U.S. occur among persons covered by Medicare, even though persons age 65 and older represent just 13 percent of the population.

We have anecdotal as well as empirical evidence that the experience of a loved one dying sometimes ends in regret at the medicalized nature of a person’s last living days.  This leads to the conclusion that EOL represents an opportunity to reduce spending in a manner that will not harm patients and which may actually improve quality of life.  In fact, there is evidence that Medicare costs can be reduced slightly via expanded use of the hospice benefit in a manner that also improves quality of life.  We know of nothing else that reduces overall spending and increases quality of life. 

Thus, reducing EOL spending seems the last remaining painless way out of the box.  Right?

Unfortunately, no.  It is doubtful that a focus on reducing EOL spending per se will result in as much savings as is often assumed, for one simple reason:  The concept of the last year of life is inherently retrospective.  You do not know when the last year of someone’s life started until it ends.  The stylized fact that leads to the assumption of wasteful EOL spending., i.e., 1 in 4 dollars spent on care in the last year of life, is based on an inherently retrospective concept that does not translate easily into the prospective decision-making that would be needed to reduce wasteful, futile or harmful spending in the last year of life. 

It is undoubtedly the case that the U.S. health care system is set up to “do everything medically possible” as a default, influenced by payment incentives embedded in how care is financed, how physicians are trained, and cultural norms and patient preferences that emphasize metaphors of “battle”, “fight” and “victory” over disease.  This orientation and the difficulty of accurate prognosis have led in part to the current financing conundrum. 

However, changes are possible.  The last 30 years have seen a cultural and professional dialogue about tradeoffs between expenditures and quality of life, leading to a general agreement that sometimes doing everything that is medically possible is not the best course, and that perhaps one could prolong a person’s life in a medical condition that could be understood as being “worse than death.”  However, acknowledging that this state is conceptually possible is not the same thing as being able to identify such a state in the midst of the emotional intensity of a family losing a loved one, and leading them to alter their health care decisions prospectively. 

The Way Forward

Medicare is unsustainable.  The only possible ways to change this are to increase the “in’s” (taxes) or reduce the “out’s” (care and/or payment rates).  What is needed to achieve sustainability is a focus on purchasing value in the Medicare program across the board, not just with respect to end of life.  It should be possible to reduce spending over the path implied by current Medicare policy while improving patient benefit and increasing value, but it will not be easy.  Reorienting Medicare policy to value-based purchasing would mean that the following questions would be integrated into everything that Medicare pays for:  “Does this extend life?” and “Does this improve quality of life?” 

If the answer to both of these questions is no, then Medicare shouldn’t be paying for it.  If the answer to at least one of these questions is yes, then some means of asking the question “Is this benefit worth the cost?” needs to be developed and made a natural part of the policy-making process of running Medicare and of an honest cultural discussion of the costs and benefits of medicine.  Obviously, there are value judgments and subjective assessments required to identify whether there are quality of life improvements, as well as to determine whether any improvements are “worth it.”  However, even agreeing that these questions are important, legitimate, and should be answered to inform Medicare policy would be a major step in the correct direction.  That will not be easy.

The recently passed Patient Protection and Affordable Care Act (PPACA) has a variety of provisions that provide a chance for us to nudge Medicare toward an increased focus on purchasing value for seniors in a way that could begin to slow the rate of cost inflation in the program.  History shows that changes in Medicare tend to filter into the health system as a whole. 

There are a variety of payment and delivery experiments enabled in the law (Accountable Care Organizations, bundling, etc.) that could have a positive effect in altering incentives that now encourage more, but not necessarily high-value, care.  Increased spending made available for comparative effectiveness research (which is now being called, patient-centered outcomes research) should help us to identify when commonly done interventions and clinical patterns do not seem consistent with purchasing value. 

The Independent Medicare Advisory Commission should help force some difficult changes that could slow cost inflation.  Finally, the Medicare Innovation Center has a chance to alter the culture of the Center for Medicare and Medicaid Services in the direction of experimentation and adopting new approaches in a more timely manner, which could revolutionize how Medicare benefit and payment policy is made. 

There are numerous technical issues and problems related to practically operationalizing these new policy innovations into the functioning of the Medicare program, especially when it comes to the emotionally charged territory of the care of potentially vulnerable individuals who are near death.  Moreover, if we seriously undertake a policy of seeking to reorient incentives and policies in Medicare, mistakes will inevitably be made.  However, the technical barriers of moving toward purchasing value are dwarfed by the cultural and political ones.  During the most recent health reform debate, a proposal for Medicare to pay for a modest palliative care consultation became “death panels.”  As anyone who has taken a glance at the financial situation of Medicare knows, there is nothing systematically limiting the amount of health care that elderly Americans are receiving. 

A Rorschach Test Measuring Perceptions Of The U.S. Health Care System

The real issue is whether there will be a revolution in how we will make medical financing decisions.  Figure 1 below shows the per capita expenditure of the U.S. on health care along with a variety of other nations that are our trading and cultural partners.  The figure in many ways is a Rorschach test for how one views the health care system in the U.S.  In factual terms, it shows that the U.S. spends about two dollars each time many of our trading partners and other high income nations spend one dollar.  However, U.S.  life expectancy at birth is about the same, or a bit worse, and other measures paint a similar story. 

When showing this information as a slide during numerous talks around North Carolina during the past year, there is a surprising reaction by the audience.  Put simply, many responded by saying, “Yeah, but those other systems stink and ours is the greatest in the world.”  There are a variety of conclusions one could draw from the information in this slide, but we think that one is a stretch, unless you were predisposed to it in the first place.  When we view this slide, our first thought is that “we aren’t getting our money’s worth for what we spend on health care.” 

The percent of GDP spent on health care in and of itself means very little.  It is simply a signal of the relative value of health as compared to other items on which we could spend our money.  However, the rate of increase in health care inflation has pushed our expenditures on health to a point of being unsustainable unless we undertake other changes in spending or taxation that we doubt are tenable.  The real question is: why aren’t we getting our money’s worth?  We believe what is needed in policy terms is more focus on purchasing value in health care, first in the Medicare program, and then throughout the system.  Purchasing based upon quality/benefit or cost alone is bad policy; cost per quality/benefit improvement is the key to defining value. 

Such a reorientation would likely reduce EOL spending – along with spending across the rest of the life cycle.  Since we cannot be sure when we are within the last year of life until it is over, the application of policies to change how we purchase care in the Medicare program must be applied across the illness trajectory and not just pigeon-holed as a last year of life issue.  We are buying Fool’s Gold if we think that EOL savings alone will provide the Medicare savings necessary to finance health reform, let alone the further cuts needed to make the program sustainable. 

It will be very hard for our nation to make the changes necessary to render Medicare sustainable.  But in one sense we are lucky:  Our fiscal situation is so grave that we really don’t have any choice but to try.  Lets get started.

Figure 1: Health care spending and life expectancy in a sample of countries (Click to enlarge.) 

Source:  OECD Health Data 2009 – Version: November 09. Graph prepared by Andrew Gelman, Applied Statistics Center, Columbia University.  Appeared in Statistical Modeling, Causual Inference, and Social Science.