TCS Daily

Winning the Health Care Olympics?

By Arnold Kling - August 31, 2004 12:00 AM

"The fact is that the mainly private U.S. health care system spends far more than the mainly public health care systems of other advanced countries, but gets worse results. In 2001, we spent $4,887 on health care per capita, compared with $2,792 in Canada and $2,561 in France. Yet the U.S. does worse than either country by any measure of health care success you care to name -- life expectancy, infant mortality, whatever."
-- Paul Krugman

This is probably the most common articulation of the problem with the U.S. health care system. That is, the United States spends more than other countries, both in absolute dollars and as a percent of GDP, yet our life expectancy is no higher. This formulation of the problem suggests that our goal should be to minimize the health care spending per capita needed to achieve the level of life expectancy prevalent in other OECD countries. I call this the Health Care Olympics.

Let me begin tongue-in-cheek by offering my own strategy for competing in the Health Care Olympics, which I call the Castration Solution. My suggestion is to castrate any male who attempts to have sex with a girl under the age of 18. I pledge my vote in this year's election to whichever of the two party's candidates endorses this proposal. In 2008, however, the issue will have declined in importance, because by that point my youngest daughter will be 19.

You might suspect that my motives for promoting the Castration Solution have little to do with the real problems in health care. I would say the same about the solutions offered by Krugman and others.

Winning the Olympics

Many people argue that we could win the Health Care Olympics through rationing of health care. In fact, as Edward Lotterman points out, in the case of government-paid health care, rationing is inevitable. It takes place through the definition of treatable diseases and reimbursable treatments.

Suppose that we were rationing health care in order to try for a gold medal in the Health Care Olympics. We could reduce spending on procedures that do little to prolong life. For example, when breast cancer is caught early, after surgery the woman's five-year survival probability might be 94 percent. Additional treatment might increase that survival probability to only 96 percent, which on average would add only 1/10 of one year to her life expectancy (5 times .02). Standard medical protocols call for giving her the additional treatment, because it reduces her risk of death by one-third. Instead, we could deny her additional treatment, because the increase in life expectancy is small relative to the expenditure.

Or consider my father, who uses expensive hearing aids and has had laser eye surgery to remove cataracts. None of this improves longevity -- it only keeps him from being deaf and blind. To win the Olympics, we ought not to spend money on his eyesight or hearing. More generally, we should stop spending on non-life-threatening ailments, such as depression and back pain, that today account for a lot of health care costs.

Another way to win the Olympics would be to try to reduce the share of spending that takes place during the last year of life, which may account for as much as 25 percent of Medicare spending.

This summer, a revered leader in our local community died after a five-year battle with cancer. The hundreds of mourners at her funeral heard how when she was first diagnosed, the doctors said that she had six months to live. The speakers pointed out how she defied the odds and how much she accomplished as a result. To win the Olympics, we should deny care to people like her.

What this example illustrates is that doctors do not know with certainty if someone is going to die over the course of the next year. Would you want to deny care to someone because they only have a 49 percent chance of living another year? You probably want to set the threshold much lower, so that if someone has even a 1 or 2 percent chance of living you give them that chance. But if you set a low threshold, then you won't save much money, so you will not win the Olympics.

Suppose that we wanted to compete in the Health Care Olympics not by cutting health care spending but by finding ways to increase longevity and reduce infant mortality. In that case, we will have to address pregnancy among low-income teens. But there is no medical procedure that can solve the Random Family problem, unless you want to talk about castration.

Overall, I think that it is a mistake to define the health care problem as the need to reduce the ratio of expenditures to life-expectancy gains. I think that the policy implications of such a definition are mostly unacceptable. If anything, the Castration Solution seems to be the least inhumane.

Hammers Without Nails

My suspicion is that Krugman and others, across the political spectrum, define the problem carelessly because they are more interested in pushing their preferred solutions. In business, the saying "to a kid with a hammer, everything looks like a nail" is used to disparage a person or a company that persistently propounds its solution (hammer) regardless of the nature of the problem (nail or otherwise). What I have come to realize is that when it comes to health care, the well-known policy proposals are hammers in search of a nail. In a subsequent essay, I will talk about other possible definitions of the health care "crisis," and examine further the links (if any) between the proposed solutions and the problem.


TCS Daily Archives