Health Care Spending: Where it Came From, Where it Went

Illich's description of social iatrogenesis and the sisyphus effect compels us to think about the large sums that are expended on people near the end of life. It is true that about 30% of Medicare spending is on people in the last year of life, who are only about 7% of all enrollees, and of this, about half is spent on people in the last 60 days of life. In fact, although I have not been able to find dollar figures, public spending on the dying elderly is even greater than that because many of them are also receiving Medicaid benefits for long term care.

It is tempting to think that much of this spending is inappropriate -- that people with no hope of regaining a decent quality of life are spending weeks in intensive care, hooked up to ventilators and IVs and feeding tubes. No doubt this does happen, but it is not the norm and it does not make a major contribution to overall health care spending. Most elderly people do not die in hospitals, but in nursing homes. That supportive care is also expensive. It is true that Medicare spending is highly concentrated in a few individuals -- 5% of enrollees account for nearly half of spending each year. However, only one fifth of these people die in that year.

Again, it is tempting to think that we could still save a good deal of money by not aggressively treating that 1% of enrollees who account for 10% of spending but die within the year anyway. The problem is that doctors aren't very good at predicting who these people are. Contrary to general perceptions, many elderly people who undergo expensive hospital care do recover and go on to lead worthwhile lives. Only in the case of terminal cancer can doctors make fairly reliable predictions of imminent death. That is why, although Medicare does offer a hospice benefit, very few people use it. It just isn't possible, in most cases, to decide with any confidence that a person only has six months to live.

Many of those expensive Medicare beneficiaries have chronic diseases such as diabetes and heart failure that can lead to expensive hospitalizations, but they live with these conditions for many years. Most people find the idea of denying them care reprehensible. For the wonkishly minded, an excellent resource full of facts and figures and references is here, from the Alliance for Aging Research.

The growth in Medicare spending is driven only in small part by the aging of the population. Most of it reflects general medical inflation, the same forces that have driven up the cost of insurance for all of us. And countries such as Sweden and Japan, that have an even higher proportion of elderly than the U.S., spend much less on medical care than we do, with better results.

How do we explain this seeming paradox?