Some Statistics On Health Care

Mark Thoma has been discussing health care a bit lately, including in yesterday’s post “Striking Health Care Fact?”, in which he responds to comments by Greg Mankiw and Tyler Cowan. Mark points out that the “striking fact” noted by Cowan (namely, that doctor salaries in the US are more than three times higher than salaries in France) accounts for just a tiny fraction of the difference in health care costs between the US and France, and that France actually has better health outcomes than the US.

I thought I’d take this chance to flesh out Mark’s assertion that France’s health care system keeps its people healthier than the American health care system – because, in fact, it’s not just France that does better than the US. It turns out that every industrialized country in the world does at least as well, if not better, than the US when it comes to health care outcomes for its people.

Here is a renewed look at some data that I wrote about a couple of years ago over at Angry Bear, directly comparing the performance of the US health care system with those of other industrialized countries.

First let’s take a look at the very young. How well does the US health care system take care of children and mothers? The following table gives a few indicators:


Notes: Infant and child mortality rates from OECD for 2003; maternal death rate from the WHO for the year 2000.

The US does worst, or tied for worst, in every measure. But what about the elderly? The following table gives details about the length and quality of life that senior citizens enjoy in various rich countries around the world.


Note: Data from year 2001. Source: WHO.

Not only is the US’s life expectancy at or near the bottom of the industrialized world, but Americans spend more of their shorter lives in poor health (as defined by the WHO).

The most striking thing about this data, however, is that Americans spend nearly twice as much as people in other developed countries spend on health care. Our closest competitor for health care costs is Switzerland, which still spends just 65% of what the US spends per capita. (Interestingly, Switzerland is also second to the US in terms of how much health care spending is paid for privately.) And yet all of these countries acheive health outcomes that are as good as or better than the US: longer lives, fewer dead babies and children, and more quality of life for their elderly. Compared dollar for dollar, there’s no contest. The productivity of the US health care system is terrible compared to other developed countries.


Notes: All data for the year 2004 except for Japan and Belgium, which are from 2003. Data from the OECD.

One last point: some have argued that the US’s poor showing on all of these measures is simply an artifact of demographic or other differences between the US and other developed nations, and is not a function of the US’s health care system. But it’s not obvious to me that such demographic differences systematically fall on the side of giving the US’s health care system more health problems to overcome. For example, the US has a higher rate of obesity; but it also has a lower prevalence of other risky behaviors such as smoking, driving without seatbelts, etc.. The US has more people living in poverty due to relatively high income inequality, but has higher overall and average levels of income than most countries.

At any rate, it turns out that this is exactly the type of question that econometric techniques were designed to answer. And the evidence seems to suggest that the US’s relatively poor health outcomes (especially when considered relative to the vast spending that the US does on its health care) are not the result of such demographic-type factors.

One example is given in a 2003 OECD Working Paper titled “The US Health System: An Assessment and Prospective Directions for Reform,” by Elizabeth Docteur, Hannes Suppanz and Jaejoon Woo. They test the influence that various demographic indicators (e.g. income per capita, fertility, education levels, and income distribution) have on infant mortality and life expectancy statistics in a panel of OECD countries. Then they use those results to estimate how much of the US’s poor results simply reflect these demographic characteristics, and how much are unexplained.

They find that such demographic factors do not explain any of the US's bad health statistics. On the other hand, the US's low level of publicly provided health care does explain about one-third of the US's low life expectancy. The conclusion is that there is something idiosyncratic about the US health care system (with some suggestive evidence that the system’s poor productivity is related to low public funding) that yields relatively poor outcomes despite its enormous costs.

In short: the average American receives mediocre-to-bad health care outcomes while paying twice as much as citizens of the rest of the developed world. And the $400 per person that high doctor salaries in the US contributes to the US’s high health care costs (as noted by Mark Thoma yesterday) does little to explain this massive discrepancy.