The American Spectator

Last week I was quoted in an AP article to the effect that on some important measures the U.S. still has the best health care system in the world. That got the attention of single-payer advocate Matthew Holt of The Health Care Blog. Holt panned my comments, to which I responded. Holt countered by pointing to an article he had written in Spot-On about measuring health care systems, and challenged me to address it. Of course I accepted.

p>These paragraphs form the crux of Holt’s argument against the notion that the U.S. health-care system is, generally, superior: br> /p>

The basic problem is that the relationship between medical care delivery and health is very tenuous. But if the free marketeers want to go down that path then they need to prove that the huge amount more money spent per head here — we spend at least 50% more than the Europeans per capita — on medical care delivers results in actual improved health outcomes.

They can’t. So they seize on cancer — a nice and very emotive topic — and claim that if you look at the numbers in a certain way we do better here than in those terrible European countries. Now the way disease is treated is bound up in centuries of national culture, and any rational health economist from Victor Fuchs on down will tell you that the process of that treatment has very little to do with the outcome of the disease. So it’s a very silly argument.

br> First, if Holt wants to jump on the free marketers for pointing to cancer data, perhaps he could also critique his fellow socialized-medicine travelers when they use the even worse measures of life expectancy and infant mortality to say that the U.S. health care system is inferior.

Nevertheless, Holt is partially correct that not every dollar we spend on health care above the Europeans is useful. Our system is dominated by third-party payers and those are inherently wasteful, especially the parts of third-party payers that administer routine expenses. However, some of the extra money that we spend is well spent.

Certainly, many factors affect the outcome of a disease that are independent of a health care system. The genetic make-up of the patient, how the patient cares for his disease, and a host of other factors make an impact. However, because a myriad of factors affect the outcome of a disease does not mean we cannot isolate the factors related to the health care system, measure them, and make cross-national comparisons.

We cannot, though, do it with the data from Organization of Economic Cooperation and Development (OECD). The cancer data that Holt refers to above comes from the OECD. At one time I used such data, but the more I looked at OECD data, the more I realized that it is not measured consistently enough across nations to make cross-national comparisons reliable. Indeed, the OECD, in conjunction with the Commonwealth Fund, concedes that that in most cases its data is not “internationally comparable” because “there is a lack of international agreement on the most promising indicators and many definitions of each indicator that could be adopted.”

SO WHAT KIND OF DATA makes for good cross-national comparisons of health care systems? In a previous paper, I laid out three conditions that I think stand up rather well:

1. The statistic must assume that the individual must have interaction with the health care system. A statistic like life expectancy is flawed because many people die with no contact with the health care system.

2. The statistic must assume that what is being measured is something that the health care system can affect. For example, the incidence of a disease assumes interaction with the health care system (an incidence of a disease cannot be known without the diagnosis of a health care professional), but it is not something that the health care system can affect. Rather, the incidence of a disease is caused by factors beyond the control of a health care system, such as diet, genetics, lifestyle, etc. Treatment of the disease, however, is something that the health care system can impact.

3. Finally, a statistic must be collected consistently across nations. While this seems simple in theory, as the OECD has shown, it is rather difficult in practice.

If OECD data does not satisfy these criteria, what sort of data does? Generally, such data is found in studies that compare only a few countries and examine very specific aspects of a health care system.

For example, a study in the journal Circulation found that there was small but significant difference in the five-year post-heart attack survival rate between the U.S. and Canada. The U.S. had a better survival rate that the researchers found was attributable to the fact that the U.S. did more bypass surgeries and angioplasties than Canada. The data collected on both sides of the border was similar. Surely, over five years, many cultural factors not related to the health care system, such as diet and tobacco use, could account for the difference. Yet these factors probably cancel each other out; smoking rates are slightly higher in Canada, but obesity is higher in the U.S. Thus, it seems that our health care system is better at treating heart attack patients.