Health Care BS

Kevin, MD links to Matthew Holt’s sophomoric attempt at slaying the dreaded free market dragon. Possessing little by way of legitimate statistical weaponry, Holt deploys tailored data against the refractory beast. An illustrative example involves bankruptcies allegedly caused by medical bills:

Perhaps health care costs don’t actually cause bankruptcies, or at least not at the 50% rate that Himmelstein and Woolhandler claim. But the 25% number comes from the reworking of the 50% number by Dranove & Millenson, which was in part funded by AHIP.

Holt does not, of course, link to the Dranove & Millenson study. That’s not surprising because he misrepresents its conclusions. As I discuss here, the authors found that only about 17% of bankruptcies can be plausibly attributed to medical bills and that the actual percentage is probably far lower.

And this is not the only example of statistical BS that can be found in Holt’s post. He also makes the following statement:

[The OECD] shows that people suffering heart attacks are much more likely to die here than in Iceland, Denmark, and Switzerland …

Beyond the absurdity of comparing U.S. health outcomes to those of three countries whose combined populations total less than 5% of ours, even someone as intellectually lazy as Holt must realize that OECD statistical data are not internationally comparable.

I’m sure there are more examples of statistical skullduggery in Holt’s post, but I got so bored with the thing that I quit reading. Still, it did raise a question that always occurs to me when adversaries of the market start playing these games: If there is substance to their position, why do they always find it necessary to misrepresent the stats?