In the course of Gladwell’s argument, he cites the RAND study’s findings that people with access to less health care suffer worse rates of hypertension. This example grossly misrepresented the RAND study, which principally concluded that access to more health care does not improve health outcomes.

Now, Gladwell is right in one sense, blood pressure was one exception (vision was the other) across 30 measures. But finding p 0.03 in across so many measures is pretty well covered by chance.

It really reads like Gladwell was data mining to flip the accepted result of this major study, which directly contradicts his point.

Gladwell says the real problem is that we aren’t very good at figuring out what health care is necessary and what care is frivolous. But even if that were true (despite RAND’s findings), it cuts both ways, against Gladwell’s argument as well as for it. It suggests that as you allow people more medical care, they don’t just spend on necessary care first, they waste more on frivolous care at the same time.

I think Gladwell’s characterization of the research is wildly off the mark, but he’s a smart guy and a clever writer, so let’s suppose I’m wrong. Even if Gladwell is right, we’re still faced with a troubling dilemma, maybe best illustrated in the extreme: how many “healing power of crystals” sessions should we pay for if it gets one more person to get a colonoscopy?

At best, Gladwell has just deemphasized moral hazard in the medical industry. At worst, Gladwell’s academic integrity looks a little suspect.

(Overcoming Bias had a piece on RAND here which informed some of my conclusions.)

Nice find.