A key component of health-care reform -- and saving our ass from going bankrupt and sick from spending too much on lousy treatments -- is establishing comparative effectiveness measures, otherwise known as "actually knowing WTF works and what doesn't."

This idea terrifies companies who don't want such objective measures. It also generates a lot of fear, partly via confusing or intentionally frightening arguments. Yet making sure we don't pay for stuff that doesn't work is key to reform -- a point made in this Times op-ed from libertarian economist Tyler Cohen, keeper of the blog Marginal Revolution.

Cohen argues that the main problem is, as he puts it,

the financial incentives for doctors and medical institutions to recommend more procedures, whether or not they are effective.

These were discussed vividly in Atul Gawande's recent New Yorker piece, and they are clearly a part of the problem. I think Cohen lets industry off a bit too easily when he says that drug-company profits aren't really part of the problem, for expenditures on drugs that either do little good or do little better than far less expensive drugs are costing us many billions as well; over the last two decades, for instance, we've spent scads of money on modern antipsychotics that cost 20 times as much as the drugs they replaced -- and only recently gathered enough data to show they work no better than the old ones.

Read the original blog entry...