I've gotten a surprising number of e-mails asking me to take down last week's New York Times op-ed that brought together John Kerry, Newt Gingrich, and, uh, baseball statistician Billy Beane, to argue that health care could do with more data. So much as I fear the obsolescence suggested by the emergence of Nate Silver and Billy Beane, I can't really knock down the op-ed. I agree with it. The next graph is one of the most famous and powerful in contemporary health policy. It comes from the Dartmouth Atlas Project which graphs Medicare spending against care quality:
See a relationship? Yeah, me neither. But it's easy to see how much we're spending for unnecessary care. The same data manifests in studies of patient outcomes: Above some base level, more treatment does not make us immortal. It does not even make us healthier. It's useless. But medicine is not a data driven profession. Doctors prescribe what they think will work. And what they think will work is informed by a well-meaning but often inchoate mixture of evidence, half-remembered studies, things they've heard, economic incentives, lawsuit concerns, etc. That's not their fault: No single human being can keep up with something as complex as 21st century medicine. Which is why the sector needs to be relying on modern health information technology that incorporates tremendous amounts of data on not only cost-effectiveness, but simple effectiveness. But it's not doing that. Your health records are printed out and kept in a manila envelope. Your prescriptions are scrawled in chicken scratch. Many of your treatments are being prescribed on the basis of evidence gathered by the company that sells them. We should be investing billions into comparative effectiveness research. It would make us healthier. It would save us money. Before we ration by denying people care, we should be cutting worthless care -- currently estimated at somewhere between 15% and 50%of all care -- out of the system. In medicine, cost control entails some hard decisions, particularly in the long-run. But that often gets conflated with the fact that it requires some easy decisions that we simply don't have the information to make. If it's true that lumbar surgeries are wildly overprescribed, parceling them out more stringently is an easy decision, not a hard one. But we need sufficient evidence to make it. Billy Beane, John Kerry, and, sigh, Newt Gingrich, are right. To step onto Beane's turf, what we have right now is a ball club funded like the Yankees that wins like the Cubs. We can do better. Update: Jon Cohn e-mails to inform me that "given this year's performances, [our health care system] is a team that's funded *and* plays like the Yankees (who, unlike the Cubs, didn't even make the playoffs)." As it happens I don't, uh, watch baseball, so that may well be right. Though in my head, I was thinking about a broader time series than just this year. Even so, Billy Beane may take my job, but he need not worry about me taking his.
See a relationship? Yeah, me neither. But it's easy to see how much we're spending for unnecessary care. The same data manifests in studies of patient outcomes: Above some base level, more treatment does not make us immortal. It does not even make us healthier. It's useless. But medicine is not a data driven profession. Doctors prescribe what they think will work. And what they think will work is informed by a well-meaning but often inchoate mixture of evidence, half-remembered studies, things they've heard, economic incentives, lawsuit concerns, etc. That's not their fault: No single human being can keep up with something as complex as 21st century medicine. Which is why the sector needs to be relying on modern health information technology that incorporates tremendous amounts of data on not only cost-effectiveness, but simple effectiveness. But it's not doing that. Your health records are printed out and kept in a manila envelope. Your prescriptions are scrawled in chicken scratch. Many of your treatments are being prescribed on the basis of evidence gathered by the company that sells them. We should be investing billions into comparative effectiveness research. It would make us healthier. It would save us money. Before we ration by denying people care, we should be cutting worthless care -- currently estimated at somewhere between 15% and 50%of all care -- out of the system. In medicine, cost control entails some hard decisions, particularly in the long-run. But that often gets conflated with the fact that it requires some easy decisions that we simply don't have the information to make. If it's true that lumbar surgeries are wildly overprescribed, parceling them out more stringently is an easy decision, not a hard one. But we need sufficient evidence to make it. Billy Beane, John Kerry, and, sigh, Newt Gingrich, are right. To step onto Beane's turf, what we have right now is a ball club funded like the Yankees that wins like the Cubs. We can do better. Update: Jon Cohn e-mails to inform me that "given this year's performances, [our health care system] is a team that's funded *and* plays like the Yankees (who, unlike the Cubs, didn't even make the playoffs)." As it happens I don't, uh, watch baseball, so that may well be right. Though in my head, I was thinking about a broader time series than just this year. Even so, Billy Beane may take my job, but he need not worry about me taking his. 
