Healthcare Reform: Ignore the New York Times; Overtreatment is Still a Big Problem

Call it the healthcare equivalent of denying global warming. Last week, two veteran New York Times reporters, Reed Abelson and Gardiner Harris, questioned the validity of research that the Obama Administration used to justify its healthcare reform plan. Led by researchers at Dartmouth Medical School, the studies of geographical variations in Medicare spending suggest that up to 30 percent of those expenditures may be wasted. While it's unlikely that the NYT's assault will lead many experts to re-evaluate that conclusion, it's telling that some revisionist critics have managed to sway the newspaper of record.

The debate over the Dartmouth Atlas of Healthcare, which details the practice variations, and the studies related to it has warmed up in recent months. Richard Cooper, a hematologist and physician workforce expert at the University of Pennsylvania, argued in a Washington Post op-ed last fall that the Dartmouth researchers and others were wrong when they said we could save 30 percent of healthcare expenditures if the high-spending areas of the country were more like the low-spending ones. Cooper pointed out that several leading experts -- including surgeon and New Yorker writer Atul Gawande, CMS Administrator-designate Donald Berwick, Dartmouth research leader Elliot Fisher, and former acting CMS chief Mark McClellan -- had acknowledged that the Dartmouth data applied only to Medicare patients and might not be extensible to those under 65. But those experts also said that the same patterns of care that keep Medicare costs low in some regions probably affect non-Medicare patients, as well.

I won't go into all of the inaccuracies and misinterpretations in the Times article, which are well covered in the response by Fisher and Jonathan Skinner. But it's frankly amazing that the NYT reporters would say that the Dartmouth research did not take the patients' relative health status into account. Not only is that untrue, but it ignores a 40-year body of work that started with John Wennberg, the pioneering epidemiologist on whose shoulders Fisher and Skinner stand.

Some years ago, I interviewed Wennberg, and here's what he told me about a 1989 study of practice variations:

"For reasons that had nothing to do with illness rates," he said, "the number of staffed beds per capita in Boston is about 60 percent greater than in New Haven, Conn. And Bostonians' rate of hospitalization for high-variation medical conditions are also about 60 percent higher.

"Bostonians aren't sicker than people in New Haven. In fact, the pattern of initial hospitalization for certain conditions is similar. For instance, our group published one paper comparing patients with hip fractures in Boston and New Haven. The hospitalization rates were the same in both places. But over a three-year period, the Boston patients had a 60 percent higher readmission rate."

In any case, the large variations in how patients are treated from one area to another, even within the same state, can't be attributed to differences in patient health status or the cost of living. For example, the Dartmouth researchers studied rates of back surgery and knee and hip replacement in south Florida and discovered huge variations that they attributed to surgeons' practice patterns. "It seems improbable... that Medicare retirees living in Fort Myers prefer back surgery 2.4 times more than residents of Miami," they wrote.

Higher health costs that do not produce better outcomes represent a waste of resources. The question is not whether we could save money if doctors in Miami or New York practiced more like physicians in Minneapolis or San Francisco. The question is how we get them to change.

Image supplied courtesy of Brave Sir Robin at Flickr.