The Dartmouth health group released a new version of the Dartmouth Atlas of Health Care today, highlighting once again the substantial variation in costs per Medicare beneficiary across the United States -- and that the higher-cost areas and hospitals do not seem to generate better health outcomes, on average, than lower-cost ones. (On the basis of this evidence, the Dartmouth researchers suggest that health care spending could be reduced by roughly 30 percent without harming health outcomes. These types of estimates suggest that, despite the substantial benefits it delivers, health care may well be the least efficient sector of our economy.) CBO recently released its own study on regional variation in health care costs. One quick aside on the New York Times article on the Dartmouth release: the quotation attributed to me is something I frequently use in speeches and testimony, but was originally coined by Professor Uwe Reinhardt of Princeton. Also note that the Times has an important story on the Medicare Health Support demonstration project, which is intended to examine whether disease management programs can help reduce health care costs. The results, at least for the programs studied in the demonstration project, are not encouraging. Advocates for the programs complain that the implementation of the demonstration project has been flawed, but the results from the project are consistent with the broad conclusions reached in an earlier CBO analysis of disease management.