As the Dartmouth Atlas of Health Care, which tracks disparities in health care costs, has increasingly taken center stage in the health care debate in Washington, University of Pennsylvania professor Richard Cooper has gained widespread media attention for charging that the Atlas improperly portrays discrepancies in regional health care expenditures.
"The Dartmouth Atlas study is shaping the health policy discussion, but is it shaping it right, or is it shaping it wrong?," Cooper said in an interview with The Dartmouth.
Dartmouth Medical School professor Elliot Fisher, the principal investigator on the Atlas Project and director for population health and policy at The Dartmouth Institute, said that Cooper has inhibited health care reform by misrepresenting TDI's work.
The Atlas research concludes that areas with higher medical spending do not necessarily have better results for patient care.
"Both geographically and conceptually, [there is] a lot of unnecessary care given to well-insured patients," Fisher said in a previous interview with The Dartmouth.
Cooper, a medical professor at the Leonard Davis Institute of Health Economics at the University of Pennsylvania, recently wrote two articles in The Washington Post criticizing the Atlas study.
Cooper argues that increased spending in poorer areas of the country is largely responsible for regional disparities in health costs, rejecting TDI's conclusion that the difference is largely due to unnecessary spending on extraneous medical procedures.
"Overall, poor people use more care," Cooper said. "They are readmitted a lot, and we know they stay sick. How big a factor is that in what the Dartmouth group is finding?"
Several Atlas researchers, though, emphasized that their analysis is adjusted for patients' income, pointing to several reports in academic journals that use statistical analysis to show that their research accounts for income-level differences, as well as other factors like race and sex.
"Dr. Cooper does not appear to understand modern statistics," economics professor Jonathan Skinner said in an interview. "Therefore, it is very difficult to debate him on substantiative issues."
Cooper criticized the Atlas for relying heavily on Medicare expenditures, as Medicare is available only to individuals over 65 years of age.
Fisher, however, said that the treatment levels for populations under and over the age of 65 are correlated.
Cooper also charged that the Atlas only looks at aggregate outcomes across regions of the country, overlooking data for specific hospitals and impoverished areas.
In an interview with The Dartmouth, Fisher took issue with Cooper's argument, noting that some of the research divides the country into five large areas, but much of the data is based upon individual patient records and several hundred regional divisions.
Cooper also found fault with the Atlas' focus on data for patients over the last two years, or last six months, of their lives.
The Dartmouth Atlas researchers maintain that their research is supported by academic professionals.
"One of the principles of the Atlas of Health Care project is that all of our research is published in academic journals, all is peer reviewed and the actual Atlas provides stories behind data that makes it accessible to policy makers," Skinner said.
Atlas supporters also contend that many of Cooper's claims are based on anecdotal evidence.
"[Cooper's] analysis identifies particular regions where poverty is high and where spending is high and goes on to say the presence of poverty would explain high spending," said Amitabh Chandra, professor of public policy at the Kennedy School of Government at Harvard University. "He is finding pockets of the U.S. where that is true, but that's not the way research is [properly] conducted."
Chandra was an assistant economics professor at Dartmouth between 2000 and 2005.
