The conventional wisdom as to why black Americans have worse health outcomes on average than whites goes like this: blacks get treated differently by doctors and nurses because of their race.

"That may be true," says Dartmouth health economist Jonathan Skinner, Ph.D., who led a recent study on the subject, "but we identify a different source of disparity, and that is that blacks go to different hospitals. . . . Most of the disparity—the overall disparity—is caused by what hospital you go to," he explains, "and what happens within that hospital," rather than how patients are treated on account of their race. Hospitals that treat mostly blacks have worse outcomes for heart-attack patients—whatever their race—than hospitals that treat mostly whites, Skinner and his colleagues reported in the October 25 issue of Circulation.

To reveal this relationship, Skinner, who
is also a professor of community and family medicine at DMS, looked at 30- and 90-day risk-adjusted mortality rates in about a million Medicare patients who were hospitalized for a heart attack between 1997 and 2001. The roughly 4,000 hospitals included in the study were divided into groups according to the racial composition of their patients. Both races received relatively poor care in hospitals that treated predominantly black patients and relatively good care in hospitals that treated predominantly white patients. For example, 90 days after a heart attack, 23.7% of patients in the "blackest" hospitals died, compared with 20.1% of patients in the "whitest" hospitals. That difference may seem small, but Skinner estimates that 1,000 fewer blacks would have died between 1997 and 2001 if they had been treated in the whitest hospitals.

Although this study looked only at heart-attack patients, one published in the December American Journal of Public Health found similar results among premature babies. Both black and white very-low-birth-weight infants were more likely to die in hospitals where over 35% of babies were black versus hospitals where fewer than 15% were black. That paper and Skinner's (both coauthored by economist Douglas Staiger, Ph.D., an adjunct professor at DMS), lead to the question of why blacks end up at worse hospitals.

Status: The answer, they found, is not socioeconomic status per se. "It's about where you live," says Skinner. But, "you shouldn't be thinking inner-city hospitals necessarily" provide poor care, he

cautions. After all, most of the country's academic medical centers, which usually provide high-quality care, are in inner cities.

Skinner suspects that hospitals in the rural, impoverished South are driving the disparities. But that is just a theory at this point and warrants more research. Skinner is also careful to point out that
his study "doesn't deny" that discrimination exists within hospitals. But it reveals a new target for reducing racial disparities in care. "Targeting quality improvements at hospitals that disproportionately serve blacks," wrote
the researchers, "could dramatically reduce the black-white disparities in care."

Jennifer Durgin