Why Racial Disparities in Health Care Persist

It's been more than a decade since Congress first officially acknowledged that this country has a problem with race and health. In 1999 the government asked the Institute of Medicine—an independent nonprofit whose reports are the gold standard for health-care policymakers—to investigate disparities in health and health care among racial and ethnic minorities. The results were damning: the ensuing study, called Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care, found that minorities had poorer health and were consistently getting lower-quality care even when factors such as insurance status and income weren't involved. They were less likely to get lifesaving heart medications, bypass surgery, dialysis, or kidney transplants. They were more likely to get their feet and legs amputated as a treatment for late-stage diabetes. Clearly, something needed to be done.

In the years since the report, the issue has gotten plenty of publicity, more reports have come out, and several agencies—including the National Center on Minority Health—have examined the problem and suggested solutions. Still, studies continue to turn up disturbing disparities. For instance, earlier this month, a paper from the Fred Hutchinson Cancer Research Center found that between 1992 and 2004, black women were up to 90 percent more likely to be diagnosed with advanced breast cancer than white women, even though rates of mammogram screening were similar for the two groups. Another recent study put the health data in financial terms and found that race-related differences in health care cost the country $229 billion between 2003 and 2006, a result that Health and Human Services Secretary Kathleen Sebelius called "just stunning and shocking."

So why, now that everyone's aware of the problem, do we still have one? Mostly, the reason is that health and race are both complicated issues to examine academically. Put them together, and constructing a study design that can tease apart all the issues and make sense of the data is an enormous challenge. In other words, we still don't really know what's causing a lot of these disparities, much less what to do about them.

Take the simple issue of how to classify people in order to study health disparities. Let's say you want to look at Hispanics. "They're a group that is linked only by being from countries that were under Spanish rule," says Thomas A. LaVeist, director of the Hopkins Center for Health Disparities Solutions. "To combine Cubans and Mexicans and everyone else into one category doesn't make a lot of sense. The populations are so different. You have a tremendous amount of variation [in health, lifestyle, and genetic heritage] that's being masked." Or take the fact that studies often put Africans living in the U.S. in the same category as African-Americans who were born and raised here. If you want to study, say, HIV rates, that catchall category becomes meaningless, says LaVeist, because the virus is so much more prevalent in Africa.

Then there's the thorny issue of causation. Almost all common health problems have two root causes: the "nature" ones (i.e., genetic factors), and the "nurture" ones, which come from one's environment. For minority health issues, both categories are complicated. Race is a notoriously inaccurate proxy for genetics, since it's such an imprecise way of describing people. Take the case of Bidil, the so-called black heart-disease drug. LaVeist posits a hypothetical question: would you give it to Barack Obama, whose mother was white? Looking at a patient's full genetic analysis would give you much more information than race does, but the era of personalized medicine is still years away.

The "nurture" category is even more complicated, because it encompasses both the social environment (how people live, their income, what they eat, how stressed-out they are) and the medical environment (whether doctors are treating them differently because of their race). It's not always possible to separate the two, says Thomas Sequist, an assistant professor of health-care policy at Harvard Medical School: "These issues aren't always so clean-cut as 'This is an issue of the social environment, and this is a problem with the delivery system.' "

Take diabetes, which is far more common in African-Americans and Hispanics than in whites. Both Sequist and LaVeist have looked at the issue extensively. Sequist's work has found that "the health system in general got much better for diabetes care" for minorities from 1999 to 2003, thanks to "cultural competency" training for doctors. In one study, physicians were treating patients of different races exactly the same way by the end of the training. But the minority patients' outcomes actually got a little worse over the same time period. Similarly, LaVeist has looked at "a community with black and white people living together, in the same conditions"—specifically, a neighborhood in southwest Baltimore—and found that "there was no race difference with diabetes."

Setting aside possible genetic factors, both these studies would seem to point to the social environment, since care was similar but the results were different. Put bluntly, the disparities appear not to be the fault of the doctors, but of the lifestyle the patients are living, often not by choice. But things aren't that simple. Other research has shown that indeed, the health-care system does play a role. Sequist says that in another set of studies, "a certain percentage of the disparities come from the fact that if you're a minority, you're more likely to be going to a community health center or a safety-net hospital," which have worse outcomes than large medical centers that serve patients in higher income brackets. Is this racial disparity the fault of the doctors at the "safety-net hospitals," then? Maybe they're treating their minority patients differently. Or maybe they just have less-sterling credentials than the doctors at major medical centers. Or maybe their poor outcomes stem from the fact that their patient population is unhealthier to begin with. The data are too tangled to say anything for sure.

All these complications make it extremely difficult to implement good policies around race and health. And yet, says Sequist, "about three or four years ago, there was a huge push to move into the phase of actually doing something about this." Cultural competency training has now become standard in many medical centers. And if the Democrats' current health-care legislation were to pass, it would implement "over three dozen provisions that offer promise" for addressing inequities, says Dennis P. Andrulis, director of the Center for Health Equality at Drexel University, who has assessed the House and Senate bills closely. (If reform doesn't pass, he adds, "I think the bills are a road map for what Congress might be looking to support in the future.")

But health-care reform probably won't be enough to change the fact that minorities are more likely to be in poor health. For that, we'll need even more sweeping social policies, says Brian Smedley, one of the authors of the Institute of Medicine report. "There's a growing recognition that we need to address environmental health hazards, that we need to improve the food options in neighborhoods and schools, to improve the availability of parks and recreation facilities in communities that are overrun with liquor stores and fast-food restaurants," he says. Those are all laudable goals, but they have to start outside the hospital.