How to fix racial disparities in medical care.

How to fix health policy.
Aug. 14 2009 10:10 AM

Color Bind

How to fix racial disparities in medical care.

Click image to expand.
Black patients often don't get as good of care as whites—but they could

Black Americans live shorter lives and have poorer health outcomes when compared with whites. Health researchers, depending on their political persuasion, explain this disparity in one of two ways, neither of which is very constructive.

Liberals generally toe the line of a detailed 2002 report from the Institute of Medicine that specifically blames doctors' "bias, prejudice, and stereotyping" for inferior medical care. A typical study from this line of thinking is an infamous 1999 Georgetown University computer-based test of chest pain management in which the only variables that changed between several scenarios were the color and gender of the patient. Doctors given the simulation were less aggressive in treating black women (though journal editors later admitted the authors "overstated" the difference).

Conservatives explain the problem differently. They believe modern disparities arise not from a patient's color but from their economic and geographic background: Minorities are sicker because they're poor and live in places without access to medical care. As Sally Satel and Jonathan Klick of the American Enterprise Institute write, treatment gaps "vary by race but not because of it."

This bickering over whether the medical system is racist takes place in many top-tier medical journals, and the conflicting data lead to some odd debates. A Dartmouth group found that black men who live in the same place and have the same insurance as whites get knee arthroscopy less frequently, which implies the black men are treated worse. The same trend exists with cardiac catheterizations.

Weirdly, black doctors were just as likely as white doctors to withhold the procedures from black patients, which means that doctors of color might be racists, too. Even weirder: The catheterization study found that death rates were actually lower among blacks, leading to another largely academic debate (subscription required) over whether blacks are genetically superior to whites (since they live longer despite worse care) or whether whites' medical care hurts more than it helps.

More to the point, neither diagnosis for disparities—blaming a doctor's racism or a patient's socioeconomic status and biology—suggests a clear template for change. So hospitals nationwide trot out "sensitivity training" for health professionals, medical schools encourage more minority applicants with some hope they'll take better care of their own, pharma companies market drugs tailored to the supposedly different biology of blacks, authorities recommend special vaccine programs for minority children, editorialists assert that medical disparities will improve only with massive changes in "unemployment, housing, transportation," and so on.

There's a more helpful way to think about racial disparities that ultimately has very little to do with either race or socioeconomic status. Beginning in 1997, health plans accepting Medicare were required to submit data about all patients with certain problems, such as heart disease, diabetes, and mental illness. The federal government used these records to create the Health Plan Employer Data and Information Set, which follows 1.8 million Americans and represents one of the largest medical databases ever made. (The data were scrubbed clean of identifying details.) This kind of information has the potential to revolutionize how we handle racial disparities in medical care.

Consider the following findings from the initial years of HEDIS: Seventy-four percent of whites and 69 percent of blacks got appropriate mammograms, 76 percent of white and 64 percent of blacks got the right blood pressure drugs, and 64 percent of whites and 55 percent of blacks with diabetes got needed eye exams. Yes, there are some disparities between the races, but they are dwarfed by a much bigger problem: Overall, one-quarter to one-third of all Medicare-insured Americans, regardless of color, aren't getting basic care.

Now consider this: Within five years of simply requiring health plans to submit data—and forcing no other intervention—HEDIS reported that 76 percent of whites and 75 percent of blacks got appropriate mammograms, 94 percent of whites and 93 percent of blacks got the right blood pressure drugs, and 72 percent of whites and 70 percent of blacks with diabetes got needed eye exams. Simply reporting of quality measures publicly made the disparities evaporate without any grand plans to deal with race. When doctors and hospitals realized they were being watched, treatment became much better for everybody.

There's more evidence that care can be improved for everybody without explicitly targeting minorities or the poor. In 1995, the Department of Veterans Affairs undertook an extraordinary re-engineering effort that included reporting the HEDIS measures as well as widespread targeted efforts to improve care for all veterans. Incredibly, despite a 40 percent increase in patient volume and no substantial increase in the VA health system's budget over the next five years, researchers found that more than 90 percent of veterans suddenly got proper care on most HEDIS measures—despite having a "low level of education, disability,or a low income." Again, the effort made no mention of racial or socioeconomic disparities in care, but it largely crushed them.

Over and over, this theme recurs: Universal quality-improvement plans coupled with publicly reported measures are the best way to cut health disparities. We need not dwell on race or class, which only causes disagreement and alienates political support. Universal screening of group B streptococcal infections during pregnancy eliminated racial disparities in newborn infections, racial disparities of deadly pneumococcal infections in children disappeared over the past decade after universal Prevnar vaccination began, and a Medicare-backed reporting plan increased proper kidney dialysis from only 46 percent among whites and 36 percent among blacks to roughly 85 percent among all comers. In fact, these kinds of race- and class-blind interventions are arguably the only ones proven to reduce disparities on a meaningful scale.

Of course, compassionate and individualized care is important. But to some extent, these programs undercut the notion that minorities just need special kinds of communication or culturally sensitive care to bridge the treatment gap. Public reporting and universal standards succeed by doing just the opposite: by making doctors treat minorities just like everybody else.

Darshak Sanghavi, a pediatric cardiologist, is a fellow of the Brookings Institution and Slate’s health care columnist. Follow him on Twitter.

  Slate Plus
Slate Picks
Dec. 19 2014 4:15 PM What Happened at Slate This Week? Staff writer Lily Hay Newman shares what stories intrigued her at the magazine this week.