Race and Medicine . . . and Law

Suppose you sometimes get heartburn after eating Indian food. Is it the sight of people in saris, or the sound of sitar and tabla playing? Unlikely. Your problem is caused by some combination of your physiology, your early-life food experience, and a substance that turns up in some Indian dishes. It's kind of complicated. But you want to avoid heartburn, so you just avoid Indian restaurants.

You could use a different strategy -- avoid foods with coriander in them, or shun long-simmering sauces -- but going by nationality is easier, because that's how people organize their food choices. The Yellow Pages in the phone book don't have a section for stew restaurants or coriander kitchens, and you take the world as you find it.

What you're doing then is using Indianness as a proxy for the real causes of your discomfort. It's not a perfect fit -- some Indian food will be fine with you, and you'll probably get heartburn from non-Indian cooking from time to time. But the Indian-food rule is good enough, you figure, as a rough and easy rule of thumb.

This kind of thinking is the justification for using race and ethnicity in medicine. We don't really know the exact combination of genetic and life circumstances that makes someone more vulnerable to, say, heart attacks than the average human. But something needs to be done, and so racial categories are pressed into service as a stopgap. As Troy Duster of New York University explains in this recent article in Science, many scientists would agree with the geneticist David Goldstein: ``Race for prescription is only an interim solution to carry us through a period of ignorance until we find the underlying causes.''

This use of racial perceptions is becoming more common in medicine and the pharmaceutical industry. For example, in the U.S. the drugmaker NitroMed is likely to get approval of the heart-disease medication BiDil later this year, making it the first medication targetted at a particular racial class -- black people. (A study showed that BiDil, though not effective in the general population, did benefit patients who were designated African-American.)

Now, using any category means ignoring other possibilities -- once you decide to avoid Indian food, you reduce your chances of discovering the real answer to your troubles, because you won't be looking for an answer any more. That's what makes people uneasy about racial medicine, as Duster explains in his essay.

After all, doctors who use race to understand their patients' problems are emphasizing one form of difference among people, by ignoring many others. BiDil's line between black and white, for example, sets aside the line between lighter and darker skinned African-Americans, who also have different rates of heart failure. The race line also ignores a real distinction among age groups (over age 65, Duster found out, the racial differences in heart failure are vanishingly small). And the distinction between geographic populations -- racial disparity is not nearly so high in Brazil, Cuba or Trinidad as it is in North America.

Why invest in racial beliefs, then, when alternative categories are available and, clearly, meaningful? The answer seems to be that race is less bother. That's how the Yellow Pages are organized. Maybe social class or age or gender or nationality might be more relevant to a disease, but race and ethnicity offer busy doctors an acceptable ratio of ease-of-use to accuracy.

The cause of Tay-Sachs disease, for example, is a mutation in the HEXA gene, which can occur in any human being. It has nothing to do with being a Louisiana Cajun. But accident of history causes this mutation to occur more often among people who call themselves Cajuns. So thinking of it as a Cajun problem is a good enough rule of thumb for a busy doctor. Test Cajun couples for Tay-Sachs carriers. Avoid Indian restaurants. Get on with life. (The HEXA mutation also turns up at higher-than-usual rates among Ashkenazi Jews, which does not mean, of course, that Cajuns are Jews, or that Cajuns and Jews are one people.)

This kind of thinking makes ethnicity a ``poor man's clue,'' but, as Sally Satel, a practicing physician, puts it in this essay, ``in the sometimes cloudy world of medicine, a poor man's clue is all you've got.''

And what's wrong with that? Nothing, on paper. But the mind has its innate biases about human kinds, and they don't mix well with this logic. One of those biases is a strong preference for the idea that the human kinds we see are real, true and eternal. To put it crudely, the mind is built so that it is hard to think of race as a stopgap, temporary category. Quite the contrary -- the mind is built so that it's easy to think of race instead as an all-powerful fact, which can't be changed, and which explains why people do what they do. And the media don't lack for people who play to this deep-seated, unconscious proclivity of the mind. Satel, for example, thinks ``political correctness'' -- denying that people are different in any way -- is a worse problem for medicine than fuzzy racial categories.

Defenders of racial thinking like to talk about political correctness because it implies that the other side is all ideologues who won't face facts. But PC is a red herring. It's critics like Duster who are grounded in science. Believing that your local version of race is true and reliable will lead to bad science and bad medicine. Bad medicine as in the case of a boy not treated for sickle-cell anemia until almost too late -- because he was white (described by Richie Witzig in this paper.) Bad science because people can't answer a question that they haven't asked, and thinking in terms of race blots out thinking in other relevant categories.

I wish I could say this era of ethnic thinking was passing, but I think, instead, that it's going to be more of a problem in the future. Already, as Duster notes, the wave of work on ethnic differences in medicine has spilled over into forensics. Look, for instance, at this description of a means to sort criminal suspects by ethnic group in the U.K.

Maybe people will understand that this sort of thing is just a housekeeping tool, helping the cops decide how to allocate resources (``no need to interview any South Asian men on this case''). More likely, I think, will be claims that something in genetic labels makes ``those people'' more crooked than ``Us.''

Scary, don't you think?