Population genomics is making rapid and remarkable advances in detailing the specific genetic variations that characterize ­people in different parts of the world. This information could provide, among other things, invaluable clues to why some medicines are more effective for particular ethnic groups. The worry is that genomic differences between groups will be misunderstood, and perhaps misused, to justify crude generalizations about races.

BiDil, a new heart failure drug that will be marketed to African Americans, doesn't help. By most estimates, it is an important medicine for heart failure, a disease that has reached epidemic levels in the United States. A clinical trial completed last year found that the pill reduced mortality among African Americans with heart failure by an astounding 43 percent. That's great news for patients and cardiologists. And the drug has gained the backing of prominent medical groups such as the Association of Black Cardiologists. But if the U.S. Food and Drug Administration approves a race-specific pill, which it could do by mid-year, it will send a confusing message about what researchers are learning from population genomics—and raise troubling questions about how drug developers and physicians will use their growing knowledge of group differences.