In the United States, where data on ‘race’ are routinely collected, researchers continue to find consistent, persistent, and usually marked disparity in nearly every indicator of wealth and health by race (Smedley et al., 2003). While there are some counterfactual findings and race-based health disparities that may be less glaring than in other countries, the overall pattern is clear. On nearly every indicator of health, whites in the United States do best and, conversely, African Americans do worst (Sacher et al., 2005). Why is it that such differences in health and disease persist among races? For well over a century, occasional debates – or more accurately, parallel arguments – have continued in scientific literatures and among various publics. Two arguments predominate: In one argument, racial differences in health are traced back to the evolutionary development of racial differences in genetics. This ‘raciogenetics’ perspective accepts that race is a viable substitute or shorthand for genetic variation between populations, and such genetic differences are assumed to be salient causes of disease as well as other racial differences. In the more recent counterargument, race-based health disparities are causally traced to variation in the ‘lived experience’ of those assigned to different racial categories. Here, lived experience refers to the totality of everyday conditions that are embedded into the fabric of social, personal, and institutional relationships. Some of these experiences are rather subtle personal interactions that communicate values based on phenotypic appearances, and others still are more deeply and profoundly personal experiences of racism. I argue that the raciogenetic explanation is both epistemologically flawed and counter to scientific facts because: (1) genetics accounts for only a small fraction of the variation in the complex diseases of interest, and, most importantly, (2) human genetic variation is extremely large within racial groups and does not easily map onto race as it is typically defined. Furthermore, in an age characterized by the celebration of genetic medicine, a compromise position in which both causal mechanisms are seen as valid is likely to have the consequence of overstating the role of genetics (Goodman et al., 2003; Sankar et al., 2004). In the following, I first outline the raciogenetic argument and then the lived experience/racism argument. I then present reasons why equating race with human genetic variation is epistemologically and scientifically flawed. I end with a critique of the compromise position and a statement of the need to clearly determine what we mean by race when it is used in public health research and practice.

The Raciogenetic Perspective

The perspective or worldview that racial differences in health are due to natural or inborn factors is an old one that extends back to at least the early nineteenth century. In the earliest (pre-Darwinian) versions of this worldview, by virtue of either assumed separate creation or separate evolutionary histories, whites were believed to be endowed with natural abilities to resist those diseases that were characteristic of ‘civilization.’ Frederick Hoffman (1896), for example, published a wealth of data on race differences in health in the United States. His influential treatise suggested that the increased morbidly and mortality of African Americans, especially in the northern U.S. cities, resulted from their collective, inborn inabilities to survive the rigors of the contemporary world. Hoffman predicted their eventual demise. At the time Hoffman was writing, many physicians felt that races were differentially susceptible to disease, and therefore, particular races were more likely to suffer from particular diseases. So, for example, once sickle cell disease was identified in African Americans, it was assumed to be a race-specific disease (Wailoo, 1997; Tapper, 1999). When Europeans began to present with symptoms of sickle cell anemia, it was assumed by physicians in the 1920s and 1930s that they must be part ‘negro.’ The possibility that sickle cell had nothing to do with race, but much to do with evolution and genetics, was not considered until the middle of the twentieth century. In the twenty-first century, the idea that germs obey the color line or that diseases are specific to one race or the other has gone out of vogue. But the remnants of geno-racial determinism persist in current ideas that races are quasiscientific units with separate disease susceptibilities. In its most common formulation, such ideas define race as a quantifiable risk factor for disease. Indeed, raciogenetics persists as a dominant explanation for variation not just in conditions such as sickle cell anemia that are caused by variation in a single gene, but also for variation in complex metabolic conditions such as diabetes and heart disease (Goodman 1997; Sankar et al., 2004). In 2005, for example, the U.S. Food and Drug Administration approved the medication BiDil for use in African-Americans, presumably because it was effective in combating congestive heart failure in this group for reasons that were thought to be, for genetic reasons, intrinsic to that group (Temple and Stockbridge, 2007). Finally, it is clear that we live in an age of genetics, a time in which genetics has taken hold as the dominant explanation for most behaviors and conditions. Certainly, genetic medicine is big business. And so it follows that genetics might explain variation in health among races (Goodman et al., 2003).