Should Medicine Be Colorblind?
The significance of race in diagnosis and medical research


Return to Contents

Anatomy of an Apology
Reflections on the 1997 presidential apology for the syphilis study at Tuskegee
(September 1999 AE)



Race has no biological reality. Biologists and anthropologists tell us that none of the ways we have carved humankind into subgroups corresponds to a stable set of biological features. Skin tone, hair texture, and the shape of the face or eyes vary among humans, but these features do not cluster in any way that would permit us to identify race biologically. If race is not biologically real, should medicine be colorblind?

Insofar as medicine is a biological science, one might think so. In P.C., M.D.: How Political Correctness is Corrupting Medicine (Basic Books, 2000), Sally Satel argues that the attention public health professionals give to race is a pernicious infiltration of political values into scientific research. Political correctness, she contends, is undermining methodological standards in the health sciences, corrupting public health initiatives, and ultimately eroding the health of individuals. In a recent essay in the Atlantic, she calls for the federal government to “cease funding into the effects of ‘powerlessness,’ ‘classism,’ and ‘racism’ on health.” Satel’s simple dichotomy, however, between value-free scientific medicine and political correctness was made problematic in the interdisciplinary Faculty Seminar on Race and Medicine, hosted by the Program in Science and Society in the spring of 2001. As a philosopher of science, I found these discussions a bountiful resource for reflection on these issues.

One of the first puzzles is that, while race has no biological reality, there are persistent correlations between race and disease. Satel and others bristle at the notion that race could cause disease. Particular pathogens cause diseases, not social constructs. To believe that race is relevant to the study of disease, they argue, is to confuse causes with mere correlations.

This conclusion is, however, an oversimplification. The correlation between race and disease can be medically significant without thinking that there is some kind of causal connection between an entity called “race” and being sick. Race is a social construct in the sense that we treat people as if similarity in skin tone or eye shape were symptomatic of some deeper similarity. What matters for medicine is not the physical similarity, but the similarity in treatment.

Racial classifications have far-reaching effects on opportunities for jobs and housing, not to mention access to health care. Differential treatment on the basis of race can mean that members of the group are exposed to similar risk factors. Suppose that housing discrimination forces disproportionate numbers of black Americans to live near industrial sites. Blacks will then be differentially exposed to health risks, and there will be a correlation between race and disease. Each individual is indeed made sick by a particular pathogen—and there could be many different exposures involved—but the similar location means that blacks are more likely to be ill than whites. Hence, there can be informative causal correlations between race and disease, even if race is not a biological entity with causal powers.

Even admitting this conclusion, one might argue, as Satel does, that attention to race, class, and other social variables has pernicious effects. Identifying race as a risk factor in disease has the effect of blaming a person’s race for his or her illness. The person is not encouraged to take responsibility for her health. Emphasizing personal responsibility is essential to good health, since only individuals have control over their health. Emphasis on race (and other social variables) serves only to distract from personal responsibility. Satel concludes that “PC” science leads to bad medicine.

The irony of this argument is that it injects a political value into medicine and public health. Satel complains that studying the correlation between race and disease is motivated by a political desire for social change. Her pc-free epidemiology would concentrate on factors within the control of the individual. This is a quest for health solutions that maximize individual autonomy, a motivation that is no less political for being unselfconscious.

Satel’s substitution of one political value for another nicely illustrates a point made by many philosophers of science. Scientific inquiry is an attempt to answer questions concerning the world around us. The questions we ask and the kind of answers we expect are not forced upon us by nature itself. Rather, our study of nature reflects extra-scientific motivations. She is insisting that public health questions be answered by appeal to factors that are within the control
of individuals. Her opponents are answering questions about disease by appeal to social factors that are not within the control of individuals. This choice of scientific questions is fundamentally a political one: should heath care interventions emphasize individual responsibility or social action?

The issue is political, but as Satel recognizes, it is not entirely political. Different emphases in public health can have measurably different heath outcomes. Empirical evidence thus bears on the question of what sort of questions we ought to pursue in public health. Satel insists that public health purged of the politics of race is better medicine. There is an important reason why this is not true: if the concept of race were eliminated from medicine, physician diagnoses would be less reliable. High prevalence of a disease in a population increases the likelihood that a member of that population has the disease. A physician must take account of this baseline probability when she makes a diagnosis. It is therefore important for a physician to know population-level generalizations about disease. Rates of lung cancer are much higher among coal miners than among university professors, so it is important to know the patient’s occupation. Since race correlates with disease, knowledge of the patient’s race is a crucial piece of information for the physician. Race is therefore an important diagnostic variable, and failure to study correlations between race and disease would lead to poor diagnosis.

The attempt to free medicine from social policy succeeds only in interjecting different values into scientific inquiry. This should be no surprise. Values and interests guide our scientific questions, so a strictly value-free inquiry is impossible. Rather than pretending that science could be purged of values, we need to deliberate on the values that undergird particular scientific inquiries. We need to reflect not only on the values themselves, but on the methodological consequences of their adoption and the practical results of the science so conducted. Race is a particularly important topic because of the continuing ramifications of slavery, discrimination, and colonization. As long as we feel these consequences in our lives, the concept of race will remain important for medicine.