By Jonathan Kahn

On June 23, 2005, the Food and Drug Administration approved a drug designed to treat heart failure in African-Americans exclusively. This race-specific drug, called BiDil, is not a new drug, but merely the combination into a single pill of two existing generic drugs that have been used to treat heart failure, regardless of race, for over a decade.

BiDil was brought to the FDA by NitroMed, a hitherto small Massachusetts biotech company with no other products on the market.  NitroMed explicitly requested race-specific FDA approval for its drug based on clinical data produced by its "African-American Heart Failure Trial" (A-HeFT) on the grounds that the trial population happened to be all African-American. 

BiDil does indeed appear to significantly help many people suffering from a heart condition. There is no scientific evidence, however, that race has anything to do with how BiDil works.  This is for the simple reason that A-HeFT enrolled only "self identified" African-Americans. With no comparison population, no legitimate claims can be sustained that BiDil works differently or better in African-Americans than in anyone else.

The FDA, however, accepted NitroMed's argument that because the trial population was African-American, the drug should be labeled as specifically for African-Americans.  This sends the troubling and unsubstantiated message that the subject population's race was somehow a relevant biological variable in assessing the safety and efficacy of BiDil. Ominously, it also gives the federal government's stamp of approval to the use of race as, in effect, a genetic or biological category.  By seeking and granting approval of BiDil as a drug solely to treat African-Americans, NitroMed and the FDA thus opened a Pandora's box of racial politics without fully appreciating the implications of what they were doing.

First, most drugs on the market today were tested almost exclusively in overwhelmingly white male populations. We do not call these "white" drugs, nor should we.  Rather, the operating assumption for approving these drugs was that the unmarked racial category of "white" was coextensive with the category "human being." A drug tested on white people was good enough for everybody.  In approving BiDil as a drug only for African-Americans, the FDA has implicitly adopted an assumption that drugs tested in black people are only good for black people.  This sends the unintended but nonetheless powerful message that black people are somehow less fully representative of humanity than are white people.

Second, given that the BiDil researchers admit their drug will work in non-African- Americans, the most plausible reason for conducting a race-specific clinical trial is that NitroMed holds the rights to a race-specific patent that will give them control over profits from BiDil until 2020. Of course, this hardly constitutes a sound scientific basis for designing a clinical trial, but it's a good economic one: if BiDil had been approved for treatment regardless of race, NitroMed's patent protection would have expired in a mere two years.1

Third, there is the problem of who "counts" as African-American. In an increasingly intermixed and complex society, one might ask just "how much" of an African-American one has to be to get the drug - 1/2, 1/4, 1/8? This starts sounding suspiciously like the blood quantum thinking of the Jim Crow era. Moreover, the label itself also refers to "black" people. Does this include dark skinned peoples from South Asia, or Australian Aborigines? The trials and label of BiDil are based on the concept of "self-identification." Self-identification, however, is a subjective social judgment.  It has nothing to do with the biological phenomena of drug metabolism and response. 

Fourth, race-specific labeling will make it more likely that non-African-Americans who might benefit from the drug will not get it. Health care providers simply may not think of prescribing it to non-African-Americans, and insurance carriers may not cover such "off-label" use.  Alternatively, what happens if someone who would typically be socially identified as white decides to self-identify as African-American in order to get insurance coverage for the drug?   Are insurance companies then going to become arbiters of racial identity?

Fifth, marketing a race-specific drug can lead to a misallocation of health care resources.  To the extent that we reduce the very real racial disparities that exist in health care to a function of genetic difference, we risk diverting political will and economic support away from addressing the pressing social, economic and political causes of racial inequality in our society.  Racialized medicine presents us with the superficially appealing and misguided message that instead of fixing injustice, we can simply fix molecules.

Finally, by approving BiDil only for African-Americans, the FDA gives the federal government's stamp of approval to the use of race as, in effect, a genetic category. Even the BiDil researchers admit that race is not a genetic category; moreover, there is no accepted biological definition of race.  Given our nation's troubled history of racial oppression, this is not something that should be taken lightly. 

At the outset, it is important to distinguish between the use of race in medical practice as opposed to racialized medicine.  It may be entirely appropriate, even necessary, to use race when tracking and addressing broad issues such as health disparities in American society. Understanding race as a social construct is entirely consistent with recognizing and addressing race-based inequalities in access to or quality of medical care in our society.  Such inequalities reflect the biological implications of the social and historical phenomenon of racial discrimination. 

Social understandings of race vary over time and across space.  In the past, the U.S. Census has included racial categories ranging from Mulatto to Hindu.  In the Jim Crow south, children of Armenian or Greek immigrants were sometimes mandated to go to schools designated for black children.  Today, someone of light brown complexion who is socially identified as "black" in the U.S. (say, for example, someone with a white mother from Kansas and a black father from Kenya) might be identified as "white" in Jamaica or Brazil.  In the U.S., the concept of "self-identification" has become the norm in assigning racial identities to individuals.  As a social practice for collecting census data this makes sense; but as a medical or scientific practice it is far more problematic. 

In medical practice what matters is our shifting understanding of the correlations between such evolving social identities and the evolving economic, political and environmental conditions to which they may be related.  For example, what are we to make of the fact that African-Americans suffer from disproportionately high rates of hypertension, but Africans in Nigeria have among the world's lowest rates of hypertension, far lower than the overwhelmingly white population of Germany?  Genetics certainly plays a role in hypertension, but any role it plays in explaining these differences must surely be vanishingly small. 

There may be occasions where race can be productively used even in genetic research, but in such cases it is very important to differentiate between using a racial group to characterize a gene versus using a gene to characterize a racial group. Thus, for example, a researcher trying to understand the genetics of diabetes may choose to study the Pima Indians in the Southwest United States because that group has a very high incidence of diabetes. This is an example of using a socially identified racial or ethnic group in order to try to characterize a gene (here for diabetes).  It is quite another thing, however, for a researcher who finds such a gene to use it to characterize the identity of Pima Indians as a group with "the gene for diabetes." The former use does not necessarily stigmatize or define a group in terms of genetics, the latter use does.

In situations such as those just mentioned, medical practitioners need not, indeed often should not, ignore race. The issue is not primarily one of whether to use racial categories in medical practice but how. Carefully taking account of race to help understand broader social or environmental factors that may be influencing health disparities can be warranted in certain situations, but it is always important to understand that race itself is not an inherent causal factor in such conditions.

In contrast, racialized medicine is premised on an implicit - and sometimes explicit - understanding of race as a genetic construct. Such an understanding is both scientifically flawed and politically dangerous. Since the inception of the Human Genome Project, much time and attention has been devoted to insuring that biological knowledge emerging from advances in genetic research is not used inappropriately to make socially constructed racial categories appear biologically given or "natural." Since Richard Lewontin's ground-breaking work on blood group polymorphisms in different groups and races in the 1970s, scientists have understood that race will statistically explain only a small portion of genetic variations. As a 2001 editorial in the journal Nature Genetics put it, "scientists have long been saying that at the genetic level there is more variation between two individuals in the same population than between populations and that there is no biological basis for 'race.'"2 More recently, an editorial in Nature Biotechnology asserted that, "Race is simply a poor proxy for the environmental and genetic causes of disease or drug response. . . . Pooling people in race silos is akin to zoologists grouping raccoons, tigers and okapis on the basis that they are all stripey."3

Politically, history teaches us that constructing races as genetically bounded and discrete categories is only one short step from constructing races as inferior and superior.   Racism feeds on biologically reductive constructions of racial difference. It is imperative to recognize the significance of race to understand and address the real and persistent health disparities that plague our country.  But these disparities are the result of social, economic and political histories of injustice. They demand social, economic and political responses. If we falsely reduce health disparities among socially defined racial groups to a function of genetic difference to be addressed through race-specific medicine, we risk diverting valuable resources and will away from developing policies and practices to confront the true causes of health disparities.

Unlike racialized medicine, which treats race as genetic, the use of race in medical practice has many legitimate and important places. Collecting broad-based epidemiological data is perhaps foremost among these. Only by using social categories of race is it possible to identify and track racial disparities in health, healthcare access and outcomes. Such information is needed to address on-going issues of racial justice in society. It may also be appropriate for individual health practitioners to take race into account under certain circumstances in trying to assess the needs of their patients. To the extent that health practitioners understand that race, as a social phenomenon, has biological consequences - such as where higher incidence of hypertension might in part be due to an array of environmental, social, or economic factors disproportionately associated with being a racial minority in the United States - it may be legitimate and important to take race into account in formulating appropriate medical interventions. 

The case of BiDil clearly raises concerns over the dangers of reifying race in a manner that could lead to new forms of discrimination. BiDil, however, is part of a much larger dynamic of reification in which the purported "reality of race" as genetic may be used to obscure the social reality of "racism." To the extent that this dynamic succeeds in reductively reconfiguring health and other types of disparity in terms of genetic difference, it casts personal responsibility and the market as the appropriate arenas for addressing differential outcomes.  It also undermines the rationale for deliberate state or institutional interventions to address discrimination.

This is not to advocate "color blind" medicine. To the contrary, there are very real health disparities in the country that correlate with race. African-Americans suffer a disproportionate burden of a number of diseases, including hypertension and diabetes. Like heart failure, these are complex conditions caused by an array of environmental, social, and economic as well as genetic factors. Central among these is the fact that African-Americans experience discrimination, both in society at large and in the health care system specifically. The question is: once society identifies these disparities in health outcomes, how does it address the underlying causes? Of course, outcomes can have multiple causes, both social and genetic. But health disparities are not caused by an absence of "black" drugs. As studies by the Institute of Medicine, among others, make clear, they are caused by social discrimination and economic inequality. The problem with marketing race-specific drugs is that it becomes easier to ignore the social realities and focus on the molecules.

For all the legitimate concerns that the genomics revolution might lead to new forms of discrimination, we must also be alert to the potential appropriation of genetics to obscure or justify existing inequalities.   


Jonathan Kahn holds a PhD in History from Cornell University and a JD from Boalt Hall School of Law.  He writes on issues in history, politics, and law and specializes in biotechnology's implications for our ideas of identity, rights, and citizenship.



1. Kahn, J. From Disparity to Difference: How Race Specific Medicines May Undermine Policies to Address Inequalities in Health Care, 15 Southern California Interdisciplinary Law Journal 105-129 (2005).

2. Editorial. Genes, Drugs and Race. 29 Nat. Gen. 239, 239 (2001).

3. Editorial. Illuminating BiDil. 23 Nat. Biotech. 903,903 (2005).

Search: GeneWatch
The use of forensic DNA databases by law enforcement around the globe is expanding at a rate that should be of great concern to civil libertarians.
View Project
Women and Biotechnology
View Project