The downside of the diversity drive

Race-based diversity or racist stereotypes? 

At many professional meetings and workshops, clinical trial investigators and coordinators crowd into presentations offering advice on recruiting and retaining women and minority subjects.

The NIH Revitalization Act of 1993 required federally funded studies to include women and members of racial and ethnic minorities among their participants, both to remedy their historic exclusion and determine whether there are differences in the way women and people from different cultures and racial backgrounds respond to new medications.

However, the recent emphasis on the inclusion of racial minorities in clinical trials as a means of addressing health disparities between whites and blacks is reinforcing erroneous beliefs about race and medicine, warns Otis W. Brawley, MD, professor of medicine, hematology, and oncology at the Emory University School of Medicine and former director of the Office of Special Populations Research at the National Cancer Institute.

"Race is not about genetics; it is a sociopolitical construct that was created 500 years ago to justify slavery," he notes. "It is a terrible, terrible way to categorize the population."

Clinical investigators should strive for a diverse subject population in order to better observe how drugs behave in different people and to ensure that people of all walks of life enjoy the same access to new treatments and the enhanced access to health care practitioners and disease monitoring that accompany them, Brawley says.

However, some investigators have taken the push for diversity a step further — advocating for proportionality of different genders and ethnicities in study populations in order to perform subset analyses on these different groups to discern any differences in response.

This practice can only serve to reinforce the erroneous belief that significant differences in biology exist in members of different races, Brawley says.

Some genetic differences do exist between people of different geographic origins, he says. However, area of geographic origin often has little to do with what racial category the person is assigned to in our society, he adds.

Sickle cell an example 

"The best example of this actually is the sickle cell gene," Brawley explains. "We in the United States, primarily because of our racist and racial attitudes and the way we have been brought up, we tend to think of sickle cell anemia as a black disease. You can even find literature that calls the specific mutation of the sickle cell a black mutation."

But the same genetic mutation can be found in natives of southern Greece, southern Italy, and southern Spain, where there is a low but significant incidence of the mutation in people normally considered white," he says.

In sub-Saharan Africa there are a lot of people who have the mutation, but in southern Africa — for instance, in Zimbabwe or South Africa — people who have the mutation are either migrants from the northern part of the continent or the children of migrants, he explains.

"So I have defined a white population with this black mutation, and a black population that never had this mutation," Brawley says. "The mutation actually came about because 1000 B.C., there was a huge malaria epidemic that encompassed the Mediterranean, southern Europe, went into the Middle East, and then went down into sub-Saharan Africa. So area of geographic origin is actually a much better identifier of this trait, called sickle cell disease, than is race."

Categories created by government office 

Our current racial categories — black, white, Asian, Pacific Islander, Native American, etc. — are re-created every 10 years by the federal Office of Management and Budget (OMB) to prepare for the U.S. census.

The new categories are issued two years before the census in the preamble to the OMB’s Directive 15, establishing how it will conduct the count of the population. The preamble always stipulates that the office is making sociopolitical categorizations.

"As an example of how it is a sociopolitical categorization is if you have a gentleman who was born in Bombay, India, and then migrated to the United States in the 1920s, that individual, according to OMB Directive 15, which has been reissued seven times over that individual’s life, that person would currently be considered to be Asian; but 20 years ago, that person would have been called Indian; prior to that, the person would have been called white," Brawley says. "These things are very political, and they have nothing to do with genetics or how we metabolize drugs."

Difference in access to care 

Disparities in access — to preventive health screenings, early diagnostic tests, regular health checkups, and early treatment — is the real culprit behind the disparities in illnesses between blacks and whites in this country, Brawley says. This fact is obscured by the furor over clinical trial subset analyses.

"It has become much easier for members of Congress to stress that they are forcing the NIH to look at these difference among the races in how drugs are metabolized and to define health disparities in genetic terms and throw $40 million at the problem through the NIH; it is easier to do that than it is to look at the real problems — disparities in care. I often like to point out that if we ever did find that one of my cancer drugs was metabolized differently in blacks than in whites, we already have data to show that blacks aren’t going to get the new drug, anyway, so it doesn’t really matter."

Disparities in access to care, and the resulting delays in diagnosis and treatment, also may partly explain the low rates of minority participation in clinical trials.

The June issue of the Journal of Clinical Oncology featured a report by Simon and colleagues1 of a study they conducted of the dynamics of clinical cancer trial accrual at a single institution.

According to that report, Brawley says, the proportion of eligible patients who refused entry was significant, as was the proportion of patients who did not have a clinical trial available or who otherwise did not qualify for the studies.

The study also found that minorities with cancer were less likely to have a clinical trial available for their disease and stage of disease and were less likely to be eligible when a trial was available.

The authors did find that similar proportions of eligible black and white patients chose to participate when trials were available.

It is particularly noteworthy, Brawley indicates in an editorial accompanying the report’s publication, that black women in the Simon study were more likely to have poor performance status and inadequate organ function.

More effort needs to be made to first ensure that black and white Americans have access to the same level of care and the same treatment before researchers go looking for differences in treatment response that they attribute to race, he says.

"A lot of people have taken what I have said to mean that Otis says blacks don’t do worse with cancer," Brawley points out. "And that is not what I am saying. I am saying I don’t think they do worse because of biological reasons, but because of sociological reasons."

To illustrate his point, he recounts the results of a research study at Cleveland-based Case Western Reserve University of white women with breast cancer.

"Now this type of study can never be done again — at least not with NIH funding — but it yielded some interesting information," he says. "The study found that white women who were poor have breast cancer at a lower age, and they have breast cancer that is more aggressive within stage — that there is something about poverty in white women that makes the breast cancer more aggressive. This is by comparing white women who were poor vs. white women who were not poor who had breast cancer. But now, when we look at black women vs. white women, we seem to be saying, Let’s see, black women have disease at a younger age; they have more aggressive disease,’ and then we immediately start saying, Maybe it is because they are black?’ Ironically, maybe it is because they are black socially, not because they are black, biologically."


1. Brawley O. The study of accrual to clinical trials: Can we learn from studying who enters our studies? J Clin Oncol 2004; 22:1-2.