Race is not a biologically determined classification. Race is a product of our social and political history. Therefore, concluded Dr. Harold Freeman, Chair of the President's Cancer Panel, "we need to look at race not as a biological indicator, but as an indicator of what happens to people socially." This assessment underscored a day of interesting and lively debate on issues of race, racism, and the use of racial classifications in cancer research and for all of science.

Making the day's deliberations somewhat unique were the diverse perspectives brought to bear on the issue of race in science--the presidentially-appointed, three member Panel, convened a group of nationally recognized experts in such disciplines as sociology, anthropology, philosophy, biology, genetics, and epidemiology to present testimony. Across all disciplines present, it was agreed that the biological concept of race is no longer tenable. Rather, race is a social construct which is a product of this Nation's social and political history. Supporting this conclusion, data were presented to the Panel showing that substantially more genetic variation is found within "races" than between them. Also, in response to changing concepts of what constitutes race, the revised UNESCO Statement on the Biological Aspects of Race, developed by the American Association of Physical Anthropologists, was shared publicly for the first time. It concluded, in essence, that the concept of a biological basis for racial classification is no longer acceptable.

If race has no biological basis, the broad question was raised to the Panel of how to characterize race so it can be applied validly in research studies designed to improve health care for different populations. Presenters vividly described the problems of classifying persons by race, particularly in a society moving towards a more multiracial identity that embraces one's entire ancestry and cultural environment. Should efforts be made to account for the amount of genetic heterogeneity in a particular population group under study? How should common cultural or behavioral traits be reported? Would geographic origin be as informative as population genetics in some cases. What are the definable populations about which scientific conclusions can be drawn. How should "race" be reported.

Currently, much research data looking at racial subsets relies only on self-reporting of race, with little information on ancestry and minimal information regarding socioeconomic status and other factors that shape individual response to disease. The current Office of Management and Budget Directive No. 15, Race and Ethnic Standards for Federal and Administrative Reporting, states that its classifications of race and ethnicity in Federal program reporting should not be interpreted as being scientific or anthropological in nature. They have been developed in response to needs for collecting standardized data to be used by Federal agencies in civil rights compliance reporting categories such as housing, budget, and other statistical reporting. They were not intended to be scientifically valid, but rather to respond to social and political needs expressed by the Executive Branch and Congress.

Rather than removing "race" as a consideration in science, more open discussion on this issue is needed in both public and scientific arenas. There are data showing clear disparities in health outcomes between different "racial" groups, with incomplete answers as to why these differences exist. Some presenters pointed to the link between race and socioeconomic status, noting that controlling for poverty appears to reduce or entirely remove the effect of "race" in some instances--more study in this area is clearly needed. Measures of socioeconomic status remain crude, focusing primarily on income, and ignoring other important economic variables, such as wealth and education that reflect social position. The lack of a uniform conceptual framework and approach to measuring economic variables greatly hampers health research.

The significance of having a racial identity can't be minimized, since we admittedly live in a society with a history of racism. For example, disparities in quality and access to health care can be attributed, in part, to racial discrimination. Evidence suggests that we incorporate biologically our social experiences. In short, our environment impacts our health. The influence of racism in this context could be significant. Little scientific research on the health consequences of racial discrimination has been done. It can not be done meaningfully until common study parameters that are truly representative of biological consequences are identified. As poignantly conveyed by one audience member, "No language currently exists to describe the experience of racism, or of growing up a minority in America...."

The Panel challenges the scientific community to review the social values that shape its scientific perspectives. Despite the fact that race is not a tenable biological classification, there are valid reasons to retain it as an indicator of health outcomes. Furthermore, science needs to reexamine its fundamental assumptions regarding race--its own biases and the social context which have shaped the intellectual process with regard to race in scientific investigation. At issue is the valid scientific use of a classification that is socially and politically determined --namely race!