WASHINGTON — Biomedical researchers should change the way they use race and ethnicity in their research, says a new National Academies of Sciences, Engineering, and Medicine report, calling on researchers to scrutinize whether the use of race and ethnicity is appropriate at each stage of their work — and explain the scientific reasoning behind their decision in any publications.  

Biomedical research spans human health and disease — from laboratory studies of animal tissue that improve our understanding of human biology, to clinical trials for new medical treatments. Race and ethnicity are used widely in biomedicine. However, the report says, racial and ethnic categories are often used inappropriately in biomedical research as proxies for biology — or as poor substitutes for factors such as genetics or environmental exposures — despite there being no genetic or biological basis for race. In some discrete cases, their use can be appropriate, for example in identifying health disparities.

Biomedical research has long emphasized exploring race at the expense of other concepts — such as racism, discrimination, and social determinants of health — which can affect health. Assumptions about race can be traced back centuries in the history of science and medicine. In more recent years, federal biomedical research grants require the use of a minimum set of racial and ethnic categories set by the White House Office of Management and Budget, in order to track enrollment of study participants and measure inclusion. But using these categories to track inclusion in biomedical research has contributed to misunderstanding and the inappropriate use of race and ethnicity data in scientific analysis.

The report calls for comprehensive evaluation from biomedical researchers to assess trade-offs based on the populations and health outcomes involved in their research, which has the potential to improve the scientific rigor of biomedical research, mitigate bias that continues to affect research and health care, and build lasting trust between the scientific community and different racial and ethnic communities. The report outlines a decision-making process that researchers can use to determine whether and how to use race in medicine, and move toward thoughtful use of race and ethnicity in research and clinical applications in the future.

Clinical calculators or guidelines that change an output based on a patient’s race or ethnicity — such as algorithms used to assess organ disease or to determine whether a patient faces a “high risk” of a specific outcome — have received criticism for contributing to health disparities and reinforcing the misconception that there are innate biological differences among racial and ethnic groups. The report says that simply removing race and ethnicity from all clinical tools, algorithms, and guidelines is complicated, however, and would not inherently resolve these issues. It recommends instead that researchers operate with transparency at each stage of the development of clinical tools, and report performance across racial and ethnic groups. Moreover, the report adds, much of the existing biomedical evidence base on which these tools are based has deep-rooted bias, which requires examination by the field and investment from biomedical research funders and sponsors to correct.

“Racial identity is complex, dynamic, and personal, and it is influenced by our perceptions of ourselves, and the perceptions others have of us. Science, on the other hand, tends toward reductionism and categorical thinking,” said M. Roy Wilson, chair of the committee that wrote the report and president emeritus of Wayne State University. “In order to better serve society, biomedical research methods must close this gap to better represent what race means in people’s lives — and produce accurate and useful scientific results.”

Changing Research Practices

The report says researchers should:

• Evaluate whether the use of race and ethnicity is appropriate or inappropriate. For example, researchers should consider the historical and social context of their research question, and consider the benefits and limitations or harms of using race and ethnicity. They should refrain from making unsupported inferences, such as relying on race and ethnicity as causes of biomedical outcomes in individuals.

• Operate with transparency. At every stage in the development of technology that may influence health — for example, clinical algorithms, AI models, or medical devices — researchers should be transparent and assess and report the performance of biotechnologies across a range of racial and ethnic groups.

• Base all inclusions or exclusions of racial and ethnic categories on scientific rationale, motivated by the research question. For example, researchers should articulate the purpose of aggregating racial and ethnic categories, or of omitting cases. The report also recommends that researchers avoid some practices, such as combining categories solely to improve statistical power.

• Define race and ethnicity in their work. Researchers should explain their approach to using race and ethnicity, including any limitations. For example, they should define which racial and ethnic categories were used, when data were collected, and how they were collected.

• Identify concepts often conflated with race or ethnicity that are relevant to their study. Race and ethnicity are often used inappropriately as placeholders for harder-to-measure factors in research, such as economic or social factors. Researchers should try to include multiple measures of these factors where possible, and not solely rely on data that asks people to self-identify their race based on the OMB race and ethnicity categories.

“This report is a call to action for biomedical research to rethink how it uses race and ethnicity,” said Victor J. Dzau, president of the National Academy of Medicine. “A system that approaches race and ethnicity thoughtfully and is based in evidence is within reach — now is the time to take the steps needed to achieve that goal.”

Measuring Multiracial Identities

The number of people who identify as multiracial in the U.S. is increasing, but there is no standard way to account for multiracial or multiethnic people in biomedical research, the report says. Even if they are included in research, they are often left out of the resulting analysis because of small sample sizes or because they are labeled as “other” and aggregated with other identities. Researchers should consider the inclusion and analysis of multiracial and multiethnic participants at each stage of the research process, especially in study design and when developing research questions. Respondents to surveys should be able to select multiple races and ethnicities, for example, and researchers should report granular data for multiracial or multiethnic respondents to the greatest extent possible.

Implementing Change

Adopting the report’s recommendations will require coordinated efforts and investment across the biomedical research ecosystem, as well as time and effort to retrain the workforce and adopt new ways of thinking. Research funders, sponsors, publishers, and journal editors should provide consistent guidelines to assist researchers in developing their work and promote the thoughtful use of race and ethnicity.

Partnering with Communities

Biomedical research that involves human participants and has social and clinical implications requires a high degree of community engagement, says the report. Researchers who use race and ethnicity data should partner with the communities relevant to their work and with community members to the greatest extent possible. When appropriate, research funders and sponsors should require a community engagement plan as part of funding applications to support partnerships between communities and research teams.

The study — undertaken by the Committee on the Use of Race and Ethnicity in Biomedical Research — was sponsored by the Doris Duke Foundation and Burroughs Wellcome Fund. The National Academies of Sciences, Engineering, and Medicine are private, nonprofit institutions that provide independent, objective analysis and advice to the nation to solve complex problems and inform public policy decisions related to science, engineering, and medicine. They operate under an 1863 congressional charter to the National Academy of Sciences, signed by President Lincoln.

Contact:

Megan Lowry, Media Relations Manager
Office of News and Public Information
202-334-2138; email news@nas.edu