National Academies calls for transforming use of racial and ethnic labels in genetics research

Saying genetics researchers inconsistently and inappropriately use racial and ethnic labels that fail to capture the complex patterns of human genetic variation, the National Academies of Sciences, Engineering, and Medicine issued a report Tuesday calling for a transformation in how such descriptors are used.

Outdated methods of grouping people may result in poor scientific results and misguided interpretations, said the report. “It is time for us to reshape how genetics studies are conceptualized, conducted, and interpreted,” the authors wrote, noting that genomic research is growing exponentially due to technological advances such as cheaper and faster sequencing.

“Genetic data are being taken up by an ever-growing slate of researchers. It’s not just a narrow concern for the study of the human genome,” Ann Morning, a professor of sociology at New York University who served on the report committee, told STAT.

Many researchers use racial and ethnic categories that arise from a federal Office of Management and Budget rule called Statistical Policy Directive 15, which was developed to standardize record-keeping by federal agencies and in the decennial census (and is currently being updated to be more flexible and inclusive.) While categories such as Black, Asian, or Hispanic may be needed for record-keeping, the report said they are not appropriate to study genetic variation because such groups are not homogenous and genetic variation within them shifts over time as people migrate and intermix.

The use of such labels, the report said, supports “a pervasive misconception that humans can be grouped into discrete, innate biological categories,” and called them “a poor fit for capturing biological diversity.” The report reaffirmed that race is a social construct, not a biological one, and strongly cautioned researchers against “typological thinking,” a view long pervasive in biology that classifies people, organisms, or species in terms of the groups to which they belong and often overlooks important variation between individuals.

The authors did not provide a new list of population descriptors to be used. Instead, they issued recommendations and best practices for researchers undertaking new genetic studies, from hunting for single genes that underlie rare Mendelian genetic disorders to conducting large genome-wide association studies to identify the many genes that increase risk for more common diseases. Researchers, they advised, should develop descriptors by tailoring them to the specific type and purpose of the study they are conducting; working in partnership with community groups to create and inform them; and being transparent about how descriptors were developed and why they were chosen.

“It’s not just an issue of semantics or what’s the right terminology to use,” Morning said. “It’s about conceptual frameworks. We need researchers to think through what they mean when they use race or ancestry.” Such work should not be “a reflex or an afterthought,” the report stated.

Morning said such improvements were necessary and would result in higher-quality science. “To do the best genomic science we’re capable of, these descriptors have to be brought up to speed with the same care and standards as the rest of the genetic enterprise,” she said.

The report was met with skepticism by Keolu Fox, a genome scientist and assistant professor at the University of California, San Diego, and co-founder of the Native BioData Consortium. Fox said far more, and deeper, work needs to be done to combat genetic racism and help the communities that had been harmed by it. “We can create as many labels as we want, but at the end of the day, who’s doing the labeling?” asked Fox, who is Native Hawaiian but does not use that term to describe himself, preferring Kānaka Maoli.

“The fact that we’re having these conversations in English — you have to see that’s a colonial way of seeing the world. These are not the names we give ourselves,” he said. Researchers, he added, should defer to communities, involving them and conducting science that would truly address disparities they face.

The report said racial descriptors should be avoided in almost all cases, except some health disparities studies that may need such categories to study harms occurring to certain populations. They also said researchers should directly evaluate environmental factors, such as exposure to toxins or air pollution, that may be relevant to their studies instead of relying on population descriptors as proxies. These changes may require genetics researchers to expand their teams to include people in other fields such as social science or environmental health, Morning said.

The authors expect resistance from some genomics researchers, who may worry that being more specific in labeling groups may decrease the statistical power of studies that require large numbers, that it will be difficult to “harmonize” or fit new studies in with older studies that used different groupings, and that incorporating environmental effects and epigenetics into studies will be challenging.

Brendan Lee, a physician and medical geneticist from the Baylor College of Medicine who serves as president of the American Society of Human Genetics (ASHG), applauded the report and said it was in line with priorities of his society to diversify both the scientists conducting research in his field and those being studied. He said the guidelines provided a framework to improve the quality of human genetic research.

He agreed that some of the goals, such as incorporating environmental factors in genetic studies, remained a challenge for the field and said that such ideas would engender much discussion among genetics researchers. “That’s a positive, not a negative,” he said, adding that it would take time for scientists to incorporate some of the recommendations in their future work. “This is the very first step,” he said. “It’s not something that can be done in a day.”

The report authors said it would be difficult for change to occur without many other players stepping in to encourage or enforce the changes. These include funding agencies, such as the National Institutes of Health, which commissioned the report; professional societies such as the ASHG, which recently apologized for its role in perpetuating scientific racism and promoting eugenics; journals, which can use the peer-review process to ensure study authors adhere to the report recommendations; and universities, which can support researchers with training, education, and resources. A number of programs at Northwestern University, UCLA, Duke, and the University of Wisconsin-Madison already are providing such services, the report said.

An NIH spokesperson said agency leaders would evaluate how to implement the report’s recommendations over the coming weeks and that it “will inform NIH’s ongoing efforts to safeguard scientific integrity in genomics and promote the responsible design of research studies so that all populations benefit from scientific advances.”

The history of how people are labeled or grouped is a fraught one, and one the report said is rooted in a long history of white supremacy and colonialism that divided humans into different racial groups, with some being seen as superior. Such thinking, the report said, persisted not only into the early 21st century with the American eugenics movement but continues today with much professional scientific training.

“Racial taxonomy becomes a familiar way of seeing and describing the world, one that is taken for granted and presumed to be ‘natural’ and objective,” the report said. “This framework has made its way unnoticed into the design and execution of scientific research.”

The report acknowledges that the many calls to improve how race, ethnicity, and ancestry are used in genetic research — some decades old — have largely gone ignored, and said many would be right to be skeptical that change will occur now. But the authors said the current atmosphere, where issues of race and racism in science are much more publicly acknowledged than they were in the past, offers hope that change can finally occur.