The National Academies of Sciences, Engineering, and Medicine (NASEM), at the behest of the NIH, has impaneled an impressive committee of experts to evaluate current and future uses of population descriptors in genomic research. This is going to require a deeply challenging examination—a communal reckoning, really—of how terms like race, ethnicity, and ancestry are perceived by diverse stakeholders in the genomic research community and the public at large. It’s not only going to be a nuanced and challenging endeavor, but it’s going to be painful.
The legacy of distrust in racial and tribal communities arising from centuries of abusive treatment based on population differences is well known. But it’s not just the horrific and, by today’s standards, glaring historic atrocities that we need to confront. NIH and NASEM wisely recognize that, as a first step, we need to confront the far subtler and even more devious role that population descriptors continue to play in current health disparities and inequities among genetically different groups. Only then can we begin to assess if there are newer and better strategies for defining genetic variation that will advance a more ethical scientific framework.
When readers encounter words like “African”, “Asian”, or “European” in a journal article describing a genomic research project, how do they interpret the study findings? Does “Asian”, for example, include individuals with Sri Lankan or Indian heritage? Should journal editors require more descriptive explanation with geographic location terms when articles are submitted? Should IRBs likewise question the use of such terms in protocols? There is actually more genomic variation among the nations of the vast continent of Africa than between Africans and Eurasians, so in genome speak “African” is not a meaningful term.
Too often, genome researchers and program officials naively or casually endorse population descriptors that embolden racial or ethnic stereotypes. This practice needs to end, full stop. If it does not, those who labor in the critical field of gene discovery will bear responsibility for promoting, under the guise of science, beliefs that are neither biologically nor evolutionarily supported.
How might we develop a reproducibly reliable approach to describing genetic diversity in genomic studies—one that is culturally acceptable and can be implemented widely across a diverse range of research institutions? When clinical researchers identify a prospective participant as “Black” or “Non-Hispanic White”, or a research participant self-identifies as such, how does this labeling process influence downstream medical decisions? Could the labeling process actually harm a patient-participant?
Take the case of A1C (glucose level) values, as an instructive example. It turns out that average A1C values differ between racial and ethnic groups; people of color tend to have higher average A1C than White people. So, if A1C level is an exclusionary criterion in a clinical trial, it has the potential to discriminate against people of color and disfavor their recruitment and enrollment. Investigators and IRBs need to understand this in order to facilitate ethical protocol design, as well as ethical study review and approval.
On April 4 the NASEM committee held its first substantive public meeting, hosting several prominent speakers with expertise in evolutionary and population genomics, sociology and bioethics. Several critical themes and questions emerged, among them:
- Understanding and treating individuals with complex diseases that contain multiple genomic variants, such as cancer and cardiovascular disease, requires consideration of population differences in some manner. Yet, “population” is not a common term in medical research, and we lack consensus on how to employ the term for purposes of sampling and analysis in genetic studies.
- Can or should we attempt to develop unified social conceptions of what it means to identify as part of a particular population? Could such consensus be achieved through population communal dialogue or public deliberation?
- Improving methods of collecting race and ethnicity data, while an initial step, will not improve science. More dedication and better methods of collecting social context data—including racial and ethnic identification stress, access to preventive medical care, clean air, and healthy food—is what it will take to really improve the science.
NASEM will be issuing a public report at the end of its inquiry. I know that I’m not alone in my excitement to see what it will say.
Carol Weil, JD, is an independent consultant in human research protections and Bioethics and an expert in the ethical, legal, and social implications of cancer and genetic/genomic research. She is a member of the Multi-Regional Clinical Trials Center of Harvard’s Return of Individual Results project team as well as the team’s subgroup on genomics. She is also a member of North Star Ethics Review Board, a non-profit IRB. From 2010-2021, she served as a program director for ethical and regulatory affairs at the US National Cancer Institute (NCI) where she advised research teams and policy officials on consent and data sharing strategies, biobank governance, community engagement initiatives, and approaches for disclosing research results and incidental findings. From 1999-2010 she was a compliance oversight coordinator and policy analyst at the US Office for Human Research Protections.
Guest contributors are valued members of our community willing to share their insights. The views expressed in their posts do not necessarily reflect those of PRIM&R or its employees.