Conflating Race and Ancestry – Interview with Lead Author Michael Bentz
The Division of Ethics interviewed Michael Bentz, lead author on "Conflating race and ancestry: Tracing decision points about population descriptors over the precision medicine research life course", recently published in Human Genetics and Genomics Advances. We asked Michael about the development of the article, its findings and his personal experience of publishing. Michael Bentz earned his MPH from
UC Berkeley in 2019, before joining the Ethics of Inclusion (EOI) study with the Division of Ethics. Michael now works at Westat as a Research Associate.
1. To start, can you describe the study- the Ethics of Inclusion: Diversity in Precision Medicine Research?
The Ethics of Inclusion (EOI) study investigated how diversity is conceptualized and operationalized across the precision medicine research (PMR) life course. We partnered with five study sites, which were PMR projects (at different stages of research) that sought to recruit a diverse sample of participants. We used qualitative methods, which included participant observation of scientific working group calls and conferences, as well as interviews with key investigators and research staff. This paper came out of a focused analysis of our data on how PMR researchers employed race and/or ethnicity in processes of sequence variant interpretation.
2. Can you tell us about your background and your role on the Ethics of Inclusion Study?
My background is in public health and I earned an MPH in 2019. While working towards my MPH, I became interested in how race and ethnicity are framed and operationalized in PMR, and biomedicine more broadly. I wrote my MPH capstone paper on how precision medicine researchers and population geneticists use and think about genetic clustering algorithms, and I joined the EOI team immediately after graduation. As a research associate on the EOI team, I was responsible for collecting and analyzing interview and observation data. I tended toward observing scientific working groups that discussed the ethics of population genetics, including how race and/or ethnicity enter the PMR research process.
3. Why is the question of how population descriptors are used in PMR important?
My interest in the topic stems from a concern with how different ways of understanding population descriptors call for different types of interventions into racial and/or ethnic disparities in health outcomes. If race and/or ethnicity is framed as a set of social categories, then it follows that health disparities along the lines of race and/or ethnicity are the consequence of social processes. This framing calls for sociopolitical interventions into health disparities. However, if race and/or ethnicity is thought or believed to overlap in some way with population level genetic differences, or patterns of sequence variants that predispose individuals to disease, then some health disparities may instead be framed as the result of genetic differences. This framing may call for pharmaceutical or clinical interventions into health disparities. This is not to say that the framings are mutually exclusive, but the ways in which we understand population descriptors and their significance direct us towards different methods for improving health.
4. What did you find in your research?
Our research found that, despite calls for PMR to shift away from the use of race and/or ethnicity, PMR researchers continue to debate the concepts’ relevance to processes of sequence variant interpretation. These debates have the effect of continuing to entrench race and/or ethnicity in PMR. In one of the examples we provide in the paper, we trace a series of working group decisions about how to stratify allele frequency data displayed in a publicly available reference database. The group sought to display the data in a way that avoided racializing genetic difference, but continually came up against challenges that stemmed from the many ways in which race is woven through PMR practices.
5. What do you see as the major challenges in shifting away from race and toward genetic ancestry in PMR?
One major challenge revealed by our data is that the onus for change is often placed on individual researchers and research teams, who tend to focus on one or two stages of sequence variant interpretation. This has the potential to result in circumscribed decisions about population descriptors that may address only the task at hand. Given that race and/or ethnicity are embedded across the stages of sequence variant interpretation, what appears to be a “fix” in one stage of research may not have the intended, or any effect, on downstream stages.
6. What changes do you propose to address these challenges?
It's important that guidance on the use of the population descriptors is comprehensive and bridges, to the degree possible, the research and clinical divide. As we mention in the article, a number of journals have issued guidelines on the use of population descriptors and a committee convened by the National Academies of Science, Engineering, and Medicine recently issued a report on the use of population descriptors in genetics and genomics research. There has also been an active conversation within the clinical realm about the use of race in biomedicine. These efforts, however, typically fall on one side or the other of the research and clinical divide. As we illustrate, decisions made early in the research process can have cascading effects on other processes, up to and including the clinical translation of findings. This suggests that guidance on the use of population descriptors may have a greater impact if it can take into account the full life cycle of PMR and its translation into the clinic.
7. Why did you and the research team choose to submit your article to HGG Advances?
We submitted the article to HGG Advances because the journal is oriented towards PMR researchers. We hope that the article might add to the current discussion amongst PMR researchers about the use of population descriptors by illustrating specific points at which decisions are made and tracing their causes and consequences. While we focus on the case of sequence variant interpretation, we hope that our findings are generalizable to other aspects of PMR as well.
8. How did the mentorship and guidance by your study team members affect your development of this article?
The article would not have been possible without the mentorship and guidance of the full team. Their insight and guidance were invaluable to the analysis, as well as the writing and submission of the article.
9. What feedback or reactions have you received about your research findings?
The article was published only recently, and we haven’t received feedback. We look forward to hearing thoughts!
10. What did you learn through this process?
This experience taught me a lot about how to move from data collection, through analysis, writing, and submission of an article. While I had published articles in the past, the analyses had been secondary and I wasn’t directly involved in data collection. It was a great experience to be involved in EOI from the early stages of data collection through to the publication of our findings.