Scientists with multiple NIH grants are overwhelmingly male and white

More and more grants from the US National Institutes of Health (NIH) are being concentrated in the hands of a small group of ‘super investigators’— a trend that could hamper efforts to increase the ethnic and gender diversity of biomedical research, a study finds.

The study¹ analysed the gender and ethnic profiles of researchers holding three or more NIH grants at the same time. Between 1991 and 2020, such investigators, whom the study dubs ‘super principal investigators’, or ‘super-PIs’, tripled as a proportion of total NIH investigators, from 3.7% to 11.3%, while the percentage of NIH funding allocated to them more than doubled, from 13% to 28%.

But this group of super-PIs has a diversity problem. Black or female investigators were less likely to be super-PIs than were white men, and Black women fared the worst: in 2020, 12 Black women were super-PIs, compared with 1,839 white men.

Collection: Diversity and scientific careers

Career stage and level of highest qualification did not play a part. “The gender, racial and ethnic disparity that we saw exists at every stage,” says Mytien Nguyen, an immunobiologist at Yale School of Medicine in New Haven, Connecticut, and corresponding author on the super-PIs study. At the early-career level, where historical bias could be expected to be less influential, Black female scientists were still the least likely to be super-PIs.

The lack of diversity among this growing group of elite NIH investigators poses “a substantial threat” to biomedical research in the United States, given the “well-documented benefits” of diverse science teams, write the study authors.

Their findings probably have “multifactorial” causes, they say, but differences in access to mentorship and patterns of grant submission could play a part. Black and female researchers are less likely to have high-impact mentors than are white or male scientists, they write. And although researchers who submit a lot of grant applications are more successful in winning at least some of those grants, Black faculty members tend to submit fewer applications than the average, they note.

Sweeping report calls for anti-bias measures in US science

Biases in grant assessment could also help to explain the disparities among super-PIs, the authors write. This is a hunch echoed by Shirley Malcolm, who directs SEA Change, a diversity, equity and inclusion initiative run by the American Association for the Advancement of Science, based in Washington DC. While reviewing grants for the National Science Foundation (NSF) in the early 1990s, Malcolm says, she encountered assumptions about who was capable of doing what that are “rooted in a system of white advantage”.

Malcolm highlights a study² of NSF grantmaking, published last year, that found that external reviewers on average scored white applicants more highly than they scored applicants from other ethnic groups. (Although reviewers were not privy to self-reported ethnic data, that study’s authors note that this could be inferred from the content of a proposal or from personal knowledge.) Malcolm says the study suggests that some of the behaviours she encountered as a reviewer are apparently continuing.

Battling bias in grant review

An NIH spokesperson told Nature that the number of concurrent agency awards is not one of the criteria that reviewers consider. However, the spokesperson says that reputational bias — which gives well-known investigators an advantage — is a concern for the agency’s Center for Scientific Review (CSR), which handles the first level of peer review for 76% of NIH grant applications. As a result, the spokesperson says, the CSR has developed training for reviewers on bias awareness and mitigation, and is working to make review panels more diverse.

Academics fight moves to defund diversity programmes at US universities

Marie Bernard, the NIH’s chief officer for scientific-workforce diversity, says that the agency has been aware for many years that female researchers, and those from under-represented racial and ethnic groups, are disproportionately low in number among its grant recipients. In an e-mail to Nature, she wrote that the CSR is changing its review process, and will move away from giving numerical scores for the ‘expertise and resources’ review criterion of applications. Instead, investigators will be rated ‘fully capable’ or ‘additional expertise/capability needed’, and research environments will be rated ‘appropriate’ or ‘additional resources needed’. The other two review criteria (‘importance of the research’ and ‘rigor and feasibility’) will continue to be assessed using numerical scores.

“The goal is to place more emphasis on the content of the science, and less on the investigator and environment. This should limit potential bias in the review process,” Bernard wrote.

A call to create funding equity for researcher-mums

Malcolm says that she would like a better idea of how holding two or more grants concurrently might influence the environment in which a researcher works. She questions whether any such researcher would produce more, or better, results than would one with fewer sources of support. And she wonders how junior scientists will be affected if funding is increasingly concentrated among a small elite.

Malcolm also posits that juggling many grants could leave researchers with less time for mentoring, for example. “I don’t think we are having these discussions with any kind of seriousness,” she says.

Bernard says that the NIH is piloting a programme that helps institutions to recruit and support a diverse group of early-career scientists. The agency is also planning a prize that recognizes institutions that have made strides towards achieving ‘inclusive excellence’.

That is an important move, says Nguyen. Historically, she says, the idea of excellence has been inherently biased, resulting in gender, racial and income disparities among recipients of prestigious medical-school awards³, for instance.

“Research has shown that diversity is required for innovation. Any award, recognition, grant or faculty promotion should employ this inclusive definition of excellence and evaluate diversity as a core component of excellence,” Nguyen says. “Excellence is the same as diversity, not the opposite.”