Published online 18 August 2011 | Nature | doi:10.1038/news.2011.485


Black applicants less likely to win NIH grants

'Unacceptable' gap in success rates heralds plan to tackle potential bias.

Only 16% of NIH grant applications from black researchers succeed, compared with 29% from whites.Punchstock

White researchers applying for grants from the US National Institutes of Health (NIH) are nearly twice as likely to win them as black researchers, reports a study funded by the agency. The finding has prompted an NIH investigation into whether its reviewers are racially biased.

"The situation is not acceptable," says Francis Collins, director of the NIH in Bethesda, Maryland. "It indicates that we have not only failed to recruit the best and brightest minds from all the groups that we need but, for those that have come, there is inequity. This is not just a problem for NIH but the whole research community."

The study, which the authors say is the first systematic investigation of racial and ethnic differences in NIH funding, is published today in Science1, along with a commentary, co-authored by Collins, committing the NIH to action2.

A team of researchers at the NIH, the University of Kansas in Lawrence and the data-mining company Discovery Logic of Rockville, Maryland, analysed more than 83,000 applications from more than 40,000 researchers to the NIH's largest grant programme for individual investigators, the R01, between 2000 and 2006.

Only 16% of applications from black researchers were funded, compared with 29% from white, a 13 percentage point difference.

No surprises

The results are shocking but not surprising, says Robert Dottin, an African-American medical geneticist at the City University of New York and founder and director of JustGarciaHill, a social-networking site that supports and promotes minorities in science. "Intuitively, African-Americans have known it," he says.

Dottin, who was not involved in the study, welcomes the NIH's response. "Many people believe that there is no problem. This study certainly says that there is and strengthens the argument that action is necessary," he says.

Even after correcting for applicants' educational background, nationality, training, previous research awards, publication record and institution, the gap remained at 10 percentage points, meaning that when these factors are accounted for black scientists are two-thirds as likely to win funding as white ones.

By contrast, an initial 4 percentage point disadvantage for Asian applicants disappeared when the sample was limited to US citizens, who can receive NIH training and probably have, on average, a better grasp of English than those from abroad. Hispanic applicants had the same success rate as whites.

"The gap for black applicants is extremely large, and very troubling in that it defies explanation," says study leader Donna Ginther, an economist at the University of Kansas.

The effect probably has many causes, she says. White scientists may have received better education and mentoring, resulting in their submitting more competitive proposals. Black researchers, Ginther notes, are significantly less likely to resubmit an unfunded application than their white colleagues, perhaps indicating a lack of mentoring.

The study found that undertaking NIH training did not equalize the success rates of black applicants.

Early bias

The study suggests that the bias arises during the NIH's initial review process. Applications that passed this review with strong priority scores were equally likely to be funded, regardless of race.

It's possible that this review process is racially biased, consciously or otherwise. Information on the race and ethnicity of applicants is not available to reviewers but it can be inferred from biographies and even names, notes Ginther.


To address the issue, the NIH hopes to increase the number of early-career reviewers from ethnic minorities and will consider giving applicants more help in preparing grants. It will also ask two advisory groups to advise it on additional steps.

In addition, the agency will carry out tests to detect racial bias in the review process, such as stripping out identifying information from applications to see whether this affects their chances of success.

"This is not one of these reports that we will look at and put aside," says Collins. 

  • References

    1. Ginther D. K. et al. Science 333, 1015-1019 (2011). | Article | ChemPort |
    2. Tabak, L. A. & Collins, F. S. Science 333, 940-941 (2011). | Article |
Commenting is now closed.