The US National Institutes of Health (NIH) is taking steps to cut down the review process for its research grants, hoping to both reduce biases and ease the burden on scientists in the process.
The NIH is the largest provider of basic research funding to US higher education, and it has struggled for years to combat persistent evidence that allocations from its $40 billion (£33 billion) annual budget do not necessarily identify the best scientific proposals but instead reinforce racial and reputational inequities.
In its latest whack at the problem, the agency is asking its advisory panel of outside academic experts to endorse trimming the judging criteria it presents to grant reviewers from five points to three to put more emphasis on project feasibility than on the applicant and institution.
NIH officials said they developed the plan after a test in which they removed from grant applications information that would clearly identify principal investigator applicants and their institutions. The result, however, still left significant race- and reputation-based disparities in grant awards, apparently because of the ability of reviewers to guess identifying elements of the applications and continue to reward those PIs they recognised.
“There was some level of a bump that white applicants were enjoying just based on their name recognition or the institution they were in,” Noni Byrnes, director of the Center for Scientific Review – the division that oversees the majority of the NIH’s peer review process – told the agency’s advisory committee to the director.
That realisation, Dr Byrnes said, led NIH leaders to decide that they needed to further “diminish the role of PI identity and institutional identity” in grant review processes.
For the bulk of its grant awards, the NIH uses a two-stage process in which outside experts whittle down lists of grant applications by about half before NIH staff make final decisions. The volunteer reviewers score on five criteria: significance, innovation, approach, environment and the suitability of the applicants.
The NIH has not yet decided the full shape of the three criteria that will remain but suggested that they will cover the importance of the proposed research and its feasibility and rigour, and perhaps include some way of measuring the innovative quality, but avoid questions about the people on the research team and its academic environment.
NIH officials preparing the change studied anti-bias efforts in a number of academic fields, Dr Byrnes said. “One thing that’s clear from all fields is that the more specific you make the review criteria, and the more you make the evaluators be specific, the less bias you’ll have in the evaluation,” she said.
Reducing the criteria also will hopefully encourage more academic experts to volunteer their time to NIH grant review teams, Dr Byrnes said. “The administrative burden can be a disincentive to serve,” she said.
Donna Ginther, a professor of economics at the University of Kansas, said the initiative looked worthwhile. “I like the idea that we should try doing something like this, and see whether or not it works,” she said.
Given that the test of the anonymisation of research applicants did not significantly help black scientists, said Professor Ginther, “changing the review criteria, reducing the number of review criteria, makes a lot of sense”.
