More rounds of grant review do not necessarily equal more funding

By Adriana Bankston, Ph.D.

The ability to make novel scientific discoveries is largely dependent on principal investigators being able to secure federal funding. This is often in the form of an R01 grant from the National Institutes of Health. According to the NIH’s 2008 policy, investigators whose de novo R01 grant, termed a true A0, was unfunded, were given only one chance to submit a revised grant application, termed an A1. Prior to this policy, PIs could revise and submit an A2 application, but the 2008 policy eliminated the A2 and deemed new A0 applications highly similar to an unfunded A1 application impermissible. The NIH revised this policy recently, now allowing for submission of an A0 grant application that is materially similar to an unfunded A1 application.

In a recent post from the NIH Deputy Director for Extramural Research, Michael Lauer’s group posted an analysis of the revised resubmission policy by identifying grants that have undergone peer review twice and have been resubmitted as new, A0 applications, which they termed virtual A2 applications. A reasonable expectation is that virtual A2 applications, having undergone previous rounds of peer review and revision, would be more successful than true A0 applications, which have received no NIH feedback. As virtual A2s are not an obvious class of application, Lauer’s group used text-mining software to identify virtual A2s. Of those submitting virtual A2 applications, over seventy percent of investigators had their A1 applications discussed at study section. However, virtual A2s had only a marginally higher priority score and percentile ranking compared to A1 applications. These small changes resulted in about 20 percent of virtual A2s being funded, similar to the success of true A0 applications.

An important observation made by Lauer’s group is that new and early-stage investigators were somewhat less likely to submit virtual A2 applications. This is an important phenomenon to understand. This trend could indicate that young researchers are able to obtain sufficient funding from non-R01 NIH mechanisms or from nonfederal sources. Alternatively, they may be dissuaded from submitting virtual A2 applications due to the low funding rates of A0 grants to begin with.

There is also a question as to what these results mean for the importance of peer review and responding to reviewers’ comments. Virtual A2 applications are treated as de novo grant submissions, as they are not linked to previous grant applications. However, given the two rounds of peer review, the expectation is that these grants would be significantly improved and their funding rates would be significantly higher compared to A0 applications. However, virtual A2s are funded at the same rate as true A0 grants. Thus, how incorporating reviewers’ comments affects the outcome of subsequent grant applications is important to understand in order to further refine grant application policies.

The analysis provided by Lauer’s group provides useful insights into NIH grant submission and resubmission practices. It will be important to track these data over time and to find ways to address the critical questions on the best policies to fund scientists.

Adriana Bankston has a strong interest in advocating for biomedical scientists and improving the biomedical research enterprise as well as educational programming, outreach and science communication. Bankston received her Ph.D. from Emory University and is currently a postdoctoral research associate at the University of Louisville and member of the National Postdoctoral Association’s Advocacy Committee. She can be reached via email ( or on Twitter (@AdrianaBankston).

The RBR Writing Program is a new program to help graduate students and postdocs receive policy writing experience. For more information, contact RBR Director Chris Pickett.