Research: NIH (Federal Gov't) Funding

This 2008 Nature article titled "Stats reveal bias in NIH grant review" describes the U.S. National Institutes of Health's grant application process, and the issues that one biostatistician sees in that system. The biostatistician, Valen Johnson of the University of Texas M.D. Anderson Cancer Center in Houston, states that the NIH "does not adequately compensate for reviewer bias". Through his study, he determined that proposal is read by 2.8 readers on average. Each proposal is evaluated by about 30 members, so an average of under 10% of those members actually read each proposal. The potential bias that comes from those 2.8 readers can make a significant impact on how government funding of medical research is distributed. After all, there are about 14,000 readers and $20-25 billion to sort out each year. While he says that "top grants were largely unaffected by reader bias, ... such bias did impact grants closer to the funding cut-off line". Johnson's solution, to give favor to proposals that cost less, is not one that I agree with, but it was certainly interesting to see a problem identified (one that I do agree with) and get a general idea of the NIH grant application process.

Google keywords: site:nature.com nih funding medical research unfair