US National Institutes of Health updates peer review system

BMJ 1999; 319 doi: (Published 07 August 1999) Cite this as: BMJ 1999;319:336
  1. Janice Hopkins Tanne
  1. New York

    The peer review system governing grant applications to the US National Institutes of Health is set to be updated over the next two years, with the aim of encouraging more innovative research.

    A report published last week by a panel chaired by Dr Bruce Alberts, president of the National Academy of Sciences, recommended changes to take advantage of opportunities in biomedical research and to keep pace with new research methods (Science 1999;285:666-7). Dr Keith Yamamoto, chairman of the Advisory Committee of the National Institute of Health's Center for Scientific Review and chairman of the department of cellular and molecular pharmacology at the University of California in San Francisco, told the BMJ that the panel's aims were to bring the review system in line with rapid advances in science, reduce conservatism, encourage risk taking, involve outside experts, and continually review the process itself. The report commented: “Many researchers fear that conservatism … and an undue requirement for preliminary data discourage innovation. In practice, the present system tends to discourage risk-taking and under-value new ideas.”

    The National Institutes of Health is the leading funder of biomedical research in the United States, providing $15.6bn (£9.7bn) this year for studies aiming to improve human health and train young scientists. Three quarters of the 40000 annual grant applications go through peer review at its Center for Scientific Review, headed by Dr Ellie Ehrenfeld, who indicated problem areas.

    The current approval procedures for grant applications were considered to be deficient on several levels. Grant applications are currently referred to integrated review groups, but no appropriate groups exist for some of the newest research fields. Applications for potentially important work may be assigned to several different study sections, “causing too much of ‘the best science’ to compete with itself,” suggested the report.

    The proposal suggests that peer reviewers should not be advocates or gatekeepers for a field; they need not be competitors of the applicant or even studying the same disease or organ system but should be “experienced researchers who are reasonably diverse in seniority, outlook, geographical location, gender, and ethnicity.” Dr Yamamoto commented: “There is concern about an old boy network but also about not taking advantage of experienced researchers. The younger investigators don't have the breadth of view that comes from being through [several research review] cycles.”

    The full report is available at

    View Abstract

    Sign in

    Log in through your institution