NIH Issues Changes To Policy On Sex/Gender Analysis In Research

Announcement: http://grants.nih.gov/grants/guide/notice-files/NOT-OD-00-048.html

The National Institutes of Health has updated its 1994 guidelines requiring inclusion of women and minorities as subjects in clinical research in light of recent criticism that funded studies often have failed to report on sex/gender differences.

Beginning with the Oct. 1 receipt date for grant applications and contract proposals, investigators requesting funds for phase III clinical trials must follow the restated policy. New language spells out steps for researchers to plan analysis of subgroup effects.

Unless prior studies strongly support no differences between subgroups, researchers must at minimum:

• Include in the application or proposal a description of plans to conduct the valid analysis of the intervention effect in subgroups;
• Include plans for analysis in the final protocol approved by the institutional review board, which evaluates protections for human subjects prior to funding; and
• Report results of subset analysis to NIH in progress reports, competitive renewal applications, contract extensions and required final reports.

NIH also encourages researchers to include the results of subset analyses in research articles submitted to journals, or at least a brief statement indicating there were no differences and describing the subsets analyzed.

The changes come in the wake of a recent General Accounting Office report applauding NIH's implementation of a 1993 congressional mandate to increase inclusion of women subjects in clinical research but saying the agency needs to do more to ensure valid analysis of intervention effects by sex.

More than half the participants in extramural clinical research studies funded in fiscal 1997 were women, the report found. Almost 62 percent of study subjects were women in fiscal 1997, and 68 percent were women in fiscal 1998. In phase III clinical studies in 1997, 74.8 percent of subjects were women.

The policy restated in the Aug. 4 NIH Guide for Grants and Contracts urges investigators to review prior evidence of any clinically important sex/gender and/or race/ethnicity differences in the intervention effect to be studied. Evidence could be from animal studies, clinical observations, metabolic studies, genetic research, pharmacology studies, observational studies, natural histories, epidemiology or other research.

If prior studies support differences of public health significance, the proposed NIH phase III clinical trial must be designed to address them by including plans for analyses to detect differences in intervention effects.

Since 1994, guidelines have required researchers to include women and minority group members in all NIH-supported biomedical and behavioral research involving human subjects. It also instructs principal investigators to assess the theoretical and/or scientific links between gender, race/ethnicity and the topic of study.

Each application for funds must address the policy and provide required information on inclusion of women and minorities and their subpopulations, or any justification for an exception to the policy. NIH also requires reviewers to consider the planned study population when assessing the scientific merit of a grant proposal.