A 10-year longitudinal evaluation of science policy interventions to promote sex and gender in health research

Background Over the past decade, the Canadian Institutes of Health Research (CIHR) has implemented multicomponent interventions to increase the uptake of sex and gender in grant applications. Interventions included mandatory reporting on applicant forms, development of resources for applicants and evaluators, and grant review requirements. Here, we aim to inform science policy implementation by describing the 10-year outcomes and lessons learned from these interventions. Methods This is a prospective longitudinal study. The population is all applicants across 15 investigator-initiated CIHR competitions from 2011 to 2019 and grant evaluators from 2018 to 2019. Quantitative data were derived from applicants’ and grant evaluators’ mandatory reporting of sex and gender integration in the grants management database. The application was the unit of analysis. Trends in sex and gender uptake in applications were plotted over time, stratified by research area. Univariate logistic regression was used to assess associations between the sex of the applicant and the uptake of sex and gender, and the latter with funding success. Qualitative review of the quality and appropriateness of evaluators’ comments informed the development of discipline-specific training to peer review committee members. Feedback was compiled from a subset of evaluators on the perceived usefulness of the educational materials using a brief questionnaire. Results Since 2011, 39,390 applications were submitted. The proportion that reported integration of sex rose from 22 to 83%, and gender from 12 to 33%. Population health research applications paid the greatest attention to gender (82%). Across every competition, applications with female principal investigators were more likely to integrate sex (odds ratio [OR] 1.60, 95% confidence interval [CI] 1.50–1.63) and gender (OR 2.40, 95% CI 2.29–2.51) than those who identified as male. Since 2018, applications that scored highly for the integration of sex (OR 1.92, 95% CI 1.50–2.50) and gender (OR 2.53, 95% CI 1.83–3.50) were more likely to be funded. Qualitative observations revealed persistent conflation of the terms sex and gender. Eighty-six percent of evaluators appreciated the tailored discipline-specific coaching. Conclusions A number of policy interventions improved sex and gender uptake in grant applications, with higher success rates observed over time for applications that integrated sex and gender. Other funders’ action plans around sex and gender integration may be informed from our experiences of the timing, type and targets of the different interventions, specifically those directed at evaluators.