Gender gap in research: a bibliometric study of published articles in primary health care and general internal medicine.

Abstract

Many studies examined gender inequalities in research, but only a few data are available for general biomedical journals. We assessed the prevalence of female first authorship in general biomedical journals and examined its variations across a number of author, article and journal characteristics. This study was nested within a larger project designed to analyze the bibliometric characteristics of scientific articles. We retrieved 767 randomly selected articles published in 2016 in high impact factor journals of primary healthcare (n = 9) and general internal medicine (n = 9). We extracted the following data: author (gender, number of publications and affiliation of the first author), paper (number of authors, number of participants and study design) and journal characteristics (journal discipline and 2015 impact factor). We compared the proportion of articles authored by women and men using univariate and multivariate logistic regressions adjusted for intra-cluster correlations. The female authorship proportion was 48% (63% for primary healthcare and 33% for general internal medicine, P-value < 0.001). In multivariate analysis, women published fewer articles (<5 versus >15 publications: OR 1.6 [95% CI 1.1-2.4]), were more often affiliated with institutions in the Western world (OR 2.2 [95% CI 1.2-3.9]), were more likely to publish qualitative studies (versus systematic reviews or experiments: OR 2.7 [95% CI 1.5-4.8]) and to publish in primary healthcare journals (OR 1.7 [95% CI 1.1-2.7]). The underrepresentation of women in articles published by general internal medicine journals, in articles from the non-Western world and in systematic reviews and trials should be addressed.