Sex Differences in Academic Productivity Across Academic Ranks and Specialties in Academic Medicine: A Systematic Review and Meta-analysis.

IMPORTANCE

Despite equal numbers of men and women entering medical school, women are underrepresented in the upper echelons of academic medicine and receive less compensation and research funding. Citation-related publication productivity metrics, such as the h-index, are increasingly used for hiring, salary, grants, retention, promotion, and tenure decisions. Exploring sex differences in these metrics across academic medicine provides deeper insight into why differences are observed in career outcomes.

OBJECTIVE

To systematically examine the available literature on sex differences in h-index of academic faculty physicians across all medical specialties and all levels of academic rank.

DATA SOURCES

Medical literature with the term h-index found in PubMed and published between January 1, 2009, and December 31, 2018, was used.

STUDY SELECTION

A PICOS (Population, Intervention, Comparison, and Outcomes), PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses), and MOOSE (Meta-analysis of Observational Studies in Epidemiology) selection protocol was used to find observational studies that published h-indexes for faculty physicians that were stratified by sex. Studies were excluded if they were review articles, retracted, or unavailable online. Ultimately, 14 of 786 studies (1.78%) met the inclusion criteria.

DATA EXTRACTION AND SYNTHESIS

Data from 9 studies across 16 specialties were examined using weighted random-effects meta-analyses. Five studies were excluded because of overlapping specialties with another study or because they were missing appropriate statistics for the meta-analysis. Four of these studies were included in qualitative synthesis to bring the total to 13 studies.

MAIN OUTCOMES AND MEASURES

The primary study outcome was the h-index.

RESULTS

The meta-analysis included 10 665 North American unique academic physicians across 9 different studies from the years 2009 to 2018. Of the 10 665 physicians, 2655 (24.89%) were women. Summary effect sizes for mean h-indexes of men and women and mean h-index difference between men and women were determined for all faculty physicians and at each academic rank. Overall, female faculty had lower h-indexes than male faculty (mean difference, -4.09; 95% CI, -5.44 to -2.73; P < .001). When adjusting for academic rank, female faculty still had lower h-indexes than male faculty at the ranks of assistant professor (mean difference, -1.3; 95% CI, -1.90 to -0.72; P < .001), associate professor (mean difference, -2.09; 95% CI, -3.40 to -0.78; P = .002), and professor (mean difference, -3.41; 95% CI, -6.24 to -0.58; P = .02).

CONCLUSIONS AND RELEVANCE

In this systematic review and meta-analysis, women had lower h-indexes than men across most specialties and at all academic ranks, but it is unclear why these differences exist. These findings suggest that future investigation should be conducted regarding the causes of lower h-indexes in women and that interventions should be developed to provide a more equitable environment for all physicians regardless of sex.