Uncovering gaps in women's authorship: A big data analysis in academic surgery

Introduction: Women are underrepresented in surgical authorship. Using big data analyses, we aimed to investigate women's representation as first and last authors in surgical publications worldwide and identify underlying predictors. Methods: We retrieved eligible surgical journals using Scimago Journal & Country Rank 2021. We queried articles indexed in PubMed from selected journals published between January 2018 and April 2022. We used the EDirect tool to extract bibliometric data, including first and last authors' names, primary affiliation country, and publication year. Countries and dependent territories were classified following World Bank income levels and regions. Women's representation was predicted from forenames using the Gender-API software. Citations were included if gender accuracy was >= 80%. Results: We analyzed 210,853 citations containing both first and last authors' forenames, representing 158 countries and 14 territories. Women constituted 23.8% (50,161/210,853) of the first and 14.7% (31,069/210,853) of the last authors. High-income economies had more women as first authors than other income categories (p < 0.001), but fewer women as last authors than upper-middle- and lower-middle-income economies (p < 0.001). The odds of the first author being a woman were more than three times higher when the last author was also a woman (OR 3.21, 95% CI 3.13-3.30) and vice versa (OR 3.25, 95% CI 3.16-3.34) after adjusting for income level and publication year. Conclusions: Women remain globally underrepresented in surgical authorship. Our findings urge concerted global efforts to overcome identified disparities.