Analysis of authorship trends in vascular surgery demonstrates a sticky surgical floor for women.

OBJECTIVE: Prior research in vascular surgery has identified significant gender disparities in leadership positions, but few data exist regarding gender disparities in vascular publications. This study aims to evaluate authorship trends by gender in the three highest impact factor vascular surgery journals. METHODS: In this bibliometric analysis, PubMed was searched for articles published in the European Journal of Vascular and Endovascular Surgery, the Journal of Vascular Surgery, and Annals of Vascular Surgery from 2015 to 2019. The web-based application Genderize used predictive algorithms to classify names of first and last authors as male or female. Statistical analyses regarding trends in authorship were performed using Stata16. RESULTS: A total of 6457 articles were analyzed, with first author gender predicted with >90% confidence in 83% (4889/5796) and last author gender in 88% (5078/5796). Overall, 25% (1223/4889) of articles had women first authors, and 10% (501/5078) had women last authors. From 2015 to 2019, there was a slight increase in the proportion of articles written by women first authors (P = .001), but no increase in the proportion of articles written by women last authors (P = .204). The proportion of articles written by women last authors was lower than the proportion of active women vascular surgeons in 2015 (8% of articles vs 11% of surgeons; P = .015), 2017 (9% of articles vs 13% of surgeons; P < .001), and 2019 (11% of articles vs 15% of practicing surgeons; P < .001). The average number of last-author publications was higher for men (2.35 ± 3.76) than for women (1.62 ± 1.88, P = .001). The proportion of unique authors who were women was less than the proportion of active women vascular surgeons in 2017 (10% unique authors vs 13% surgeons; P = .047), but not in 2015 (9% unique authors vs 11% surgeons; P = .192) or 2019 (13% unique authors vs 15% surgeons; P = .345). Notably, a woman last author was associated with 1.45 higher odds of having a woman first author (95% confidence interval, 1.17-1.79; P = .001). CONCLUSIONS: Over the past 5 years, there has been no significant increase in women last authors among top-tier journals in vascular surgery. Women remain under-represented as last authors in terms of proportion of published articles, but not in terms of proportion of unique authors. Nevertheless, women last authors are more likely to publish with women first authors, indicating the importance of women-led mentorship in achieving publication gender equity. Support for women surgeons through grants and promotions is essential not only for advancing last authorship gender equity, but for advancing junior faculty and trainee academic careers.