Authorship Gender Composition in the Ophthalmology Literature from 2015 to 2019

Despite progress toward gender equality, female academic ophthalmologists continue to face considerable challenges. Although ophthalmology ranks third among surgical specialties in female representation,¹ 72% of US residency program directors and 90% of department chairs are male.^2,3 This study assessed the ophthalmology literature to track female authorship trends over 5 years, focusing specifically on first and last author positions given their relative importance for promotion and tenure.⁴

The research adhered to the tenets of the Declaration of Helsinki and the need for Institutional Review Board approval was waived by the University of Miami Institutional Review Board. Four comprehensive and 5 subspecialty ophthalmology journals were chosen for analysis (Table S1, available at www.aaojournal.org). The first and last authors from articles published between January 1, 2015, and December 31, 2019, were analyzed for gender. Gender-API (Munich, Germany, available at https://gender-api.com/), a gender inference tool that returns gender assignments based on first name, was used. Names returning a score of less than 90% confidence on Gender-API were manually verified via an exhaustive Google search (Google Inc., Mountain View, CA). Authors with unknown genders were excluded. When only 1 author was listed, they were assigned to the first author cohort and excluded from the last author group.

MATLAB R 2020a (MathWorks, Natick, MA) was used to download articles from PubMed, extract author names and their author position, and perform statistical analysis. Additional analyses were done using SAS version 9.4 software (SAS Institute, Inc, Cary, NC).

The study analyzed 17100 articles with 31591 authors: 8775 unique male authors and 5224 unique female authors. Men published more articles per person (mean, 2.44 ± 3.66; median, 1; interquartile range, 1−2) compared with women (mean, 1.95 ± 2.77, P < 0.001; median, 1; interquartile range, 1−2, P < 0.001). The average percentage of female authors overall was 32%, and the proportion of female first authors (38%) was higher than the proportion of last authors (27%, P < 0.001) (Table S1, Fig S1, available at www.aaojournal.org). The majority of journals had a significant gender association between first and last authors (Table S2, available at www.aaojournal.org). The gender association for general journals combined (odds ratio, 1.4; confidence interval [CI], 1.27−1.54; P < 0.001) was significantly less than that of a similar study performed more than a decade ago (odds ratio, 2.19; CI, 1.96−2.46; P < 0.001).⁵

The proportion of female first authors significantly increased over time for all journals (P = 0.007) and subspecialty journals (P = 0.007), but not general journals (P = 0.071) (Fig 2). The proportion of female last authors significantly increased for general, subspecialty, and all journals (P = 0.007, 0.025, and 0.025, respectively). There was no significant difference in the mean rate of increase between female first authorship (4.3%, CI, 0.9−7.7) and last authorship (4.8%, CI, 3.3−6.3).

Figure 2.

Female authorship trends from 2015 to 2019 in all, general, and subspecialty journals by first and last author.

Female authorship increased in both first and last author positions over 5 years, but gender disparities remain. Female first authorship is below 40%, whereas female last authorship is approximately 25%. Furthermore, our study found that women on average published significantly fewer articles per person than men. This suggests that not only is there a significantly greater number of men publishing, but also men are more prolific than women.

There was no significant difference in the rate of increase between female first and last authorship, contrasting with previous descriptions of a steeper rise in female first authorship compared with last authorship in the ophthalmic literature between 2002 and 2014.^5,6 These older findings may reflect the increase in female residents and trainees over those years,⁵ and it is possible that we identified a subsequent echoing increase in last/senior authors in our later analysis because these previous first authors have since advanced in their careers. For subspecialty journals, female authorship in both first and last positions also significantly increased and increased by similar proportions. This newfound trend suggests that although there may have been a lag in the natural migration of women up the academic ladder to senior author positions, particularly in subspecialties, this gap may be closing as women take on more senior roles.

One factor implicated in the persistence of gender disparities is an implicit gender association. Our analysis suggests authors are more likely to collaborate with people of the same gender, as female first authors in our study were more likely to have a female last author and vice versa.⁷ This association also stands true for men and is significant because last, or senior, authors often serve as mentors for first authors. These findings align with previous results showing authorship gender associations in general ophthalmology journals⁵; however, we found gender association to be significantly lower in our study, suggesting that although gender associations still exist, these have lessened over the past decade.⁵

This study has limitations, including limiting analysis to first and last authorship, analyzing US-based journals but not excluding international authors, using a previously validated online tool for gender identification, having the inability to identify gender in a small subset (<5%) of analyzed authors, and using binary gender classifications. Additionally, if the author changed their last name during their academic career, they would be considered as 2 different authors. This has the potential to affect the number of unique authors and publications per unique author noted in this study. Last, the gender association findings do not account for ophthalmologists who have not published manuscripts in the last 5 years.

This study provides promising evidence that the rise in female last authorship is finally increasing at a similar rate to female first authorship. The gender authorship association also suggests that senior women have positively impacted this progress by mentoring junior female authors. This gender association, combined with the well-documented lower number of senior women faculty compared with more junior faculty, may be yet another contributor to the bottleneck in academia, where growing numbers of women complete training, but far fewer advance to the highest academic ranks. However, our analysis provided some evidence that this is changing and that more men are taking on mentorship roles for young women. These findings support recommendations that target robust mentorship programs, encouraging gender-blind collaborations and strategies to support the promotion of women to senior-level positions.