Trends in Gender Distribution Among Authors of Research Studies in Pediatrics: 2015-2019.

Abstract

In a previous report on the gender distribution of authors of articles published in leading pediatric journals, researchers found disparities in the proportion of articles written by women relative to men, with some closing of the gender gap from 2001 to 2016.1 In 2016, 63% of pediatricians were women; 58% of the publications had a female first author, but only 38% had a female senior author.1,2 Additionally, reports have found that female first authors are associated with female authors in the senior author position and female gender of the first or senior author is associated with a higher number of collaborating departments and institutions.1,3–6 However, to our knowledge, this report is the only study on this topic in the field of pediatrics, assessing authorship gender distribution in 5-year intervals through 2016.1The Pediatrics editorial board is committed to addressing underrepresentation of all types, including gender differences in authorship. To begin to understand this issue, we evaluated regular articles published in Pediatrics within the past 4.5 years for female representation in authorship, the degree to which the gender of first and senior author is correlated, and other measures of collaboration (eg, the number of participating institutions and total number of authors). Regular articles are defined as “original research contributions that aim to inform clinical practice or the understanding of a disease process…includ[ing] but are not limited to clinical trials, interventional studies, cohort studies, case-control studies, epidemiological assessments, and surveys.”7 These more recent findings will be used to inform future journal policies to address implicit bias in the review process and inform academic departments about the need to support women in academics.We reviewed research articles published in Pediatrics from January 2015 to July 2019, extracting the number of authors and institutions. The gender of the first and senior author was determined from the following sources, in sequence: the author’s institutional Web site, ResearchGate, LinkedIn, or Doximity. Only 1 affiliation was counted per author. Differences in categorical variables were evaluated with χ2 tests, and median values of nonparametrically distributed continuous variables were compared by using Mann–Whitney U tests. P < .05 was considered statistically significant.From January 2015 to July 2019, there were 1231 published research articles, involving 9444 authors. Overall, 61% (756) had a female first author, and 44% (547) had a female senior author. Gender could not be identified in 2% (22) of the first authors and 1% (13) of the senior authors. The gender distribution of first and senior authors over the study period is shown in Fig 1. There was no change in the proportion of female first or senior authors over the study period (P = .25 and P = .63, respectively). The number of coauthors (median: 7 [interquartile range (IQR) 5–9]) and number of collaborating institutions (median: 3 [IQR 2–4]) per article did not change over the study period (P = .90 and P = .70, respectively).First authors were more likely to be female in articles with female senior authors than in those with male senior authors (69% [377 of 547] vs 56% [375 of 671]; P < .001). Senior authors were more likely to be female in articles by female first authors than in articles by male first authors (50% [377 of 756] vs 36% [164 of 453]; P < .001).Female first authors wrote articles involving a lower number of collaborating institutions compared with male first authors (median: 2 [IQR 1–4] vs 3 [IQR 2–4]; P = .03). Female senior authors also wrote articles with a lower number of collaborating institutions compared with male senior authors (median: 2 [IQR 1–4] vs 3 [IQR 2–4]; P = .02). There was no difference in the median number of collaborating authors on the basis of the gender of first or senior authors (P = .62 and P = .98, respectively)We noted 2 important findings regarding articles published in Pediatrics: articles with female first authors were more likely to have female senior authors, and articles with female first or senior authors had fewer collaborating institutions.The proportion of female first authors publishing in Pediatrics over the past several years appears representative of the proportion of female physicians in the field of pediatrics overall.1,8,9 In contrast, the proportion of female senior authors is lower than the proportion of female physicians in the field. This may be related to factors discussed in past reports including, but not limited to, fewer women in senior positions, such as division chiefs or leaders of academic collaboratives, implicit gender bias in grant awarding, and midcareer burnout.10–12 Nonetheless, the pipeline of future female senior academic pediatricians is promising on the basis of our first author findings, if we can preserve the successful transition of these first authors to the senior authorship role. Our finding that articles with a female first author were more likely to have a female senior author is consistent with previous publications.1,3–5 This may highlight the importance of female research guidance for female young investigators, promoting what may be an important strategy for supporting more junior female faculty. Our finding that articles with a female first or senior author involve fewer centers is in contrast with a previous report regarding radiologists.6 Future work is needed to better understand if female pediatricians have less opportunity for multisite collaboration.This study is limited by the restriction to a 4.5-year study period in a single journal, and only published original investigations. We assumed first authors are junior and senior authors serve as mentors. We did not consider authors’ subspecialty of practice. The submission platform during the study period did not elicit information about sex, gender, or race and/or ethnicity. We categorized author gender on the basis of the pronouns used in their profiles on institutional or professional Web sites but were not able to assess whether specific gender preferences were accurately represented. We were also not able to ascertain the gender for 1.4% of the authors. The intersectionality of race and ethnicity and gender identification likely adds additional complex layers into the role bias may be playing in these findings.With this study, we suggest that the current pool of junior faculty publishing as first authors in Pediatrics is representative of the current percentage of women pediatricians in the United States. With the findings, we underscore the importance of ensuring that female junior faculty progress in their career pathway to increase the pool of female physicians in academic leadership roles and expand the number of female senior authors engaged in research and subsequent publications.