The impact of COVID-19 on rheumatology health-care workers and their scientific output: heavy lies the crown on women in rheumatology

Interest in gender equity has increased in the past decades, as the share of women in the global medical workforce has grown, including within the field of rheumatology. For example, 30·2% of US rheumatologists were women in 2005, and 40·8% in 2015, with estimates projected to reach 57% in 2030.1Deal CL Hooker R Harrington T et al.The United States rheumatology workforce: supply and demand, 2005–2025.Arthritis Rheum. 2007; 56: 722-729Crossref PubMed Scopus (154) Google Scholar, 2Bolster MB Bass AR Hausmann JS et al.2015 American College of Rheumatology Workforce study: the role of graduate medical education in adult rheumatology.Arthritis Rheumatol. 2018; 70: 817-825Crossref PubMed Scopus (19) Google Scholar However, women's representation in the overall US rheumatology workforce is not reflected in academic medicine, where, in 2014, women were less likely to be full or associate professors than men (odds ratio 0·78, 95% CI 0·62–0·99).3Jorge A Bolster M Fu X et al.The association between physician gender and career advancement among academic rheumatologists in the United States.Arthritis Rheumatol. 2021; 73: 168-172Crossref PubMed Scopus (16) Google Scholar The COVID-19 pandemic imposed changes in most aspects of life, reversing some of the previous progress made in gender parity. Several authors have described gender disparity in scientific authorships that was associated with the COVID-19 pandemic.4Brubaker L Women physicians and the COVID-19 pandemic.JAMA. 2020; 324: 835-836Crossref PubMed Scopus (91) Google Scholar A decline in women as first authors during the first phase of the pandemic has also been observed in the field of rheumatology.5Lerchenmüller C Schmallenbach L Jena AB Lerchenmueller MJ Longitudinal analyses of gender differences in first authorship publications related to COVID-19.BMJ Open. 2021; 11e045176Crossref PubMed Scopus (35) Google Scholar In our study, we aimed to comprehensively examine research output in the field of rheumatology by gender (women vs men) and to explore underlying causes of alleged authorship disparity. The presentation of gender as a binary variable was based on the body of existing bibliometric research and the availability of artificial intelligence-powered tools that use author names to infer gender. Although not reflective of the current understanding of gender, the Gender API service we used has been reported to be highly accurate (>98%) for binary analysis of gender.6Sebo P Performance of gender detection tools: a comparative study of name-to-gender inference services.J Med Libr Assoc. 2021; 109: 414-421Crossref PubMed Scopus (37) Google Scholar We did a literature search in PubMed and the preprint servers medRxiv and bioRxiv to compare authorship before and during the beginning of the COVID-19 pandemic: from January, 2020, to June, 2020, we searched for publications using rheumatic diseases “AND COVID-19/SARS-CoV-2”, whereas from January, 2019, to June, 2019, we searched for publications using rheumatic diseases only, without the search term “AND COVID-19/SARS-CoV-2” (appendix pp 1–2). We included 2173 authors from 296 papers published in 2019, and 1343 authors from 201 papers published in 2020 (appendix p 2). Publication topics differed substantially between 2019 and 2020, with a focus on use of rheumatological therapeutic options for COVID-19 in 2020, often published in high-impact journals (appendix pp 3, 6). Women were generally under-represented across all authorship positions in our 2019 sample, with the most extreme disparities for last authors (82 [28%] of 290 were women), and corresponding authors (45 [31%] of 143 were women; figure A, B). The gender gap increased in 2020 for first, last, and corresponding authors, with a significant decline in women first authors (133 [46%] of 289 in 2019 vs 71 [35%] of 201 in 2020; p=0·023), especially of original research (67 [43%] of 156 in 2019 vs six [18%] of 33 in 2020, p=0·0089; figure A, B; appendix p 3). To further investigate the underlying causes of this disparity and potential differences between women and men in burden experienced during the pandemic, we developed an English-language online survey assessing academic output, and professional and environmental factors before the pandemic, during the early phase (January–June, 2020), and thereafter (appendix pp 10–15). The survey was distributed to identified corresponding authors, European rheumatology societies, and individual contacts via snowball sampling. We collected 180 complete responses, 94 (52%) of which were from women (appendix pp 7–9). Respondents were from 28 countries (appendix p 8). Women respondents were significantly younger than men (median age range 30–39 years vs 40–49 years, p=0·0039), and more likely to live alone (18 [19%] of 94 vs nine [10%] of 86, p=0·028). Regarding overall academic output, women reported significantly fewer total publications (p=0·0002), and fewer first (p=0·0017) and last authorships (p=0·0003; appendix p 4). During the first phase of the pandemic, women had significantly fewer last authorships (p=0·011), and after the first 6 months, men reported a higher number of publications (p=0·027; appendix p 4). Asked about general experience of burden and potential causes thereof, 108 (60%) of 179 respondents appraised the burden as substantial or intolerable irrespective of gender, predominantly due to modified patient care (figure C, D). However, the underlying causes differed, with women reporting family care significantly more often than men (p=0·024; figure D). This disparity was more pronounced among respondents with children younger than 12 years, of whom 23 (77%) of 30 women and ten (34%) of 29 men cited family care as a burden (p=0·0011; appendix p 5), although both women and men reported devoting additional hours per week to childcare. Additionally, women reported a larger increase in hours spent on household duties compared with men (p=0·038; figure E), and 39 (41%) of 94 women were still doing one or more additional hours of household work in November, 2020 (figure F). Irrespective of gender, 77 (57%) of 134 individuals felt that co-workers provided the most support to cope with managing the situation at work; only eight (6%) individuals acknowledged support from official institutions, such as governmental agencies. Consequently, 96 (59%) of 164 individuals would like to see increased future support from official institutions and 72 (44%) of 164 individuals from their superiors. To our knowledge, this is the first study investigating initial impacts of the COVID-19 pandemic on the scientific output of rheumatologists, which also analyses possible underlying causes. Our findings align with data reporting a declined proportion of women first authorships—the only authorship category with a gender balance before the pandemic7Bagga E Stewart S Gamble GD Hill J Grey A Dalbeth N Representation of women as authors of rheumatology research articles.Arthritis Rheumatol. 2021; 73: 162-167Crossref PubMed Scopus (17) Google Scholar—in the first phase of the pandemic, and provide further evidence that this decline occurred primarily in original research publications. In our survey results, we found that a greater burden of family and domestic responsibilities on women could underlie this effect, consistent with other research on gender inequities during the COVID-19 pandemic.8Krukowski RA Jagsi R Cardel MI Academic productivity differences by gender and child age in science, technology, engineering, mathematics, and medicine faculty during the COVID-19 pandemic.J Womens Health (Larchmt). 2021; 30: 341-347Crossref PubMed Scopus (138) Google Scholar, 9Staniscuaski F Reichert F Werneck FP et al.Impact of COVID-19 on academic mothers.Science. 2020; 368: 724Crossref PubMed Scopus (98) Google Scholar Notably, overall burden experienced during the first year of the pandemic was similar for both women and men, with the majority classifying it as substantial. Likewise, altered patient care was the main concern reported by most respondents irrespective of gender. Although the survey's snowball sampling method and the small sample size reduce the generalisability of our findings and are limitations, our study provides a snapshot of scientific publishing during the early phase of the pandemic. Additionally, we showed a more nuanced view of how women and men personally experienced related burdens during a global stress response. Further research is needed to fully investigate these topics. We hope that the lessons learned from this pandemic will help us avoid these disparities in future crises. MRA and CK contributed equally. CD and RF-S contributed equally. All authors contributed to the study concept and design. MRA, CK, CD, and RF-S did the systematic literature review. MRA and CK extracted authors' gender. KT, TS, CD, and RF-S designed the questionnaire, processed the online survey, and collected the data. MRA, CK, EM, and TS did the statistical analyses. MRA, CK, and RF-S wrote the first draft of the manuscript, which was finalised by EM, TS, and CD. All authors read and approved the last version of the manuscript. TS reports grants and personal fees from AbbVie and Roche, and personal fees from Sanofi, Takeda Pharmaceuticals, and Novartis, all outside of the submitted work. CD has received research grants by Eli Lilly, Pfizer, and UCB; consulting fees from AbbVie, AstraZeneca, Eli Lilly, Janssen, UCB, and Vifor Pharma; payments for honoraria for lectures, presentations, speakers bureaus, manuscript writing, or educational events from AbbVie, AOP Orphan, AstraZeneca, Bristol Myers Squibb, Eli Lilly, Janssen, Galapagos, Merck Sharp Dohme, Novartis, Pfizer, Roche, Sandoz, UCB, and Vifor Pharma; support for attending meetings from AbbVie, AstraZeneca, Bristol Myers Squibb, Eli Lilly, Novartis, and Pfizer; and support for travel from AbbVie, AstraZeneca, and Pfizer, all outside of the submitted work. RF-S has received speaker honoraria from AbbVie, AstraZeneca, and Novartis, and consultant honoraria from AbbVie, AstraZeneca, and GSK. All other authors declare no competing interests. There was no funding source for this study. Download .pdf (2.12 MB) Help with pdf files Supplementary appendix