Intended for healthcare professionals

Editorials

The gender gap in first authorship of research papers

BMJ 2016; 352 doi: https://doi.org/10.1136/bmj.i1130 (Published 02 March 2016) Cite this as: BMJ 2016;352:i1130
  1. Kathryn M Rexrode, associate professor of medicine and faculty director of Office for Women's Careers 1 2
  1. 1Harvard Medical School, Boston, USA
  2. 2Center for Faculty Development and Diversity, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, USA
  1. krexrode{at}partners.org

Parity of authorship should be a priority for journals, universities, and funding agencies

Women have been attending medical school in numbers equal to or greater than those of men since 1996 in the United Kingdom,1 and in nearly equal numbers in the United States since 2003.2 However, despite parity in medical school admissions, substantial gender differences in rank and leadership remain in academic medicine. The progress of women attaining first and senior authorship of original research publications, the currency of success for promotion in academic medicine, will directly translate into equity of academic advancement for women in the future.

In a linked paper (doi:10.1136/bmj.i847),3 Filardo and colleagues examined the prevalence of female first authorship among original research articles published over the past two decades (1994-2014) in six high impact general medical journals. This study is notable for its rigor, examining six issues a year for all selected journals throughout the 20 year study period and using online searches of authors to identify gender in those with ambiguous names. The authors report some good news—an overall increase in the prevalence of female first authorship from 27% in 1994 to 37% in 2014. However, they also report that female first authorship plateaued in later years, with no further gains between 2009 and 2014. Of additional concern was the decline in female first authorship from 2009 through 2014 compared with 1994 through 1998 noted in two of the six journals considered (The BMJ and the New England Journal of Medicine).3 This study provides worrying evidence of a failure to further advance research authorship by women.

Some people may ask whether we truly need diversity in authorship. Increasing the prominence of women and minority groups currently under-represented in medicine and research brings new perspectives that expand the scope of research and practice. It is no coincidence that women are more likely than men to conduct research on women’s health or gender disparities. Women comprise only 30% of research principal investigators funded by the National Institutes of Health,4 and lower funding rates among women are also evident at the European Research Commission.5 Correcting the serious loss of human capital resulting from unequal participation by, and success of, women in research should be an urgent priority.

What can we do to eliminate the gender gap in research and publication? Authorship is necessary for career progression and is also a symptom of success; it is the culmination of career development, mentorship, funding, and support. Gender inequity must therefore be tackled at the level of journals, universities, and funding agencies by training leaders to understand unconscious bias and by developing and implementing institutional policies that promote gender equity.

Filardo and colleagues found significant differences in the prevalence of female first authorship among high impact general medical journals.3 What factors might explain these differences, and how can journals correct or at least reduce these gender disparities? Journals could start by examining and reporting (systematically) their acceptance rates for research by gender of author, editor, and reviewer to test whether unconscious bias may be affecting publication of research authored by women.6 Masking reviewers to the identity of authors may minimize the effect of unconscious gender (and other) biases and could be evaluated. The genders of the editor-in-chief, other editors, and editorial board members may also influence the kind of topics given priority at particular journals. These factors can also affect the diversity of reviewers and the likelihood of submissions from women first authors.

Journals, universities, and funding agencies must consider how to reduce the internalization of bias in our social environments. We have good evidence that such bias operates in academic medicine and science.7 8 Interventions to reduce implicit gender bias have been shown to lead to demonstrable differences in actions that promote gender equity and an unbiased departmental climate.9

Institutional commitment and gender equity policies at universities and funding agencies ultimately make a substantial difference to women’s career success. Taking a lead from the social sciences, Boyle and colleagues advocate five steps for universities: publish the gender breakdown in key areas, including promotions, appointments, and rewards; embed gender equity issues into working practices; support women’s career progression through the development of promotion criteria that focus on quality rather than quantity; engage men in championing gender equality, including shared parental leave; and celebrate women’s achievements equally, in a public way.10

Eliminating gender bias in reviewing submissions for research funding is an attainable goal. The European Research Commission’s Scientific Council has launched initiatives to improve gender equity that aim for balance in research teams and decision making, as well as integration of gender analysis in research and innovation.11 Funding agencies can equalize representation on review panels, ensure that reviewers have been trained in minimizing unconscious bias, and publish data by gender on funding applications, success rates, and monetary allocation.10

The equal representation of women in research matters for science, for patients, and ultimately for public health. Filardo and colleagues’ study ought to encourage universities, funding agencies, and journals to examine their policies, renew their commitments to gender equity, and reduce the enduring and damaging symptom of lagging female authorship.

Footnotes

  • research, doi: 10.1136/bmj.i847
  • Competing interests: I have read and understood the BMJ policy on declaration of interests and declare the following interests: none.

  • Provenance and peer review: Commissioned; not externally peer reviewed.

References

View Abstract