Sex disparity in academic medical discourse
BMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g4536 (Published 15 July 2014) Cite this as: BMJ 2014;349:g4536
- Nicholas H Neufeld, resident in psychiatry1,
- Bhanu Sharma, graduate student in rehabilitation science2,
- Alexander McGirr, resident in psychiatry3
- 1Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada
- 2Graduate Department of Rehabilitation Science, University of Toronto
- 3Department of Psychiatry, University of British Columbia, Vancouver, British Columbia, Canada
- alexander.mcgirr{at}alumni.ubc.ca
Abstract
Nicholas Neufeld and colleagues look at whether there is sex inequality in the letters published by medical journals
There is sex inequality in science—in over 80% of countries, women are less likely than men to become scientists.1 Even in wealthy nations, women in academia face an uphill battle from an early stage in their careers. They contend with the presumption of inferior competence,23 with lower salaries,4 and they have less research funding.5 It is unsurprising, therefore, that there is a higher rate of attrition as women progress through academia.6
This sex inequality is reflected in scholarly publications. A recent bibliometric study examining over 5 million publications across disciplines showed that fewer than 30% of authors were women.7 The same study also revealed that women were one third as likely as men to serve as first authors, and their work was less likely to be cited.7
Academic medicine is not immune to this sex inequality. Grants awarded to women are lower than those awarded to men.8 Women also have lower salaries, an effect that is not accounted for by productivity, specialty, rank, or working hours among men.9 Further, as in other fields, the published medical literature is, with the exception of psychiatric journals, predominantly written by men.7 This is not attributable to a dearth of women in medical faculties, as this bias occurs despite data from the United States indicating that female medical graduates have been more likely than males to assume medical school faculty positions.10
Research
Research is resource intensive and so involvement in it may be limited by the funding constraints facing women, but published correspondence in medical journals may be less affected by these funding constraints. Clearly, correspondence submitted for publication is subject to considerable selection pressures as a result of the myriad forms of potential comment and criticism. But this sanctioned debate may reflect the stakeholders in academic medical discourse more than the publication of research studies.
Letters to the editor
We were interested in the sex distribution in the debates in medicine. We wanted to characterise debate across medical specialties in some of the most important studies in medicine. To this end, we assessed letters to the editor in the three highest impact general medical journals according to Thomson Reuters’ impact factor—namely, the Lancet, the New England Journal of Medicine (NEJM), and the Journal of the American Medical Association (JAMA).
We systematically assessed a total of 773 letters published in these journals from 1 July 2013 to 31 December 2013. These comprised 575 letters, 195 responses, two journal responses, and one retraction, with global representation.11 To avoid the potential limitations implicit in algorithmic approaches,7 we manually coded the sex of each corresponding author. To do this we searched institutional websites, personal research web pages, and media reports. In six cases, the sex of the corresponding author could not be identified.
Consistent with the broader sex inequality in medical journals, the sex distribution of corresponding authors was 80.4% male and 19.6% female. There was also no significant variation across journals (Lancet 19.2%, NEJM 17.5%, and JAMA 24.7% female).
As trainees aspiring towards careers in academic medicine, we were also interested in the sex distribution among our peers. Using the same methodology, we identified corresponding authors who were graduate students, medical students, postdoctoral researchers, or clinical fellows. To our surprise, a similar sex disparity was observed among trainee corresponding authors (20.0% female; 5/25) despite comparable numbers of men and women in medical schools, according to the American Association of Medical Colleges12 and the British Medical Association’s equal opportunities committee.13
This sex inequality is disconcerting and may have other influences on the culture of debate in medicine, notably on collaboration. Indeed, our data suggest that female lead authors are more collaborative than men. Communications with women as corresponding authors had an average of 2.86 authors, while those with men as corresponding authors had an average of 2.34 authors (Mann-Whitney Z=2.49, P<0.05). Women were also less likely to write publications alone, with 20.1% of female led contributions being single authored, compared with 30.4% of male led contributions (χ2=6.22, P<.05).
We would caution against interpreting our findings to reflect bias at the editorial level.14 Analysis of grant submissions to the National Institutes of Health revealed an analogous sex disparity in funded grants, but further analysis showed there were comparable ratios of submitted to accepted grants for men and women.5 It seems more likely that the sex imbalance in submissions reflects challenges limiting participation by women, such as child care.415 Further, implicit and explicit sex biases may lead female faculty members towards educational, clinical, or administrative academic roles.11
New initiatives
Several promising initiatives have emerged to address the sex disparity in scholarly journals, such as the European Association of Science Editors gender policy committee, aimed at creating journal standards.16 Broader and more comprehensive action may be required, however, to change the cultural and institutional forces contributing to the conspicuous absence of women from academic medical debate. Indeed, our disconcerting findings with respect to trainees suggest entrenched sex biases that may be operating from early stages of training.
The irony of three men calling for additional interest and research on the sex disparity in academic medical discourse has not escaped us, but we believe that the debates in medicine suffer from an under-representation of a major stakeholder.
Footnotes
Competing interests: We have read and understood the BMJ policy on declaration of interests and declare that we have no competing interests.