January of 2005, at a conference sponsored by the National Bureau of Economic Research, Lawrence Summers, then the president of Harvard, sought to address the lack of tenured female faculty in university science departments. After promising to “provoke” his audience, Summers said that a discrepancy in “intrinsic aptitude” between men and women—a biological asymmetry—might be partly to blame. At the time, I was a pre-med student at Harvard, majoring in biochemistry. Surrounded by brilliant women in my classes and in the lab, I didn’t place much weight on Summers’s remarks. But, in the debates that ensued, it occurred to me that, beyond the basics, I wasn’t actually familiar with the biological differences between women and men—and I suspected that many of my colleagues weren’t, either. Sex and gender (the former refers to biology, the latter to social attitudes and behaviors) were rarely discussed in the context of scientific research.
As I learned later, such discussions are important; in medicine, they can be a matter of life and death. Too often, though, they are neglected. Women die from cardiovascular disease at higher rates than men but have historically been underrepresented in clinical trials. Men sustain a third of osteoporosis-related hip fractures but are widely underdiagnosed and undertreated, perhaps because osteoporosis research tends to focus on women. The problem extends to pharmaceuticals, too. Of the ten drugs that were recalled in the United States between 1997 and 2000, eight were found to pose greater health risks to women than to men.
Over the years, policymakers have proposed various solutions. In 1993, the National Institutes of Health established guidelines to promote the inclusion of women as subjects in clinical research, and effective last year the agency added a requirement that preclinical research include sex as a biological variable. In 2010, the N.I.H.’s Canadian counterpart began requiring that all grant applicants indicate whether sex and gender were accounted for in their study, and if not, to provide a rationale. Meanwhile, the N.I.H., the European Commission, and other organizations have taken steps to increase the number of women pursuing scientific research. But could the two goals—improving the quality of research and increasing the diversity of researchers themselves—in fact be related?
That’s the question behind a new study in the journal Nature Human Behaviour. Led by the Stanford-trained sociologist Mathias Nielsen, a team of Danish and American researchers surveyed a million and a half papers from around the world, published between 2008 and 2015, and sorted them by authorship, paying particular attention to papers on which a woman was listed as the first author (signalling that she made major contributions to the study) or the last author (signalling that she likely designed the study). Nielsen and his collaborators also recorded the over-all proportion of women in each author group. According to their analysis, the more involved women were in the creation and execution of a study, the more that study accounted for “gender-related and sex-related factors.”
The findings address two important pieces of the sex-and-gender problem, both of which, Nielsen told me, “are important to emphasize, for different reasons.” The first is that women remain generally marginalized in medical research, accounting for just thirty-five per cent of all the authors that Nielsen’s team surveyed and a mere twenty-seven per cent of last authors. This, in turn, may have implications for the validity and applicability of the research. “Most universities or research organizations are on board to try to increase the number of women, and they think that that’s somehow going to solve all the problems,”Londa Schiebinger, one of Nielsen’s co-authors and his former mentor at Stanford, said. “But universities typically don’t understand how to fix the knowledge aspect. And I think that’s what’s so important about our study: it shows a link between the participation issue, the diversity on teams, and excellent knowledge outcomes.”
The identification of a link does not imply cause and effect, of course. But it does suggest that a more diverse author group contributes to a more successful research methodology. In an opinion piece published earlier this year in theProceedings of the National Academy of Sciences, Nielsen, Schiebinger, and others refer to this as the “innovation dividend.” They cite a study from 2010 that found that a team’s ability to perform various tasks effectively—solving visual puzzles, making moral judgments—is directly related to its demographic makeup. In particular, teams with more female participants achieved “greater equality in participation,” perhaps because women “exhibit higher levels of social perceptiveness” than their male counterparts. Tellingly, this boost in “collective intelligence” was less pronounced in teams made up entirely of women. Diversity, it seems, was the key.
Another study, published in February in jama Internal Medicine, further illustrates the importance of sex and gender both as research variables and as mediators in how medical professionals think and interact with others. The study examined mortality and hospital-readmission rates for more than a million elderly Medicare beneficiaries, finding that those who had been treated by female physicians had consistently better outcomes than those treated by men. The study did not propose an explanation for this disparity, though it did note that female doctors “are more likely to practice evidence-based medicine, perform as well or better on standardized examinations, and provide more patient-centered care.” Far from assigning an “intrinsic aptitude” to anyone, the study raises useful questions about how women and men differ, and how thesedifferences affect patient care. And this, ultimately, is the main reason to study sex and gender—because, for both patients and professionals, they matter.
At a time when women are still struggling for equal pay, for freedom from sexual harassment and domestic violence, and for greater representation in business, politics, science, and medicine—in short, for the same rights that most men have long enjoyed—it may seem counterintuitive for scientists to emphasize gender- and sex-related differences. To Schiebinger, though, this is an area where good politics and good medicine meet. “Democratic societies have always kind of assumed that you need sameness in order to have equality,” she told me. “And intellectually, that’s just not the case. We should be able to have differences and still achieve political equality.”