Recalled drugs, unsafe products and even environmental chaos are just some of the consequences of research that doesn’t consider sex and gender, says Londa Schiebinger. That’s why Schiebinger, who studies gender and science at Stanford University in California, is helping funders convince researchers to analyse the effect of these factors in their studies.
On 25 November, the European Commission — one of the world’s largest research funders — said that it aims to make sex and gender analysis mandatory in the research it funds through its €85-billion (US$100-billion) Horizon Europe programme, which is set to begin in 2021. Its policy would apply to all disciplines, except topics for which the commission decides such analyses wouldn’t be relevant, for example, pure mathematics. The commission will ask researchers to consider these factors at every stage of their work — from study design to data collection and analysis.
The move strengthens a policy the commission began to implement in 2013. By 2020, it asked research applicants in about one-third of fields to account for sex (biological characteristics commonly used to classify people as male, female or intersex) and gender (socially constructed roles, norms and identities, not necessarily binary or aligned with a person’s sex) in their research. It was one of the first funders outside of health research to do so. But fewer researchers than expected did the analyses.
The strengthened policy is a result of recommendations made in the commission’s second report on Gendered Innovations, published today and produced by a 25-person expert group that Schiebinger chaired. The report provides guidance for how researchers can incorporate sex and gender analysis across the gamut of research topics Horizon Europe will fund, from finance to agriculture.
Nature spoke to Schiebinger about the group’s work.
How do you convince people of the need for sex and gender analysis in research?
Our iconic example of failure when you don’t do this analysis is that between 1997 and 2001, ten prescription drugs were withdrawn from the US market, and eight of those were more dangerous for women than for men. When drugs fail, you’re losing a lot of money and people are suffering and dying. From preclinical trials in cells and animals through to human clinical trials, you have to collect data on males and females and analyse them separately. It’s also important to look at age and genetic ancestry, and at drug use in pregnant women.
This kind of analysis is not about women — it’s about getting the research right, and it benefits everyone. In osteoporosis, for example, men have been neglected because it is mostly seen as a disease of menopausal women.
Why have researchers been slow to embrace these analyses?
Awareness is growing, but by and large, researchers don’t know how to do this kind of analysis very well. I’m kind of shocked and pleased that in my lifetime change is coming about, as when I started out in the 1980s, nothing was happening.
This is your second report on the topic; the first was in 2013. What’s changed?
The biggest change in our focus is on intersectionality. That means not just considering race and ethnicity and how this intersects with sex and gender, but also what’s the impact of age, geographic location and socio-economic status. The other one is to think of gender less in binary terms, and to think about gender-diverse people. We’re also giving much more field-specific guidance.
What mistakes do researchers make in these analyses?
The biggest mistake is simply ignoring sex, gender and intersectionality. But another is to not distinguish between biological sex and sociocultural gender. Many European languages, such as German, don’t even have a word for gender, but use the English. You also have to realize that gender is very specific to ethnicity, age group and the culture. Researchers need to be aware that they need to get the right variables, collect their data correctly and do the analysis well.
What needs to be done outside of funding agencies?
Some peer-reviewed journals have policies asking for sex and gender analysis, if it’s relevant in that field. But I would love to see some engineering journals get involved. Another major problem is that universities don’t teach this type of analysis in core technical classes in science and engineering, or even in medicine. In engineering, we don’t design crash-test dummies to consider the frail bones of older people, women in particular.
There are some movements — for example, Harvard University and Stanford now have a course called Embedded EthiCS in computer science. They understand that artificial intelligence can cause so many problems when it leaves groups of people out. They are asking students to think about social issues and outcomes when they’re learning about algorithms.
How important is it to analyse the effect of sex and gender in the coronavirus pandemic?
Essential. A case study on the COVID-19 pandemic in our report finds that there are sex differences in how people respond to the virus, in viral reproduction, viral receptors and antibody production — but that gender is also very important. We see that many more men are dying and this has to do with gender norms and behaviours — for example, more men smoke, and their hand-washing rates are generally lower. And women make up the majority of health-care workers, so they will be exposed more.
Are there areas of research where people might be surprised that sex and gender analysis is essential?
It’s very important to do sex analysis of some marine organisms, because for some, sex is determined by temperature. If the ratio of males to females or hermaphrodites gets out of whack, that can lead to extinctions. Our report includes a fascinating study from Australia, where they found that the turtles in the north of the Great Barrier Reef were 99% female, whereas in the cooler southern part, it was about 67% female. So it’s important that we understand how global warming is skewing the sex ratios, so that we can efficiently manage ecosystems.