Lab member Ann Caroline Danielsen is a graduate student at the Harvard T.H. Chan School of Public Health.
Lab member Ann Caroline Danielsen is a graduate student at the Harvard T.H. Chan School of Public Health. By Ryan N. Gajarawala

Investigating COVID-19 Data Through A Feminist Lens

Through their work, the lab hopes to illuminate how gendered experience within society interacts with biology — and it believes that it is critical to consider these variables in intersection, as opposed to isolation, with regards to COVID-19.
By Shelby N. Tzung and Maya M. F. Wilson

In February, the Chinese Center for Disease Control and Prevention reported that men are about twice as likely to die as women from COVID-19. Since then, researchers have been clamoring to understand the perceived sex disparity in COVID-19 cases and mortality rates. In order to explore this relationship further, some clinicians have experimented with using estrogen patches and progesterone injections to boost the immune responses of infected men.

Without commenting on the efficacy of these experiments, Harvard’s GenderSci Lab posits that the emphasis on biological sex differences as the cause for this disparity may be misguided.

Beginning in March, the lab launched their COVID-19 Project, which aims to counter the misconception that gender/sex disparities in COVID-19 case and mortality rates are solely caused by biological factors.

A self-identified “feminist laboratory,” the GenderSci Lab challenges biological essentialism — the notion that biological factors alone determine gendered difference. To avoid the perpetuation of this misunderstanding, the lab uses the term “gender/sex” in their research as opposed to using either term on its own; they aim to steer conversation away from the traditional tendency to separate the biology of sex and the sociality of gender and instead toward a more holistic integration of those interconnected variables.

Through their work, the lab hopes to illuminate how gendered experience within society interacts with biology — and it believes that it is critical to consider these variables in intersection, as opposed to isolation, with regards to COVID-19.

“Sex is not a free floating variable,” says History of Science and Studies of Women, Gender, and Sexuality professor Sarah S. Richardson, the founder of the GenderSci Lab. “It’s situated within the direct biological milieu, the organ, the tissue, but also the species and the environmental and social context. We know that that variable has different meaning across those contacts and that those sorts of comparisons so often overlook, and obscure, really the important differences.”

Richardson founded The GenderSci Lab in 2018 with the goal of bringing together “gender scholars and biomedical scientists to build and validate new theory and methods for the intersectional study of gender/sex in human populations,” the lab’s website says. More specifically, its work centers on innovations in the methodology of operationalizing gender, countering bias, and engaging with the implications of data on policy.

In tracking COVID-19 data, the lab has found ample evidence to suggest that, at the very least, further investigation is needed to fully understand how the interaction of social variables leads to varying case and mortality rates across social groups. Probing the generalization that men are dying at higher rates than women, the lab has discovered significant variance in the gender/sex disparity of case and mortality rates both between different U.S. states and between racial groups.

The lab, for example, found that women and men in North Carolina were dying at approximately the same rate. However, when they looked specifically at the African-American population, they found that women were testing positive for COVID-19 at a rate of 58 percent, compared to 42 percent in men.

This finding highlights one of the main issues with sex essentialism. “The harm is that it may be that the vulnerability relates to a small group of people with a particular trait, but in a particular population, it’s differentially distributed across the sexes,” Richardson explains. Neglecting intersectionality within genders risks overlooking highly vulnerable groups.

Historically, “the hype in sex difference research” has been given too much weight, Richardson claims, and headlines about the current pandemic similarly overemphasize biological essentialism.

“As we were reading these articles and talking amongst ourselves, there was almost this energy — we couldn’t help but think and talk about what this meant given our background and research,” explains another of the lab’s founding members, Marion C. Boulicault, a graduate student at MIT whose work focuses on feminist philosophy in science. “We were worried that the emphasis on the biological was causing researchers and public policy experts to miss important cues from the social world,” she continues. “Social conditions are much more apt to being intervened on than biological conditions.”

Starting the project was not an easy undertaking; data collection on COVID-19 cases in the U.S. was (and largely still is) sparse, sporadic, and negligent in addressing the intersection of different social variables. As a reaction to the lack of useful data, the GenderSci Lab launched an offshoot of the COVID-19 Project — the U.S. Data Report Card, which assigns grades to different states based on how well they report COVID-19 data. The grades are based on the reporting of age, gender/sex, comorbidity status, and race/ethnicity.

As of June 26, only 11 states were reporting comorbidities among fatalities, even though data suggests that 93 percent of COVID-19 deaths are associated with pre-existing conditions. The lab hopes the report card will function as a tool for accountability, encouraging states to be more thorough in their reporting of data.

Lab member Ann Caroline Danielsen, a graduate student at the Harvard T.H. Chan School of Public Health, comments, “without a really fine-grained level of data, it’s hard to then craft a public health response that is sensitive to the needs of people that need it the most and to also understand what the sources of those vulnerabilities are.”

The need for “fine-grain” data is illuminated when gender/sex mortality rates are compared across different states and between different racial demographics. In Massachusetts, women are actually more likely to die from COVID-19 than men, contradicting the idea that men are inherently more vulnerable. Of course, there are also other states like New York and New Jersey where men suffer significantly higher mortality rates than women. What this variance in data suggests is the likely influence of more local and social factors, the lab claims — biology alone would leave little room for these differences between states.

Another finding that has spurred the lab to advocate for more investigation into the interaction of social variables is the difference in case and mortality rates between racial groups. Not only are Black and Latinx Americans being infected at much higher rates than white Americans, but the gender gap within these groups also varies.

For example, data from Georgia and Michigan — the only two states currently reporting COVID-19 mortality data disaggregated by both gender/sex and race/ethnicity — shows that the gender gap within the Black population is much more significant than that of the white population, in addition to the Black population as a whole contracting the virus at higher rates. Further, although Black men are dying at significantly higher rates than Black women, the mortality rate of Black women is still much higher than that of white men.

The generalization that men are dying at higher rates than women, [x explains], may exclude Black women from potential policy solutions, an alarming possibility that the lab hopes to prevent through their data collection and awareness-building efforts.

Richardson has a theory about where and why generalizations based on gender/sex arise: “It’s a pattern that we see frequently,” she says. There is “a major effort by some of the largest funders of bio science research to emphasize sex as a biological variable,” she explains. “If you’re a biomedical researcher, you want a molecule, you want a biological target, you don’t want someone to tell you it relates to social structures. You can’t give a medicine for that.”

But this tendency toward biological essentialism is dangerous, Richardson claims, explaining that it “contributes in a pernicious way to broad-based gender ideologies that suggest that men and women are different, and that this is why we have the inequalities that we have in society.”

Though the lab has been able to generate some hypotheses about COVID-19’s interaction with gender/sex, the researchers want to emphasize that their main objective has been to build awareness for the potential inaccuracies posed by biological essentialism, as well as the need for comprehensive data across multiple intersections of social and biological factors. They have published a guide for responsible media communication about COVID-19 and a teaching module that can be used to spur discussion about the gender/sex disparity misconception.

The goal is progress, not perfection. “There’s never going to be a single perfect way to collect data. You have to make the kind of choices that I think are unavoidably based on values,” Boulicault says. “[The kinds of values that] we, as the lab, think are important to consider about in data collection, analysis, and visualization are about making sure that the data is presented in a way that makes the concerns of communities who are both marginalized — and therefore more vulnerable — more visible,” she says. “We make choices about data that reflect the kind of open society that we want to be.”

The Scoop