In May 2014, the NIH released a new policy mandate requiring equal representation of “sex” in all preclinical research (the research on animals and cellular materials that occurs before clinical trials in humans). I am part of a Harvard-based working group of scientists and science studies scholars that spent the last year reviewing the evidence for this new policy. In a recent opinion piece in the Proceedings of the National Academy of Sciences, we argue that while this policy is well intended, if the goal is to address women’s health inequities, it is ultimately unlikely to be effective. Here, at the invitation of Gap Junction Science, I delve into some of our criticisms in a more substantial way for a feminist science studies audience.
First, a bit of history. Once upon a time, white males were taken as the normative “species type” and clinical medical trials largely included only white men as participants. As a result, occasionally women using medications would have serious and unanticipated adverse reactions to drugs that had proved safe in men. In the 1980s feminists and/or medical scientists began agitating to improve the representation of women in medical research. This led the NIH to mandate the inclusion of women in clinical trials and to establish the Office of Women’s Health Research in the early 1990s. Yet today, despite adequate representation of women in clinical trials, women continue to disproportionately report adverse drug reactions. These drug reactions are a serious public health concern, as they are responsible for up to 5% of emergency room visits and roughly 100,000 deaths a year in the US.
Why, when women are approximately equally represented in clinical trials, do women continue to report a higher percentage of these reactions? According to proponents of the new NIH mandate, it is largely due to the unequal representation of females in preclinical research; i.e., with animal models and cellular material. Similar to human studies of yore, in some areas of preclinical research, research material is predominately male, and the donor sex of cellular material is often not reported. The new mandate suggests that unequal representation in preclinical materials is lurking behind the health disparity seen in men and women’s adverse drug reactions. NIH policies must serve human health and the explicit policy aim of the new mandate is to improve health outcomes between men and women in the realm of adverse drug reactions.
In our opinion piece, we argue that this policy is not likely to achieve that goal for two primary reasons. First, the non-hypothesis-driven study of sex differences in all preclinical research lacks conceptual clarity about just what sex is. Animal and cellular materials that poorly model human sex run the risk of generating non-replicable or irrelevant findings. For some preclinical materials, such as cell lineages, it is unknown whether XX and XY lineages are valid models of sex differences in humans (see this older post by Stacey Ritz!). For example, a non-negligible percent of XY cell line material has lost the Y chromosome during repeated replication. To what degree do these materials model human male biology? Animal models of sex difference are also challenging – in social animals housing condition produces significant variability in study outcomes, and male animals are often housed at lower densities than females. When modeling human sex differences in the lab, it is not enough to simply include XX and XY tissues or male and female animals and then attribute any differences found to intrinsic sex. Depending on the disease or condition under investigation, researchers analyzing sex-related factors may need to consider whether hormone exposures are relevant, the appropriate age or developmental stage to discern differences, and other variables. A broad mandate requiring study of sex in preclinical materials overlooks these considerations; instead preclinical sex difference models should be hypothesis-driven and validated with respect to the research question at hand.
Second, human sex-linked health disparities may be attributable to sex, gender, and the interaction of the two; focusing solely on sex variables, the mandate presents an impoverished approach to advancing scientific understanding of health disparities between men and women. Highlighting the need for research on gender alongside sex is a critical contribution of feminist science studies scholars. The case of health disparities in adverse reactions is an excellent example of the need for greater intellectual and institutional commitment to sex-gender research, and here I’ll develop this point in greater detail than our short PNAS piece allowed.
We know that gender influences how women and men interact with the medical system. For instance, women are more likely than men to regularly visit a doctor. The reporting of adverse drug reactions is voluntary, and fascinating recent data released by the FDA indicates that for women, nearly half of the adverse drug reactions on record are self-reported to a doctor, while for men, the majority of reactions are reported by a healthcare professional or other third parties. Similar to women’s greater likelihood to regularly visit a doctor, these data could indicate a greater propensity by women to report the experience of an adverse drug reaction.
Polypharmacy is another gender-related variable influencing higher rates of adverse drug reactions in women. Women are more likely than men to be on multiple medications, a major risk factor for adverse drug reactions. It is possible that the majority of serious drug reactions could be avoided simply by adhering to known information regarding contradicting prescriptions. Oral contraceptives are involved in significant numbers of adverse drug reaction cases. Recent research also indicates that SSRIs (a popular type of antidepressant) increase the risk of serious gastrointestinal adverse effects when combined with non-steroidal anti-inflammatory drugs (such as aspirin and ibuprofen). Nearly 1 in 4 US women take antidepressants, more than twice the rate of men, and we have long known that doctors are more likely to diagnose women than men with psychological diseases, even given the same clinical presentation.
Furthermore, there is evidence that some of the sex differences in adverse reactions are due to differences in outcomes such as allergic skin reactions. Allergic skin reactions are especially interesting to consider in light of known sex and gender differences. Outside of the context of adverse drug reactions, women generally report more allergic dermatitis reactions than men, a disparity linked to differences in exposure and contact factors, rather than intrinsic sex differences. Research on eczema has also shown that women experience greater distress from active eczema episodes, and that men are more likely to leave their eczema untreated. How these gender-inflicted differences in exposure factors, distress experience, and treatment probability in non-drug associated allergic dermatitis relate to the greater reported incidence of allergic dermal adverse drug reactions in women is unknown.
This is not to say there are no sex-intrinsic differences in adverse drug reactions, including those potentially related to drug metabolism. On average, there are known differences in enzymatic activity between men and women, though these are not consistent in all studies and may vary by age and ethnicity. We emphatically support basic and preclinical research on these types of sex differences. In our article we specifically argue for more validation research on how animal and cellular materials can model human sex differences. But we believe that the wholesale introduction of sex as a variable into all preclinical research will produce results of dubious meaning and introduce conceptual muddle into understanding sex and gender disparities in health outcomes. In an atmosphere of limited funding, this mandate will capture institutional resources and energy, which come at an opportunity cost to research on other types of questions. On the question of why women have greater rates of adverse drug reactions, we sorely need studies that examine how both sexed and gendered factors interact in the lives of men and women to create this health disparity.
Heather Shattuck-Heidorn is a Ph.D. candidate in Human Evolutionary Biology, with a secondary field in Women and Gender Studies, at Harvard University. Her research focuses on immune function in humans, and sex and gender in science. You can contact her at email@example.com, HEB Department, 11 Divinity Ave. Cambridge, MA 01239.