Three Years In: “Sex as a Biological Variable” Policy in Practice - and an Invitation to Collaborate

By: Annika Gompers

The GenderSci Lab has been a leader in engaging with the NIH’s 2016 policy requiring consideration of sex as a biological variable (SABV) in all NIH-funded preclinical research on vertebrate animals and human cells and tissues(1, 2) The stated aim of the policy is to “improv[e] the health of men and women,” as the “over-reliance on male animals and cells may obscure understanding of key sex influences on health processes and outcomes.”(3) Three years later, what have been the impacts of the policy on scientific research?

To answer this question, this year I conducted in-person, semi-structured interviews with nine basic science researchers from three different laboratories on the East Coast of the United States that use animal and tissue models to study metabolic disease. I transcribed the full interviews and then conducted thematic analysis of the data using the NVivo software to help organize my coding.

The researchers I interviewed were unsure how to actualize the SABV policy in practice. In the context of translational preclinical research, immovable practicalities - availability of materials and, to some extent, cost - constrained scientists’ ability to consider the potential variability associated with sex as a biological variable, as mandated by the NIH. This is an important finding that sheds light on the realities of translating the construct of “sex as a biological variable” into research practice.

Preclinical research: Managing complexity to increase generalizability

In the US biosciences, there is a growing push to consider and account for variability in order to make biomedical findings generalizable to all (or at least more) humans; this is the motivation behind the NIH SABV policy.(4) Much of the scientific community applauded this policy and agreed with its premise that including sex as a biological variable will drive discovery of sex differences that may be important to health.(5, 6 , 7) Many of the researchers I interviewed echoed these sentiments, emphasizing the importance of considering differences in the population - primarily sex - in order to ensure that findings are “universally applicable,” “common to all,” or “generalizable to the entire population.”

However, adding sex as a biological variable increases complexity. Historians, philosophers, and sociologists of science have shown that a central element of successful scientific research is the reduction of complexity in order to produce coherent findings.(8, 9, 10, 11) Science is about making sense of a complex world; reduction of this overwhelming complexity via idealization is therefore necessary (and is indeed ubiquitous) in order for science to function. Idealizations are simplified models that allow scientists to understand phenomena, such as when population geneticists assume infinite population sizes.(12) Preclinical research in particular must rely on idealizations because it is by nature translational. That is, it uses simplified animal and cellular models of human systems and conditions with the aim of generalizing to human applications. For basic scientists, translation is already and always in mind, motivating them to use idealized models that will enable valuable findings, such as using C. elegans to study neurological disorders, or mice to study diabetes. In the face of considerable complexity, and with the aim of developing therapies for human disease, scientists decide to selectively exclude unpredictable and uncontrollable material from their data.(13, 14)

Sex-based biology is a particularly complex field of scientific inquiry. Feminist scholars have long noted that sex and gender are intertwined in humans and both affect health, (15, 16, 17, 18, 19) making studies focused on the impacts of sex on health a tricky endeavor. These difficulties are exacerbated when studies are based on animal and cell models, which are idealized models of human biology and especially of human sex.(20, 21) The “sex” of cells and tissues refers to the presence of XX or XY chromosomes, but this is just one part of sex determination and differentiation in humans. Animal models do not well represent factors that are relevant to the health of human males and females, such as extended life span and menopause. And none of these models can account for the lived experience of gender in humans.(22) The scientists I spoke with recognized the tenousness of the translation between their findings and human health, and that human sex and gender are not very well modeled by animals and cells. Adding sex as a biological variable to standardized research models in science may actually decrease rather than increase their generalizability by mandating a focus on sex differences where they may not be relevant.

Pragmatics constrain which variables can be considered in scientific research

In my interviews, researchers expressed uncertainty as to how to implement the NIH requirement to consider sex as a biological variable, pointing to a variety of immovable materialities. Scientists repeated again and again that their experiments were in large part constrained by practicalities of various sorts. These practicalities were often used to explain why they don’t or can’t account for sex in their research.

For example, for scientists who work with human donations, availability of tissues or cells is a huge factor. One of the labs I studied relies on pancreatic cells from deceased human donors, which are a scarce resource with unpredictable availability. As the researchers put it, “We just get what we’re offered,” “It’s so rare that we get donors, we take everything,” “It’s irregular and difficult to plan,” “We pretty much just take them whenever they’re available,” and, “The problem is, we know the gender of the donor, but we cannot control it.” One lab studied adipocytes from discarded human fat tissue, which primarily comes from cosmetic surgery. This means that these scientists disproportionately use female cells, since women constitute the majority of cosmetic surgery patients. And even when scientists use animal models instead of tissues from human donors, the availability of mice can be limiting. As one Principal Investigator (PI) said, “From a purely practical standpoint, a lot of times it’s like, okay, what mice do you have available? Are they male or female? And then, you know, we just do it in whatever mice we have enough to do it in.” The inconsistent and limited availability of human tissues and mouse cohorts constrains the ability of the scientists reliant on these materials to balance sex in their experiments.

Cost was also cited as a significant practical concern, as most of the participants recognized that research costs can rise when they have to balance the sex of preclinical materials. One PI described the considerations:

 In all fairness, if I do a study using only male animals, it costs one amount of money – if I want to use both male and female, that’s going to double the expense, and who’s going to pay for it? ... So it’s really a balancing act – how do you get the most bang for your buck, as a researcher, if you want to answer an important question? And, you know, if you don’t do a sufficient number of people or, you know, of any one thing, you learn nothing. So, you know... those are the issues we need to address, not whether or not we need to test all of these things – we do. But, how we are going to handle the cost?

This PI, along with most of the other scientists, recognized that addressing variability is important but must be balanced with the need to produce usable, translational findings, and the fundamental need to fund the research. Several scientists concurred that the cost of incorporating more variables is burdensome; one graduate student stated, “I think it’s definitely a legitimate concern, especially for smaller labs that have trouble getting funding,” and another said, “I suspect that we’re not going to be quite as efficient if we do everything twice.” However, it should be noted that some participants expressed that cost should not prevent researchers from incorporating SABV into research practices, commenting: “I think that the benefits still outweigh the costs,” “It’s probably worth it,” and “It’s not terribly onerous.”

While my participants recognize that considering sex can be important for the generalizability of results, in practice it is often impractical if not impossible to incorporate sex as a variable into biomedical research. Such a finding is consistent with scholars who have long looked at science as practice(23) and observed how practicalities - as mundane as the availability of materials - are often central to the reduction of complexity into “doable problems.”(24)

Next steps: SABV in practice

 This pilot study was designed to impartially investigate scientists’ thoughts and approaches to SABV in preclinical research. A deeper understanding of how operationalizing sex as a biological variables works - or doesn’t work - at the level of these pragmatics is essential.

The GenderSci Lab is beginning a major research project to examine how SABV is implemented in practice. We hope to work closely with a small sample of labs to learn more about their fine-grained research practice and what it looks like or would look like to consider SABV in preclinical research, as the NIH policy intends. Additionally, we’ll be looking at grant applications, protocols, publications, and other research documents to better understand the practicalities of considering sex in preclinical biomedical research. We warmly invite interested investigators to reach out to the GenderSci Lab to discuss collaborating on the next phase of this research.

References:

[1] Clayton, J. A., & Collins, F. S. (2014). Policy: NIH to balance sex in cell and animal studies. Nature News, 509(7500), 282. https://doi.org/10.1038/509282a

[2] Richardson, S. S., Reiches, M., Shattuck-Heidorn, H., LaBonte, M. L., & Consoli, T. (2015). Opinion: Focus on preclinical sex differences will not address women’s and men’s health disparities. Proceedings of the National Academy of Sciences of the United States of America, 112(44), 13419–13420. https://doi.org/10.1073/pnas.1516958112

[3] National Institutes of Health. NOT-OD-15-102: Consideration of Sex as a Biological Variable in NIH-funded Research, (2015).

[4] Ibid.

[5] Klein, S. L., Schiebinger, L., Stefanick, M. L., Cahill, L., Danska, J., de Vries, G. J., … Zucker, I. (2015). Opinion: Sex inclusion in basic research drives discovery. Proceedings of the National Academy of Sciences of the United States of America, 112(17), 5257–5258. https://doi.org/10.1073/pnas.1502843112

[6] Mazure, C. M. (2016). Our evolving science: studying the influence of sex in preclinical research. Biology of Sex Differences, 7(1). https://doi.org/10.1186/s13293-016-0068-8

[7] Shansky, R. M., & Woolley, C. S. (2016). Considering Sex as a Biological Variable Will Be Valuable for Neuroscience Research. Journal of Neuroscience, 36(47), 11817–11822. https://doi.org/10.1523/JNEUROSCI.1390-16.2016

[8] Latour, B., & Woolgar, S. (1979). Laboratory life: the construction of scientific facts. Princeton, N.J.: Princeton University Press.

[9] Nowotny, H. (2005). The Increase of Complexity and its Reduction: Emergent Interfaces between the Natural Sciences, Humanities and Social Sciences. Theory, Culture & Society, 22(5), 15–31. https://doi.org/10.1177/0263276405057189

[10] Wynne, B. (2005). Reflexing Complexity: Post-genomic Knowledge and Reductionist Returns in Public Science. Theory, Culture & Society, 22(5), 67–94. https://doi.org/10.1177/0263276405057192

[11] Potochnik, Angela. (2017). Idealization and the Aims of Science. Chicago: University of Chicago Press.

[12] Ibid.

[13] Nowotny 2005.

[14] Wynne 2005.

[15] Butler, J. (1993). Bodies that matter: on the discursive limits of “sex.” New York, New York ; London: Routledge.

[16] Grosz, E. A. (1994). Volatile Bodies: Toward a Corporeal Feminism. Indiana University Press.

[17] Fausto-Sterling, A. (2000). Sexing the body: gender politics and the construction of sexuality (1st ed.). New York, NY: Basic Books.

[18] Krieger, N. (2003). Genders, sexes, and health: what are the connections—and why does it matter? International Journal of Epidemiology, 32(4), 652–657. https://doi.org/10.1093/ije/dyg156

[19] Fausto‐Sterling, A. (2005). The Bare Bones of Sex: Part 1—Sex and Gender. Signs, 30(2), 1491–1527. https://doi.org/10.1086/424932

[20] Ritz, S. A. (2016). Complexities of Addressing Sex in Cell Culture Research. Signs: Journal of Women in Culture and Society, 42(2), 307–327. https://doi.org/10.1086/688181

[21] Richardson et al. 2015.

[22] Ibid.

[23] Pickering, A. (1992). From science as knowledge to science as practice. In Science As Practice and Culture (pp. 1–26). Chicago: University of Chicago Press.

[24] Clarke, A. E., & Fujimura, J. H. (1992). What Tools? Which Jobs? Why Right? In A. E. Clarke & J. H. Fujimura (Eds.), The Right Tools for the Job, At Work in Twentieth-Century Life Sciences (pp. 3–44). https://doi.org/10.1515/9781400863136