Impossible “Choices”: The Inherent Harms of Regulating Women’s Testosterone in Sport

The 2018 regulations are substantively similar to the 2011 regulations. Analysis of the 2011 regulations thus remains relevant (see, for example, Karkazis et al. 2012; Schultz 2012; Viloria and Martínez-Patiño 2012; Cooky and Dworkin 2013; Sönksen et al. 2015; Bavington 2016; Karkazis and Jordan-Young 2018).

An IAAF “fact sheet” states: “The IAAF’s new regulations are based on a range of published research, expert review and most importantly, evidence collected over 15 years. The evidence and data, some of which is not able to be shared publicly due to confidentially [sic], but has been shared with the Court of Arbitration for Sport (CAS), shows elevated testosterone levels give athletes the biggest performance advantage in the events from 400m to 1 mile. As wide a range of evidence as possible has been made available where it does not breach individual confidentiality” (IAAF 2018b, 1). In fact, a single study by IAAF researchers provides the primary published evidence for this claim (Bermon and Garnier, 2017). The study’s methods have been extensively critiqued (Karkazis and Meyerowitz-Katz 2017; Sönksen et al. 2018; Menier 2018; Franklin, Ospina Betancurt, and Camporesi et al. 2018) leading independent researchers to request that the IAAF release the study’s raw data for re-analysis (Pielke 2018). Following the release of a subset of those data, the independent researchers calculated errors in the data ranging from 17–33 per cent for four of the regulated events (400m, 400H, 800m, and 1500m) and called for the study to be retracted (Pielke 2018; Longman 2018). Days before the independent researchers submitted their re-analysis, Bermon and colleagues released their own re-analysis of the data (Pielke 2018; Bermon et al. 2018).

Just as this article went to press, IAAF-affiliated researchers published a review article with their evidence for this claim (Handelsman, Hirschberg, and Bermon 2018). They argue that there is a “reproducible dose-response relationship between circulating testosterone and muscle mass and strength as well as circulating hemoglobin in both men and women. These dichotomies largely accounts [sic] for the sex differences in muscle mass and strength and circulating hemoglobin levels resulting in at least an 8–12% ergogenic [performance] advantage in men” (Handelsman, Hirschberg, and Bermon 2018, 2). Conceding the evidence is “incomplete,” the researchers nevertheless conclude it is “highly likely that the sex difference in circulating testosterone of adults explains most if not all the sex differences in sporting performance” (Handelsman, Hirschberg, and Bermon 2018, 22). Noting that the data on women with intersex variations is “sparse and mostly uncontrolled” (Handelsman, Hirschberg, and Bermon 2018, 22), they extrapolate that elite women athletes with naturally high testosterone will have a male-typical advantage (~10 per cent) over their fellow competitors. There is no evidence to support this claim.

Throughout this paper we use intersex variations in lieu of the more contested nomenclature “differences of sex development” and “disorders of sex development.”

Women can also challenge a decision made under the regulation, such as a finding of material androgenizing effect.