Female athletes with high testosterone levels are facing even more scrutiny

Testosterone can boost performance but others call international rules "unfair."

caster semenya

As scientists, athletes and governing bodies try to grapple with what exactly testosterone means in the world of female athletics, the fate of several athletes hangs in the balance. Testosterone has become the hormone that, in some cases, literally defines the category of women’s sports. That means that every new study, including one released this Monday, can seriously impact how people of all gender identities compete in the future.

This new paper published in the British Journal of Sports Medicine delivers a rigorously designed study supporting the idea that additional testosterone is advantageous to women. In a sample of 48 athletic women (not elite athletes) testosterone supplementation did increased lean muscle mass and allowed them to run for 21.17 seconds longer during a high intensity running exercise compared to a placebo group.

That’s an 8.5 percent increase in performance — and that’s a game-changing difference in elite sports, as lead study author Angelica Lindén Hirschberg, Ph.D., told Inverse.

“The study shows the magnitude of performance enhancement by testosterone,” Hirschberg explains. “Testosterone levels increased more than 4 times but were still much below the male range. The improvement in endurance performance by the increased testosterone levels was more than 8%, this is a huge effect in sports!”

Caster Semenya
Caster Semenya, middle, races in Berlin in 2011. 

This study falls directly into a very complicated conversation around testosterone in women’s sports. Two of the study authors, including Hirschberg, are members of the International Association of Athletics Federations (IAAF), the governing body that has been tasked with defining who is and is not allowed to compete in the women’s category in track and field at the highest levels. The study was also financed by grants from the International Athletics Foundation, which was founded by the IAAF.

As Hirschberg explains these testosterone-centric results are “highly relevant” to that process. However, some critics argue that they really shouldn’t be.

How can this study be applied to female sports?

Earlier this year, the highest court in international sports, the Court of Arbitration for Sport (CSA), ruled on the case of Caster Semenya, a South African middle-distance runner who has won multiple world championships and claimed gold at the Rio Olympics.

Semenya has a condition called hyperandrogenism, which means that she produces more testosterone than the average woman. The IAAF has since argued that her higher testosterone levels should have prevented her from competing in the female category for her signature event, the 800 meters. 

To compete in the female category for several middle distance events, the IAAF states that female athletes who have differences in sexual development that allow them to produce levels of testosterone in the male range must have testosterone below (5) nmol/L. Women like Semenya, who have naturally higher testosterone, have to reduce their levels by taking drugs for up to 6 months before competition.

Semenya challenged that rule but in April the CSA accepted the IAAF’s line of thinking, which is that the very purpose of the “female category” in sports is to “prevent athletes who lack that testosterone-derived advantage from having to compete against athletes who possess that testosterone-derived advantage.” In its executive ruling, the CSA stated:

“The Panel found that the DSD Regulations [Differences of Sex Development] are discriminatory but that, on the basis of the evidence submitted by the parties, such discrimination is a necessary, reasonable, and proportionate means of achieving the legitimate objective of ensuring fair competition in female athletics in certain events.”

Meanwhile, this new study does suggest that there is a “testosterone-derived advantage.” When the women received testosterone cream, their levels increased from about .9nmol/L to about 4.3 mol/L on average, compared to a placebo group. Their lean muscle increased by 923 grams for the testosterone group, and by only 135 grams for the placebo group.

However, Peter Sonksen Ph.D., a retired professor of endocrinology whose published work discussing the “medical and ethical concerns” of IAFF’s treatment of hyperandrogenism, says this study isn’t a great analog to cases like Semenya’s. He argues that it’s not all that surprising to see an increase in performance because these women have naturally “normal” levels of testosterone — and that difference between natural and not is what matters.

“Giving T [testosterone] to normal women is bound to have anabolic effects — [but] the women with hyperandrogenism do not respond ‘normally’ to T, they are ‘Testosterone resistant,” he tells Inverse. “The women with raised testosterone all have an abnormality in the activation of testosterone so they can’t get the (doping) benefits that normal women do.”

Importantly, no one was accusing Semenya of testosterone doping — she comes by the extra testosterone naturally. But Sonksen’s point is that this study may not actually shed light on how testosterone is actually affecting the way that athletes like Semenya race. Though it may end up informing and supporting the CSA’s May ruling, it may not be as well suited to the task as it seems.

The science behind the IAAF’s ruling

Sonksen was not involved in the current research but has previously criticized other studies used to bolster the IAAF’s testosterone ruling in the past. There is an argument that the IAAF’s ruling is inherently discriminatory, and Sonksen, amongst others, questions the science that has allowed these rules to exist.

For instance, one key study study that speaks to the existence of the “competitive advantage”, was led by one of the co-authors of this current paper, Stéphane Bermon, Ph.D. It concluded that female athletes who have significantly higher testosterone levels have competitive advantages in several middle-distance events based on testosterone levels and competition results from the 2011 and 2103 IAAF World Championships.

However, in a blog post written for the  British Journal of Sports Medicine’s website, several bioethicists critiqued that paper calling it “severely flawed.” 

caster semenya, running,
Caster Semenya wins the women's 800 Meters during the Nike Prefontaine Classic at Stanford University Palo Alto, CA. 

Sonksen says he is unimpressed with the state of the science behind the rules that the IAAF has applied to female athletes, calling it “rubbish of dubious ethical value.”

“Sadly IAAF has poor scientific advisors and have ‘painted themselves into a corner’ with their stupid and unfair rules,” he summarizes.

Using testosterone to define women’s sports

Still, as this current study can attest, there is demonstrable evidence that testosterone is linked to athletic performance.

Because the CSA, at least for now, has ruled that testosterone (or lack thereof) is a defining feature of women’s sports, results like these will continue to hold powerful sway. However, when it comes to deciding whether the testosterone advantage is inherently unfair and something that the IAAF should protect against, even these study authors argue that hormones aren’t the only thing that we should take into account.

“It has certainly increased our scientific knowledge,” Hirschberg says. “But how it will be applied is another question because the Regulations must consider also other aspects than the scientific one.”

Partial Abstract:

Results: Serum levels of testosterone increased from 0.9 (0.4) nmol/L to 4.3 (2.8) nmol/L in the testosterone supplemented group. TTE increased significantly by 21.17 s (8.5%) in the testosterone group compared with the placebo group (mean difference 15.5 s; P=0.045). Wingate average power, which increased by 15.2 W in the testosterone group compared with 3.2 W in the placebo group, was not significantly different between the groups (P=0.084). There were no significant changes in CMJ, SJ and knee extension. Mean change from baseline in total lean mass was 923 g for the testosterone group and 135 g for the placebo group (P=0.040). Mean change in lean mass in the lower limbs was 398 g and 91 g, respectively (P=0.041).

Conclusion: The study supports a causal effect of testosterone in the increase in aerobic running time as well as lean mass in young, physically active women.

Editor’s Note: This article has been updated to better reflect the characterization of the IAAF’s rules as they apply to female athletes with Differences in Sexual Development.