Skip to Main Content

This week track and field athletes from all over the world are gathering to compete in the World Championships, an event second only to the Olympics in its level of prestige. Two of the competitors, South African Caster Semenya and Indian Dutee Chand, will represent their countries while on a quest for gold and glory that started last summer in Rio. But their future careers, and those of other women like them, are again in question as the sport’s governing body attempts to reinstate a limit on female athletes’ testosterone levels.

The limit dates back to 2011, when the International Association of Athletics Federation (IAAF), which oversees track and field events, first created the rule. Naturally produced high testosterone, the group ruled, provided female athletes with an unfair advantage akin to doping — and so, to compete, a woman over the limit would have to lower her testosterone, through medication or surgery, or prove that she was not sensitive to its effects.


That rule was put on pause after a challenge was brought in 2015. The arbitrators gave the IAAF a two-year window to provide evidence as to why the rule should persist, or else it would be permanently struck down.

Now, just as the races begin in London, the track authorities have taken their starting positions for a renewed legal fight, filing new evidence with the court that they say supports a testosterone limit — and setting up a complicated battle about sex, physiology, and ethics destined to play out on the largest stage in sports.

A running controversy 

The case in consideration by the Court of Arbitration for Sport centers on Chand. The 21-year-old runner was a rising 100-meter star in India in 2014, when the Indian affiliate of the IAAF barred her from competing. Blood tests showed that her testosterone was above the 10 nanomoles per liter limit, the level deemed to be the lower end of the “male range.”

In 2015, she appealed that decision to the Court of Arbitration for Sport — an international high court that settles sports-related disputes. The court heard from various ethical and scientific experts. Ultimately, its ruling hinged on whether female athletes with high testosterone — a condition called hyperandrogenism — experience a boost in performance equivalent to the average athletic performance gap between men and women. It quantified that gap as a 10-12 percent difference.


On July 27, 2015, citing a lack of evidence for that magnitude of performance boost, the court temporarily suspended the policy. However, the arbitrators gave the IAAF two years to submit more evidence to support their claim.

In recent months the IAAF has done just that, according to a recent press release and an expert witness who testified in 2015. Now, both sides will have an opportunity to respond to evidence in a back-and-forth that could drag on for months, and will culminate in the court deciding whether or not to strike down the policy once and for all.

Caster Semenya
South Africa’s Caster Semenya competes at the 2015 IAAF World Championships in Beijing in August 2015. OLIVIER MORIN/AFP/Getty Images

The science of sex

The problematic mapping of gender onto biology is a story as old as sports: Ever since women have been able to participate, athletic organizations such as the IAAF and the International Olympic Committee have struggled with how to define the female category, and how to decide who is eligible to compete in it. Over the years, these organizations have deployed policies requiring female athletes to undergo everything from genital examinations to chromosome testing to prove their eligibility to compete as women, and even mandated that women who failed the tests undergo medical or surgical treatments to regain their eligibility, raising serious ethical questions in the process. And while such widespread “sex-testing” was officially abandoned in the 1990s, elite female athletes continued to be flagged, often based on physical appearance, and subjected to physical examinations and other medical tests to confirm their eligibility.

The reason for a testosterone policy in the first place centers on the relationship between testosterone and biological sex. Testosterone is a key hormone for the development of male reproductive organs, and also promotes increased muscle and bone mass and the growth of body hair, among other things. Consequently, most testosterone is produced in the testes, and men tend to have a lot of it, generally between 10 and 40 nm/L. But women have testosterone too, typically in the 0.5 to 2 nm/L range, produced by the ovaries and adrenal gland. Testosterone over 2 nm/L is often considered hyperandrogenism, though there is no hard and fast cutoff.

According to Dr. David Cohen, an endocrinologist at Beth Israel Deaconess Medical Center not involved in the case, not all hyperandrogenism is created equal. The most common hyperandrogenism is called polycystic ovary syndrome (PCOS), in which the ovaries produce excess testosterone, and which is unlikely to increase testosterone above the 10 nm/L level. Rather, Cohen said, women above the cutoff who aren’t doping likely have one of two things going on: a tumor in the ovaries or adrenal gland, or what’s called a disorder of sexual differentiation — a genetic anomaly. One such anomaly, for instance, is 5 alpha reductase deficiency, a rare genetic condition in which a woman can have XY chromosomes and internal male gonads that produce excess testosterone. Disorders of sexual differentiation, previously known as intersex conditions, highlight the inherent messiness of trying to strictly define biological male and female: Scientifically speaking, not everyone fits neatly into one sex category or the other.

Under the testosterone policy, female athletes over the limit had to lower their testosterone levels to compete, which can generally be accomplished with medication or surgery, depending on the source of the hormone. Semenya continued competing, though the details of if and how she lowered her testosterone are not known. Neither she nor Chand has publicly discussed the cause of her hyperandrogenism.

Testosterone and sport

One of the strongest new pieces of evidence that the IAAF has in its arsenal is a study published just last month. In early July, two IAAF scientists published a paper in the British Journal of Sports Medicine that looked at the testosterone levels in roughly 2,000 elite male and female track athletes who competed in the 2011 and 2013 IAAF World Championships.

The researchers, funded by the IAAF and the World Anti-Doping Agency, divided male and female athletes into low, medium, and high testosterone groups based on results from routine pre-meet blood tests, and compared their athletic performance.

Their main conclusion, touted in an IAAF press release, was that women in the high testosterone group had a significant competitive advantage over women in the low group in a smattering of events, namely the 400 meter, 400-meter hurdles, 800 meter, hammer throw, and pole vault. The advantage ranged from 1.8 to 4.5 percent.

That, the IAAF says, is evidence in support of its assertion that women with high testosterone see a boost in performance. And although 4.5 percent falls well below the 10-12 percent range that the court established, Dr. Angelica Hirschberg, a professor and gynecologist at Karolinska University Hospital who testified for the IAAF in 2015, points out that for elite athletes, even 3 to 4 percent is a big difference that “can decide if you will get a medal or not.” Additionally, the highest testosterone group of women in the study averaged below the 10 nm/L cutoff, so she said an even higher testosterone level could translate into more of an advantage. “We can only speculate that probably if you have a larger group of women with testosterone in the male range, the difference in physical performance would be greater,” Hirschberg said.

Cohen, however, sees some problems with the methodology of the study. Perhaps most notably, the study states that some women in the high testosterone group artificially achieved those high levels by doping — which he said “muddles” the findings. “What they injected may be a more powerful testosterone or may be a weaker testosterone, or they may have done it yesterday or last week or last month,” he said. “You can’t interpret these numbers as the function of endogenous hormone levels when there are exogenous hormone levels included.”

He also explained that a woman’s testosterone level fluctuates considerably during her menstrual cycle, something the study did not control for. And the paper also double-counted 17.3 percent of the women who competed in both 2011 to 2013, which, according to Cohen, is “going to skew the data an incredible amount.”

For Cohen, the paper can only suggest an association between hormone levels and physical performance at a single snapshot in time. “It doesn’t tell me anything about what was going on a week ago, a month ago, a year ago, during training,” he said. “It doesn’t tell me anything about whether it’s endogenous or exogenous, and it doesn’t tell me cause and effect.”

Hirschberg has also done research since the testosterone limit was suspended. Her paper, published in late June 2017 and also submitted as evidence for the 2017 hearing, looked at the precursors of testosterone in 106 Swedish women Olympic athletes. It found that, compared to non-athletes, the athletes had higher levels of these testosterone precursors, and also had more muscle mass and better “explosive” athletic performance.

However, like the IAAF paper, that study can only establish an association between hormone levels and athletic performance, and it cannot determine whether increased hormone levels directly caused this improved performance.

These papers are certainly not the only evidence IAAF has gathered (Hirschberg said other submissions will include individual cases of athletes with testosterone levels over the cutoff), but for Cohen, they are not convincing.

“If the goal is to prove that women with endogenous androgen levels have a higher degree of physical performance when compared with women of lower endogenous androgen levels, I don’t think these studies prove that, and they definitely don’t prove causation,” he said. “And I definitely don’t think they prove that there’s an unfair advantage between women.”

Dutee Chand
Dutee Chand, India’s 100-meter champion, was barred from competing in 2014. Rafiq Maqbool/AP

Ethics and sex

While the court is choosing to focus on the science, it is impossible to ignore the ethical arguments from both sides, particularly when it comes to defining what is fair.

Much of the IAAF argument hinges on the idea that a policy is needed to make competitions fair for other female athletes, and without it, female sports will decline.

Hirschberg pointed out that several female athletes testified for the IAAF in 2015 (while others have been vocal in the media) and IAAF’s goal is to level the playing field for all women so they stay motivated to compete. “You can speculate about what will happen with female sport if we don’t have any regulation about this,” she said. “If we can beforehand guess who will win, then perhaps the audience will lose interest.” She added that even one or two high-testosterone women in each event could have an effect.

However, Dr. Katrina Karkazis, a medical anthropologist and bioethicist at Stanford University who testified for Chand’s side in 2015, disagreed. “We haven’t had a regulation for two years and women’s sports look just fine to me,” she said. “We’ve got women breaking women’s records. I don’t think anything blew up in any crazy way.” And while Semenya has excelled on the world stage since the policy was suspended, winning 800-meter gold in the 2016 Olympics, Chand has been far from dominant: She goes into the World Championships ranked 103rd in the world in the 100 meter.

Which brings up another question of fairness: is it fair to create a policy that singles out testosterone when so many other factors contribute to overall athletic performance? Hirschberg says yes. “It’s not all about testosterone, of course not,” Hirschberg said. “But since we have today two main categories, we have female sport and we have male sport, we have to have regulations about who can compete as a woman and who can compete as a man.” Hirschberg and the IAAF believe testosterone is the main factor driving the performance difference between men and women, and should thus be the basis of the policy.

Georgiann Davis, a medical sociologist at the University of Nevada, Las Vegas, who coauthored a 2012 critique of the policy, doesn’t buy into this logic. She pointed out that elite athletes have many kinds of extreme physiologies that contribute to their success, and, moreover, athletes in countries like the U.S. have access to top-notch training facilities that don’t exist in other places. “I understand the desire to create fairness for sports,” she said, “But at what point do you start saying, wait a second, this is starting to get a little ridiculous?” Cohen put it even more simply: “You can’t say, oh, we’re not going to allow sumo wrestlers over 300 pounds, or we’re not going to allow basketball players to be over 6’6”.”

Then there is the question of whether it is fair for the policy to only target women. The IAAF argues that there is a scientific basis, namely scientific evidence that indicates women, but not men, with high testosterone levels have a significant performance advantage over their peers. Others, however, believe a policy that targets only female athletes is inherently sexist. “I think that the only way it would be fair to regulate athletes [like] Caster and Dutee would be if we also had testosterone testing for men, and men that were in atypically high ranges had to take estrogen to lower their ranges down as well,” said Hida Viloria, chairperson of the Organization Intersex International, an intersex advocacy and support group that has supported these athletes since 2010.

Moreover, as much as the IAAF asserts that the policy focuses on testosterone, not sex, Viloria pointed out that it still questions a woman’s eligibility to compete as a woman, hearkening back to the days of widespread sex testing for female athletes. “I think the elephant in the room is that even though on record these sporting bodies keep acknowledging that these athletes are women, they keep trying to make regulations based on the conception that they’re not,” s/he said.

Dr. Adrian Dobs, an endocrinologist at Mayo Clinic not involved in the case, can see both sides. “If this is biological, if this is what they have, then it may be an advantage, but so are a lot of things an advantage. Good for them, so to speak,” she said. “But if these women really have a medical problem that they want to ignore because it gives them an advantage on the playing field, then in many ways they should be treated.” Conditions that cause hyperandrogenism have been associated with health issues such as infertility, cardiovascular disease, and insulin resistance.

And finally, perhaps the most fundamental question of all: Even if female athletes with hyperandrogenism do have a unique and significant performance advantage, is it automatically unfair? At the end of the day, Karkazis believes that question is a social and cultural one. “It really is an open question about whether or not something is fair or unfair, leaving aside the science of it,” she said. “The science could still say there’s a link between [testosterone] and performance and we could still say, and that’s fine, it shouldn’t be understood as unfair.”

The ethics of the situation are complicated to say the least, and while the court won’t officially consider ethical arguments as it begins the arduous process of sifting through new evidence, it’s hard to believe ethics won’t come into play — especially with the futures of several elite female runners hanging in the balance.

But for this year at least, as athletes take to the track, women will compete as women, men will compete as men, and testosterone will be left on the sidelines.

  • It seems the issue here is not just about testosterone levels, its defining what is a woman. Testosterone has a large impact on athletic potential. Yes women can have higher levels of testosterone, but when “intersex women” higher levels of testosterone is actually caused by internal testes and they have no womb/ovaries, they could hardly be considered biologically female. That sounds harsh but its biological reality. Testes (usually developed from normal chromosomes) along with high testosterone levels are the primary attributes that distinguish males from females.

    There is not the same need for a male cutoff (unless from doping) because men are not naturally limited in testosterone the way women are. This is the reason for the two categories in the first place, not just the levels which vary, but its role in development of skeletal structure etc. would definitely have a big impact.

    Its not a question of “why shouldn’t they run just because of a genetic advantage?!”. Its a question of; is this “advantage” (and other physical distinctions) drastic enough that they don’t belong in the “women’s” category? I would say in some scenarios I’d that answer would be; yes

  • What about they can compete with male athletes? Of course, in that case they would end up just being average/mediocre athletes. After all, the effect of testosterone is not so clear right? Surely, not the reason why many world records were established by female athletes doped with it in the 70-80s. Please ….

  • There are so many things wrong with using a testosterone cut-off.
    1. Is there any research that shows the effect of testosterone on male sports abilities. I know that many people think that there is and take testosterone. If there is, there should be a male cut-off as well.
    2. In many of these cases, there is a small genetic change that is causing higher testosterone. Excluding women because of genetic happenstance is unfair. It is as unfair as not covering pre-existing conditions. It also means that people who had very high levels of growth hormone as children should not be allowed to play sports which provide advantages to height. If someone gets Tommy John surgery because they injure a ligament should they be excluded? (And note that TJS is elective and not a genetic part of a person.) Let the women run.

Comments are closed.