The so-called "sex gap" in testosterone — the typical difference between men and women in blood levels of the hormone — shouldn't be used to determine who is and isn't a female athlete, according to a commentary published Thursday in the journal Science.
Testosterone is the sole marker used to categorize men and women in professional sports. The rules of the two major international sports organizations say that in cases where a female athlete's sex is in question, those with unusually high levels of testosterone aren't allowed to participate in women's competitions.
But that policy is deeply flawed, according to the authors of the new report. When it comes to elite athletes, they say, extraordinary biology is always the norm.
"It's a cultural anxiety," Katrina Karkazis, a medical anthropologist at Stanford University and lead author on the new paper, told BuzzFeed News. "This actually isn't really a scientific question — it's a social question on how to think about human variation."
There have been only two large studies, both published last year, looking at testosterone levels in elite athletes. At face value, they seemed to have come to wildly different conclusions.
The first, funded by the the International Olympic Committee (IOC) and the World Anti-Doping Agency, analyzed testosterone levels of 234 female Olympic athletes and found that 13.7% were higher than the typical range for women.
The other study, run by the International Association of Athletics Federations (IAAF), looked at testosterone levels in 849 elite female athletes in track and field, and concluded that just 1.5% had high testosterone.
In the new paper — which is contested by some researchers — Karkazis and her colleagues argue that the IAAF study was biased to support its existing policy. Its numbers were so low, they say, because it did not include female athletes with intersex conditions — those who have biological traits that could be considered male or female — who often have high testosterone levels.
When contacted by BuzzFeed News, the IOC and the IAAF declined to comment specifically on Karkazis' paper. An IOC representative noted that it plans to discuss developments on this issue at a meeting in November.
Testing for "true" sex has long been a controversial part of elite sports competitions.
In 1968, the International Olympics Commission began testing for sex based on chromosomes — XX for female and XY for male. Before that, so-called "sex impostors" had to undergo demeaning physical inspections in the nude, so the new lab protocols were seen as "simpler, objective and more dignified."
But as it turned out, chromosomal testing was not so simple. Exceptions to the rule cropped up all over the place. In 1999, the IOC dropped the requirement, acknowledging that when it comes to sex, things didn't always sort neatly into two piles.
"We've known for a long time that the number of elite female athletes with a Y chromosome is about 1 in 420, while in the general population it's 1 in 20,000," Eric Vilain, a medical geneticist who researches disorders of sex development at UCLA, told BuzzFeed News. "So we know there is no strict biological line between men and women. The question is, do we throw our hands in the air?"
The issue was further complicated in 2009 when 18-year-old South African runner Caster Semenya — who had lived her whole life as a girl but sported what The New Yorker called a "breathtakingly butch" physique — was barred from competing after fellow athletes publicly questioned her sex. Semenya was forced to undergo a physical examination, genetic tests, and a hormone exam. In what exploded into a "painful" public airing of a bigger debate, leaked medical results showed that Semenya had both male and female anatomy.
While she was eventually allowed to compete, in 2011 the IAAF instituted a policy saying that females with testosterone levels above the "normal" range — anything higher than 10 nanomoles per liter in their blood — would be barred from competing.
If female athletes who test above this ceiling want to compete against other women, they must lower their testosterone with hormone therapy or surgeries. The IOC has similar policies. (Interestingly, IOC rules state that women with high testosterone are allowed to compete against men.)
The testosterone policy is a source of significant debate since it's being used as the single defining factor in a biological question that's still riddled with confusion.
A recent case involving teenage Indian sprinter Dutee Chand — who was excluded from competing under the IAAF's testosterone rule — is challenging the policy at the Court of Arbitration for Sport in Switzerland, the so-called "Supreme Court of sport." She appealed the policy in March and still awaits the court's decision.
Some argue that the new paper is making a false argument in service of the larger aim of taking down the testosterone policy.
Vilain, who advised the IOC on its testosterone policy, finds many problems with Karkazis' arguments. Comparing the two studies is like comparing apples and oranges, he said. The IOC study, for example, did not toss out dopers — an indeterminate but potentially sizable portion of their sample that could weigh it further on the side of high testosterone. What's more, unlike the IAAF test, the IOC study drew blood after a race — despite the fact that testosterone is known to rise after vigorous exercise.
Beyond whether testosterone can be used as a marker to decide who is "male" and "female," researchers don't fully understand the link between testosterone and athletic performance. The IOC study found, for example, that 16.5% of elite male athletes have testosterone below the typical male range.
"Testosterone is certainly important for muscle growth," Jakob Vingren, associate professor of exercise physiology at the University of North Texas, told BuzzFeed News. But the link to performance is more tenuous. "If we give you more, your performance will improve. If we take it away, there will be a negative effect. But if we take you and 10 other people and line you up and measure your natural testosterone levels and compare it to your performance? There's a very weak correlation there."
Vilain acknowledges that testosterone is an imperfect way to decide who is "male" and who is "female." The real debate, he said, is whether we should use biological markers at all, or instead rely on how an athlete chooses to identify.
For Karkazis, the fact that women like Semenya and Chand identify as women — and are legally recognized as such — means that they should be allowed to compete. "These are women that challenge people's deeply ingrained ideas about what kinds of bodies women are supposed to have."