PIEN HUANG, HOST:

The Olympics end this weekend, and it's been a fun couple of weeks, with celebrations of all the incredible things that elite athletes can do. Boxer Imane Khelif of Algeria won gold in the women's welterweight final on Friday. Another boxer, Lin Yu-ting of Taiwan, also won gold earlier today in the featherweight division. But despite their success, both boxers have been at the center of some ugliness recently. The International Boxing Association claims that these boxers are not female, and that they have an unfair advantage in fights because of, quote, "their hormonal imbalance." But this claim was condemned by the International Olympic Committee, which defended the boxers' right to compete in the women's category.

This isn't the first time this issue has cropped up in the world of athletics. NPR's Embedded podcast has a new series about sex testing and elite sports with CBC in Canada. It's called Tested, and it focuses on track and field. The sport's international governing body, World Athletics, has created regulations covering athletes with differences of sex development or DSDs. Some of these athletes have higher levels of testosterone than what's considered, quote, "normal" for women. Well, World Athletics claims that this gives them an unfair advantage, but what makes an advantage in sports unfair? In this excerpt from Tested, host Rose Eveleth digs into that question.

ROSE EVELETH, BYLINE: Advantage - it's an interesting word when it comes to sports, because in some ways, it's what sports are all about, right? Who is faster or stronger or smarter? And the idea of advantage is the foundation of DSD policies, this claim that some women with DSd have an advantage over other women. But what does advantage actually mean?

Step into my laboratory, will you? Imagine that the walls of this purely hypothetical lab are lined with little vials, each labeled with tiny, precise script. And within each of these test tubes is some element of athletic advantage. If we combine them, we can create athletic alchemy, the perfect athlete.

First, you're going to want the vials labeled time and money. Those are key to hire coaches, eat right, travel for competition, have the best training facilities, the newest shoes. Next, let's add something a little bit more ineffable - mentality, determination, that single-minded focus. Our perfect athlete beaker is now half full. But on top of all this stuff, you need the body.

Depending on what sport this imaginary athlete might compete in, you'll want to tweak things like height, weight and body proportions. For example, Michael Phelps has an incredibly long torso and comparatively short legs, perfect for swimming. There's also a whole section of this lab's wall of vials dedicated to genetic mutations that an elite athlete might want. There are at least 20 different genetic factors that researchers have identified as being potentially correlated as being potentially correlated with athletic performance.

So go back to the wall and grab the vial labeled EPOR, an acronym for erythropoietin receptor. This gene determines how good the body is at making red blood cells. People with this mutation can carry more oxygen in their blood, which helps with aerobic exercise. A Finnish skier named Eero Mantyranta had this exact mutation, and he won seven Olympic medals. And while you're over there, grab the vial labeled ACTN3. That one will be handy, too. People with this mutation have a slight advantage in powerful sprinting events.

You might also grab smaller tweaks from our wall of beneficial mutations - ACE insertion/deletion, angiotensinogen, AMPD1, homeostatic iron regulator, leukin six, (inaudible) nitric oxide, activated receptor alpha, alpha receptor gamma, uncoupling protein two. Some athletes with these kinds of mutations are viewed as icons in their sport, like that Finnish skier I mentioned. Here's Morgan Campbell, a sportswriter at CBC.

MORGAN CAMPBELL: We just sort of celebrate him as a medical marvel, right? Whereas in any - in a different context, we would say he is a natural-born cheater. It all depends on, you know, who's lucky enough to get born with the right set of characteristics that fit, you know, with where we decide to draw the line between somebody with a genetic advantage or somebody who we decide is born a cheater.

EVELETH: But why is it that some kinds of biological advantages are fine, and others require a whole new rule to be written to remove the alleged advantage? The answer to this question, according to those in favor of regulations, is simple. We don't divide sports by blood oxygen or fast-twitch muscles, but we do divide sports by sex. And so these folks argue advantages that might be connected to sex are fair game.

In 2015, the Court of Arbitration for Sport actually agreed with this idea that sex-based advantages might warrant regulation. But they said that before World Athletics could actually put rules into place, they had to prove just how big this advantage was. If it was big, like 10 or 12%, then sure, that might call for rules like this. But if it was small, like two or 3%, then that's a lot harder to justify because that would put the advantage in the realm of those other things we just discussed.

World Athletics lost their case in 2015 because at the time, they didn't have any evidence to actually show how big this advantage might be. But the ruling was provisional. It suspended the testosterone regulations for two years. And it said that sports officials had those two years to go off and find evidence to justify their policies. This directive alone raised some red flags among researchers following this topic.

ROGER PIELKE: And it'd be a little bit like, you know, a regulatory agency saying to a tobacco company, hey, you got two years, go gin up some research and tell us that smoking doesn't cause cancer.

EVELETH: That's Roger Pielke, a professor of science policy at the University of Colorado Boulder, with a special interest in how data is used to shape policy. CAS asked World Athletics to answer a fairly straightforward question. How much of an advantage do athletes with DSDs have? And the ideal study you would do to answer this question, it's pretty simple. You'd want to compare the testosterone level and performances of DSD athletes with non-DSD athletes. But that's not the study World Athletics did.

PIEKLE: There are no studies, peer reviewed or not peer reviewed, of the relative performance of female athletes with certain DSD conditions as compared to female athletes without those conditions. There's just no studies.

EVELETH: Instead, they did something slightly different. In 2017, two World Athletics researchers published a paper in the British Journal of Sports Medicine. The study looked at the testosterone level of all athletes, regardless of their DSD status. And the paper argued that they did actually find evidence that higher testosterone levels meant better performance, but only for some events, mostly the so-called middle distances - the 400, the 400 hurdles and the 800.

But critics of World Athletics noted a few things about this study. They pointed out that its authors were not independent researchers. They were both associated with World Athletics. And experts like Roger also found the actual results kind of weird. Why would testosterone only impact middle-distance events?

PIEKLE: I don't know. My bulls*** detector kind of went went nuts. Ding, ding, ding, ding. Let's go look at this data and figure out what's going on.

EVELETH: Roger made his misgivings public on his blog. Other researchers piled on as well. One called the study a mess. And then in April of 2018, World Athletics announced a new policy.

(SOUNDBITE OF ARCHIVED RECORDING)

SEBASTIAN COE: The first decision this morning was a very, very big decision.

EVELETH: World Athletics President Sebastian Coe said the policy was based, at least in part, on the study it had done.

(SOUNDBITE OF ARCHIVED RECORDING)

COE: We were asked by the Court of Arbitration for Sport to provide the evidence regarding the magnitude of this advantage, which we now have. And the council has been taken...

EVELETH: And Coe said that this was one of the organization's, quote, "toughest subjects."

(SOUNDBITE OF ARCHIVED RECORDING)

COE: And I want to make one point really clear, crystal clear upfront. This is not about cheating. No HA athletes have cheated. This is about our responsibility as a sports international federation to ensure, in simple terms, a level playing field. It is our sport, and it is up to us to decide the rules and the regulations.

EVELETH: The new rules restored the eligibility restrictions. Once again, women with high testosterone would have to lower it in order to run in the women's category. And the rules only applied to some of the events in which the paper found an advantage, those middle-distance races. Roger Pielke had been hoping to get a look at the paper's data, and a few months after the new policy was released, the study's authors finally allowed him to see some of it. And when they looked closely, Roger and his collaborators found all kinds of errors.

PIEKLE: I mean, it was stunning. There was between, I think the numbers are like 17 and 32% of the data was erroneous. So they had duplicated data. They had phantom data that didn't exist anywhere. They duplicated athletes. They had athletes who've doped. Hugely problematic.

EVELETH: Eventually, the authors of the original study published a response to the critiques, admitting that there were some issues with the data and offering up a new analysis. And in that new analysis, the results changed. Only one of the running events cited in the original maintained the level of advantage they claimed. But still, World Athletics stuck with its regulations - no changes, corrections or updates.

HUANG: That was Rose Eveleth, host of the new podcast series Tested from NPR's Embedded and CBC in Canada. And we should say World Athletics declined interview requests for the podcast. In an email to Rose, a representative wrote that World Athletics quote, "has only ever been interested in protecting the female category. If we don't, then women and young girls will not choose sport," end quote. You can listen to all six episodes of Tested now. It's available in the Embedded podcast feed wherever you get your podcasts. Transcript provided by NPR, Copyright NPR.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.