The Jessica Hardy Doping Case - Something Just Doesn't Make Sense

My scientific background leaves me more inclined to trust laboratory results than people, and I'm no more inclined to give athletes the benefit of the doubt in doping cases than anyone else who's been paying attention over the last couple of decades. When I heard that Jessica Hardy had tested positive for a banned substance at the Olympic Trials, and most likely will not get to swim in the olympics, I wasn't really surprised. Swimming hasn't been plagued with the same sort of doping scandals that other sports have seen, but it would be shocking if there weren't at least a few cheaters out there waiting to get caught. That's why there are tests. Someone tested positive? Toss them off the team and move on.

But when I took a few minutes to read the full story, something didn't make sense. I looked at a couple of more stories, and the situation made even less sense. At this point, I'm hopelessly confused, but I'm going to keep writing this anyway. If you keep reading, one of two things will probably (hopefully) happen: either you'll be able to spot something I missed, and unconfuse me in the comments, or you'll join me in confusion and the hope that someone else will be able to clear this one up for us.

Here are the facts, as far as they've been reported so far:

A drug test was conducted on Jessica Hardy following her 4th place finish in the 100 M freestyle at the Olympic Trials on July 4th. There's an international standard that governs how these tests are carried out, so we know that the sample was immediately divided into two sub-samples, both of which were sealed while the swimmer watched. The test panel that was carried out on the first sample came back positive for the presence of clenbuterol. The second sample was tested, and the positive was confirmed.

Clenbuterol is a banned anabolic agent and stimulant. It's a long-acting beta-2 andrenergic agonist, with a half-life of about 36 hours. In the US, it's approved only for the treatment of respiratory disease in horses, but it's often abused as a weight-loss drug. It's also been used by anabolic steroid abusers as a way to maintain muscle mass when they are not taking the steroids.

In addition to the test that was conducted on July 4th, Hardy was also tested following her win in the 100 M breaststroke on July 1st, and after her 2nd place finish in the 50 M freestyle on the 6th. Those two tests came back clean. And that's where things get really confusing.

I'm not trying to argue that it makes no sense for her to have taken a banned substance after winning a race and making it onto the team. It doesn't, but I've seen far too many allegedly intelligent people do really stupid things for that particular argument to fly.

No, what doesn't make sense here is the timing. The negative-positive-negative results on the three tests during the trials can only be explained if she took the drug after the first test, before the second test, and long enough before the third test that the dose had entirely cleared her system. I haven't been able to find a reliable source for the length of time that clenbuterol is expected to remain detectable, but I did find multiple pro-steroid use websites that say that the detection window is 2-4 days. That seems reasonable, given the 36-hour half-life.

It's possible that she took the drug on the 4th, and that it cleared her system before she was screened on the 6th, but that requires that the drug cleared her system in essentially the absolute minimum time expected. But if she was taking the drug for the short-term benefits (which is the only reason to take the drug during a competition), it makes no sense - from a cheating standpoint - for her to have taken the drug on the 2nd or 3rd. She swam no events at all on the 2nd, and she had no finals on the 3rd.

This whole thing is a mystery to me, and as much as I hate mysteries, I don't think this one's likely to be solved or end well. It's very, very, very unlikely that the results are a false positive. The two sample screening addresses that issue well. As unlikely as the scenario for intentional use seems in this case, the confirmed positive lab test places the burden of proof firmly on the athlete, which means that it's highly unlikely that she's going to be swimming in China on the 10th. I'm not sure that we're ever going to know what really went on in this case. But if you've got any ideas, please let me know.

More like this

As I mentioned earlier, the current issue of Nature has a perspectives article by Donald Berry, a statistician at the MD Anderson Cancer Center that addresses problems with the current system for testing athletes. I more or less agree with the overall conclusion of the article - there needs to be…
Floyd Landis, most recent winner of the Tour de France, has tested positive for testosterone use: Landis denied cheating and said he has no idea what may have caused his positive test for high testosterone following the Tour's 17th stage, where he made his comeback charge last week. But he aims to…
I got a request to collect participants for an online survey on science fiction — take a look and help out if you want. It's long, and a little depressing: it's a list of science fiction movies and TV shows, and you're supposed to rate their scientific accuracy. I think I'm rather picky about that…
The things Zuska has to say often make me uncomfortable, and the way she says them often irritates the hell out of me. But I still read her blog, because she almost always makes me think about things in ways I hadn't before. Her post about the recent Lindsey Vonn Sports Illustrated cover is a…

The two sample screening addresses that issue well.

Almost - there is still a chance that both tests will be false positives. Are the figures for the specificity of the test used in this case available? That and the number of tests done per year need to be thrown at Bayes' theorem.

"The two sample screening addresses that issue well."

I've wondered about the reliability of the two sample screening as well. Exactly what does that mean? Are both samples tested by the same person in the same laboratory using the same equipment? What measures are in place to minimize the possibility that a mistake or contamination during the testing of the first sample is not simply repeated during the testing of the second sample?


As best I understand it the second sample can be tested at any approved lab that the athlete nominates. If I recall they can also nominate an observer for the process.

By Matt Penfold (not verified) on 30 Jul 2008 #permalink

For what it's worth here is an explanation. It is possible that she has taken clenbuterol under a regimen calculated to not reach the cutoff/detection level of the doping test. So if her levels fluctuate just below the cutoff/detection level, one can imagine that on rare occasions she would run a positive. I dont buy the false positive argument. Rather, I think the other two tests are false negatives.

By notgoingtobeijing (not verified) on 30 Jul 2008 #permalink

The World Anti-Doping Agency test standards are available online.

It looks like the standard for clenbuterol detection appears to be an initial immunoassay, followed by mass spec and/or chromatography for confirmation. That's the standard for each of the two samples. In addition, three aliquots are tested on the A sample, and the lab is required to run both positive and negative controls with each batch.

The B sample is tested by the same lab, but all the sample handling steps must be conducted by a different tech. The athlete and/or a representative are allowed to oversee B sample testing.

As far as the possibility of false negatives on the other tests go, I thought about that. Thing is, there's no specified cutoff level for clenbuterol, and the lab must be able to detect concentrations as low as 2 ng/ml. I'm not a pharmacologist, but I have to wonder if it's even possible to construct a regimen for using this drug that has any hope of accomplishing anything at those dosing levels.

I wonder even more why an athlete smart enough to construct such a regimen would continue use during a competition period, when you know you're going to be tested multiple times.

I'm not saying that Hardy's innocent, by any stretch of the imagination. Just that the results still seem really strange.

Basically, the half-life of clenbuterol in the human body is one and half day. And it takes 4 half-life to bring the total amount down to one sixteenth of the intial concentration. In other words, you will still have 6% of clenbuterol in your body after six days. It is a drug that stays in the body for a "long" time.

If it is intentional use, nobody would use it during the week of competition bacause it will not clear from the body fast enough to avoid detection.

If it is accidentally ingested as a contaminant in a supplement, the sequence of the negative-positive-negative would not be an issue.

Different drug but related story. Italian cyclist Ricardo Ricco who tested positive at the Tour de France for the EPO analog CERA admits he took the drug after two weeks of denials. Note that he only tested positive twice out of 10 tests. Clearly, too many false negatives. My hunch is that the tests are too conservative in general and some are obviously willing to take a chance. I wonder how many sports are testing for CERA. Apparently, the test is new and not optimal yet, which could also explain the 80% false negative rate.

As to clenbutenol, I have no idea how much 2 ng/ml is. What about aiming at having 1 ng/ml levels. Would that be enough to get a boost? I would think so especially in combination with other drugs.

By notgoingtobeijing (not verified) on 30 Jul 2008 #permalink

A "negative" test result does NOT mean the sample was free of clenbuterol. Instead, it means the sample did not yield results that, per WADA procedures and standards, would be considered a "positive" test.

If the amount of clenbuterol in a person's system is near the threshold for what counts as a "positive" test, or is detectable as such, then it would not be surprising to have really any sequence of mixed negative and positive results from different samples (as noted above by "NotgoingtoBeijing").

Since athletes who are intentionally doping are presumably eager to avoid detection, it's not too much of a stretch to think they'd attempt to have the levels of any banned compounds be below detectable/"admissible" thresholds by competition (or masked to those levels by other agents). As such, the expectation would again be that multiple tests might well yield different results.

There is also empirical evidence of this in other cases: notably that of Floyd Landis in the Tour de France (different compound, but same "sandwiching" of a positive test between negatives).

So to me, the test results make sense just fine. If they don't to you, then realize that your standard for detecting doping has effectively become a sequential run of identical test results (and I'm not taking about the "A" and "B" aliquots, but rather multiple samples). Such a standard would greatly please the many drug cheats in sports, who already seem to have the edge over the testers.

The "contaminated supplement" hypothesis is actually far less plausible here: you'd have to assume that she started taking a new supplement in mid-competition, or a new batch that just happened to be contaminated, but then stopped and and and ....

Remember that all these arguments have been made by pretty much every athlete who has tested positive, a number of whom have subsequently confessed, such as Marion Jones.

I know your post had some thoughtful disclaimers, but please be careful when -- as a scientist -- you comment on breaking news issues in other areas of science. After all, we've seen how wrong a biochemist can get evolutionary biology (Behe and intelligent design), no sense returning the favor with an evolutionary biologist muddying the waters of biochemistry and forensic drug testing.

If you want to see the human cost of this stuff, check out the blog of Tara Kirk, the third-place finisher in the Olympic Trials race that Ms. Hardy won

The concentration of drugs in the urine depends on hydration of the person. The drugs will concentrate if the person is dehydrated. That could push the drugs from undetectable levels up to detectable levels.

As for the amount of clenbuterol needed for its anabolic effects, it does not require much. Basically, the typical regimen is a prolonged period (multiple weeks) with a very low daily dosage.

Doing some quick calculations, it would seem that 2ng/ml would be about what you would expect in someone taking clenbuterol. Dosages are in the range of 50,000-100,000ng and a human body has about 40 liters of water. Fluctuations in hydration would obviously affect concentration.

By michael fugate (not verified) on 30 Jul 2008 #permalink

#8: "I know your post had some thoughtful disclaimers, but please be careful when -- as a scientist -- you comment on breaking news issues in other areas of science. ... no sense returning the favor with an evolutionary biologist muddying the waters of biochemistry and forensic drug testing."

Oh for crying out loud.

He laid out the situation well, still had questions, asked for others thoughts.

I fail to see the possible danger or fatal harm to 'science'. Or discussing topics science-related.

#11: "Oh for crying out loud. He laid out the situation well, still had questions, asked for others thoughts. I fail to see the possible danger or fatal harm to 'science'"

The danger is that scientific testing of samples is the number one way of catching drug cheats, especially if positive tests lead to public condemnation (since that impacts sponsorship dollars, the main source of income for athletes in "Olympic" sports). An ill-informed post on a science website that calls into question a particular positive test, claiming the results to be mysterious and impossible to resolve when they are neither, undermines the credibility of the testing process. People can just say "oh, well, even 'scientists' are confused by the test results, I guess we shouldn't worry about them."

As for "fatal harm to 'science'", I never said or implied anything like that.

But let's be clear: people's careers are at stake based on whether these particular test results "make sense" or not. So while commentary is still fair game, I was merely suggesting caution, especially since the field of science is quite different than the poster's own. I stand by that.

As I said, I think that the tests clearly indicate that the substance was present in the samples. Given the care that's taken with chain of custody, I also have no reason to doubt that the sample tested is the same one that she gave after finishing the race.

Just so we're crystal clear, I am not and have not questioned the validity of the testing procedures. In fact, if I thought that there was any question about the validity of the testing, I wouldn't think that this whole thing was so difficult to explain. I'd say, something like "clear false positive, she's being railroaded", and let it go at that.

I'm saying that the test results don't make sense because the positive-negative-positive pattern that popped up during the competition makes no sense from a cheating perspective. If the drug was being abused for its anabolic properties, she most likely wouldn't be taking it during competition, and the drug's half-life makes it unlikely that she'd still be testing positive at all that far into the competition. If she was taking it for the short-term respiratory effects, she should have pinged the test on the 6th, as well.

As far as expertise goes, I'd love to hear from someone who has some and can point me toward the appropriate readings. That's one of the things I was hoping for when I decided to post my concerns.

"If the drug was being abused for its anabolic properties, she most likely wouldn't be taking it during competition,"
Wrong. To quote Victor Conte (=Mr. BALCO): "We used THG [an anabolic steroid] immediately after the competition. Because of the exertion and the micro-tears, you need to facilitate healing and recovery." Especially if you are competing in multiple events, each with heats, as Ms. Hardy was.

"If she was taking it for the short-term respiratory effects, she should have pinged the test on the 6th, as well."
Wrong. Numerous comments, including mine, have explained that taking a drug does not always lead to a "positive" test. The hydration level of the athlete, the effectiveness of any masking agent(s) that are being used, and whether they've managed to calcuate the "right" dosage (enough to help, but not so much as to exceed the tests' threshold) can all vary from sample to sample, in addition to the variation inherent in these tests (e.g., the "A" and "B" aliquots don't always return the same results). Plus there is the empirical evidence, as pointed out by NotgoingtoBeijing, of confessed drug cheat Ricco, who only "pinged" two tests out of ten.

You originally said it didn't make sense, but asked for explanations. Fair enough. But multiple explanations have been given, quite clearly. So maybe either show how these explanations are inadequate, or consider that maybe the results do make sense.
- #8

With regard to the long-term use point:

Clenbuterol is an anabolic agent, but it's not a steroid. From what I've been able to find, it's mostly used by bodybuilders to maintain muscle bulk when not actively on the steroids. The one article I found that is from a clearly reputable source (Greydanus & Patel, Pediatr Clin N Am 49 (2002) 829 855) indicates that use would normally cease prior to competition.

In addition, if you look at the Wikipedia article for clenbuterol (linked in the main post), you'll see that this is a drug that actually slows down glycogen storage. That's not exactly something that you'd be looking for in a drug for the recovery period.

I'm willing to be proved wrong on this one. If you've got a source that identifies this specific drug as a recovery enhancer, please point me at it. I've looked, and I can't find one.

As far as the false negative point goes, the situation with the biker is very different. He was taking EPO, which is a much more difficult drug to test for - it's basically an artificial form of a substance normally found in the blood, so the doping screening needs to be able to distinguish the two. That's a procedure that's going to be much more prone to false negatives than detecting clenbuterol - a substance that doesn't naturally occur in the human body, and is therefore going to raise flags at any concentration.

I'd also like to point out that assuming that the test for clenbuterol is prone to massive (66%) false negatives would seem to cast a great deal of doubt on the clinical testing procedures. You know, the ones you were concerned that I was casting doubt on by asking questions in the first place.

Something I should have said earlier:

I really feel for the swimmers (including Tara Kirk) who should have been on the team instead of Hardy given the test results. USA Swimming's conduct in this case has been absolutely inane. If they didn't have final clean results for all the swimmers, they should have included alternates.

"I'd also like to point out that assuming that the test for clenbuterol is prone to massive (66%) false negatives would seem to cast a great deal of doubt on the clinical testing procedures. You know, the ones you were concerned that I was casting doubt on by asking questions in the first place."
I wasn't worried about you casting doubt on negative tests -- there is already tremendous and well-founded doubt about whether "testing negative" = "being clean." Look at all the speculation regarding Lance Armstrong, to pick just one example. This doubt exists in part exactly because the "false negative" rates of the tests are indeed high, in part by design so as to avoid stripping an innocent person of their career and livelihood. Plus people (dopers) are intentionally trying to achieve "false negatives".
I was worried about you casting doubt on a POSITIVE test. Given the need to design the tests to err on the side of the negative, and the numerous measures taken by some athletes to beat the tests, it's really hard to catch someone. So when they do, when someone actually tests positive on an A and B sample, the last thing the world needs is muddled analysis of how that positive "just doesn't make sense" by someone with zero experience in the field.

As for literature on clenbuterol as a recovery agent:
1. Dopers are notoriously lax on submitting their findings to peer-reviewed publications.
2. Legit researchers would have a very hard time getting human subjects approval to experiment with clenbuterol as an aid for athletic recovery, and little motivation to do so since it's banned.
3. It really doesn't matter what you think the "proper" doping usage of clenbuterol is, or what I think. It's what the people who use it think. To argue that the test is off because it "wouldn't make sense" for the athlete to be using it assumes that the athlete a) thinks like you do, and b) is being sensible. Neither of those need be the case.

I see.

I've found nothing - peer reviewed or otherwise - that suggests that clenbuterol is abused as a recovery aid. In addition to this, I've also pointed out that one of the physiological effects of clenbuterol is inhibition of glycogen storage. Apparently, I wasn't clear enough about this before, so let me state it more clearly: inhibition of glycogen storage is more likely to slow recovery than speed it. You want to get glycogen back into your cells before your next event. You don't want to slow that process.

Your response is that you have no knowledge of any source that says clenbuterol is used as a recovery aid, but that it's still reasonable to assume that this was the intent, because dopers do dumb things. And that I've been engaging in muddled arguments.

As far as your point about running down positive results goes, let me state this as clearly and unequivocally as I can:

I. think. there. was. clenbuterol. in. Hardy's. system. at. the. time. of. the. drug. test.

I do not think that the positive is the result of poor procedures or a false positive.

The series of results still seems strange, particularly if the explanation for the presence of the drug in Hardy is intentional substance abuse.

"The series of results still seems strange, particularly if the explanation for the presence of the drug in Hardy is intentional substance abuse."

I am surprised you still persist in calling the results strange. Anon's and the other posts addressed this clearly. A negative-positive-negative is the sort of pattern you would expect if she has intentionally taken the drug. If she has cheated she would be totally stupid to take a high enough dose that would definitely yield a positive test. Rather the only logical cheating pattern would be a regimen aimed at keeping her below the threshold. Because of her natural fluctuations (dehydration etc...) and other variables like the test itself, catching her with levels above the threshold would be a rare event. Thus, the test pattern makes sense.

I am also surprised as how much the media is making of her explanation of tainted supplements. WADA is clear on this. The athlete is responsible for whatever they ingest. End of story.

By notgoingtobeijing (not verified) on 30 Jul 2008 #permalink