3rd September 2021

The burrito and the ban: Shelby Houlihan’s four year sanction

In June, the Court of Arbitration for Sport (CAS) banned US middle distance runner Shelby Houlihan for four years on the eve of the US Olympic Trials, ending her hopes of taking the field at the postponed Tokyo 2020 Olympics. Houlihan blamed a dodgy burrito for an adverse analytical finding (AAF – or ‘positive test’) for nandrolone (19-norandrosterone). The 44 page Decision (PDF below), published by CAS this week, reveals that as Houlihan failed to conclusively prove that her AAF was caused by the burrito, she was judged to have intentionally doped and was sanctioned with a four year ban.

The CAS didn’t rule that the probability that a contaminated pork burrito led to her AAF was ‘very close to zero’, nor did it claim that her explanation ‘simply cannot be accepted’, as has been reported. These were submissions put forward by World Athletics in its case against Houlihan.

In fact, the CAS Panel outlined that it ‘finds it is possible but improbable that the meat of an uncastrated boar ended up in the burrito that the Athlete ate. Therefore, the Panel finds that the Respondent did not establish on the balance of probabilities that the burrito she partially consumed on 14 December 2020 contained boar offal.’ Perhaps simply put, it didn’t believe her story. 

The Bachelorette & the food truck

A burrito…

Houlihan claimed that she had ordered a beef burrito at 7:30pm on 14 December 2020, ten hours before her 15 December 2020 AAF. She claims she was mistakenly given a pork burrito containing non-castrated boar meat, which caused her AAF.

Houlihan met fellow athletes Karissa Schweizer and Courtney Frerichs, as well as her sister Lindsey, at 7:30pm on 14 December 2020 to eat burritos from a food truck and watch The Bachelorette TV show – apparently a Monday night tradition. ‘All witnesses that were present at the dinner testified or declared that the burrito was extremely greasy’, reads the Decision. None of them finished their burrito. 

Houlihan reported her AAF at 6am on 15 December 2020. She reported negative tests on 22 November 2020 and on 23 January 2021. 

The levels of 19-norandrosterone found in Houlihan’s A sample were 6.9ng/mL and 7.8ng/mL in her B sample. WADA’s TD2021NA Technical Document recognises that consumption of non-castrated male pigs can cause AAFs for 19-norsteroids. 

A Study performed in 2000 found that 19-norandrosterone and 19-noretiocholanolone concentrations in urine reached 7.5ng/mL ten hours after consumption of 310g of non-castrated boar meat. This tallies with Houlihan’s consumption of her burrito ten hours before her AAF.


As usual with such cases, the devil is in the detail. The CAS Decision turned on Houlihan’s contention that she must have been given a pork burrito rather than a beef burrito by mistake. The CAS Decision features an exhaustive analysis of whether the meat used by the food truck at Beaverton, Oregon, could have contained non-castrated boar, as argued by Houlihan’s team. 

Houlihan’s team didn’t contest that the food truck normally buys its pork from an IBP/Tyson plant, which doesn’t operate a boar kill plant. Houlihan’s team did establish that the pork stomach meat used by the food truck on 14 December 2020 originated from a frozen batch purchased in September 2020 which ‘was likely to have been affected by the summer disruption to the pork industry caused by the Covid-19 pandemic’. It wasn’t the usual meat used by the food truck.

Irrespective of this, the CAS accepted evidence from Prof. Christiane Ayotte that the concentrations of 19-norandrosterone in Houlihan’s sample were incompatible with the ingestion of boar meat. In addition, despite acceptation disruption to the food truck’s normal supply chain caused by Covid-19, it accepted evidence from World Athletics that the only way that a non-castrated boar could have entered the pork supply chain was if it was a cryptorchid, i.e. a boar with undescended testicles.

WADA’s TD2021NA clarifies that when a Laboratory finds that a sample from an athlete reports a finding for 19-norandrosterone between 2.5ng/mL and 15ng/mL, additional analysis using Gas chromatography combustion isotope ratio mass spectrometry (GC/C/IRMS) analysis must be performed. As Houlihan’s samples reported 19-norandrosterone at 6.9ng/mL and 7.8ng/mL, GC/C/IRMS analysis was performed in order to determine if the 19-norandrosterone was exogenous (i.e. external, and not formed internally).

Houlihan’s A sample delta-delta (δ13C value) values were 3.8 (i.e. -23‰ for 19-norandrosterone and -19.3‰ for pregnanediol) and her B sample values were 3.8 (-23‰ for 19-norandrosterone and -19.2‰ for pregnanediol). This won’t mean much to anybody other than scientists. However, in short, the Laboratory determined that this meant her 19-Norandrosterone positive was not the result of endogenous (internal) formation of the steroid, and was consistent with exogenous (external) use.

However, confusingly for non-scientists, Houlihan’s values appear to fall within the TD2021NA guidelines that indicate boar meat consumption may be a factor. ‘Following consumption of the edible parts of non-castrated male pigs, concentrations of excreted 19-NA in urine are usually in the low ng/mL range (< 10 ng/mL), although higher concentrations have been exceptionally reported’, they read, referencing a 2009 Study conducted by Claudine Guay and Christiane Ayotte, one of the experts called to testify in Houlihan’s case. ‘The origin of the urinary 19-NA may not be established by GC/C/IRMS analysis, since the varying diets of migrating wild boars lead to dissimilar δ13C values which may range between -15 ‰ and -25 ‰. Therefore, if the consumption of edible parts of intact pigs is invoked by an Athlete as the unlikely origin of a 19-NA finding, this may be established based on the pharmacokinetics of 19-NA excretion. Profiles of 19-NA and 19-NE excretion following oral ingestion will have a different time course than following an injection of 19-norsteroids.’

In other words, GC/C/IRMS analysis is of no use in determining whether consumption of non-castrated male pigs leads to a positive test for 19-norandrosterone in an athlete. However, if an athlete does claim that consumption of non-castrated male pigs led to their AAF, then a pharmacokinetics study may establish if this is correct (note that the above guidance predicates this as ‘unlikely’).

Despite Houlihan’s claim that her burrito may have contained pork meat, no pharmacokinetic study was carried out. Houlihan’s team contested that this should have meant that her positive test was reported as an atypical finding (ATF) rather than an AAF, and she should therefore not have been charged with an ADRV.

This was dismissed by the CAS Panel. It pointed out that a pharmacokinetic study was not explicitly required, and it was up to the Laboratory to determine whether the GC/C/IRMS analysis is consistent with exogenous use of 19-norandrosterone. It accepted that Houlihan’s δ13C values were inconsistent with the guidelines for human endogenous urinary steroid excretion spelled out in TD2021NA. 

These read as follows: ‘To reject the hypothesis of endogenous or in-situ 19-NA formation based on the application of GC/C/IRMS analysis (i.e. to report the finding as an AAF), the following criterion shall be met:
• The │∆δ13Cvalues between two (2) ERCs and 19-NA, i.e. |Δδ13C| = |δ13CERC – δ13C19-NA|, is greater than (>) 3 ‰ (refer to the TD IRMS [12]).’

See. Anti-doping is simple. If it’s on the Prohibited List, don’t take it. 

Joking aside, Prof. Ayotte argued that the GC/C/IRMS analysis showed that Houlihan’s δ13C values for 19-norandrosterone were -23.0‰, which was a difference greater than 3% over 19-norandrosterone reference compounds (ERC) for androsterone (-19.0‰) and pregnanediol (-19.3‰). Taking the guidance above, this means that endogenous formation of 19-norandrosterone must be ruled out. As such, it had to be external in origin.

However, remember the 2009 Study conducted by Ayotte and mentioned in the TD2021NA? This involved six specimens from three volunteers who had consumed uncastrated pig offal from a Canadian slaughterhouse, which were analysed using GC/C/IRMS. The δ13C values ranged from -19.77‰ to −22.43‰.

This means that one of the three people in a Study used to underpin the TD2021NA would have reported an AAF, since the difference between the −22.43‰ δ13C value and the 19-norandrosterone ERC for androsterone (-19.0‰) and pregnanediol (-19.3‰) was greater than 3%. Yet despite this apparent inconsistency, the CAS Panel found that ‘the carbon isotope signature of the Athlete’s A- and B-Samples is neither consistent with the carbon isotope signature of commercial pork in the United States’.

The above Study shows that North American pig offal could cause an AAF under WADA’s own Guidelines. In addition, as highlighted in the TD2021NA, consumption of wild boar could cause the δ13C values exhibited by Houlihan. Yet despite this and despite accepting that the food truck was not using its usual supply chain, the CAS Panel accepted that the GC/C/IRMS findings reported excluded this possibility.

The CAS Panel also accepted evidence that the corn-rich diet of US pigs couldn’t cause an AAF for 19-noandosterone. The Panel was led by the fact that Prof. Ayotte quoted a 2020 Study led by Frank Hülseman as proving that eating boar meat resulted in ‘no measurable 19-NA in their urine samples’.

However, the same Study also warned: ‘The fact of varying δ13C values of wild boar could also be problematic for people living in countries with a high consumption of C4-plants like the United States or Southern Africa. Human endogenous δ13C values in these countries are enriched in 13C compared with Germany, and for these, there is the possibility of adverse analytical findings after the consumption of the meat of 19-norsteroid producing C3-fed boars.’ It continues: ‘Not only the consumption of wild boar’s offal in the hours preceding a doping control test but also the consumption of wild boar meat may result in an atypical or even positive test result, albeit the urinary NorA concentrations are expected to be lower than after consumption of wild boar’s offal. Both athletes as well as anti- doping laboratories and authorities should still be aware of this aspect.’

Houlihan consumed her burrito ten hours before her positive test. As previously mentioned, a Study performed in 2000 found that 19-norandrosterone and 19-noretiocholanolone concentrations in urine reached 7.5ng/mL ten hours after consumption of 310g of non-castrated boar meat. 

Is she lying?

As media coverage has outlined, there are many people who won’t believe Houlihan’s story. They will argue that it is a convenient tale spun by a doping cheat. It is important to acknowledge that this is a real possibility. Very few athletes admit to doping until after their career is over and they’ve found a profitable media deal.

Since Houlihan failed to prove the origin of the 19-norandrosterone, she failed to demonstrate that ADRV was unintentional. The CAS therefore ruled that her ADRV must be considered intentional, and sanctioned her with a four year ban from 14 January 2021 – the date of her provisional suspension.

Houlihan took a polygraph (lie detector) test…

Blaming her AAF on a burrito could have been an invention. Houlihan was aware of this, and attempted to convince the CAS Panel that she is telling the truth. The CAS Panel found that tests on 6cm of Houlihan’s hair; a polygraph (lie detector) test; plus the submissions of athletes and colleagues regarding her good character were not enough to prove that her ADRV was not intentional. 

Houlihan’s argument that injection of 19-norandrosterone involves a nine month detection window was not disputed by World Athletics. It argued that Houlihan’s AAF was consistent with use of oral nandrolone. This was disputed by Houlihan’s team on the basis that a one-off use of oral nandrolone would be useless in enhancing her performance. This was not disputed by World Athletics in the CAS Decision, however it continued to argue that her AAF was consistent with oral use.

Paul Greene, who represented Houlihan, is an experienced anti-doping lawyer. He would have told Houlihan that if she admitted the ADRV, she would have received a one year reduction to her ban – as did two athletes recently caught doping in competition.

Houlihan chose not to. Suppose she is telling the truth. If we accept her story, Houlihan doesn’t know for certain what was in her burrito. She only knows that it wasn’t the usual meat she was accustomed to being served by the food truck, and none of the party that had convened to watch The Bachelorette finished their greasy food. Ten hours later she tested positive, consistent with the findings of a Study performed in 2000.

She had ordered a beef burrito. She assumes she was given a pork burrito by mistake. What if it was beef? 

A CAS Panel also refused to believe Alex Wilson’s argument that meat eaten in the US was contaminated…

Alex Wilson missed the Tokyo 2020 Olympics after the Ad Hoc division of the CAS agreed with the AIU and WADA that the Swiss Olympic Committee had misapplied the rules of Anti-Doping Switzerland by asking if his AAF for trenbolone – another steroid – could have been caused by contaminated beef eaten in a Las Vegas restaurant. An institute in Strasbourg analysed a sample of hair from his beard, and concluded that epitrenbolone was ‘most likely’ to have entered his body via food intake.

Earlier this year, WADA recognised that meat contamination is an issue. It issued guidance in the form of a Technical Letter changing the evidential burden of proof regarding AAFs for clenbuterol, ractopamine, zeranol, and zilpaterol. Perhaps scientists could explain why these four steroids are a problem and others aren’t. If producers are adulterating meat to promote cattle growth, then presumably they aren’t picky about what they use.

In a more recent case, Brenton Rickard said that testing methods have improved to the point where they have surpassed contamination protocols. Although his case involved pharmaceutical products, the same is more than likely to be correct for meat contamination protocols – especially during Covid-19.  

“You just kind of realise proving your innocence is a challenge”, he told ABC. “Proving that was impossible, basically”. 

Whether you believe Houlihan’s burrito story or not, Rickard’s comment encapsulates the issue that should concern athletes. Anti-doping guidelines are drawn up in consultation with scientists who work for Laboratories that depend on accreditation from WADA. At CAS, non-scientist Arbitrators are asked to briefly become scientific experts by the sporting bodies that use these Laboratories, whose scientists are often called as witnesses in support of the anti-doping guidelines they created. It is not hard to see how one could argue a conflict of interest.

Houlihan reported negative tests one month either side of her positive. The levels of 19-norandrosterone were consistent with consumption of boar meat. There were questions as to whether the food truck was using its usual meat supply source. Studies appeared to indicate that meat contamination in North America could produce findings similar to those evident in Houlihan’s sample. 

Yet despite all of the above, the CAS Panel accepted the assertion that she was intentionally doping. Rickard is not the only athlete who has found proving their innocence impossible at CAS. Gordon Gilbert, Dematre Pena, Alex Wilson and many others have failed to convince CAS Panels.

Whether you believe their explanations is, of course, up to you. But when WADA’s own science is apparently ignored in a case likely to end an athlete’s career, then it would appear that anti-doping has a serious problem.

You may also like...

Pin It on Pinterest

Share This