Don't miss a moment from Paris-Roubaix and Unbound Gravel, to the Giro d’Italia, Tour de France, Vuelta a España, and everything in between when you join Outside+.
By Agence France Presse
Anti-doping tests used at the Olympics and other major sporting events are too often based on faulty science and statistical methods that can yield erroneous results, a researcher charged Wednesday in a leading scientific journal.
Donald Berry, an expert in biostatistics at the University of Texas, used the case of American cyclist Floyd Landis to point up flaws in anti-doping procedures, but cautioned that the problems he uncovered apply across the board to lab tests designed to ferret out athletes who cheat by using performance enhancing substances.
Landis was stripped of his 2006 Tour de France cycling race victory after a drug test showed he had taken synthetic testosterone to boost his endurance.
The cyclist has maintained his innocence, but in June the International Court of Arbitration for Sport dismissed his final appeal to overturn the test results.
Writing in the British science journal Nature, Berry argues that the tests performed by the French national anti-doping laboratory (LNDD) that condemned Landis to ignominy and barred him from competition for two years were “non-informative” and potentially subject to error.
While Berry does not have an opinion as to the cyclist’s guilt or innocence, he is highly critical of what he called the “inherent flaws” in current testing practices.
“The situation in drug-testing labs worldwide must be remedied,” he said. “Cheaters evade detection, innocents are falsely accused and sport is ultimately suffering.”
The same type of tests that ruin careers would be rejected by regulatory agencies such as the U.S. Food and Drug Administration as scientifically unsound for diagnostic tests to detect disease, he told Agence France Presse in an interview.
International Olympic Committee (IOC) medical director Patrick Schamasch, contacted ahead the article’s publication, would not comment directly on the study but said: “What we are doing in the area of doping is the most advanced in terms of certitude.”
He compared the level of precision used in anti-doping analyses to legal standards used in assessing medical data.
At issue are two kinds of intertwined error, Berry argues.
One is a type of reasoning sometimes called the “prosecutor’s fallacy”, which conflates the odds that something could happen by chance alone with the probability of innocence or guilt.
If there is only a one-in-100,000 likelihood, for example, of a random DNA match for a crime suspect, one might be tempted to conclude that the chances he is innocent are the same — one-in-100,000, or 0.001 percent.
But suppose the suspect lives in a city with million residents. That means about 10 other people have the same DNA profile, and that any one of them ? absent other evidence ? could have been at the crime scene. Suddenly the suspect has gone from almost certain guilt to likely innocence.
A woman in Britain and another in the Netherlands were both convicted within the last decade of multiple murder on the basis of such reasoning, though the first case was overturned in 2003 and the other is under review.
In the case of Landis, who had no previous record of doping violations, the chances that the positive result could result from anything except cheating ? a lab error, an abnormally high natural occurrence of testosterone ? were dismissed as not credible.
The problem, says Berry, is that for the actual process used by doping labs there is no body of scientific data to show just how rare “false positives” or “false negatives” really are, and that such data is essential for interpreting lab results.
World Anti-Doping Agency (WADA) guidelines say that any ratio of two specific types of testosterone in the body higher than four-to-one is suspicious, and should trigger further tests.
“It is proper to establish threshold values such as these, but only to define a hypothesis. A positive test criterion requires further investigation of known samples” and individual variation, Berry said.
A study published last month highlights just how variable testosterone test results can be from one individual to the next.
Researchers at the Karolinska Institute in Stockholm gave a single dose of testosterone to three groups of men who had either one, two or zero copies of a particular gene, known as UGT2B17.
More than 40 percent of the men missing the gene went undetected by the doping test, while those with two copies of the gene showed testosterone levels 20 times higher.
Once statistical adjustments were made for the genotypes, the differences disappeared.
Berry also pointed to the need for careful sample handling, advanced technician training, and precise instrument calibration during the lab analysis of the urine samples upon which the doping tests are based.
“The process in unlikely to be error free,” he said, pointing out that the American Arbitration Association that ultimately ruled against Landis initially threw out the result of the French lab screening due to improper procedures.
He also called for greater transparency, including standard testing procedure, unambiguous criteria for determining positive results, and blinded experiments.