PARIS (AP) A long-delayed study funded by the World Anti-Doping Agency says one third of athletes may have knowingly doped shortly before they competed at the 2011 track world championships - although few of them were caught at the time.
The peer-reviewed study, published in the journal Sports Medicine, estimated that doping was even more widespread at the Pan-Arab Games in 2011, with at least 45 percent of competitors thought to have doped in the 12 months before the regional multi-sports event.
The researchers said a total of 2,168 athletes at the two events participated in anonymous questionnaires upon which the study was based. The volunteers were offered the choice of replying to the question: ''Have you knowingly violated anti-doping regulations by using a prohibited substance or method in the last 12 months?''
After discounting some answers, from athletes who responded so hastily that they may have misunderstood the survey instructions or not carefully considered their response, the team of nine researchers from Europe and the United States came up with estimates of doping prevalence among athletes at the two events: 30 to 31 percent at the world championships in Daegu, South Korea, and 45 to 49 percent at the Pan-Arab Games.
They said those findings may still have underestimated the extent of cheating.
''There are many reasons to suspect that we may have undershot the true values,'' Harrison G. Pope Jr., one of the authors, told The Associated Press in a telephone interview.
Pope is a professor of psychiatry at Harvard Medical School.
Other experts who did not take part in the study said doping may now be less widespread than it was in 2011 - thanks to improved detection methods and following more recent doping scandals involving Russian athletes, in particular.
Michel Audran, director of France's WADA-accredited anti-doping laboratory, noted progress track has made against doping, with more than 100 athletes caught by a biological passport program that tracks competitors' blood and other readings over time for tell-tale signs of doping.
''It's a snapshot of the time,'' Audran said of the study. ''In my opinion, it has diminished a lot since then.''
Olivier de Hon, manager of scientific affairs at the Netherlands Anti-Doping Authority, also said he has ''good hopes'' that doping is less widespread than in 2011. He said the study's methodology was sound.
''It is an estimate but it's a pretty good estimate - within a 10 percent range of what was likely the truth at that time,'' de Hon said.
''It's a pity it took so long to publish,'' he added. ''It was really new when they conducted it.''
The authors said the delay of nearly six years between collecting the raw data and publication this week was due to negotiations with track's governing body, and between WADA and the governing body.
''I don't really know where in the system the delay occurred,'' Pope said. ''What actually happened behind the scenes was quite murky.
''I do know that we wanted to be sure that we had thoroughly satisfied everybody.''
An earlier, but subsequently revised, draft of the study was published by the British parliament in 2015.
The IAAF's new anti-doping unit said that while it could not comment on the study's accuracy, it welcomes ''any research on the prevalence of doping.''
The Athletics Integrity Unit added that it has ''no doubt'' that ''significantly'' more athletes are doping than are caught by drug testing.
''As a newly-established body, the AIU is confident that as it builds its investigations and intelligence capability to complement its testing program, the Unit will be able to better detect doping and, ultimately, be able to narrow that gap,'' it said.