Researchers from the University of Toronto analyzed the publications from 164 randomized Phase III clinical trials and found that a third were reported positively despite not meeting the primary objective, by emphasizing other, less important outcomes. “These reports were biased and used spin in attempts to conceal that bias,” the authors wrote. Some studies even changed the primary objective halfway through, possibly because early results suggested the trial would otherwise fail.
The researchers also found evidence of bias in the reporting of toxic side effects in two-thirds of the trials, particularly in trials that found the treatment was effective for its primary objective. In these cases, high toxicity findings were omitted from the abstracts and conclusions, and instead buried in the “small print” of the article.
The authors conclude that journal editors, regulators, and reviewers should apply a more critical eye to trial reports. Meanwhile, Richard Francis, research manager at UK-based charity Breakthrough Breast Cancer, told The Telegraph: “We strongly believe that making clinical trial data freely available would decrease the potential for spin by allowing more researchers to investigate the results.”