Pharmaceutical companies are under increasing pressure to release previously hidden data on how well their drugs work. The primary push for much greater transparency has come from the Cochrane Collaboration, an international network of experts based in Oxford, England; the British Medical Journal, and the European Medicines Agency, which recently proposed that, starting next year, clinical trial data be released once a drug is approved for marketing.

The saga of Tamiflu, an anti-flu drug produced by Roche, a multinational company based in Switzerland, exemplifies the difficulties of extracting information vital to public health. In mid-2009, a Cochrane researcher was hired by the British and Australian governments to update his previous evaluation of Tamiflu, which had concluded that it reduced the risk of complications from the flu. The information was important because the drug was being stockpiled in the belief that it would help people survive a potentially severe epidemic of swine flu.

As described in a Times story, eight of 10 studies that supposedly attested to its effectiveness had never been published. In December 2009, after failing to get the missing data from Roche because of a refusal to sign a confidentiality agreement, the Cochrane researchers reported that Tamiflu had not been shown to reduce complications like pneumonia or hospitalizations. The British Medical Journal reported on its own investigation showing that Roche had hired ghost writers to author Tamiflu articles, some of whom felt pressure to highlight the positive.

The double punch created a sensation and ultimately led Roche to agree to give the Cochrane team the data it wanted. Roche says it will release detailed clinical data not just on Tamiflu but on other drugs. Another big company, GlaxoSmithKline, has announced that it will share detailed data from all clinical trials since 2000. That is an important move that should become standard for an industry whose credibility is at stake.

From an editorial in the New York Times