Intended for healthcare professionals


Drug companies advised to publish unfavourable trial results

BMJ 2003; 326 doi: (Published 29 May 2003) Cite this as: BMJ 2003;326:1163
  1. Debashis Singh
  1. London

    New guidelines aimed at increasing transparency and encouraging the responsible and ethical reporting of industry sponsored clinical trials have been published in Current Medical Research and Opinion (2003;19: 149-54), a peer reviewed international journal that focuses on evidence based clinical studies, research investigations, and experimental treatments.

    Although not commissioned by the pharmaceutical industry itself, the article presents the first set of guidelines developed by members of the industry (as well as journal editors and academics) for the industry itself.

    The “good publication practice” (GPP) guidelines seek greater transparency by addressing two main themes: publication bias and the relationship between drugs companies and academic investigators.

    The guidelines call on companies to publish results of all their clinical trials of marketed products, including trials with unfavourable findings—something drug companies have been criticised for failing to do in the past.

    The guidelines also recommend practical steps to reduce publication bias and redundant publication, such as multiple presentations of the same trial results.

    A unique feature of the guidelines—one not covered by existing guidelines from journal editors—is that they address the role of professional medical writers and, especially, writers' interactions with authors. Professional writers work for the drugs industry and are used to ease the publication of trials.

    The guidelines are voluntary and aspirational rather than obligatory, and there is no mechanism for policing them. Of the 75 pharmaceutical companies that have so far been informed of the guidelines, only six have endorsed them.

    Even endorsing the guidelines does not put the drug company under any obligation to follow them and to publish trials with unfavourable results. However, they can be seen as a first step in establishing a common standard and are intended to complement other guidelines, such as the uniform requirements for submission of manuscripts to biomedical journals issued by the International Committee of Medical Journal Editors and the CONSORT guidelines for accurate reporting of randomised control trials.

    Liz Wager, a member of the team that developed the guidelines, said, “We want the guidelines to be a ‘one stop shop’ for people working in or for drug companies. Until now you had to look all over the place for relevant information. Even the best guidelines did not cover all the issues. We hope that good publication practice will become widely accepted and implemented throughout the industry.”

    Professor Joe Collier, editor of the Drug and Therapeutics Bulletin, said, “The more open the industry the better. This is an aspirational agreement, and I suspect there may be loopholes in the arrangement, with companies not publishing material when it doesn't suit them.”


    View Abstract