Intended for healthcare professionals

Education And Debate

Cochrane at crossroads over drug company sponsorship

BMJ 2003; 327 doi: https://doi.org/10.1136/bmj.327.7420.924 (Published 16 October 2003) Cite this as: BMJ 2003;327:924
  1. Ray Moynihan, journalist (raymond.moynihan{at}verizon.net)1
  1. 11312 21st Street NW, Apt 4, Washington, DC 20036, USA

    Can independent scientific evaluation be sponsored by vested interests? The Cochrane Collaboration's decision on this question at its annual colloquium next week could have a big effect on its future

    As it turns 10 this month, the Cochrane Collaboration is at a crossroads, with its supporters deeply divided over whether to embrace drug company sponsorship. In just a decade the collaboration has fundamentally changed the way questions are asked and answered about what works and what doesn't in health care. Its rigorous methods for systematically reviewing and synthesising the best evidence have underpinned a collective move towards evidence based decision making among health professionals, policy makers, and consumers. Now, against a backdrop of growing global entanglement between doctors and drug companies, the collaboration must decide whether it can accept industry funding without compromising its reputation for integrity and independence.

    Big decisions in Barcelona

    Delegates to next week's annual Cochrane Colloquium in Barcelona will debate the thorny question of conflict of interest at the opening plenary session and in a series of high level meetings.

    When the collaboration last got together in Melbourne in March, two senior professors of medicine ended up in a serious discussion about the issue, pre-empting the forthcoming conflict in Barcelona. One thought that the organisation might not be able to remain viable without industry support, whereas the other believed that the inevitable industry influence would destroy Cochrane's credibility.

    Drummond Rennie, codirector of the Cochrane Center in San Francisco and a deputy editor of JAMA, thinks industry funding would be a catastrophe for Cochrane. “It would be far better that some reviews are not done than reviewers accept money from interested parties,” he said. In contrast, Peter Tugwell, a member of Cochrane's international steering committee and professor of medicine at the University of Ottawa, is comfortable with reviews being funded by multiple sources, including companies with an interest in the treatments being reviewed, although he is also keen to see an open debate on the subject. “I'd like to have industry as part of the mix because they're important actors to have at the table along with the non-profit foundations and the public funding agencies.” For Rennie that direction holds real dangers: “All the evidence shows that corporate money talks, and it's nonsense to think that reviewers won't be influenced consciously or unconsciously.”

    Evidence on industry sponsored research

    The growing body of evidence that industry funding is associated with favourable outcomes was synthesised in two large systematic reviews published earlier this year. Bekelman et al found a “significant association between industry sponsorship and pro-industry conclusions” in biomedical research,2 and Lexchin et al showed that studies sponsored by pharmaceutical companies were four times more likely to have outcomes favouring the sponsors' products than studies funded by other sources.3 Importantly, there is no evidence that industry funding produces studies of lower methodological quality. Rather, the authors of these reviews speculate that industry influence can affect both the research agenda and the trial design, as well as suppress or delay unfavourable findings.


    Embedded Image

    It is this sort of evidence that Professor Lisa Bero plans to introduce when she chairs the plenary session in Barcelona. A leading expert on the link between industry funding and research outcomes, Bero has the respect of both sides. Preparing for the forthcoming debate, she discovered that only a tiny number of Cochrane reviews disclose funding from a single company and only a small proportion have some degree of industry funding. She is therefore wary of the argument that certain groups within the collaboration will fail without commercial support. “I think it's a complete red herring” she said.

    Evidence showing an association between funding and research outcomes is drawn largely from analysing studies with single sponsors rather than consortia of multiple sources. Nevertheless, Bero is worried that the risks of any corporate sponsorship of reviews may outweigh its benefits. “Cochrane reviews are so well known as being an independent source of information. I don't think there should be any support for a review from a sponsor with a vested interest in the outcome.” One of her biggest concerns is perception. “We want to be seen as a definitive source for systematic reviews that people can trust.”

    Sponsored reviews?

    In order to preserve that trust, the collaboration's rule-book currently stipulates that “direct funding from a single source with a vested interest in the results of the review is not acceptable.”4 Yet the Cochrane site already contains two reviews of migraine drugs funded largely by Pfizer, which manufacturers one of the drugs.5 6 The funding is disclosed in a short note attached to the reviews, virtually acknowledging a breach of Cochrane policy but explaining the decision to publish was made because the Pfizer contract specified “all data would be made available to reviewers and that the reviewers had the right to publish the results without restriction.

    Despite the disclaimer, the publication sparked a strong reaction from some parts of the collaboration, particularly the consumer network.7 “I think it debases the currency,” said the managing editor of Cochrane's consumer website, Hilda Bastian. “The danger is we may end up with more medicines based evidence than evidence based medicine. It doesn't matter whether individual reviewers are influenced or not, no one can trust that they are not.”

    Lead researcher on both the migraine reviews, Andrew Moore, argues that systematically reviewing the raw patient data held by drug companies is the best way of finding out how well drugs work, and he strongly defends the sponsorship. “Our view is we've been open. We had explicit rules to ensure independence.”

    Industry views

    The idea of individual companies sponsoring reviews, whether solely or as part of a group of funders, has also attracted opposition from an unexpected quarter: the pharmaceutical industry itself. In 1996, the Association of the British Pharmaceutical Industry wrote to the UK Cochrane Centre expressing strong support for the organisation and the concept of preparing, maintaining, and disseminating systematic reviews. Their letter stated “we would consider it inappropriate for individual companies to be associated in any way with the funding or sponsorship of the activities of any review group.”

    One of the two signatories to that letter was Frank Wells, the previous medical director of the association, who now chairs a drug development company. Wells still believes that individual companies should not be associated with any review of a therapeutic product but thinks it is acceptable for the association to make a small annual grant to the national Cochrane centre to conduct training workshops for reviewers. For Wells, this sort of collective support is a model for future funding because companies have no chance to influence reviews. “It demonstrates the willingness of the pharmaceutical industry to say Cochrane reviews are independent,” he said.

    What is the Cochrane Collaboration

    An international organisation that prepares, maintains, and makes accessible, systematic reviews of the evidence about the effects of healthcare interventions.

    • 10 000 active participants

    • 1700 completed Cochrane reviews

    • 1300 more reviews under way, at published protocol stage

    • An estimated 10 000 reviews envisaged

    • Research in more than 80 countries

    • A register of more than 300 000 studies

    The other signatory of the 1996 letter was current medical director, Richard Tiner, who is now more open to the idea of individual companies sponsoring review activities. “The industry as a whole believes it's a vital, very important organisation,” he told me, explaining drug companies regularly draw on Cochrane reviews in their dealings with health authorities, “and its independence is absolutely important. But personally I see no problem with companies providing some support for a review–as long as it's clear where the money comes from and what role, if any, the company has played.”

    For its part, Pfizer welcomes the possibility of funding more Cochrane reviews. Roy Sutherwood, a spokesperson for the company, suggested corporate sponsorship could be a more reliable source than short term public grants. “Longer term funding might be more beneficial for Cochrane, provided that the necessary accountability and transparency measures are in place to ensure real independence.” The company, however, would not comment on the idea of collective funding mechanisms until it had seen the specifics.

    Mechanisms for sponsorship that maintain independence?

    Clearly Cochrane's dilemma is part of a global debate over growing entanglement between the pharmaceutical industry and the health professions–across education, clinical practice, and research–and discussions about where financial ties are appropriate and disentanglement is desirable8 Accepting sponsorship and maintaining independence is possible, according to plenary speaker Cindy Mulrow, deputy editor of the Annals of Internal Medicine. “If a mechanism could be established whereby industry collectively gives money to a general fund that could then be used for any type of review, I wouldn't be opposed to that.” Mulrow acknowledges that industry funding could taint Cochrane's reputation but argues that other sources of bias in research are also of concern–whether driven by the desire to maintain individual reputations or the implicit agenda of non-profit foundations or public agencies. But Peter G⊘tzsche, director of the Nordic Cochrane Centre, believes “No other sources of bias even come close.” A former industry insider, G⊘tzsche will propose a prohibition on industry sponsored reviews when he speaks at the Barcelona meeting.

    Summary points

    The Cochrane Collaboration is divided over drug company sponsorship of systematic reviews

    Supporters say Cochrane needs commercial support and could benefit from company perspectives

    Opponents say sponsorship will influence research agendas and damage independence and integrity

    Mechanisms are being floated for industry to collectively sponsor Cochrane activities and avoid perceptions of influence

    One important group, until now silent in this debate, are the public agencies who currently provide a large part of the collaboration's core funding. In England, the Department of Health's funding is signed off by the director of research and development, John Pattison. “It seems perfectly legitimate for Cochrane to consider industry funding, and it's healthy to have this open debate,” he said before adding that the proposition did raise some worrying issues. “The worry would be that he who pays the piper calls the tune, and we ended up with a situation where it was industry who were determining the nature of the work done, or worst of all, the possibility that it might have some influence on outcomes.”

    Pattison reserved judgment about whether industry funding would damage public support for the collaboration. “If the perception is that by taking industry money this would deter public funders from funding because they thought there was a loss of independence–that's a legitimate point of view, but I want to see the strength of that argument.” Although declining to advocate a particular path, Pattison did say that mechanisms to minimise influence would be highly desirable if Cochrane decides to accept industry funding. “It's very much better for any monies to come through a generalist association rather than a specific company–because then you would worry about a hidden agenda. To put it into some activity independent of specific reviews would also be another legitimate way to ensure separation.”

    Alongside the debate about industry funding, sits the even more difficult question of the individual financial ties of researchers who produce Cochrane reviews. Where does disclosure end as a management strategy and disqualification begin? In both debates, no matter how deep the complexity or the conflict, the delegates in Barcelona can take some small comfort from knowing they are by no means alone. Indeed, they will generate considerable global interest if they can design a foolproof mechanism for accepting sponsorship and ensuring independence. Perhaps they might even patent it.

    Footnotes

    • Contributors and sources RM has been reporting on medicine and health care for seven years and is a visiting editor with the BMJ. This article is based on interviews with those quoted plus Iain Chalmers, Mike Clarke, Phil Wiffin, Mark Gibson, and others who spoke on condition of anonymity.

    • Competing interests None declared.

    References

    View Abstract

    Log in

    Log in through your institution

    Subscribe

    * For online subscription