Data row sparks research debate

November 25, 2005

An urgent debate on the responsibilities of academics when authoring research papers was demanded this week after it emerged that university researchers collaborating with a pharmaceutical company had published findings without full access to the drug trial data on which the conclusions were based.

The case has highlighted a divide between academics and industry over what is considered acceptable practice when companies sponsor academics to produce joint research papers.

The pharmaceutical industry argues that it is standard practice to limit academics' access to the results of their multimillion-pound drug trial databases, but leading scientists argue that academics must have complete access to all findings for there to be genuine research partnerships. They say researchers should put their names to publications only if they have had full, unfettered access to data and can independently verify the conclusions.

Sir David Weatherall, emeritus regius professor of medicine at Oxford University, said: "There needs to be a discussion about this issue; people need to know what role academics are playing when they are putting their names to published papers."

Documents obtained by The Times Higher reveal that research findings on Procter & Gamble Pharmaceutical's osteoporosis drug Actonel were released under the name of Sheffield University researchers despite the fact that the academics had not carried out their own independent analyses of the firm's drug trial data. The academics, from Sheffield's Bone Metabolism Research Unit, published a journal article with the incorrect declaration that "all authors had full access to the data and analyses".

The company initially refused requests to allow the academics to carry out their own independent analyses of the data, and when one of the research team finally gained supervised access to some data, he expressed doubts about the conclusions being drawn.

P&G this week said that it was "standard industry practice" to limit external access to raw drug trial data generated at great expense. A spokesman said it was "typical" for P&G's internal statisticians to carry out analyses developed in collaboration with external researchers.

"Occasionally, the researcher is given temporary and limited access to the data, to perform the analyses directly," she said.

She added that all the Sheffield researchers had had "sufficient access" to "be confident and comfortable with the conclusions they state in scientific communications".

Guidelines on the conduct of clinical trials from the Pharmaceutical Research and Manufacturers of America state that "all authors, whether from within a sponsoring company or external, will be given the relevant statistical tables, figures and reports needed to support the planned publication".

But they are clear that, "as owners of the study database, sponsors have discretion to determine who will have access to the database" of drug trial results.

The Association of the British Pharmaceutical Industry said data access would be subject to individual contracts between researchers and companies.

But Professor Weatherall said that if academics had only a limited role to play in analysis of results, then this should be made clear. He said restrictions on academics' access to data might be common practice "but that doesn't mean it is correct practice".

Stephen Evans, professor of pharmacoepidemiology at the London School of Hygiene and Tropical Medicine, said a company should allow an independent statistician the opportunity to rerun analyses if this is requested by academic partners.

Concern that academics were lending prestige and credibility to industry-sponsored research without full access to data was highlighted four years ago when journal editors issued an unprecedented joint statement on the issue.

In autumn 2001, 13 of the world's leading medical journals simultaneously published an editorial warning that they had sometimes published "studies where the declared authors have not participated in the design of the study, had no access to the raw data and had little to do with the interpretation of the data".

The International Committee of Medical Journal Editors published guidelines declaring that academic authors should state that they "had full access to all of the data" and that they "take complete responsibility for the integrity of the data and the accuracy of the data analysis".

But Richard Horton, editor of The Lancet , said: "To ensure completely independent science, we must require all research to be analysed by independent scientists - scientists without ties to the pharmaceutical sponsors of research."

The journal editors are meeting in January to consider tougher restrictions for publications.

You've reached your article limit.

Register to continue

Registration is free and only takes a moment. Once registered you can read a total of 3 articles each month, plus:

  • Sign up for the editor's highlights
  • Receive World University Rankings news first
  • Get job alerts, shortlist jobs and save job searches
  • Participate in reader discussions and post comments
Register

Have your say

Log in or register to post comments