Published online 26 October 2010 | Nature | doi:10.1038/news.2010.564

News

Questions raised over medical journals' financial ties to industry

Warning flag raised over profitable reprinting of industry-funded papers.

Medical journals seeking to expose the financial ties of their authors ought to look in the mirror, the authors of a new study contend.

JournalsJournals are under scrutiny over their potential conflicts of interest.Getty Images

They find that industry-funded clinical trials boost the prestige of medical journals, which also stand to profit from selling reprints of such papers.

"The best journals have their own interest in publishing industry-sponsored drug trials because it will lead to various incomes, for example reprints," says Peter Gøtzsche, director of the Nordic Cochrane Centre in Copenhagen and an author on the study, which appears online today in PLoS Medicine1.

Gøtzsche's team measured the citations of 1,353 randomized controlled clinical trials published in 1996–97 and 2005–06 by the New England Journal of Medicine (NEJM), the Journal of the American Medical Association (JAMA), The Lancet, the British Medical Journal (BMJ), Archives of Internal Medicine and Annals of Internal Medicine.

Industry-funded trials tended to rack up more citations than trials not supported by industry, the team found. As a result, these studies boosted the journals' impact factors — a figure roughly equal to the average number of citations each article gains that is used widely across academic publishing as a measure of a journal's prestige.

For instance, had they not accepted any trials receiving at least some support from industry, the NEJM 's impact factor would have dropped by 13% for 1996–97 and by 15% for 2005-06, The Lancet 's by 6% and 11%, and JAMA 's by 5% for each period, Gøtzsche's team estimates. The BMJ's impact factor was the least influenced by industry-sponsored trials.

Money-spinners

Because journals sometimes make a lot of money selling reprints of their articles to pharmaceutical companies, there is a potential conflict of interest, the authors argue. The Lancet told them that, between 2005 and 2006, it earned 41% of its income selling more than 11 million reprints. Meanwhile, the BMJ indicated that 3% of its income from that period came from the sale of nearly one million article reprints. The publishers of the NEJM, JAMA, Archives of Internal Medicine and Annals of Internal Medicine told Gøtzsche's team that it is their policy not to disclose financial information.

Gøtzsche and his colleagues say that there are a number of reasons why industry-sponsored trials might be more highly cited than others — a finding echoed by previous work2,3. Drug companies often publish numerous follow-on studies and reviews that cite major clinical trials. Industry-led trials also tend to be larger than independent trials, which could lead to increased citations, the authors note.

“The editors who work here don't have a clue what the business side is doing, and I would prefer it stay that way.'”

Richard Horton
The Lancet

Harvey Marcovitch, a director of the Council of Science Editors and author of a commentary on the paper4, thinks that journals' conflicts of interest deserve closer attention, but he doubts that major medical journals, staffed by professional editors, accept papers to push reprint sales. He does worry that smaller medical journals, particularly those in fields such as psychiatry that publish numerous drug studies, could be influenced by reprint sales[. Marcovitch currently works as a freelance associate editor at the BMJ.

Richard Horton, editor-in-chief of The Lancet, agrees that there is potential for medical journals to be influenced by the drug industry. His own journal addresses this issue by walling off editorial decisions from business operations, he says.

Complete disclosure

To guard against financial conflicts, Gøtzsche says that journals should disclose the income they receive from reprints and other sources. His co-author Andreas Lundh also recommends that journals publish the income generated from individual papers in the same way that some publish citation statistics along with each paper.

ADVERTISEMENT

Christine Laine, editor of the Annals of Internal Medicine, said in a statement that the journal would consider disclosing its income from reprints. A statement from the BMJ noted that it already provides information about its revenue sources.

The NEJM declined to comment, as did the American Medical Association, which publishes JAMA and Archives of Internal Medicine.

Horton says that editors at medical journals should work to draft a joint policy on disclosing their own conflicts of interest. However, he worries that reminding journal editors about the revenues generated from industry could make them more reactive to such interests. "The editors who work here don't have a clue what the business side is doing, and I would prefer it stay that way," he says. 

  • References

    1. Lundh, A., Barbateskovic, M., Hróbjartsson, A. & Gøtzsche, P. C. PLoS Med. 7, e1000354 (2010).
    2. Shams I (2007) PLoS ONE 2(5): e403. Kulkarni, A. V., Busse, J. W. & Shams, I. PLoS ONE 2, e403 (2007).
    3. Conen, D., Torres, J. & Ridker, P. M. Circulation 118, 1321-1327 (2008). | Article | PubMed
    4. Marcovitch, H. PloS Med. 7, e10003555 (2010).

Comments

If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

  • #60872

    Another aspect of free access would be to require scientific publishers to electronically share the authors' data with their readership. I think we are only allowed to see the results after analysis by the author. Frequently, the author's analysis is poorly done or inappropriately done.

Commenting is now closed.