Medical journals are doing poorly in adhering to their own guidelines on disclosing financial conflicts, said Anu Gupta from Yale University at last week's meeting.
The Uniform Requirements for Manuscripts to Biomedical Journals recommend that all published studies should include information on sources of funding, financial conflicts of interest of the authors, and specific descriptions of “the type and degree of involvement of the supporting agency.” Over 500 journals, including the BMJ, subscribe to these requirements.
Gupta and her fellow contributors examined whether these requirements were met in 268 randomised controlled trials published by the Annals of Internal Medicine, BMJ, JAMA, the Lancet, and the New England Journal of Medicine. Just over a third were supported wholly or in part by industry, and only 9% failed to give the source of funding. In the trials supported by industry a third did not provide any information on the authors' relations with industry.
The type and degree of the involvement of the funding source was disclosed in only 8% of cases, and all these disclosures were in the Annals of Internal Medicine. The other journals, including the BMJ, failed completely to disclose the nature of the involvement. The journals did not need, said Gupta, to introduce new requirements on disclosure of involvement of sponsors—as they did last week (15 September, p 588)—rather, they needed to implement the guidelines they had.
Frank Davidoff, former editor of the Annals, explained that he had been sensitised to this issue after one set of authors repeatedly failed to tone down their conclusions despite editorial requests. When Davidoff phoned to ask why, they explained that the unidentified sponsors didn't want them to do so.
Footnotes
Congress on peer review in biomedical publication Reports by Richard Smith BMJ