How honest can a medical journal be?

‘Financial conflicts of interest threaten patient care, taint medical information and raise costs. They create deception, impair physicians’ judgement and reduce their willingness to be their patients’ advocates.’1 Strong stuff, offered recently by Jerome P Kassirer, former Editor-in-Chief of The New England Medical Journal. Is Dr Kassirer's indignation justified? Patient care may be threatened when new drugs are promoted on the basis of flawed research. For instance, studies sponsored by pharmaceutical companies are around four times more likely to have outcomes favouring the sponsor's product than studies with other funding, apparently owing to biased choices of comparator drugs (seriously inferior, or at unusually low doses), and to decisions not to publish less favourable results.2, 3 In 1998, three major studies without sponsorship from the pharmaceutical industry found that women taking third-generation oral contraceptives had a slightly higher and significant risk of venous thrombosis than those using older pills. Three studies sponsored by the newer drugs’ manufacturers, however, found no such excess risk. The companies insisted – in a wide range of publications and in promotions to prescribers – that the studies finding higher risks suffered from bias and confounding.4 These criticisms were refuted by reanalyses and medicine regulators, and warnings about the risk were added to prescribing material and patient information leaflets for third-generation contraceptive pills. But, in the meantime, how many doctors’ clinical judgement was impaired by misleading information and, more importantly, how many women on third-generation pills had thromboses? Many doctors who lack the time and academic interest to read original research rely instead on review articles and clinical guidelines to keep them up to date. But such articles are increasingly becoming a form of product placement. Drugs are not the only products promoted in this way: others include surgical devices5 and, of course, tobacco.6 One author ran into trouble when the New York Times pointed out that a review article he co-wrote for Nature Neuroscience– on treating mood disorders – did not mention his financial competing interests in some of the treatments he reviewed favourably.7 The author said he had not been asked by the journal to declare any conflicts of interest, and the journal confirmed this, so the blame did not rest entirely with the author. How clean, then, are editors’ hands when it comes to conflicts of interest? Pretty dirty, according to Richard Smith, former editor of the BMJ, because editors do not try hard enough to encourage disclosure and transparency. Indeed, Smith accuses journals of being little more than pharmaceutical marketing departments,8 a sentiment echoed by Richard Horton, editor of The Lancet, who says many editors are simply laundering information for the industry.9 The International Committee of Medical Journal Editors (ICMJE) sets the uniform requirements for presenting and submitting manuscripts. The committee holds authors of research papers, editorials and review articles responsible for disclosing all financial and other interests – including personal relationships, academic competition and intellectual passion – which might bias their work (http://www.icmje.org/#conflicts). It also spells out how to handle conflicts of interest among reviewers and editors. The World Association of Medical Editors (WAME) echoes the ICMJE requirements and provides editors with a forum for debating these and other editorial policies (http://www.wame.org). And many medical journals, including the big five general ones –The New England Journal of Medicine, JAMA, Annals of Internal Medicine, The Lancet and the BMJ– follow the requirements and make them clear to their authors. Yet, too many others do only the bare minimum. They ask authors to follow the uniform requirements, but actively implement only the most basic rules on declaration of financial competing interests relating to submitted original research papers. Relatively, few journals insist that authors should describe in their manuscripts the role of sponsors in study design and execution and in the writing and submission of papers. Fewer still publish statements of the conflicts of interest of journal staff and editorial board members. Some editors believe that the only realistic way to deal with conflict of interest is to create a climate of real openness and trust. If authors, reviewers and editors feel free enough to disclose any interests, however, large or small, then readers can make up their own minds about possible bias. Taking this a step further, the BMJ now has a clear and comprehensive policy on transparency that covers all of the issues around conflict of interest (which we prefer to call competing interests, a less pejorative term) and more at http://www.bmj.com. For instance, the BMJ has had for 7 years a policy of open peer review, with all reviewers signing their reviews and no additional comments being made to the editor behind authors’ backs. Some journals have set financial thresholds above which they consider authors – at least of review articles, commentaries and editorials – to be unacceptably conflicted. For both the Canadian Medical Association Journal (http://www.cmaj.ca/cgi/content/full/171/11/1313?etoc&eaf) and The New England Journal of Medicine that threshold is $US 10 000, made in the year preceding submission of the manuscript. For nearly a decade, the New England Journal was even tougher: it refused to consider reviews and editorials written by anyone with any financial interest at all in companies (or their competitors) that made products discussed in such articles. In the end, though, the editors ran out of authors and had to be more realistic.10 Given all of this, are any journals honest enough? Not according to Merrill Goozner, formerly professor of journalism at New York University and chief economics correspondent for the Chicago Tribune, now director of the Integrity in Science project. Mr Goozner argues that editors cannot publish what they do not know about: journals’ definitions of financial conflict of interest are too narrow, giving authors loopholes by which they can avoid disclosure. And, even when authors do declare important conflicts, editors often fail to publish them – either through unduly liberal interpretation or gross oversight.11 Insisting on disclosure of all financial arrangements with private firms within the past 3 years, whether or not those arrangements are directly related to the subject of the article, would help authors and editors to clean up their act, asserts Mr Goozner. So would sanctions, such as a 3-year ban on publication within the pages of a journal in which an author failed to disclose a conflict of interest. Are any of us ready for this? Trish Groves plays a key role in developing the BMJ's editorial policies, particularly its transparency.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.