The correspondence section of this month's issue discusses a controversy that arose because a review article that was published in Nature Neuroscience last year did not disclose an author's competing financial interests. The story was originally reported in the New York Times, and led to a series of follow-up articles and editorials in the popular press, many of them critical of the authors and of the journal's policies. We take these concerns seriously, and we are now extending our existing disclosure policy to include review articles.
We should begin by stating for the record that the authors of the review, Charles Nemeroff and Michael Owens of Emory University, did not violate our editorial policies. Although one might argue that it would have been prudent to volunteer a disclosure statement, given the extent of their financial interests (which included a patent, stock options and consulting fees from companies whose products were discussed in the review), the fact remains that under our policies they were not obligated to produce a disclosure, and we did not ask them to do so. There is nothing to suggest that they have sought to conceal their financial interests in the past; indeed Nemeroff has published a detailed disclosure statement in his capacity as the editor of Neuropsychopharmacology. Whether the authors' financial interests have influenced the content of an article is often difficult to judge, and we do not presume to do so on this occasion; readers can form their own opinions based on the exchange of arguments beginning on page 999.
Nevertheless, this episode underlines the importance of financial disclosure for maintaining public confidence in science, and we (along with the other Nature journals) have now changed our policy on disclosure for review articles. For the past two years, we have been publishing statements of competing financial interests for all primary research papers, and henceforth we shall request similar disclosures for reviews as well. What should be disclosed? Our website provides detailed guidelines on what constitute financial interests (http://www.nature.com/neuro/info/cfi)—including funding, employment, stocks, consultant fees and patents—but the underlying principle is 'anything that could embarrass you'. As this episode shows, the potential for embarrassment is considerable. It is also avoidable; if the review article had included a disclosure statement, there would have been no news story.
The argument for extending our existing disclosure policy to reviews is strong. Studies of the clinical literature have concluded that industry funding is associated with pro-industry results, so there is a clear prima facie cause for concern. One can argue that because review articles are inherently selective and opinionated, they provide more scope for bias than do reports of research results. Moreover, there have been clear examples of abuse, in which academic authors have been paid by pharmaceutical companies to put their names and credibility to reviews produced by ghost writers employed to boost company products. The most compelling argument for disclosure, however, is to remove suspicion. When scientists (particularly those receiving public funding) offer their professional expertise without disclosing potential financial benefits to themselves, it threatens to undermine public trust, not simply in a particular paper or journal, but in the integrity of the scientific enterprise as a whole.
The main purpose of our disclosure policy, therefore, is to maintain the credibility of the material we publish, in the eyes of the scientific community and the public. We believe our current policy has worked well for primary research, and that extending it to reviews will further strengthen the journal. Like all the Nature journals, we require each paper to state whether the authors do or do not have competing financial interests, and authors can expect to be held publicly accountable for the truthfulness of what they say. The disclosure rate for primary research papers in Nature Neuroscience last year was 5.5%. Although many other biomedical journals have formal disclosure policies, few papers carry any disclosure statement (about 0.5% according to one survey), leaving readers uncertain whether there was nothing to disclose or whether the authors simply neglected to mention their competing interests despite the journal's stated policy. It is tempting to speculate that the difference reflects authors' reluctance to tell lies of commission as opposed to lies of omission.
Of course financial interests do not invalidate research studies or review articles. Although the Center for Science in the Public Interest calls for “rejection of submissions where authors' conflicts are incompatible with integrity in science” (see their letter on page 1001), we believe such cases are very rare. One can argue for more stringent safeguards in clinical trials, but for basic research of the type that appears in Nature Neuroscience, we do not believe editors should ever reject a paper simply because the author has a financial stake in its contents. Our aim is not to conduct a moral crusade, and it should be obvious to even the 'purest' academics that the practical benefits of academic research cannot materialize without the help of industry.
As Zach Hall notes in his book review (page 1003), the public interest is not served by stigmatizing commercial research. Academia and industry are increasingly intertwined, particularly in the US, and such partnerships can offer significant benefit to scientific progress. The challenge is to manage the relationship in a way that does not undermine academic values such as open communication, prompt publication and the perceived integrity and objectivity of the scientific community. Many of these concerns are best addressed by public disclosure, which journals are uniquely well placed to promote. Journals cannot eliminate all tensions arising from commercialization of academic research, of course, and ultimately these issues must be resolved in the marketplace of ideas. Our ambitions are more modest; we will judge our policy successful if it keeps us—and our authors—on the science pages of the newspapers rather than alongside the business scandals.
About this article
British Food Journal (2009)
The social and cultural shaping of medical evidence: Case studies from pharmaceutical research and obstetric science
Social Science & Medicine (2006)
Journal of Affective Disorders (2006)
Australasian Psychiatry (2005)