Declaring conflicting interests

In June, the Journal ran a debate triggered by critical correspondence to a paper in The Lancet by Hickie and Rogers (Boyce and Malhi, 2012; Hickie and Rogers, 2011). The letters published in The Lancet all declared that the claims made of agomelatine’s efficacy or tolerability could not be supported by available data (Barbui and Cipriani, 2012; Carroll, 2012b; Howland, 2012; Jureidini and Raven, 2012; Lloret-Linares et al., 2012; Serfaty and Raven, 2012). One of the letters explicitly noted that acceptance of benefits from agomelatine’s manufacturer by the article’s authors may have biased their interpretation of the evidence (Lloret-Linares et al., 2012). Two others clearly implied that the review may have been penned for ‘promotional effect’ (Carroll, 2012b) or that Elsevier might gain improper financial benefit from its publication (Jureidini and Raven, 2012). Each comment ended with a declaration that no author had a conflict of interest.

We felt the apparent lack of conflicting interests was counter-intuitive. It seemed unlikely that so many academics would have sufficient expertise to provide objection to Hickie and Rogers’ review, yet themselves have no involvement with industry that might be regarded as a possible source of conflict. The Lancet’s ‘Information for Authors’ document states that ‘conflicts of interest exist if authors … have financial or personal relationships with other people or organisations that could inappropriately influence (bias) their actions’ (The Lancet, 2011). We ran a simple internet search of each author’s name to look for evidence of any relationship that might be a source of bias.

Despite declaring no conflict of interest, one or two authors in each of three of the letters was currently receiving, or had recently received, industry support for their research (Cornu et al., 2011; Serfaty et al., 2010; United States National Institutes of Health, 2011a, 2011b, 2011c). In each case the company involved produced a product with claimed efficacy in mood disorders. Authors of two letters did not disclose in The Lancet correspondence, but had disclosed elsewhere that they recently received ‘honoraria’ or ‘compensation’ from pharmaceutical companies active in neuroscience (Carroll et al., 2012; Cornu et al., 2011).

The Lancet also holds that potential sources of bias include, not only financial and personal relationships, but ‘academic competition’ and ‘intellectual passion’ (The Lancet, 2011).

Two authors obviously have an ‘intellectual passion’ for this issue, which they too did not declare or did not regard as a conflict. One was the immediate past chair of Healthy Skepticism (Healthy Skepticism, 2011); another blogs extensively about the effects of pharma influence on academic psychiatry (Carroll, 2012a).

Experts in a field will very frequently hold relationships or passions that could influence their opinions or, at least, create a reasonable apprehension of bias. If the value of disclosure policies is to be maintained, it is essential, even in correspondence, that expert authors adhere to journal guidelines and fully disclose such conflicts.