Conflict of interest in medical journals
- John Dowden
- Aust Prescr 2015;38:79-81
- 1 February 2015
- DOI: 10.18773/austprescr.2015.001
Competing interests are everywhere. Everyone has a range of interests and these interests have the potential to conflict with each other. These conflicts are of particular concern in medical publishing because biased information can have adverse effects on practice. The competing interest may be personal, academic or intellectual, but most attention is paid to direct financial conflicts of interest.1 For many medical journals, particularly those focused on therapeutics, the influence of the pharmaceutical industry has to be considered.
There is evidence of the widespread influence of the industry. A systematic review found that 23–28% of academic investigators receive industry funding, and industry-funded studies are likely to produce pro-industry conclusions.2 An Australian study of 1500 clinicians found that while only 6% had been paid by industry, 23% had served on an industry advisory panel, 52% had accepted travel sponsorship and 96% had accepted gifts.3 The gifts that have been commonly offered to Australian medical specialists include food and items for the office or for personal use.4
Conflicts of interest may be hidden or not reported. A review of 29 meta-analyses of 509 drug trials found that the authors' financial interests were only disclosed in 26% of the trials. None of the meta-analyses reported on the financial links between authors and industry in the trials they analysed.5 A 2011 review of guidelines listed by the National Health and Medical Research Council found that only 15% had published conflict of interest statements.6 While many Australian universities have policies on conflicts of interest, few require their staff to make regular declarations of their interests.7
Asking authors to declare their interests over the previous three years is one way medical journals identify competing interests. The International Committee of Medical Journal Editors has produced a standard form authors can use.1 Since 1996, Australian Prescriber has been asking authors to declare any conflicts of interest. This policy was later extended to include the specialist referees who review the articles. The members of the Editorial Executive Committee have to make annual declarations of their interests in accordance with the policies of our publisher NPS MedicineWise.
The International Society of Drug Bulletins (ISDB), of which Australian Prescriber is a founder member, encourages its members to have policies on conflicts of interest. Members without their own policies can use an adapted version of the conflict of interest form produced by the International Committee of Medical Journal Editors. There is, however, now a view within ISDB that this is insufficient to prevent the publication of possibly biased information. A proposal that member bulletins should not publish material written by authors with competing interests is being considered. This would include the editorial team as well as external authors. While only publishing articles written by authors with no competing interest is a noble aim, is it practical?
In the 1990s, the New England Journal of Medicine decided that authors of its editorials and review articles should have no financial interests in the companies whose products are discussed in the journal. This policy had to be revised in 2002 because of the difficulties in finding authors with no conflicts of interest. In a two-year period the journal was only able to publish one article about a new drug therapy.8 If finding authors with no conflict of interest is difficult in the USA, with its huge population, how hard will it be in Australia? With limited access to other sources of funding, it is highly likely that anyone involved in researching new drugs in Australia will have received some support from a pharmaceutical company.
During 2014 Australian Prescriber published 35 editorials and articles. In 11 of these, one or more authors declared an interest. Should we be as concerned about an author who declares funding from the National Health and Medical Research Council as we might be about someone who obtains research funding from a pharmaceutical company? What about an author who works in an academic institution that holds a global licence for a product? Should we exclude someone who is an adviser to the Therapeutic Goods Administration, but has also been an adviser to industry? There are many possible questions about potential conflicts of interest, but the Editorial Executive Committee believes that those 11 articles should still have been published.
While publishing declarations of interest at the end of articles may not solve all the difficulties of competing interests, it informs readers. Journal readers are quick to comment if their perceptions about a conflict of interest differ from those of the authors.9-12
The Editorial Executive Committee does not think it should refuse to deal with people who may be very knowledgeable about a treatment because they have participated in industry-funded research. Often their expertise is the source of the conflict. Although assessing conflicts of interest can be difficult, the Editorial Executive Committee believes that the disclosure and peer-review processes of Australian Prescriber mitigate the risk of bias.
Competing interests are everywhere, but they can be managed.
John Dowden is Editor of Australian Prescriber.
Editor Australian Prescriber , Editor Australian Prescriber