- Journal List
- BMJ
- v.323(7322); 2001 Nov 17
- PMC1121658
Declaring competing interests
Editor—Hussain and Smith have provided evidence of a small, but increasing proportion of articles declaring competing interests in a sample of selected medical journals.1 Krimsky, however, in a study including over 60 000 articles from a much wider representation of scientific journals, found that competing interests were reported in less than 1% of articles, despite the policies endorsed by the journals.2
Both these data and those by Hussain and Smith are in striking contrast with a study that analysed 789 articles written by authors from Massachusetts universities and reported that, in about one out of three cases, at least one author had a vested interest in research.3 Even though the rates of disclosure are increasing as a result of more stringent editorial policies, they are still unacceptably low.
Furthermore, disclosure of competing interests is not even requested by many medical journals and scientific meetings, particularly in Europe.3 The main scope of many scientific meetings currently is apparently to sell the participants to the sponsor.4 It is curious, however, how disclosure, even when requested, is seldom endorsed to its full consequences. Few journals require disclosure of competing interests by reviewers or ask their editors to disclose their own interests. This could be particularly important in view of the increasing emergence of special interest groups, self selecting academic oligarchies with strong corporate ties that have key roles in academic settings, medical journals, and research agencies.4
Disclosure of competing interests is simply a first, necessary, but insufficient step for a better credibility of medical information. It is not even practised. Other strategies are required, as described in detail elsewhere.4 A substantial problem in developing effective strategies towards a better control of competing interests in clinical medicine lies in the lack of recognition that this problem cannot be solved by medical editors alone.
References
- BMJ. 2001 Nov 17; 323(7322): 1187. »
- Types of competing interests would be of interest
Types of competing interests would be of interest
Editor—Lack of reporting of competing interests, as outlined by Hussain and Smith in their article, is probably most disturbing when it pertains to articles that have a major influence on decision making and clinical practice.1-1 For example, the low rate of reporting of competing interests in letters is unfortunate but probably not that disturbing, since letters rarely have a major impact on clinical practice, even if they express strong opinions. Editorials and original research are probably more influential, but the subject matter should also be taken into account. It makes a difference if the paper describes a randomised trial, a prognostic model, a case series, or a basic science report without any direct clinical promise.
We are therefore interested to know the rates of reporting of competing interests for specific categories of original papers and editorials at “high risk”—for example, those pertaining to randomised trials, decision analyses, cost effectiveness analyses, and meta-analyses of therapeutic and preventive interventions.
We examined the reporting of competing interests in guidelines of therapeutic and preventive interventions published in prestigious journals.1-2 The reporting rates were disappointingly low even for these influential publications. We would also be interested to know the specific competing interests reported in the 52 articles that mentioned something.1-1 Were they serious enough, or were the serious conflicts still undeclared? In our evaluation, most of the specific conflicts that were actually being declared entailed minor issues, such as travel expenses, speakers' honorariums, consultation, and such like.1-2 We hardly found authors (other than company employees) who had acknowledged being on the payroll of a company or owning company stock, despite the fact that both of these situations are probably not uncommon among high profile experts. Information on the types of competing interests disclosed is necessary to judge whether this sensitive issue has improved recently.
Finally, it would be interesting to know whether there are specific variables, other than year of publication, associated with reporting of competing interests—for example, the number of authors or the source of funding for each publication.1-2
Facebook
Twitter
Google+