Evaluation of conflicts of interest among systematic review authors on pharmacological therapies for alcohol use disorder: A cross-sectional study

Abstract

Introduction and aims: The high prevalence and economic burden of alcohol use disorder (AUD) requires methodologically sound research to guide treatment decisions. Systematic reviews (SR) are fundamental to clinical decision making as they collate results of all studies for a given topic and provide summaries of the clinical evidence. Bias resulting from industry relationships can compromise the validity of SRs. Thus, we investigated financial conflicts of interests (FCOI) within SRs involving the pharmacologic treatment of AUD. Design and Methods: We conducted a systematic search of PubMed, Embase, and the Cochrane Database to retrieve SRs published between September 2016 and December 2019. Screening and data extraction were performed in duplicate, blinded fashion—collecting characteristics of the reviews including the authors’ FCOI statements. The results and discussion sections were rated as favorable or unfavorable toward the treatment drug. A stepwise method was used to identify undisclosed FCOIs for all SR authors. Results: Nineteen (of 51) authors representing 7 (of 12) SRs were found to have FCOI. Among reviews with conflicted authors, 3 of 7 (42.9%) results sections were favorable toward the treatment drug, whereas 6 of 7 (85.7%) conclusions were rated as favorable. Discussion and Conclusions: More than one-third of SR authors and nearly two-thirds of studies were found to have FCOIs. Our investigation shows that financial ties are frequent among SR authors on AUD pharmacotherapies and that discussion sections often favored the drug for which the sponsor provided payments.

Keywords:

• Financial conflict
• alcohol
• alcohol use disorder
• alcohol use therapy

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

No financial or other source of support was provided during the development of this protocol. Vassar reports grant funding from the National Institutes of Health, the U.S. Office of Research Integrity, and Oklahoma Center for the Advancement of Science and Technology, all outside the present work. All other authors have nothing to report.