Invited article: Conflicts of interest for authors of American Academy of Neurology clinical practice guidelines

Background: Clinical practice guidelines (CPGs) shape clinical care worldwide but are prone to potential error and bias due to conflicts of interest (COI). Objective: To explore the extent and scope of American Academy of Neurology (AAN) guideline author reported COI and implications for management; and to review process of AAN guideline COI management to highlight challenges, establish comparative benchmarks, and identify areas to be improved. Methods: Authors of AAN clinical practice guidelines with an active membership panel completed a COI reporting form. Authors were asked to report current interests including the 1 year prior to the date of completing the form. Interests include personal income relationships (consulting, speaker's bureaus, advisory boards), equity (stocks/stock options), patent/royalties, research, clinical practice, fiduciary interest in a company, and expert testimony. Comparisons were made between the two committees that oversee CPG development at the AAN: the Quality Standards Subcommittee (QSS) and the Therapeutics and Technology Assessment (TTA) Subcommittee. Results: There were 50 CPG with an average of 8.5 authors per CPG. There were a total of 425 available authors, 351 of whom completed a COI reporting form (83% response rate). Forty-six of the 50 guidelines had at least one author with a COI. The most commonly reported COIs were research-related (45% of authors), clinical practice-related (42%), and personal income relationships (33%). Authors of QSS guidelines were more likely to have personal income COIs with pharmaceutical and medical device companies (39% vs 24%, p < 0.01), whereas authors of TTA guidelines were more likely to have clinical practice-related COIs (50% vs 38%, p < 0.05). A minority of authors had individual COIs exceeding >$25,000 or had multiple interests (> 10) that overlapped with content of the guidelines. Conclusion: Conflicts of interest are common for authors of American Academy of Neurology clinical practice guidelines across many domains of personal and professional interests. More research is needed to improve the methods to identify and quantify the types of conflicts and their potential biasing effects on selecting guideline topics, grading research evidence, and formulating practice recommendations.