Conflict of Interest and Risk of Bias in Systematic Reviews on Methylphenidate for Attention Deficit Hyperactivity Disorder

Abstract

Abstract Background Systematic reviews (SR) are crucial to evidence-based medicine, but associated conflict of interest remains under-investigated. We set out to compare disclosure of financial conflicts of interest (fCOI) of authors with publically available information and risk of bias (RoB) in SR on the effect of Methylphenidate for treatment of Attention Deficit Hyperactivity Disorder. Methods A librarian searched Medline, Cochrane Library, Embase and PsycInfo in December, 2020. PRISMA guidelines were adopted for selection. Two blinded researchers independently searched open websites for information on fCOI for all authors of all included SR. ´ Disclosure on fCOI was extracted from full-text versions of SR. RoB was judged using Cochrane´s tool for SR. Results Disclosure on fCOI was missing for 2/44 (5%) included SR. For 15 (34%) SR, at least one author disclosed fCOI, and confirmed by open websites in all cases. Authors of 27 (61%) SR disclosed no fCOI, but discordant information was publically available for 8 (30%). In all, 23/44 (55%) of SR were associated with fCOI. RoB was high for 37/44 (84%) SR; 18 (49%) with and 19 (51%) without fCOI noticed on open websites. For the remaining 7 SR with low risk of bias, the corresponding figures were 6 (86%) with and 1 (14%) without fCOI. Conclusion The findings indicated an underreporting of fCOI corresponding to one third of SR disclosing no fCOI, methodological shortcomings were common, and 43/44 (98%) SR were compromised by either high RoB, or by fCOI. Similar investigations on other topics for comparison are warranted.