The Effect of Financial Conflict of Interest, Disclosure Status, and Relevance on Medical Research from the United States.

Abstract

Financial interactions between industry and healthcare providers are reportable. Substantial discrepancies have been detected between industry and self-report of these conflicts of interest (COIs). Our aim was to determine if authors who fail to disclose reportable COI are more likely to publish findings that are favorable to industry than authors with no COI. In this blinded, observational study of medical and surgical primary research articles in PubMed, 590 articles were reviewed. Reportable financial relationships between authors and industry were evaluated. COIs were considered to have relevance if they were associated with the product(s) mentioned by an article. Primary outcome was favorability, defined as an impression favorable to the product(s) discussed by an article and determined by 3 independent, blinded clinicians for each article. Primary analysis compared Incomplete Self-Disclosure to No COI. Two-level multivariable mixed-effects ordered logistic regression was used to assess factors associated with favorability. A 69% discordance rate existed between industry and self-report in COI disclosure. When authors failed to disclose COI, their conclusions were more likely to favor industry partners than authors without COI (favorable ratings 73% versus 62%, RR 1.18, p = < 0.001). On univariate (any COI 74% versus no COI 62%, RR 1.11, p = < 0.001) and multivariable analyses, any COI was associated with favorability. All financial COIs (disclosed or undisclosed, relevant or not relevant, research or non-research) influence whether studies report findings favorable to industry sponsors.