Funding sources and financial disclosures, and their relationship to study outcomes and level of evidence in the Journal of Shoulder and Elbow Surgery

Abstract

Hypothesis/background: Concern exists regarding the reliability of published manuscripts due to influence of industry funding and author financial conflicts of interest (COI). We aim to determine whether COI affect the outcome of a research study or the level of evidence (LOE).

Methods: We reviewed 244 consecutive original articles in Journal of Shoulder and Elbow Surgery from January 2014 to December 2014. Articles included only those available in the printed journal. For LOE, 178 articles from the Shoulder and Elbow section were used (basic science articles were excluded). COI was determined by comparing financial disclosures and stated funding sources to the study content.

Results: COI were present in 44 of 244 articles (18%); of these, 24 (55%) had positive outcomes. Of the 200 without COI, 128 (64%) had positive outcomes. This difference in proportions was determined to be significant (P = .007). COI were present in 27 shoulder and elbow articles; of these, only 1 was LOE I or II (4%). Of the 151 without COI, 34 (23%) were LOE I or II. This difference in proportions was determined to be significant (P = .023).

Conclusion: We found that Journal of Shoulder and Elbow Surgery articles with COI are neither more likely to have positive outcomes nor higher LOE than those with no COI. Although the χ² analysis found a statistically significant relationship between COI and study outcomes, the study outcomes were more often positive in articles without COI. This is contrary to previously published analyses that found outcomes to be more positive in articles with COI.

Keywords: Conflict of interest; financial disclosure; funding; level of evidence; medical ethics; study outcome.