Industry Bias in Randomized Controlled Trials in General and Abdominal Surgery: An Empirical Study

Abstract

Background: Industry sponsorship has been identified as a source of bias in several fields of medical science. To date, the influence of industry sponsorship in the field of general and abdominal surgery has not been evaluated.

Methods: A systematic literature search (1985-2014) was performed in the Cochrane Library, MEDLINE, and EMBASE to identify randomized controlled trials in general and abdominal surgery. Information on funding source, outcome, and methodological quality was extracted. Association of industry sponsorship and positive outcome was expressed as odds ratio (OR) with 95% confidence interval (CI). A χ test and a multivariate logistic regression analysis with study characteristics and known sources of bias were performed.

Results: A total of 7934 articles were screened and 165 randomized controlled trials were included. No difference in methodological quality was found. Industry-funded trials more often presented statistically significant results for the primary endpoint (OR, 2.44; CI, 1.04-5.71; P = 0.04). Eighty-eight of 115 (76.5%) industry-funded trials and 19 of 50 (38.0%) non-industry-funded trials reported a positive outcome (OR, 5.32; CI, 2.60-10.88; P < 0.001). Industry-funded trials more often reported a positive outcome without statistical justification (OR, 5.79; CI, 2.13-15.68; P < 0.001). In a multivariate analysis, funding source remained significantly associated with reporting of positive outcome (P < 0.001).

Conclusions: Industry funding of surgical trials leads to exaggerated positive reporting of outcomes. This study emphasizes the necessity for declaration of funding source. Industry involvement in surgical research has to ensure scientific integrity and independence and has to be based on full transparency.