Conflicts of Interest Among Authors of Systematic Reviews and Meta-analyses Investigating Interventions for Melanoma: Cross-sectional Literature Study (Preprint)

Abstract

BACKGROUND Background: Previous studies have highlighted the potential influence industry relationships may have on the outcomes of medical research. OBJECTIVE Objectives: We aimed to determine the prevalence of author COI in systematic reviews focusing on melanoma interventions, as well as determine whether the presence of these COI were associated with an increased likelihood of reporting favorable results and conclusions. METHODS Methods: This cross-sectional study included systematic reviews with or without meta-analyses focusing on interventions for melanoma. We searched MEDLINE and Embase for eligible systematic reviews published between September 1, 2016 and June 2, 2020. COI disclosures were cross-referenced with information from the CMS Open Payments Database, Dollars for Profs, Google Patents, the United States Patent and Trademark Office (USPTO), and previously published COI disclosure statements. Results were quantified using descriptive statistics and relationships were evaluated by Fisher's exact test. RESULTS Results: Of the 23 systematic reviews included in our sample, 12 (12/23; 52%) had at least one author with a COI. Of these reviews, seven (58%) reported narrative results favoring the treatment group and nine (75%) reported conclusions favoring the treatment group. Of the 11 systematic reviews without a conflicted author, four (36%) reported results favoring the treatment group and five (45%) reported conclusions favoring the treatment group. We found no significant association between the presence of author COI and the favorability of results (p= 0.53) or conclusions (p= 0.15). CONCLUSIONS Conclusions: Author COI did not appear to influence the outcomes of systematic reviews regarding melanoma interventions. Clinicians and other readers of dermatology literature should be cognizant of the influence that industry may have on the nature of reported outcomes, including those from systematic reviews and meta-analyses.