RESUMO
OBJECTIVE: Our study examines the association between the favorability of percutaneous coronary intervention (PCI) and/or coronary artery bypass surgery (CABG) and the presence of conflicts of interest (COIs) among authors. METHODS: We used the "Citing Articles" tool on the New England Journal of Medicine website to identify editorials on the use of PCI/CABG for stable ischemic heart disease. Authors were rated as "supportive," "neutral," or "critical" of these interventions based on the content of their editorials. COIs for each author were identified using past publications found on Scopus, PubMed, or a general internet search. RESULTS: A total of 606 articles were identified, and data were extracted from 56 of them. Among the 149 authors, 64 (43.0%) had a COI. Of these 64 authors, 19 (29.7%) disclosed their COI, while 45 (70.3%) did not. Overall, among authors with a COI, there was no association between disclosed and undisclosed COIs and the authors' view of PCI/CABG [χ2 (2, N = 64) = 1.63, p = .44]. If an author was associated with Medtronic, Abbott, or Boston Scientific, they were more likely to favor PCI/CABG if they had an undisclosed COI relative to authors who disclosed COIs [χ2 (1, N = 31) = 5.04, p = .025]. Authors publishing in a cardiology journal were more likely to view PCI/CABG favorably relative to those publishing in a general medicine journal [χ2 (2, N = 62) = 7.17, p = .028]. CONCLUSION: Editors should adopt policies to counteract the unbalancing effects that COIs have on medical opinions and evidence.
