Improving Transparency in Clinical Research
Adopting a standard nomenclature for conflicts of interest would increase ease of sharing and may improve understanding of conflicts' influence on research.
Medical journals should require article authors to include conflict-of-interest statements as a way to enhance transparency, according to the authors of a research letter published in JAMA.
Quinn Grundy, PhD, RN, from the Charles Perkins Centre at the University of Sydney in New South Wales, Australia, and colleagues conducted a cross-sectional study of all articles published in journals in accordance with International Committee of Medical Journal Editors policies between January 1, 2016, and December 31, 2016.
Articles were randomly sampled to reach a prespecified sample of 1000 articles, which included primary research articles, commentaries, editorials, narrative reviews, systematic reviews, and meta-analyses.
Disclosures were classified as positive when at least 1 author reported a conflict of interest, negative if all authors reported they had no conflicts, and missing if there was no disclosure statement. Altmetric score was used to estimate media attention. The primary outcome was prevalence of positive disclosures, with confidence intervals (CIs) calculated using the Clopper-Pearson exact method.
A total of 1002 articles from 269 journals were included in the study. Overall, 22.9% (95% CI, 20.3%-25.6%) of articles included a positive conflict-of-interest statement, 63.6% (95% CI, 60.5%-66.6%) had a negative disclosure, and 13.6% (95% CI, 11.5%-15.9%) had no statement.
The prevalence of positive disclosure statements varied according to article type and focus. Although 31.4% of commentaries, editorials, and narrative reviews had a positive conflict-of-interest statement, only 19.8% of primary research articles had one (95% CI, 26.1%-37.1% and 16.9%-23.0%, respectively; P <.001).
Furthermore, among primary research articles, 26.6% of those focused on drugs, devices, or surgical procedures had a positive conflict-of-interest statement compared with 15.4% of those that did not (95% CI, 21.4%-32.3% and 12.1%-19.3%, respectively; P <.001).
In addition, articles with positive conflict-of-interest statements were published in journals with a higher median impact factor, at 6.0 compared with 2.7 for articles without (P <.001), and received higher median Altmetric scores (3.7 vs 0.5; P <.001). The greater media attention is likely to amplify the effect of articles with conflicts of interest.
The authors suggested that to improve transparency, journals should require that manuscripts include both disclosures and funding statements.
Grundy Q, Dunn AG, Bourgeois FT, Coiera E, Bero L. Prevalence of disclosed conflicts of interest in biomedical research and associations with journal impact factors and altimetric scores. JAMA. 2018;319:408-409.