Article type
Year
Abstract
Title: Conflict of interest disclosure in systematic reviews of surgical interventions and devices: a cross-sectional survey
Background: A conflict of interest (COI) is acknowledged as an important source of bias in the design, conduct and reporting of studies. Surgery, with its wide use of medical devices and surgeons' preference, is a field with high potential for COI. In this study, we tried to investigate the frequency and types of conflicts of interest of systematic reviews including RCTs assessing surgical interventions.
Methods: A methodological survey was conducted by searching for systematic reviews (SRs) involving randomized controlled trials of surgical interventions published in 2017 via PubMed. Information regarding general characteristics, source of funding, COI disclosure and the type of COI were gathered. A multiple logistic regression analysis was conducted to explore factors associated with the reporting of COI disclosure in SRs.
Results: 163 systematic reviews were finally identified in 2017, almost half of SRs (55.8%) reported their funding resource and 90.1% reported authors' COI disclosure. Among 147 systematic reviews that provided COI disclosures, 33 (20.2%) declaimed at least one author reporting any type of COI. Cochrane SRs were more likely to reporting funding source (81.4% versus 51%, P=0.003) and at least one type of COI than non-Cochrane SRs than non-Cochrane SRs. SRs reported funding and significance of primary outcome were associated with better reporting of COI (adjusted OR=12.6, 95% CI 1.5 to 102.3; adjusted OR=3.79, 95%CI 1.29 to 11.09).
Conclusion: Although most of SRs stated their COI disclosure, the majority of them were no COI disclosure and less information regarding non-financial COI was found. Disclosure should be required before the publication of a systematic review and be better reported regarding the nature of interests. Additionally, more efforts from editors and peer-reviews in medical journals should be made to verify the accuracy and integrity of conflict of interests.
Ethics approval and consent to participate: not applicable
Background: A conflict of interest (COI) is acknowledged as an important source of bias in the design, conduct and reporting of studies. Surgery, with its wide use of medical devices and surgeons' preference, is a field with high potential for COI. In this study, we tried to investigate the frequency and types of conflicts of interest of systematic reviews including RCTs assessing surgical interventions.
Methods: A methodological survey was conducted by searching for systematic reviews (SRs) involving randomized controlled trials of surgical interventions published in 2017 via PubMed. Information regarding general characteristics, source of funding, COI disclosure and the type of COI were gathered. A multiple logistic regression analysis was conducted to explore factors associated with the reporting of COI disclosure in SRs.
Results: 163 systematic reviews were finally identified in 2017, almost half of SRs (55.8%) reported their funding resource and 90.1% reported authors' COI disclosure. Among 147 systematic reviews that provided COI disclosures, 33 (20.2%) declaimed at least one author reporting any type of COI. Cochrane SRs were more likely to reporting funding source (81.4% versus 51%, P=0.003) and at least one type of COI than non-Cochrane SRs than non-Cochrane SRs. SRs reported funding and significance of primary outcome were associated with better reporting of COI (adjusted OR=12.6, 95% CI 1.5 to 102.3; adjusted OR=3.79, 95%CI 1.29 to 11.09).
Conclusion: Although most of SRs stated their COI disclosure, the majority of them were no COI disclosure and less information regarding non-financial COI was found. Disclosure should be required before the publication of a systematic review and be better reported regarding the nature of interests. Additionally, more efforts from editors and peer-reviews in medical journals should be made to verify the accuracy and integrity of conflict of interests.
Ethics approval and consent to participate: not applicable