Individual and institutional financial conflicts of interest reported by authors of randomized controlled trials: a systematic survey

Article type
Year
Authors
Hakoum MB1, Noureldine H2, Jouni N2, Abou-Jaoude EA3, Lopes LC4, Guyatt G5, Akl EA1
1Clinical Research Institute, American University of Beirut Medical Center, Lebanon
2American University of Beirut, Lebanon
3School of Medicine and Biomedical Sciences, University at Buffalo, USA
4Pharmaceutical Science Master Course, University of Sorocaba, Brazil
5 Department of Clinical Epidemiology and Biostatistics, McMaster University, Canada
Abstract
Background: Systematic reviewers are expected to collect the financial conflicts of interest (COI) disclosures of authors of included studies. These disclosures usually lack important details that would allow the judgment of their significance.

Objectives: The objective of the study is to survey reports of randomized controlled trials (RCT) for the characteristics of individual and institutional financial COI disclosures.

Methods: We are using standard systematic review methodology to survey reports of clinical RCT papers published in any of the 119 Core Clinical Journals in 2015. We categorized the types of disclosed financial COI as grant, employment, personal fees, non-monetary support, drug or equipment supplies, patent, stocks, and other types. We will collect data on general characteristics of the RCTs, the reported funding, and the characteristics of the COI disclosures including type, source, relation of the source to the trial subject and funder, the duration, and the monetary value. We will also collect data on the characteristics of authors that report the COIs, including authorship rank, title, affiliation, and gender. We will conduct descriptive and regression analyses.

Results: We included 108 RCT papers with authors reporting financial COIs. Of the 108 RCTs, 96% had the first author affiliated with an institution from a high income country and 57% were on a pharmacological intervention. All RCTs reported being funded, of which 58% were funded by a private for profit source. Eighty-five per cent of papers provided COI disclosures as narrative statements in the main document and 20% as ICMJE forms available online or upon request. We identified 818 authors disclosing a total of approximately 3000 financial COIs. The data collection phase is ongoing and we will present additional results at the Colloquium.

Conclusions: The findings of our study will support the development of future guidance for authors of systematic reviews to record the individual and institutional financial conflict of interest disclosures of authors of included studies.