Verification of the accuracy and completeness of disclosures of Conflict of Interest in the medical literature: a systematic survey

ID: 

18832

Session: 

Long oral session 21: Issues in systematic review methods

Date: 

Saturday 16 September 2017 - 11:00 to 12:30

Location: 

Meeting Room 1.44

All authors in correct order:

Hakoum MB1, El-Rayess H2, Abou Ghaddara H2, Bejjani M3, Haddad S2, Schunemann HJ4, Guyatt G4, Akl EA2
1 Clinical Research Institute, American University of Beirut Medical Center, Lebanon
2 Department of Internal Medicine, American University of Beirut Medical Center, Lebanon
3 Faculty of Medicine, American University of Beirut, Lebanon
4 Department of Health Research Methods, Evidence, and Impact, McMaster University, Canada
Presenting author and contact person

Presenting author:

Maram Hakoum

Contact person:

Maram Hakoum
Elie Akl
Abstract text
Background: There is evidence that conflict of interest (COI) disclosures are not always accurate or complete. Reasons for incomplete or inaccurate disclosure include carelessness, missed detection, inappropriate assessment of potential risks, and intentional under-reporting.

Objectives: The first objective of this systematic survey is to assess the prevalence of inaccurate or incomplete COI disclosures in the medical literature as reported in studies done to date on this issue. The second objective is to summarise the methods that studies reported in their efforts to assess the accuracy and completeness of COI disclosures.

Methods: We will conduct a systematic survey to identify studies that either employed or discussed methods or conceptual approaches to verifying COI. We ran a systematic search of the following electronic databases: Medline and EMBASE. In addition, we searched for studies mentioning the following COI-related databases: Open Payments Data, Dollars for Docs, Association of the British Pharmaceutical Industry (UK database) and Danish Health and Medicines Authority. Reviewers will complete calibration exercises and will work in duplicate and independently on study selection and data extraction. For methodological studies, we will describe their general characteristics and findings in terms of completeness and accuracy. We will describe the methods described in discussion papers in narrative and/or tabular formats, as appropriate.

Results: Our search identified 8076 citations. The study-selection and data-extraction phases are ongoing and we will present results at the Summit.

Conclusions: Currently, there are no guidelines on when and how to verify COI disclosures. The ultimate aim is to develop a standardised methodology for the assessment of the accuracy and completeness of COI disclosures in the medical literature.