Unreliable Threats: Conflicts of Interest Disclosure and the Safeguarding of Biomedical Knowledge

Medical epistemology lately has seen a strengthening of the view that the construction of evidence should be sensitive to the social context in which it is produced. A poignant illustration of this is the undue influence of the pharmaceutical industry on research results and reporting. I challenge a particular application of this view by examining a common practice in the medical and scientific community: mandatory author disclosure of conflicts of interest (COIs) in published articles. In illustrating problems with COI disclosure policies in biomedical publishing, including unappreciated shortcomings of the scant empirical data supporting mandatory disclosure, I hope to demonstrate that the value given to journal COI disclosure policies as a way to protect the reliability of medical evidence might well be misplaced. Rather than extract away the "messy" details of the real world, the analysis is ultimately more responsive to how medical knowledge is produced and disseminated.