Disclosures of conflicts of interest in psychiatric review articles

To characterize disclosures of conflicts of interest in review articles in psychiatry, we identified 285 reviews from 10 high-impact journals in psychiatry and 2 in general medicine. Disclosures were reliably coded as biotechnology/pharmaceutical/other material interests, nonprofit/government, communication companies, or other. The authors in both types of journals frequently reported industry ties. However, the reviews in the psychiatric journals were significantly less likely to include industry-related disclosures (32% of the reviews; 18% of the authors) compared with the general medical journals (64% of the articles; 40% of the authors). The most common types of industry-related disclosures were for consulting, research support, and speaking fees. Disclosures seemed to be of limited utility in helping readers assess possible biases because the nature and the extent of the relationships being disclosed were often unclear. Efforts to screen out authors with significant financial relationships pertaining to the topic under review may be more effective than are disclosures in protecting the integrity of the medical literature.