Disclosing conflicts of interest in patient decision aids

BACKGROUND In 2005, the International Patient Decisions Aid Standards (IPDAS) Collaboration developed quality criteria for patient decisions aids; one of the quality dimensions dealt with disclosure of conflicts of interest (COIs). The purposes of this paper are to review newer evidence on dealing with COI in the development of patient decision aids and to readdress the theoretical justification and definition for this quality dimension. METHODS The committee conducted a primary systematic literature review to seek published research addressing the question, "What is the evidence that disclosure of COIs in patient decision aids reduces biased decision making?" A secondary literature review included a systematic search for recent meta-analyses addressing COIs in other spheres of health care, including research and publication, medical education, and clinical care. RESULTS No direct evidence was found addressing this quality dimension in the primary literature review. The secondary review yielded a comprehensive Institute of Medicine report, as well as four relevant meta-analyses addressing disclosure of COIs in health care. They revealed a broad consensus that disclosure of COIs is desirable in such areas as research publication, guideline development, medical education, and clinical care. CONCLUSIONS The committee recommends the criteria that are currently used to operationally define the quality dimension "disclosing conflicts of interest" be changed as follows (changes in italics): Does the patient decision aid: • report prominently and in plain language the source of funding to develop or exclusively distribute the patient decision aid? • report prominently and in plain language whether funders, authors, or their affiliations, stand to gain or lose by choices patients make after using the patient decision aid? Furthermore, based on a consensus that simple disclosure is insufficient to protect users from potentially biased information, the committee recommends that the IPDAS Collaboration consider adding the following criterion when the IPDAS consensus process is next conducted: “Does the patient decision aid: • report that no funding to develop or exclusively distribute the patient decision aid has been received from commercial, for-profit entities that sell tests or treatments included as options in the patient decision aid?”