Teaching Conflicts of Interest and Shared Decision-Making to Improve Risk Communication: a Randomized Controlled Trial.

BACKGROUND: Risk communication is a core aspect of a physician's work and a fundamental prerequisite for successful shared decision-making. However, many physicians are not able to adequately communicate risks to patients due to a lack of understanding of statistics as well as inadequate management of conflicts of interest (COI). OBJECTIVE: To evaluate the effects of an integrated curriculum encompassing COI and shared decision-making on the participants' risk communication competence, that is, their competence to advise patients on the benefits and harms of diagnostic or therapeutic interventions. DESIGN: A rater-blind randomized controlled trial with a 30 (± 1)-week follow-up conducted from October 2016 to June 2017 at two German academic medical centers. PARTICIPANTS: Sixty-three medical students in their fourth or fifth year. INTERVENTIONS: Participants received either a newly developed 15-h curriculum or a course manual adapted from teaching as usual. MAIN MEASURES: Primary outcome: change in risk communication performance in a video-observed structured clinical examination (VOSCE). KEY RESULTS: Participants were 25.7 years old on average (SD 3.6); 73% (46/63) were female. Increase in risk communication performance was significantly higher in the intervention group with post-intervention Cohen's d of 2.35 (95% confidence interval (CI) 1.62 to 3.01, p < 0.01) and of 1.83 (CI 1.13 to 2.47, p < 0.01) 30 (± 1) weeks later. Secondary outcomes with the exception of frequency of interactions with the pharmaceutical industry also showed relevant improvements in the intervention as compared with the control group (d between 0.91 and 2.04 (p < 0.001)). CONCLUSIONS: Our results show that an integrated curriculum encompassing COI and risk communication leads to a large and sustainable increase in risk communication performance. We interpret the large effect sizes to be a result of the integration of topics that are usually taught separately, leading to a more effective organization of knowledge. TRIAL REGISTRATION: The trial is registered in the International Clinical Trials Registry with the trial number DRKS00010890.

Conflicts of Interest in Medicine. A Systematic Review of Published and Scientifically evaluated Curricula.

Objective: Conflicts of interests resulting from interactions with pharmaceutical companies are pervasive in medicine and can result in an undue influence on physicians' decision-making. The objective of this systematic review is to analyze published and scientifically evaluated curricula for medical students and residents regarding such conflicts of interest. We begin by describing the covered topics and teaching methods; afterwards we analyze the quality of the curricula using the published data on their evaluations and comparing the content with content recommended for such curricula. Methods: We searched Pubmed, PsycInfo, EMBASE, OECD, WISO, SOWI and googlescholar up to and including the 5th of September 2016. Publications describing curricula for residents or medical students on the topic of conflicts of interest in medicine and evaluating them for their effects on the participants' learning were included. We analyzed the covered topics and the teaching methods used and compared them with recommendations by the American Medical Students' Association (AMSA) and Health Action International (HAI). Results: The literature search resulted in 20 publications that fulfilled our search criteria. In five trials, a control group was used, in no trial the participants were randomized to intervention or control group. 16/20 published curricula primarily covered marketing strategies by pharmaceutical companies, especially the interaction with pharmaceutical sales representatives (PSRs). Most curricula only covered a limited number of topics recommended by AMSA/HAI. The most frequent teaching method was a group discussion, which was used in 18/20 curricula; all curricula used at least one interactive teaching method. The evaluation of the curricula was heterogeneous in results as well as design. Some publications described a change of attitudes toward a stronger skepticism regarding interactions with pharmaceutical companies. Four publications described improved knowledge, one publication described a change in behavior toward a reduction of the acceptance of gifts. Conclusion: The trials conducted to this date regarding curricula on conflicts of interests are methodologically flawed and the described curricula lack important topics beyond marketing strategies of pharmaceutical companies. In addition, there are no data so far on the sustainability of the courses' effects on participants' behavior. It is therefore necessary to develop a model curriculum that covers a broader variety of topics and to evaluate it using a well thought-out methodology to create a foundation for the further improvement of teaching conflicts of interest in medicine.