Impact of Scientific Versus Emotional Wording of Patient Questions on Doctor-Patient Communication in an Internet Forum: A Randomized Controlled Experiment with Medical Students

Background: Medical expert forums on the Internet play an increasing role in patient counseling. Therefore, it is important to understand how doctor-patient communication is influenced in such forums both by features of the patients or advice seekers, as expressed in their forum queries, and by characteristics of the medical experts involved. Objective: In this experimental study, we aimed to examine in what way (1) the particular wording of patient queries and (2) medical experts' therapeutic health concepts (for example, beliefs around adhering to a distinctly scientific understanding of diagnosis and treatment and a clear focus on evidence-based medicine) impact communication behavior of the medical experts in an Internet forum. Methods: Advanced medical students (in their ninth semester of medical training) were recruited as participants. Participation in the online forum was part of a communication training embedded in a gynecology course. We first measured their biomedical therapeutic health concept (hereinafter called "biomedical concept"). Then they participated in an online forum where they answered fictitious patient queries about mammography screening that either included scientific or emotional wording in a between-group design. We analyzed participants' replies with regard to the following dimensions: their use of scientific or emotional wording, the amount of communicated information, and their attempt to build a positive doctor-patient relationship. Results: This study was carried out with 117 medical students (73 women, 41 men, 3 did not indicate their sex). We found evidence that both the wording of patient queries and the participants' biomedical concept influenced participants' response behavior. They answered emotional patient queries in a more emotional way (mean 0.92, SD 1.02) than scientific patient queries (mean 0.26, SD 0.55; t(74) = 3.48, P<.001, d=0.81). We also found a significant interaction effect between participants' use of scientific or emotional wording and type of patient query (F-2,F-74=10.29, P<.01, partial eta(2)=0.12) indicating that participants used scientific wording independently of the type of patient query, whereas they used emotional wording particularly when replying to emotional patient queries. In addition, the more pronounced the medical experts' biomedical concept was, the more scientifically (adjusted beta=.20; F-1,F-75=2.95, P=.045) and the less emotionally (adjusted beta=-.22; F-1,F-74=3.66, P=.03) they replied to patient queries. Finally, we found that participants' biomedical concept predicted their engagement in relationship building (adjusted beta=-.26): The more pronounced their biomedical concept was, the less they attempted to build a positive doctor-patient relationship (F-1,F-74=5.39, P=.02). Conclusions: Communication training for medical experts could aim to address this issue of recognizing patients' communication styles and needs in certain situations in order to teach medical experts how to take those aspects adequately into account. In addition, communication training should also make medical experts aware of their individual therapeutic health concepts and the consequential implications in communication situations.