This career development award will enable me to develop the necessary skills to conduct and lead biomedical informatics interventions that address crucial health communications issues throughout healthcare. Poor healthcare communication is a serious public health issue that is linked to medical errors, avoidable patient harm and death, suboptimal health outcomes, and low patient satisfaction(1-11). The scientific aims of this health informatics research are to: 1) investigate the mediating influence of computer based nonverbal communication training using virtual human software?MPathic-VR; 2) refine a conceptual model of nonverbal communication; and 3) develop and test an automated assessment of nonverbal communication in the virtual human software. Virtual humans?intelligent, computer-based conversational agents?enable healthcare professionals to develop communication skills through an engaging and active learning experience: practicing communication behavior; receiving assessment feedback on their cognitive, verbal, and nonverbal behaviors; and then practicing again incorporating feedback. A recently concluded trial with the virtual human software found evidence of improved verbal and nonverbal communication skills relative to a computer-based control module condition. However, previous analyses of virtual human software data did not specifically investigate how the software led to improvement in nonverbal communication skills.
Aim 1 of this study is to understand how nonverbal behavior improved by conducting a secondary analysis of virtual human trial data using mixed methods. Because nonverbal communication is an important, but often unaddressed, aspect of communication assessment, Aim 2 will develop a new conceptual model of nonverbal communication through interviews with 20 providers. Based on those results, Aim 3 will add an automated nonverbal communication behavior assessment and feedback mechanism to MPathic-VR and conduct a validity check with 30 medical students. My overall career goal is to become an independent investigator focused on applying computer and information sciences to improve dissemination of biomedical information in health settings, which I will accomplish by achieving four training goals: 1) analyze verbal and nonverbal communication, 2) analyze dyadic data, 3) apply virtual human technologies to communication interventions, and 4) obtain advanced skills in grant writing. The mentoring team is experienced with informatics-based interventions, health communication investigations, and mixed methods research. In addition, the University of Michigan and the Department of Family Medicine offer a world-class research infrastructure with strong support for early career investigators through defined programs and courses. Completing this research and the proposed training goals is critical to developing a rigorous R01 intervention study. This work is also necessary for developing a novel scenario within the MPathic-VR system that can be tested in a future intervention, and it will provide the knowledge needed to advance interactive training in verbal and nonverbal communication.
Patient-provider communication is a major public health issue linked to increase risk for medical errors, poor outcomes, and lower patient satisfaction. This research seeks to develop an automated assessment of nonverbal communication in virtual human software so that nonverbal communication may be incorporated into simulations with the goal of enhancing provider communication skills.
