Modifying the Roter Interaction Analysis System to study provider-patient communication in telemedicine: promises, pitfalls, insights, and recommendations.

This paper suggests modifications to the Roter Interaction Analysis System (RIAS)--the most widely used measure for assessing provider-patient communication during conventional face-to-face consultations--for use in telemedicine. The RIAS, which describes and categorizes communication behaviors, is used to quantify communication events, which may then be correlated with patient, provider, and system attributes and health outcomes. Most of the changes suggested here add new coding subcategories to characterize technology-related utterances and to provide opportunities for global assessments of the overall technology environment within which provider-patient interactions took place. There are also general issues raised that interaction analysis researchers should consider when studying provider-patient communication in a telemedicine context. These relate to nonverbal behavior, multiple participants, missing information, and validity and reliability. In addition to comparing telemedicine to in person consultations, a modified RIAS could be used to compare televideo consultations to each other, across different specialties and technical specifications. A modified RIAS would accommodate not only differences in the current technology environment, but also changes in the way providers and patients communicate over time. The more we know about what interaction patterns lead to best outcomes, the more emphasis can be placed on developing training programs and other interventions to enhance patient-provider interactions in telemedicine.