Adapting the Roter interaction analysis system for telemedicine: lessons from four specialty clinics

Four specialists (a child psychiatrist, an oncologist, a cardiologist and a rheumatologist) conducted telemedicine clinics using videoconferencing at a bandwidth of 128-384 kbit/s. The consultations were videorecorded. The coded interactions from the first two patients recruited from each of the four telemedicine clinics were analysed. Tapes were coded by two raters. We adapted the Roter interaction analysis system (RIAS) for the telemedicine context. Utterances were coded into socio-emotional and task-focused categories. There were 1055 utterances in total. Providers generated significantly more utterances across categories than patients. In the patient-provider interactions, only 2% of the total utterances related to the technology. The predominance of socio-emotional utterances compared with task-focused utterances for providers was contrary to our expectations. Further studies are required to establish the reliability of the adapted RIAS measure and to increase understanding of telemedicine communication patterns.