Emergency physicians and nurses were trained in telemedicine techniques in two emergency departments, one rural (low volume) and one suburban (high volume). Fifteen patient complaints were selected as appropriate for the study. Of 122 patients who met the inclusion criteria, 104 (85%) consented to participate. They were randomized to control and experimental groups. The suburban emergency physician diagnosed and treated the control patients. Experimental patients presenting to the high-volume emergency department were evaluated and treated by the telemedicine nurse in person and the rural emergency physician via the telemedicine link. Immediately before discharge all telemedicine patients were re-evaluated by the suburban emergency physician. Data collected on each patient included: diagnosis; treatment; 72 h return visits; need for additional care; and satisfaction of patient, physicians and nurses. There were no significant differences (P > 0.05) for occurrence of 72 h return visits, need for additional care or overall patient satisfaction. The average patient throughput time (from admission to discharge) was 106 min for the telemedicine group and 117 min for the control group. Telemedicine was a satisfactory technique for the chosen group of patients in the emergency department and was acceptable to the participants.