OBJECTIVE To identify specific components of veterinarian-client-patient communication during clinical appointments in companion animal practice. METHODS Cross-sectional descriptive study. METHODS A random sample of 50 companion animal practitioners in southern Ontario and a convenience sample of 300 clients and their pets. METHODS For each practitioner, 6 clinical appointments (3 wellness appointments and 3 appointments related to a health problem) were videotaped, and the Roter interaction analysis system (RIAS) was used to analyze the resulting 300 videotapes. Statements made during each appointment were classified by means of a communication framework reflecting the 4 essential tasks of the appointment (ie, data gathering, education and counseling, relationship building, and activation and partnership). RESULTS 57% of the veterinarians contacted (50/87) and 99% of the clients contacted agreed to participate in the study. Mean duration of the appointments was 13 minutes. Typically, veterinarians contributed 62% of the total conversation and clients contributed 38%. Fifty-four percent of the veterinarian interaction was with the client, and 8% was with the pet. Data gathering constituted 9% of the veterinarian-to-client communication and was primarily accomplished through closed-ended questioning; 48% of veterinarian-to-client communication involved client education and counseling, 30% involved relationship building, and 7% involved activation and partnership (the remaining 6% constituted orientation). CONCLUSIONS Results suggest that the RIAS was a reliable method of assessing the structure, process, and content of veterinarian-client-patient communication and that some veterinarians do not use all the tools needed for effective communication.