Past research in continuity of medical care has generally examined its determinants or its consequences. A model is proposed which combines these lines of research and incorporates a continuity measure as a process and specifying variable. The validity of this model is tested with a set of data consisting of 1,410 persons with major illness episodes. Results show that the model is supported, at least in part. Variables that significantly influence the number of ambulatory visits to a physician include the patient's payment method, the severity of his illness and his referral status. Moreover, the practice type of the usual physician emerges as a significant determinant of physician utilization when the level of continuity is controlled. The use of this more comprehensive model in future studies on continuity is indicated.