Patient compliance with drug therapy is an important component in the successful management of chronic diseases. The objective of this study was to determine the mean compliance ratio for calcium channel blocking agents in a 2-year study of prescription claims records. The effect on compliance relating to the number of daily doses, length of drug therapy, therapeutic use, and patient age also was investigated. We analyzed prescription drug claims records from an insurance program with prescription drug benefits for state employees, teachers, and retirees in the state of Oklahoma. A total of 105,559 calcium channel blocker prescription claims representing 9807 treatment regimens were made between July 1990 and June 1992. A stratified random sample of 2377 treatment regimens was drawn from this population. The mean compliance ratio for all patients was 78.22%. The number of daily doses and the length of the treatment regimen were significant factors in the compliance model (P = 0.0001). The age of the patient was not significant. Differentiating therapeutic use by symptomatic (angina pectoris) or nonsymptomatic (hypertension) indications also was not significant. Once-daily treatment regimens were found to provide the highest mean compliance ratio (84.86%). This was followed by twice-daily administration, with a mean of 79.88%; three times daily, 75.16%; and four times daily, 73.06%. Multiple comparisons analysis indicated a statistically significant difference between each dosing regimen with the exception of three-versus four-times-daily administration. The highest mean compliance ratio (101.51%) was found with treatment lengths of 2 months. The mean compliance ratio declined as the length of therapy increased. The decline in patient compliance was most pronounced with treatment lengths between 2 and 4 months. The mean compliance ratio was lowest for treatment lengths of 16 months (71.00%), although there was no significant difference between treatment lengths of 6, 12, 16, or 20 months. The deterioration of compliance over time ceased with treatment lengths exceeding 16 months. Based on these data, prescribers should attempt to minimize or simplify the frequency of the dosing regimen when possible. Discussions with the patient involving medication compliance may be of greater value during follow-up visits with the physician or pharmacist when noncompliant behavior becomes most pronounced.