AIM: To assess compliance with a drug regimen of two doses a day compared with one a day. PATIENTS AND METHODS: A prospective crossover study was set up in a general practice environment to compare compliance on a drug regimen of once a day versus twice a day. Data were collected by electronic monitoring in 113 patients with hypertension or angina pectoris. All patients were prescribed slow-release nifedipine twice a day during the first month and then crossed to a single daily dose of amlodipine for another month. RESULTS: Compliance, defined as the proportion of days on which the correct dose was taken, improved in 30% of patients (95% confidence interval 19-41%; P < 0.001) when the patients were switched from twice a day to once a day, but at the same time there was a 15% increase (95% confidence interval 5-25%; P < 0.02) in the number of patients with one or more no-dose days. Approximately 8% of patients displayed low compliance, irrespective of the dose regimen. Actual dose intervals were used to estimate the extent and timing of periods with unsatisfactory drug activity for various hypothetical drug durations of action. CONCLUSIONS: The apparent advantage of a single daily dose in terms of compliance appears to be clinically meaningful only when the duration of activity extends beyond the dose interval in all patients.