The long-term efficacy and safety of amlodipine (2.5 to 10 mg) once daily was compared with that of hydrochlorothiazide (HCTZ) (25 to 100 mg) daily in 139 patients with mild-to-moderate hypertension. The study was a randomized, open-label, parallel comparison of 50 weeks' duration. Patients were randomized in a 2:1 ratio (amlodipine n = 92: HCTZ n = 47). Atenolol was added at week 12 if monotherapy was inadequate. At week 12, the mean reductions for supine and standing systolic and diastolic blood pressure values with amlodipine were found to be -15.2/-12.3 mmHg and -14.0/-11.6 mmHg respectively, as compared to -15.5/-11.1 mmHg and -16.1/-10.1 mmHg after treatment with HCTZ. The percentage of patients responding to treatment at week 12 was 74% on amlodipine and 70% on HCTZ. The addition of atenolol in those patients not adequately controlled on monotherapy produced additional mean reductions in supine and standing systolic and diastolic pressures in both the amlodipine-atenolol group and in the HCTZ-atenolol group. The incidence of adverse effects was 47% with amlodipine and 26% with HCTZ at 12 weeks. Overall, six patients were discontinued because of side effects while receiving amlodipine monotherapy and one from the HCTZ monotherapy group; none were discontinued because of side effects on combination therapy. Laboratory test abnormalities were reported by 16% of amlodipine-treated patients compared with 63% of patients on HCTZ. The antihypertensive effects of amlodipine and hydrochlorothiazide appeared to be comparable and were maintained during long-term therapy.