The present study of primary prevention in white men aged 40 to 64 years attempts to investigate whether a beta-blocker given as initial antihypertensive treatment would lower total mortality to a greater extent than thiazide diuretics. Patients were randomized to metoprolol (n = 1609, 8110 patient-years) or a thiazide diuretic (n = 1625, 8070 patient-years). The median follow-up time was 4.2 years. The mean dose of metoprolol was 174 mg/d, and of thiazide diuretics, 46 mg/d of hydrochlorothiazide or 4.4 mg/d of bendroflumethiazide. Identical control of blood pressure was achieved using a fixed therapeutic schedule. Total mortality was significantly lower for metoprolol than for thiazide diuretics because of fewer deaths from coronary heart disease and stroke. Total mortality was also significantly lower in smokers randomized to metoprolol. The benefit demonstrated in patients treated with metoprolol seems to have important implications for clinical practice.