In addition to efficacy and safety, the cost of therapy has become an increasingly important factor to consider when selecting drugs to treat patients with mild-to-moderate hypertension. However, acquisition prices alone do not determine the total cost of therapy. To better assess total costs, we conducted a systematic, retrospective, cost-minimization analysis of drugs used to treat 673 patients with newly diagnosed, mild-to-moderate (> 95 to < 110 mmHg) diastolic hypertension between the years 1985 and 1992. Patients included in the study had started antihypertensive monotherapy, and a minimum of one dose titration was required before adding another antihypertensive agent to the regimen. A patient had to have a diastolic blood pressure of < or = 90 mmHg while undergoing therapy to be included in the analysis. Drug classes included diuretics, beta-adrenergic blockers, centrally acting alpha 2-agonists, alpha 1-adrenergic blockers, calcium channel blockers, and angiotensin-converting enzyme (ACE) inhibitors. Costs, adjusted to 1992 price levels, were analyzed for 32 individual agents for each of the following five cost variables: initial drug acquisition, supplemental drug acquisition, laboratory monitoring, clinic visits, and treatment of side effects. Mean total costs per patient for all five variables by drug class were $895 for beta-blockers, $1043 for diuretics, $1165 for centrally acting alpha 2-agonists, $1243 for ACE inhibitors, $1288 for alpha 1-blockers, and $1425 for calcium channel blockers. However, costs within each class varied considerably. Acquisition cost was often a poor predictor of the total cost of treatment. Therefore, acquisition cost must be considered in conjunction with a number of outcome variables to assess the true cost of antihypertensive therapy.