The antihypertensive effect of debrisoquine (20 mg/day), methyldopa (100 mg/day) and propranolol (160 mg/day) was compared to that obtained with a placebo in a controlled trial carried out by a group of 14 internists. Forty-eight patients with uncomplicated essential hypertension were included. Mefruside (25 mg/day) was first given alone for 6 weeks ("open phase" of the trial) and to this diuretic was then added in double-blind fashion and randomized sequence a placebo or an active drug. Each of the 4 blind phases lasted 4 weeks. At the end of the "open phase", blood pressure in seated position averaged 168/111 +/- 19.6/13.5 mm Hg (mean +/- SD). A significant blood pressure decrease was observed after 4 weeks of treatment with the placebo as well as with the investigated compounds. With the placebo blood pressure was reduced to 158/102 +/- 19.6/13.5 mm Hg (p less than 0.001). The magnitude of the additional blood pressure decrease induced by the active drugs was relatively small and varied from 4 (debrisoquine) to 10 mm Hg (methyldopa, p less than 0.01) for the systolic and from 3 (debrisoquine, p less than 0.05) to 5 mm Hg (propranolol, p less than 0.05) for the diastolic. The percentage of patients with systolic pressure of less than or equal to 140 mm Hg and with diastolic pressure of less than 90 mm Hg during administration of either drug was not greater than 40 to 20% respectively. Propranolol appeared to be better tolerated than the other antihypertensive agents. These rather disappointing blood pressure results suggest that the efficacy of antihypertensive agents in private practice cannot be extrapolated from studies carried out in specialized hypertension clinics.