A survey was conducted in a cohort of 235 general practitioners (GP) selected by Sofres Medical who were representative of the French medical population, to measure the percentage of patients with hypertension, treated hypertensives and patients with controlled hypertension. Data were collected over 1 week of office consultation. Practitioners were initially instructed to use the same type of mercury sphygmomanometer, equipped with pneumatic cuffs of different sizes. Three consecutive blood pressure (BP) measurements were made and the last two were recorded. Practitioners had to carry out their own survey over a period of 1 week on all patients > 18 years of age who visited their offices. Patients were considered as hypertensive (HP) if the mean of the two recorded BP measurements was > or = 140/90 mm Hg or if they were taking antihypertensive drug treatment. Three cutoff points were used to define controlled hypertension: < 140/90 mm Hg (overall population of HP), < 160/95 mm Hg (HP < 65 years of age), and < 160/90 mm Hg (HP > or = 65 years of age). Among 12,351 patients (mean age, 48.6 years; women, 58%), 5020 were HP, (41%) of whom 2035 were without treatment (41%) and 2985 were receiving antihypertensive drug treatment (59%). Two hundred-thirty patients (4.6%) remained at high risk with moderate or severe hypertension (BP > or = 180 [systolic] or 105 [diastolic] mm Hg), ie, 1 patient/week/GP. The study confirms the high prevalence of hypertension in general practice and shows that 7 of 10 patients have an acceptable control of their BP (< 160/95 or < 160/90 mm Hg according to age) but only 24% of treated HP achieved the target of a BP level < 140/90 mm Hg, representing 28% of the 18 to 64 year old group and 21% of the elderly group. French GP did not choose an optimal control, and the medical community is waiting for answers to crucial questions, ie, does optimal BP control significantly improve the absolute cardiovascular risk? How far should blood pressure be lowered?