Ambulatory recorded blood pressure (BP) has been shown to correlate better with secondary target-organ damage than casual blood pressure, and has demonstrated that ambulatory BP cannot be predicted on the basis of casual BP measurements by the physician. To assess whether these recordings have practical repercussions on diagnosis and treatment, 182 patients (74 female/108 male) aged 16 to 84 years (mean 51.9 +/- 14.8 SD), all referred by their physicians to the investigation unit of the hospital, underwent ambulatory BP measurements using the portable Remler M2000 BP recorder. The mean systolic blood pressure was above 140 mm Hg in 96 patients (53%) and the mean diastolic blood pressure above 90 mm Hg in 81 (45%). -3 to 9 months later, multiple choice questionnaires were sent to the 34 physicians who had referred patients and had meanwhile received recording results without comments. The questions were designed ascertain the indication for BP recording in each patient and the changes undertaken in the light of the results. In 27% of the cases the assumed diagnosis was not confirmed and in 40.8% of the cases the planned or prescribed treatment was obviated or withdrawn. This percentage corresponds to the data previously published describing 30 to 40% of patients as hypertensive only in the physician's office. This latter percentage shows that ambulatory BP recordings contribute to the practical assessment of patients with mild to moderate elevation of office BP.