BACKGROUND In a non-selected group of hypertensive patients with a new diagnosis at primary care, blood pressures obtained in the office (oBP) are compared to ambulatory ones (aBP). White coat hypertension (WCH) and white coat phenomenon (WCP) are estimated to evaluate a white coat effect on such population. PATIENTS AND METHODS An ambulatory blood pressure monitoring (ABPM) was performed to 91 hypertensive patients (55 females) from 21 to 70 years-old, with consecutive diagnosis in 4 office rooms in an health center (oBP mean > or = 140 mmHg for systolic blood pressure and/or 90 mmHg for diastolic blood pressure during three measures at least). WCH was defined by systolic aBP < or = 135 mmHg and diastolic < or = 85 mmHg, simultaneously, and WCP by oBP-aBP differences > or = 20 and/or 10 mmHg (for respective systolic and diastolic blood pressure). RESULTS oBP systematically exceeded aBP with wide variations for each subject (difference mean +/- SD: 15 +/- 13/7 +/- 9 mmHg for systolic/diastolic), sex (female: 19 +/- 12/11 +/- 9, male 8 +/- 11/2 +/- 9) and depending on the observer who made the measure. WCH was detected in 27 patients (22 females), and WCP in 47 (37 females), so 30% (confidence interval [CI]: 21-40%) and 52% (CI: 41-62%) respective prevalences are supposed. CONCLUSIONS Within primary care, white coat effect causes a substantial hypertension sobrediagnosis as quantitative as qualitatively. Confirming measures made by nurses and additionally, ABPM for women, seem to lessen this effect.