Background
: Hypertension is a common, chronic disease that poses as a main risk factor of coronary artery disease. Therefore, it requires accurate diagnosis. This study attempted to examine the problem of misclassification and accurate diagnosis of hypertension in primary care settings and to consider the relationship between variability of blood pressure and number of measurements.

Methods : Among the 158 patients with initially high blood pressure who visited health screening center of one university hospital from May to November, 199,97 persons who corresponded to the following conditions were chosen for this study. We classified hypertension as stage 1 and 2. The subjects were not previously diagnosed as hypertensive and had no past history of use of antihypertensive medication. Blood pressure was measured according to 1999 WHO/ISH Hypertension Guideline, and two or more measures were performed at each visit on five separate occasions at one week intervals.

Results : The mean of initial blood pressure was 159.6 mmHg in systole, 95.3 mmHg in diastole. The mean of subsequent blood pressure was 155.6 mmHg, 146.1 mmHg, 143.4 mmHg, 138.7 mmHg in systole and 92.5 mmHg, 88.4mmHg, 87 mmHg, 85.1 mmHg in diastole, which showed the tendency to be lower. In both systole and diastole, the mean differences between first and second measurements, second and third measurements were significant, but insignificant between third and forth measurements, and forth and fifth measurements.
We divided systolic and diastolic pressures ito two subgroups according to stage 1, 2 classification of hypertension. In stage 1 subgroup, the means of blood pressures were lower from 151.3 mmHg to 135.4 mmHg in systole, from 95.1 mmHg to 85.3 mmHg in diastole, but there were not significant. In stage 2 subgroup, the mean blood pressure was lower from 169.7 mmHg to 142.5 mmHg in systole, form 105.4 mmHg to 87.8 mmHg in diastole, and the mean differences between first and second measurements, second and third measurements were significant, but not significant between third and forth measurements, and forth and fifth measurements.

Conclusion : Blood pressure tends to be checked significantly lower until subsequent third measurements, but not thereafter. We think that more studies to find out how many blood pressure measurements are needed for diagnosing hypertension in consideration of patient's blood pressure level and risk factors.