Blood pressure stability of normotensives and mild hypertensives in different settings.

Generalizability theory was used to examine the stability of blood pressure (BP) measurement in normotensives and mild hypertensives. Three to six readings at home or at work provided adequate reliability for the same day in each setting. Under structured laboratory conditions, two to three BP measures taken on each of 2 to 3 days for systolic and diastolic BP provided conservative estimates that were generalizable across days. Finally, generalizations across settings called for five or more measurements taken in at least two settings.