We reviewed the course of 1,076 patients with essential hypertension whose condition had been initially evaluated with both ambulatory BP (ABP) and office BP (OBP) measurements. During the period of follow-up (mean, five years), fatal cardiovascular events occurred in 75 patients, and nonfatal events occurred in 153. Each patient was classified according to the difference between the mean observed ABP at entry and that predicted from the mean OBP at entry by means of an equation for the linear regression of ABP on OBP. Life-table analyses demonstrated a significantly greater estimated cumulative ten-year incidence of both fatal and nonfatal events among patients with higher than predicted ABPs than among those with lower than predicted ABPs. Because OBPs were comparable in the two groups, we conclude that ABP was an important determinant of clinical outcome.