Associations between various attended automated office blood pressure estimations and all-cause and cardiovascular mortality: Minhang study.

Abstract

The method of evaluating office blood pressure (OBP) varies greatly among different guidelines. We performed a cohort study to compare the association of various directly transferred attended automated OBP (AOBP) estimations with all-cause and cardiovascular mortalities. Overall, 475 181 sets of OBPs from 35 622 participants aged 35 years or older were extracted from the electronic health record of the Xinzhuang town hospital in the Minhang District, Shanghai, China. Each set of OBPs contained three consecutive AOBPs that were transferred directly to the electronic health record. The mean of three OBPs, mean of the last two OBPs, and alternative average OBP were calculated. The difference between the first and average OBPs changed along with the calendar month, and it was highest in December (5.3/2.1 mmHg) and lowest in July (3.8/2.0 mmHg). The subjects older than 80 years of age displayed the largest discrepancy in the blood pressure control rate according to the first OBP or average OBP (12.1%). During the 3.9-year follow-up, 1055 deaths occurred. The alternative average SBP was associated with both all-cause [hazard ratio: 1.07, 95% confidence interval (CI): 1.04-1.11] and cardiovascular (hazard ratio: 1.17, 95% CI: 1.11-1.23) mortalities. The uncontrolled alternative average OBP remained significantly associated with an increasing risk of all-cause (hazard ratio: 1.24, 95% CI: 1.09-1.40) and cardiovascular (hazard ratio: 1.53, 95% CI: 1.25-1.86) mortality, but not the average of the last two or mean of three readings. We observed an obvious discrepancy in the OBP level and OBP control rate according to different AOBP estimations. The alternative average OBP seemed to be more powerful in predicting both all-cause and cardiovascular mortalities than the average of the last two or mean of three readings.