Comparison of the accuracy and errors of blood pressure measured by 2 types of non-mercury sphygmomanometers in an epidemiological survey

Abstract

A few studies have compared auscultation and oscillometric devices with the mercury sphygmomanometer (MS) reference values for blood pressure (BP) measurement in an epidemiologic survey. Four trained observers recorded BP measurements from 766 subjects from general Korean population in 2014 and 2015. Measurements were repeated 3 times for each device, alternately using an MS and 2 electronic devices (EDs; Greenlight 300 vs Omron HEM-907), together with a randomized device sequence. The BP measurement difference was defined as BP measured by MS minus BP obtained by ED, and the absolute error as the absolute value of the difference. Mean differences in systolic BP (SBP) were -0.52 and -0.62 mmHg and those of diastolic BP (DBP) were -0.78 and 6.23 mmHg (P < .01) in the Greenlight and Omron device group, respectively. The concordance correlation coefficients were 0.97 and 0.94 for SBP and 0.95 and 0.76 for DBP in the Greenlight and Omron group, respectively (P < .05). Kappa values for the Joint National Committee 7 BP classification were 0.84 and 0.74 for Greenlight and Omron group, respectively. The prevalence of normotension, prehypertension and hypertension were 53.5%, 33.9%, and 12.5% with the MS and 59.8%, 29.0%, and 11.2% with the ED in the Omron group (P = .03, McNemar test), whereas they were insignificant in the Greenlight group. The Greenlight 300 may be a good alternative to the MS, and the Omron HEM-907 has good accuracy in SBP measurement. Due to the measurement error in DBP, Omron HEM-907 was inferior to the Greenlight device.