Automated BP readings with BP Tru reduce zero preference

Preference for numbers ending in zero may bias blood pressure measurements affecting clinical decision making. Use of an automated device should reduce this phenomenon. As part of a continuous quality improvement project the last blood pressure recorded was taken from 50 patients in four subspecialty nephrology clinics at an academic health sciences centre. In three of the clinics one nurse practitioner trained to estimate blood pressure to the nearest two mmHg, assesses patient's blood pressure with a mercury sphygmomanometer and records it onto the medical record. In the fourth clinic a BP Tru is used to measure blood pressure (blood pressure is taken automatically 6 times, the first result is discarded and the last five averaged with terminal digits displayed from 0 to 9). Each practice had from 400–1250 total charts to choose from. Charts were selected using a random number table to identify the last three numbers of the hospital file number. In the nurse measured clinics, 51% of the systolic and 65% of the diastolic readings ended in zero, whereas only 32% and 26% respectively of the reading in the automated clinics ended in 0 (c 2(1 d.f.) = 5.4, p=0.02 for SBP and (c 2(1 d.f.) = 23.1, p<0.0001 for DBP) Mean blood pressure in the nurse measured clinics was 136.6/79.9 mmHg vs the automated BP clinic 130.2/76.4 mmHg (t(197 d.f.)=2.05, p=0.04 for SBP and t(197 d.f.)=1.85, p=0.07 for DBP). Thus, an automated blood pressure device helped to reduce the preference for a systolic or diastolic blood pressure reading ending in 0 compared to measurements by a nurse. The average blood pressure was lower in the automated group reflecting either reduced white coat effect, better blood pressure control or both. In conclusion, BP measurement with the BP Tru helps to reduce potential bias in blood pressure measurements and may help achieve lower overall blood pressures in clinic patients.