Reliability and device objectivity in oscillatory blood pressure measurement - a measurement error analysis to inform...

The implementation of different hypertension thresholds (e.g. ESC/ESH: 140:90 or ACC/AHH: 130:80) substantially increased hypertension prevalence donation of pharmacological treatment across various populations. Under narrow margins of only 10mmHg, measurement accuracy and reliability become decisive for clinical decision-making. However, detailed information on systematic and random errors naturally occurring in clinical practice under repeated testing conditions remain widely limited.To provide practically relevant insights, 65 participants were recruited for visited the lab four times with one un-standardized arrival measurement to mimic oftentimes observed clinical real-world scenarios followed by four standardized measurements performed in adherence with current guidelines. Tests were performed with two certified blood pressure devices, each on two testing days to quantify intra- and interday, as well as interdevice measurement errors via relative reliability and systematic and random measurement errors.Insufficient pre-measurement rest periods caused clinically relevant systematic blood pressure enhancements. Within a subsequent 10-18 min rest, the blood pressure was reliably reduced with about 10 mmHg on both days and both devices. However, the Boso Medicus testing device showed consistently higher blood pressure values with 13mmHg compared to the Welch Allyn monitor.Although measurements could be replicated inter-day in both devices, substantial interdevice discrepancies and clinically relevant random errors were observed, that can cause false-positive hypertension diagnoses, especially when pre-measurement rest was insufficient. Results underscore the relevance of performing several blood pressure assessments (per day and on different days) to account for random measurement errors and reduce the risk of inappropriate clinical decision-making and unnecessary treatment escalation.