Effect of exercise-induced short-term changes in blood pressure and heart rate on smartwatch measurement accuracy

Abstract

Blood pressure (BP) is a key vital sign that many people around the world are increasingly monitoring continuously using smartwatches. It remains unclear whether smartwatches can fully replace traditional oscillometric cuff-based devices under real-world conditions. The aim of this study was to verify the accuracy of BP and heart rate (HR) measurements using smartwatches compared to a standard patient monitor before and after a short-term exercise. Twenty-seven participants underwent an experimental part divided into three phases: pre-exercise, exercise, and post‑exercise. BP and HR were measured simultaneously with a Samsung smartwatch Watch Active2 or Watch3 and a patient monitor Datex-Ohmeda S/5 two times before and four times after exercise. The results showed a slight deterioration in the accuracy of smartwatch blood pressure measurements after exercise compared to values measured before exercise (-0.3 ± 6.3 vs. 2.9 ± 8.8 for systolic BP and 1.0 ± 4.8 vs. 1.9 ± 5.7 for diastolic BP given as mean ± standard deviation). Furthermore, smartwatches underestimated high BP readings and overestimated low BP readings. On the other hand, measuring HR with smartwatches proved to be very accurate even after exercise (-0.8 ± 3.7 vs. 0.0 ± 2.5). The study confirmed the accuracy of smartwatch heart rate measurements but questioned the accuracy of smartwatch BP measurements taken at values far from the calibration point.