Accuracy Of Motionsense Hrv For Assessing Sedentary Behavior And Physical Activity

Abstract

MotionSense HRV (MS-HRV) is a wrist-worn accelerometery-based sensor that is paired with a smartphone to examine health behaviors such as stress response, heart rate, and physical activity (PA). However, little information is available on the validity of MS-HRV in estimating PA. PURPOSE: To evaluate the accuracy of MS-HRV for assessing PA and sedentary behavior (SED) in adults, using the most widely utilized accelerometer-based activity monitor, ActiGraph GT9X (GT9X), as a criterion measure, during free-living conditions. METHODS: 19 adults (Female: 58%, Age: 30.9 ± 13.7 yrs, BMI: 30.3 ± 4.1 kg·m2) wore the MS-HRV on non-dominant wrist and AG on dominant hip simultaneously for seven consecutive days. The MS- HRV is composed of a smartphone and wristband that is accelerometer and LED biometric sensor to measure PA unobtrusively. Raw acceleration data from both MS-HRV and GT9X were processed using GGIR package that summarizes multiday raw acceleration data to the amount of time (min/day) spent in SED and moderate-to-vigorous PA (MVPA) using Euclidean Norm Minus One (aka. ENMO). Pearson correlations and Bland-Altman (BA) plots were used to examine the relationship and agreement between MS-HRV and GT9X. Equivalence test was used to compare the 90% confidence intervals (CI) of the estimates from the MS-HRV with the respective equivalence zone (EZ; ± 10% of the mean estimates) from the GT9X. RESULTS: The correlations between MS-HRV and AG were high for both SED (r = 0.95, P <.01) and MVPA (r = 0.89, P <.01). BA plots illustrated no variance difference in SED estimates (Pitman’s test: r = -0.16, P = 0.95), but significantly different variance in MVPA (r = -0.91, P = .02) from two devices. The estimates of SED and MVPA from the MS-HRV (SED Mean (90% CI): 237.6 min/day (198.5 - 276.8)); MVPA: 21.5 min/day (16.4 - 26.6) were not significantly equivalent to those from the GT9X (SED Mean (EZ)): 257.6 min/day (231.8 - 283.4); MVPA: 8.7 min/day (7.8 - 9.5)). CONCLUSIONS: The MS-HRV accelerometer yielded comparable estimates of SED, but significantly higher estimate of MVPA when compared with the GT9X accelerometer. The observed difference in MVPA estimate could be due to the difference in device placement location (wrist vs. hip). Therefore, subsequent research that compares the estimates from two devices worn at the same location is warranted.