Wearable trackers that detect sleep offer users a way to track their sleep quality and patterns without the use of expensive equipment. Few studies have tested the validity of these trackers on sleep measure. PURPOSE: To examine the validity of the Actigraph GT9X (AG), SenseWear Mini Armband (SW), Basis Peak (BP), Fitbit Charge HR (FB), Jawbone UP3 (JU), and Garmin Vivosmart (GV) for estimating sleep variables as compared with a sleep diary. METHODS: 78 healthy individuals participated in the study. Group 1 (n= 38) and wore the AG, SW, BP, and FB or Group 2 (n = 40) and wore the AG, JU, and GV. Monitors were worn on the non-dominant arm for 3 nights and a sleep log was completed. Sleep variables were total sleep time (TST), time in bed (TIB), sleep efficiency (SE), and wake after sleep onset (WASO). Pearson correlation, mean absolute percentage errors (MAPE), equivalence testing, Bland-Altman plots, and ANOVA were used to assess validity compared with the diary. RESULTS: Overall, monitors that showed the greatest correlation with the sleep diary for TST were the JU and FB (effect size= 0.09 and 0.23, respectively). The greatest correlation with the sleep diary for TIB was seen with the SW, GV, and JU (effect size= 0.09, 0.16, and 0.07, respectively). SE and WASO showed very poor correlation with the log. Measures for equivalence testing confirmed the success of the JU, SW, FB, and GV for measureing TIB and TST. CONCLUSION: The FB, SW, JU, and GV could be valid measure of TST and TIB. The monitors are not valid regarding wake times during sleep. Further research is needed to validate these monitors with polysomnography.
Lee, Jung-Min; Kim, Youngwon; Kwon, Yi-Suk; Derrick, Timothy R.; and Welk, Gregory J., "Calibration of Built-in Accelerometer Using a Commercially Available Smartphone" (2014). Research Presentations. 13.