Comparison of the Fitbit® charge and polysomnography for measuring sleep quality in children with sleep disordered breathing

BACKGROUND: Due to the cost and inconvenience of polysomnography (PSG), wrist-worn accelerometers have been explored as an alternative method to measure sleep efficacy in children with obstructive sleep apnea (OSA) or sleepdisordered breathing (SDB). We compared the measurement of sleep quality with the Fitbit (R) charge (Fitbit Inc., San Francisco, CA, USA) compared to PSG in children presenting for sleep studies. METHODS: Children ages 3 to 18 years presenting for PSG with persistent SDB symptoms were enrolled. During PSG, the Fitbit (R) charge (Fitbit Inc.) was placed on the wrist and time-synchronized with sleep laboratory devices, which were worn while a single-night PSG was performed. Bias and concordance in measurements of total sleep time (TST), total wake time (TWT), and number of awakenings were assessed using paired t-tests, sign-rank tests, and Lin's concordance coefficient. RESULTS: The study cohort included 22 patients (9 boys and 13 girls; 9 +/- 3 years). TST was significantly overestimated using the Fitbit (R) charge (Fitbit Inc.) (difference in means: 30 min; 95% confidence interval [CI] of difference: 3-58; P=0.031), while TWT was underestimated (difference: 23 min; 95% CI: 4-42; P=0.018). All measures showed a lack of concordance between the Fitbit (R) charge (Fitbit Inc.) and PSG. CONCLUSIONS: The current prospective study confirms that the Fitbit (R) charge (Fitbit Inc.) overestimates time spent asleep compared to PSG in children with OSA/SDB symptoms, limiting the validity of sleep monitoring with wearable activity trackers appears in these patients.