Comparison of multichannel and single channel wrist-based devices with polysomnography to measure sleep in youth.

STUDY OBJECTIVES To compare sleep parameters produced by the Fitbit Charge 3 (Fitbit) and Actigraph GT9X accelerometer (Actigraph) to polysomnography (PSG) in youth. METHODS Youth (n=56, age=9.2±3.3 years) wore a Fitbit and Actigraph on their non-dominant wrist concurrently with PSG during an overnight observation at a children's sleep lab. Total sleep time (TST), sleep efficiency (SE), wake after sleep onset (WASO), sleep onset, and sleep offset were extracted from Fitabase and Actilife with the Sadeh algorithm, respectively. Bland-Altman plots were used to assess the agreement between wearable devices and PSG. RESULTS Seventy-nine percent of participants were diagnosed with obstructive sleep apnea. Compared to PSG, Fitbit and Actigraph underestimated TST by 6.1 min (absolute mean bias [AMB] = 27.7 min) and 31.5 min (AMB = 38.2 min), respectively. Fitbit overestimated SE by 3.0% (AMB=6.3%) and Actigraph underestimated SE by 12.9% (AMB=13.2%). Fitbit overestimated WASO by 18.8 min (AMB=23.9 min) and Actigraph overestimated by 56.1 min (AMB=54.7 min). Fitbit and Actigraph underestimated sleep onset by 1.2 min (AMB=13.9 min) and 10.2 min (AMB=18.1 min), respectively. Fitbit and Actigraph overestimated sleep offset by 6.0 min (AMB=12.0 min) and 10.5 min (AMB=12.6 min). Linear regression indicated significant trends with Fitbit underestimating WASO and SE at higher values. CONCLUSIONS Fitbit provided comparable, and in some cases better sleep estimates with PSG, compared to Actigraph. Findings support the use of multichannel devices to measure sleep in youth. Additional studies are needed in healthy samples, over several nights, and in free-living settings.