Validation of sleep-wake estimation from thigh-worn accelerometers against polysomnography in adolescents with and without mental disorders.

BACKGROUND: To validate the estimation of sleep-wake patterns using thigh-worn accelerometers against polysomnography -the gold standard for sleep assessment-in adolescents with and without mental disorders from the Danish High Risk and Resilience Study, addressing the feasibility of simpler methods in clinical populations. METHODS: 146 adolescents (ages 15-17) underwent one night polysomnography and concurrent thigh-worn accelerometry. Sleep-wake parameters-total sleep time, sleep efficiency, sleep onset latency, wake after sleep onset (WASO), and number of awakenings -were compared using ActiPASS software. Reliability was evaluated in the full sample and in subgroups of participants fulfilling and not fulfilling criteria for an Axis I mental disorder. Sensitivity and specificity were estimated based on cut-off values. RESULTS: ActiPASS underestimated total sleep time by 51.7 min, sleep efficiency by 1.4%, and number of awakenings by 13.7, and overestimated sleep onset latency by 23.7 min and WASO by 25.7 min compared to polysomnography. Intraclass correlation coefficients (ICCs) were high for total sleep time (0.88), moderate for sleep efficiency (0.55), sleep onset latency (0.69), and WASO (0.64), but low for number of awakenings (0.39). Group-level ICCs were marginally higher in the "No Mental Disorder" group. Sensitivity was perfect and specificity high for total sleep time < 360 min. Sensitivity was high and specificity moderate for WASO > 50 min and sleep onset latency > 30 min. CONCLUSIONS: Thigh-worn accelerometers showed potential for monitoring of sleep-wake patterns and screening for sleep disturbances in adolescents with mental disorders, demonstrating moderate to high reliability for key sleep-wake metrics.