Estimating adolescent sleep need using dose-response modeling

Study Objectives: This study will (1) estimate the nightly sleep need of human adolescents, (2) determine the time course and severity of sleep-related deficits when sleep is reduced below this optimal quantity, and (3) determine whether sleep restriction perturbs the circadian system as well as the sleep homeostat. Methods: Thirty-four adolescents aged 15 to 17 years spent 10 days and nine nights in the sleep laboratory. Between two baseline nights and two recovery nights with 10 hours' time in bed (TIB) per night, participants experienced either severe sleep restriction (5-hour TIB), moderate sleep restriction (7.5-hour TIB), or no sleep restriction (10-hour TIB) for five nights. A 10-minute psychomotor vigilance task (PVT; lapse = response after 500 ms) and the Karolinska Sleepiness Scale were administered every 3 hours during wake. Salivary dim-light melatonin onset was calculated at baseline and after four nights of each sleep dose to estimate circadian phase. Results: Dose-dependent deficits to sleep duration, circadian phase timing, lapses of attention, and subjective sleepiness occurred. Less TIB resulted in less sleep, more lapses of attention, greater subjective sleepiness, and larger circadian phase delays. Sleep need estimated from 10-hour TIB sleep opportunities was approximately 9 hours, while modeling PVT lapse data suggested that 9.35 hours of sleep is needed to maintain optimal sustained attention performance. Conclusions: Sleep restriction perturbs homeostatic and circadian systems, leading to dose-dependent deficits to sustained attention and sleepiness. Adolescents require more sleep for optimal functioning than typically obtained.