Abstract
Early childhood represents a period of profound developmental changes for sleep and circadian biology. Although the relationship between sleep and circadian timing has been well characterized in older populations, such data in young children remain limited. Here, we provide fundamental data on the relationship between endogenous circadian phase and sleep timing in a sample of preschool-aged children. Participants were 49 healthy children ages 3.1 to 6.0 years (M = 4.44 years, SD = 0.69 years, 27 female). After 7 days of maintaining a consistent, parent-selected sleep schedule, children completed an in-home, dim-light circadian assessment. Saliva samples were collected in 30-min intervals throughout the evening to determine the timing of children's dim-light melatonin onset (DLMO). Children's DLMOs occurred an average of 35.0 ± 35.3 min before their bedtimes, with parent-selected bedtime occurring before DLMO for 18.4% of children. Children with later DLMOs had significantly later bedtimes (r = 0.65), sleep onset times (r = 0.74), midsleep times (r = 0.74), and wake times (r = 0.66) (all p < 0.001). For every hour later that DLMO occurred, average bedtime and sleep onset time were 28.0 and 33.4 min later, respectively. In addition, children with later DLMOs had higher scores on a parent-reported measure of chronotype (r = 0.56, p < 0.001), indicating greater eveningness. No association between DLMO time and sleep duration or social jetlag was observed. These data extend previous findings in toddlers, demonstrating a consistent relationship between circadian phase and sleep timing, as well as chronotype, throughout early childhood.