Longitudinal Study of Infant Sleep Development: Early Predictors of Sleep Regulation Across the First Year

An important developmental task for infants over their first few years of life is to learn to settle to sleep with a reasonably short latency, maintain sleep through the night and coordinate with family sleeping and waking schedules. A child who can reliably do this is exhibiting self-regulated sleep. Otherwise, children's sleep may have to be other (non-self) regulated to some degree and they may exhibit pediatric sleep disturbances (e.g., extended sleep latency, and/or frequent nightwaking); these are reported by 36-45% of parents of infants between ages four to 12 months.

To answer the question: Can infant and parent factors observed at 1 month of infant age predict which infants will have regulated sleep at 6- and 12-months of age? Prediction from 1 month has not previously been investigated.

In a prospective longitudinal study, the mothers of 52 typically developing infants completed 6-day sleep diaries at 1, 3, 6, 9 and 12 months from which a composite sleep score (CSS) was derived for each child at each month. Diary reliability was assessed once (for 54% of families) using all-night videosomnography.

At 6 months, CSS scores were distributed bi-modally and thus differentiated into two groups by an empirically observed CSS cutoff score, with a majority (56%) of infants classified as self-sleep regulated (S-R) and the rest as non-self sleep-regulated (NS-R). At 12 months, 72% could similarly be classified as S-R, while 28% exhibited some continuing sleep disturbance. Discriminant function analysis investigated the predictors of S-R vs NS-R group membership at 6 and 12 months from parent and child variables recorded at 1 month. Parent presence at sleep onset and less total infant sleep time predicted group membership at 6 months with 94% classification accuracy, and parental presence at sleep onset and frequency of infant night wakings predicted group membership at 12 months with 85% accuracy. At 1 month, parents of infants later classified as NS-R at 6 and 12 months had higher frequencies of all settling activities than parents of those later classified as S-R.

Variables measured at 1 month that predicted sleep status at 6 and 12 months were parental presence at sleep onset, frequency of infant night waking and total infant sleep time. The overall frequency of parent settling activities at 1 month also clearly differentiated the two sleep groups at the older ages. Parenting behaviours are modifiable factors and thus may have the potential for preventing pediatric sleep disturbances in children.