A longitudinal study of child sleep in high and low risk families: Relationship to early maternal settling strategies and child psychological functioning

Sheridan, A., Murray, L., Cooper, P.J., Evangeli, M., Byram, V., Halligan, S.L. (2013). A longitudinal study of child sleep in high and low risk families: Relationship to early maternal settling strategies and child psychological functioning. Sleep Medicine, 14:3, 266-273.

Hypothesis

To investigate whether sleep disturbances previously found to characterize high risk infants: (a) persist into childhood; (b) are influenced by early maternal settling strategies and (c) predict cognitive and emotional/behavioral functioning

Design: Longitudinal Study

Variables Measured, Instruments Used

1. Child Sleep: Infant Sleep Questionnaire (ISQ) completed at 12-months and 18-months.
2. Mother’s perception of child sleeping problems: At five years answered questions about frequency and severity of child sleep difficulties.
3. Infant Wake Time and Bed Time: recorded in a sleep diary at 12-months and 18-months, repeated again at five years for a five day period.
4. Five Year Actigraphy: During 50day period, children wore Actiheart physiological monitoring devices around the clock.                                                             
5. Child Cognitive Ability: Wechsler Preschool and Primary Scale of Intelligence – Revised (WPPSI-R) completed at five years.                                                               
6. Child Adjustment: Child Behavior Checklist (CBCL) completed by parent and teacher at 5 years.
7. Maternal Settling to Sleep Strategies: Self-report Parental Interactive Bedtime Behavior Scale (PIBBS) completed at four weeks, seven weeks, 12 weeks and 12-months           

Participants

• N=121 recruited
• Participant ages: four weeks to five years old
• Location: Royal Berkshire Maternity Hospital in Reading, UK
• Eligibility: Selected from previous study and screened via a 20 item questionnaire indexing level

of psychosocial adversity

• SES: Unknown                          

Limitations                                                      

1. The sample reduced from 121 at recruitment to 98 at five year follow-up (81% retention rate). Path analyses were based on the entire recruitment sample (i.e., n = 121) with missing data modelled using full information maximum likelihood estimation, the recommended approach for minimizing the introduction of bias due to missing data.                                       

Finding(s)

1. Early maternal over-involvement in infant settling is associated with less optimal sleep in children, which in turn, is related to child adjustment.
2. The findings highlight the importance of supporting parents in the early development of good settling practices, particularly in high-risk populations.                                                                   
3. Significant gender differences were identified in relation to logged CBCL aggression scores, sleep period according to diary report and actigraphy sleep duration. Gender was controlled for in subsequent analyses, as necessary.                                                       
4. Previously reported associations between psychosocial adversity and sleep quality in infants in the current sample were maintained in childhood.                                                   
5. Even once risk group effects were taken into account, maternal settling strategies in the first months of life predicted sleep quality at five years.
6. Consistent with previous research, there was some evidence of associations between child sleep quality and levels of anxiety/depression and aggression in children; the possibility of links between sleep characteristics and IQ received less support.
7. The current findings are consistent with previous reports of links between family stress or low SES and poorer sleep quality in children.                                                                                    
8. Maternal active involvement in settling the infant in the first 12 weeks of life was a significant predictor of child sleep characteristics at five years, and contributed to the association between psychosocial adversity and sleep disturbance at five years.
9. Poorer five year sleep was associated with concurrent child anxiety/depression and aggression