BioPsychoSocial Development Laboratory

Objectives To assess variations in children's cognitive and language test scores according to observed dyadic maternal sensitivity and the child's gestational age, after adjustment for known confounders. Study design Systematic review and a one-stage individual participant data meta-analysis of harmonized data from 7 birth cohorts across 5 countries (n = 2560) using mixed effects linear regression. Maternal sensitivity and child cognition and language test scores were z -standardized relative to contemporary cohort-specific term-born controls. Models were adjusted for child sex, multiple birth, small for gestational age, age at outcome assessment, mother's education, mother's age at the child's birth, single mother status, and child neurodevelopmental impairment. Results Maternal sensitivity was associated with subsequently assessed cognition (0.20; 95% CI, 0.14-0.26; P < .001) and language (0.15; 95% CI, 0.08-0.22; P < .001). A significant interaction between gestational age and sensitivity on cognition (−0.01; 95% CI, −0.02 to −0.01; P = .001) was observed; the association between sensitive parenting and later cognition was stronger among children born at younger gestational ages. Sensitivity analyses adjusting for children's neurodevelopment impairments produced similar findings. Conclusions High maternal sensitivity is a potential protective factor for cognitive and language functioning for children born preterm, especially among those born at earlier gestational ages.