Associations between expressive and receptive language and internalizing and externalizing behaviours in a community-based prospective study of slow-to-talk toddlers

BACKGROUND: Evidence suggests that language and social, emotional and behavioural (SEB) difficulties are associated in children and adolescents. When these associations emerge and whether they differ by language or SEB difficulty profile is unclear. This knowledge is crucial to guide prevention and intervention programmes for children with language and SEB difficulties. AIMS: To determine whether receptive and expressive language skills are associated with internalizing and externalizing behaviours in slow-to-talk toddlers. METHODS & PROCEDURES: In a community-based prospective study of 200 slow-to-talk children, language was measured at 24 and 36 months using Preschool Language Scale 4th Edition and at 48 months using Clinical Evaluation of Language Fundamentals-Preschool 2nd Edition. Internalizing and externalizing behaviours were measured by parent report at each age. Longitudinal data were analysed using repeated-measures regression, with up to three observations per child. Robust standard errors were used to account for non-independence of measures within participants. The shape of the associations were examined by fitting quadratic and cubic terms. The effects of confounders on the associations were examined. OUTCOMES & RESULTS: Receptive language had a negative linear association with internalizing behaviours after adjusting for confounders (β = -0.16, 95% [CI = -0.26, -0.07], p = .001); and a negative curved association with externalizing behaviours after adjusting for biological confounders (βquadratic = 0.08 [0.01, 0.15], p = .03, βcubic = -0.04 [-0.07, -0.02], p = .001), attenuating after adjusting for environmental confounders (βquadratic = 0.06 [-0.01, 0.13], p = .09, βcubic = -0.03 [-0.06, -0.003], p = .03). The curvature suggests that the negative association with externalizing behaviours only existed for children with either very low or very high receptive language scores. After controlling for confounders, there was no evidence that expressive language scores were associated with internalizing (β = -0.08, 95% [CI = -0.17, 0.01], p = .10) or externalizing behaviours (β = 0.03, 95% [CI = -0.09, 0.18], p = .61). Tests of interaction revealed no evidence of a differential association by age. CONCLUSIONS & IMPLICATIONS: In 24-48-month-old slow-to-talk children, lower receptive language scores were associated with higher internalizing behaviours. The magnitude of the association was small. For children with very poor receptive language scores, lower receptive language skills were associated with higher externalizing behaviours. Young children with low receptive language abilities may be at risk of internalizing difficulties; those with very low receptive language skills may be at particular risk of externalizing difficulties. This has clinical implications for interventions for young children with receptive language difficulties