Verbal and nonverbal outcomes of toddlers with and without autism spectrum disorder, language delay, and global developmental delay

Background and Aims: Children with autism spectrum disorder (ASD) exhibit a heterogeneous clinical phenotype with wide variability in their language and intellectual profiles that complicates efforts at early detection. There is limited research examining observational measures to characterize differences between young children with and without ASD and co-occurring language delay (LD) and global developmental delay (GDD). The first aim of this study was to compare early social communication measured in the second year of life in children diagnosed at age 3 with ASD, developmental delays (DD), and typical development (TD). The second aim was to compare early social communication in six subgroups of children: ASD, ASD+LD, ASD+GDD, LD, GDD, and TD. Our third aim was to determine the collective and unique contributions of early social communication to predict verbal and nonverbal developmental outcomes at three years of age for children with and without ASD. Methods: Analyses of covariance controlling for maternal education were employed to examine group differences in social communication in 431 toddlers recruited through screening in primary care. Multiple linear regression analyses were conducted to evaluate associations between the Communication and Symbolic Behavior Scales (CSBS) Behavior Sample composite standard scores and Mullen Scales of Early Learning T scores for children with and without ASD. Results: Distinct patterns of early social communication were evident by 20 months. Children with TD differed significantly from children with ASD and DD on all three CSBS Behavior Sample composites. Children with ASD had significantly lower scores than those with DD and TD on the social and symbolic composites. Among the six subgroups, all three composites of the CSBS Behavior Sample differentiated children with TD from all other subgroups. Children with ASD+GDD scored significantly lower than all other subgroups on social and symbolic composites. Patterns of social communication emerged for children with and without ASD, which held among subgroups divided by developmental level. The CSBS Behavior Sample social and symbolic composites contributed unique variance in predicting developmental outcomes in both groups. The speech composite contributed unique variance to expressive language, receptive language, and visual reception in children without ASD, and contributed uniquely to expressive language only for children with ASD. Conclusions: The CSBS Behavior Sample, an observational measure for children aged 12-24 months, detected social communication delays and explained a significant amount of variance in verbal and nonverbal outcomes a year later in this large sample of young children grouped by ASD diagnosis and developmental level.