Predictors of Pragmatic Language Functioning in Younger Siblings of Children with Asdpredictors of Pragmatic Language Functioning in Younger Siblings of Children with ASD

Background:  Pragmatic language impairments are a hallmark of children with autism spectrum disorder (ASD; Simmons et al., 2014), and are sometimes observed in their parents (Hurley et al., 2007; Landa et al., 1992) and younger siblings (Ben-Yizhak et al., 2010; Miller et al., 2015). Limited research has examined predictors of later pragmatic language functioning (Gillespie-Lynch et al., 2015).

Objectives:  To examine pragmatic language abilities in school-aged siblings of children with ASD, and identify early predictors of these abilities.

Methods:   Between 8 and 12 years of age, 34 younger siblings of children with ASD (high-risk, HR) and 10 low-risk (LR) controls were assessed with the Pragmatic Rating Scale–School-Age (PRS-SA; Landa), scored based on behaviors observed during the Autism Diagnostic Observation Schedule (ADOS; Lord et al., 2002, 2012), Module 3 or 4. All participants also had been assessed at least twice between 14 and 36 months (14/18,24/30,and 36 months), with the ADOS and Mullen Scales of Early Learning (MSEL; Mullen, 1995). Based on 36-month assessments and clinical impressions, high-risk siblings were classified as ASD (HR-ASD,n=15), broad autism phenotype (BAP; HR-BAP,n=8), and typically developing (HR-TD, n=11). Nonparametric analyses compared PRS-SAscores across HR-ASD, HR-BAP, HR-TD, and LR controls; early predictors of later pragmatic language abilities were examined using Pearson’s correlations and linear regression.

Results:  A Kruskal-Wallis Chi-Square test revealed significant between-group differences in PRS-SA scores, Χ(3)=33.28,p<.001. Follow-up Mann-Whitney tests revealed:

• no significant difference between LR controls (M=6.50,SD=3.60) and HR-TD (M=5.64,SD=2.58),Z=-0.57,p=.57

• significantly lower (better) PRS-SA scores for HR-TD than HR-ASD (M=31.87,SD=8.17),Z=-4.29,p<.001, and HR-BAP (M=17.75,SD=7.30),Z=-3.36,p=.001

• significantly lower PRS-SA scores for HR-BAP than HR-ASD,Z=-3.13,p=.002.

Pearson’s correlations were found between PRS-SA scores and 24-month:

• ADOS Communication+Social scores, r= .60,p<.001,r²=.36.

• MSEL Receptive Language, Expressive Language, and Early Learning Composite scores, r= -.60,p<.001,r²=.36; r= -.51,p<.001,r²=.26; and r= -.55,p<.001,r²=.30, respectively.  

Linear regression analyses revealed that the MSEL Receptive Language score had a unique negative effect and ADOS Communication+Social scores had a unique positive effect on PRS-SA scores, (b=-.35,SE=.15), t₍₃₆₎= -2.40,p=.02 and (b=.99,SE=.41), t₍₃₆₎=2.43,p=.02, respectively. The model with both predictors accounted for a significant amount of variance in later pragmatic language functioning, R²=.45,F(2,39)=15.70,p<.001,R²_adjusted=.42.

Conclusions:  The present data confirm poorer pragmatic language abilities in school-aged high-risk siblings with ASD and characteristics of the BAP, but not in siblings developing typically. These findings support recommendations to subdivide unaffected siblings into those who do and do not demonstrate BAP characteristics when examining pragmatic language (Ben-Yitzhak et al., 2011). Findings also suggest that autism-related social-communication and early learning abilities at 24 months are predictive of later pragmatic language functioning in high-risk siblings and low-risk controls. Continued investigation into the pragmatic language abilities of younger siblings of children with ASD and predictors of these abilities is warranted.