Social communication disorder outside autism? A diagnostic classification approach to delineating pragmatic language impairment, high functioning autism and specific language impairment

BackgroundDevelopmental disorders of language and communication present considerable diagnostic challenges due to overlapping of symptomatology and uncertain aetiology. We aimed to further elucidate the behavioural and linguistic profile associated with impairments of social communication occurring outside of an autism diagnosis. MethodsSix to eleven year olds diagnosed with pragmatic language impairment (PLI), high functioning autism (HFA) or specific language impairment (SLI) were compared on measures of social interaction with peers (PI), restricted and repetitive behaviours/interests (RRBIs) and language ability. Odds ratios (OR) from a multinomial logistic regression were used to determine the importance of each measure to diagnostic grouping. MANOVA was used to investigate differences in subscale scores for the PI measure. ResultsGreater degrees of PI difficulties (OR=1.22, 95% CI=1.05-1.41), RRBI (OR=1.23, 95% CI=1.06-1.42) and expressive language ability (OR=1.16, 95% CI=1.03-1.30) discriminated HFA from PLI. PLI was differentiated from SLI by elevated PI difficulties (OR=0.82, 95% CI=0.70-0.96) and higher expressive language ability (OR=0.88, 95% CI=0.77-0.98), but indistinguishable from SLI using RRBI (OR=1.01, 95% CI=0.94-1.09). A significant effect of group on PI subscales was observed (=1.38, F(4, 56)=19.26, p<.01) and PLI and HFA groups shared a similar PI subscale profile. ConclusionsResults provide empirical support for a conceptualisation of PLI as a developmental impairment distinguishable from HFA by absence of RRBIs and by the presence of expressive language difficulties. PI difficulties appear elevated in PLI compared with SLI, but may be less pervasive than in HFA. Findings are discussed with reference to the proposed new category of social communication disorder' in DSM-5.