Examining the Role of Social-Communication in Explaining Treatment Gains for Adolescents with ASD Following the PEERS® Social Skills Intervention

Background: Social communication skills, including social interaction, social competence, speech style and language pragmatics, are strongly associated with long-term outcomes for youth with autism spectrum disorder (ASD) (Howlin et al. 2004; Ingersoll, 2011). Deficits in social communication are often the earliest identifiable indicators of ASD, and several early intervention models target this domain (Ingersoll, 2013; Kasari, 2015). However, research has not previously examined the influence of social communication on the development of social skills in adolescence.

Objectives: This study examines the role of social communication in predicting social skills ratings following the implementation of the UCLA Program for the Education and Enrichment of Relational Skills (PEERS^®), an evidence-based social skills curriculum in both school-based and outpatient mental health settings.

Methods: Participants included 186 adolescents with ASD referred for social skills training in outpatient and school settings. Among the outpatient sample, participants included 80 adolescents (males=68; females=12) 11-18 years of age (M=13.8, SD=2.25) School based participants included students (n = 80 males; n = 26 females) in 6^th – 12^thgrades attending a non-public school for students with ASD. Adolescents, parents, and teachers completed pre and post-test measures, including the Social Skills Improvement System (SSIS; Gresham & Elliot, 2008) and the Social Responsiveness Scale (SRS; Constantino & Gruber, 2005). Following the completion of baseline measures, adolescents participated in a 14-week manualized teacher-facilitated intervention then completed post-test measures. A linear regression analysis was conducted to examine the relation between baseline measures of social-communication (SRS) and the SSIS post-treatment.  

Results: Data analysis shows that social communication (SRS) at baseline is significant in explaining the variance in social skill acquisition (SSIS) following intervention. Adolescent social communication scores (SRS) both significantly predicted b = -.61, t(70) = -6.92 p<.000 and explained a significant proportion of variance in social skills as rated by parents R² = .41, F(1,70) = 47.92, p£.000  in the school-based sample. In the same sample, social communication both significantly predicted b = -.82, t(104) = -6.89, p<.000 and explained a significant proportion of variance in social skills R² = .32, F(1,104) = 47.34, p£.000 as rated by teachers following intervention. Social communication also significantly predicted social skills scores as rated by parents in the clinic sample, b = -.37, t(80) = -3.51, p<.001 and also explained a significant proportion of the variance R² = .14, F(1,78) = 12.32, p<001. However, social communication was better able to account for outcomes in the school-based sample than the clinic sample. 

Conclusions: Data analysis revealed that social communication significantly predicted social skills scores and also explained a significant proportion of variance in social skills scores following intervention in both school-based and outpatient settings.  This research demonstrates the importance of social communication in the acquisition of social skills following intervention and should be further evaluated. These results may provide additional support for early intervention targeting the domain of social communication to support social skills intervention and boost treatment outcomes for social skills acquisition.