Positive Self-Esteem As a Predictor of Decreased Problem Behaviors in Adolescents with ASD Following the UCLA PEERS® Intervention

Background: Individuals with Autism Spectrum Disorder (ASD) are known to have social deficits and elevated problem behaviors (Macintosh & Dissanayake, 2006). Problem behaviors may be categorized as either externalizing (e.g., aggression or hyperactivity/inattention) or internalizing (e.g., social withdrawal, anxiety, or depression) (Gresham & Kettler, 2010), both of which are frequently observed in youth with ASD (Bauminger et al., 2010). Yet, positive self-esteem or self-concept during adolescence has been found to protect against internalizing and externalizing behaviors in non-clinical samples of adolescents (Piers et al., 2002; Ybrandt, 2008). The Program for the Education and Enrichment of Relational Skills (PEERS^®) is an evidence-based, parent-assisted social skills intervention shown to improve social skills and decrease problem behaviors in adolescents with ASD (Laugeson et al., 2009; Laugeson et al., 2012); however, the influence of self-esteem on problem behaviors following this treatment has yet to be investigated. 

Objectives: This study examines the extent to which baseline self-esteem predicts decreased problem behaviors in adolescents with ASD following completion of the UCLA PEERS^®intervention.

Methods: Eighty-three adolescents with ASD from 11-18 years of age (M=14.17, SD=1.70) and their parents participated in weekly 90-minute group treatment sessions for 14-weeks as part of the UCLA PEERS^® for Adolescents social skills intervention. To assess baseline self-esteem, adolescents completed the Piers Harris Self-Concept Scale, Second Edition (PHS-2; Piers et al., 2002) prior to treatment. The PHS-2 includes six subscales measuring self-concept. To assess changes in problem behaviors following treatment, parents completed the Social Skills Improvement System (SSIS; Gresham and Elliot, 2008) pre- and post-intervention. The SSIS assesses problem behaviors including externalizing, bullying, hyperactivity/inattention, internalizing, and autism spectrum problem behaviors. Pearson correlation coefficients were calculated to examine the relationship between baseline PHS-2 and change in SSIS-Problem Behaviors following treatment.

Results: Results indicate that higher baseline standard scores on the PHS-2 Behavioral Subscale significantly predict a decrease in the instances of bullying others per parent report on the SSIS Bullying subscale (p<.05) following treatment. In addition, results reveal that higher baseline PHS-2 Behavioral Subscale scores predict a decrease in parent-reported overall problem behaviors on the SSIS Problem Behaviors Scale (p<.1) and a decrease in externalizing problem behaviors on the SSIS Externalizing Subscale (p<.1) at trend levels following treatment. Results did not reveal any other significant relationships between baseline PHS-2 subscales and change in scores on the SSIS Problem Behaviors subscales.

Conclusions: Findings suggest that adolescents with ASD reporting higher self-esteem prior to treatment may be less likely to exhibit problem behaviors following the PEERS^® intervention. In particular, adolescents endorsing more positive behaviors prior to treatment are less likely to engage in bullying behaviors towards others following intervention. Inversely, adolescents endorsing more problematic behaviors prior to treatment are more likely to engage in bullying after intervention. Additionally, greater behavioral problems prior to treatment may result in more externalizing and problem behaviors for adolescents with ASD at a trend level. These findings emphasize the benefit of identifying predictors of treatment outcome to foster the evolution of targeted social skills interventions for adolescents with ASD.