Evidence-Based Friendship Training for Adolescents with Autism Spectrum Disorders: Replication and Follow-up Study of the UCLA PEERS Program

Background:

Social deficits and poor friendship quality are common areas of impairment for youth with Autism Spectrum Disorders (ASD), yet the majority of social skills intervention studies for this population have focused on improving the social competence of younger children in the lower ranges of functioning.  Few evidence-based social skills interventions exist aimed at improving friendships for higher functioning teens with ASD. 

Objectives:

This study represents the second randomized controlled clinical trial of the UCLA PEERS Program, an evidence-based parent-assisted social skills intervention aimed at improving friendship skills among high-functioning middle school and high school adolescents with ASD.  Treatment outcome, maintenance of treatment gains, and predictors of treatment success were examined.

Methods:

Twenty-eight participants and their parents were randomly assigned to a treatment with follow-up or delayed treatment control condition.  Participants attended weekly 90-minute group sessions over a 14-week period. Targeted skills included: conversational skills; electronic communication; appropriate humor; developing friendship networks; peer entry and exiting skills; good host/guest behavior during get-togethers; good sportsmanship; changing bad reputations; and strategies for handling teasing, bullying, rumors/gossip, and arguments. Skills were taught through didactic instruction using concrete rules and steps of social etiquette in conjunction with role-playing exercises.  Teens practiced newly learned skills during behavioral rehearsal exercises and parent-assisted weekly socialization homework assignments.

Results:

Outcome data reveal that, in comparison to a delayed treatment control condition, teens in the treatment condition significantly improved in their knowledge of social skills (p < .0001), showed an increase in teen-reported hosted get-togethers (p < .05) and parent-reported hosted get-togethers (p < .05), and demonstrated improvement in overall social skills (p < .005), according to parent reports on a standardized measure of social functioning (SSRS). 

Data further suggests that treatment gains were maintained at a three-month follow-up assessment.  Of the 12 participants with follow-up data, 58.3% reported increases in get-togethers maintained at follow-up (p < .05). Correlations between teen and parent reports were significant for hosted (p < .05) and invited get-togethers (p < .01).  

Age and baseline parent reported social skills predicted treatment response.  Mean age was 14.4 years (SD = 0.5) for responders and 15.8 years (SD = 0.8) for non-responders (t₁₁ = 3.49, p < .01). Mean baseline social skills score was 42.0 (SD = 7.4) for responders and 32.0 (SD = 5.2) for non-responders (t₁₁ = 2.58, p < .05). Other demographic variables including gender, ethnicity, school placement, IQ, and adaptive functioning did not predict treatment success.

Conclusions:

This study represents one of the largest of its kind in the treatment literature for adolescents with ASD.  Findings suggest that the use of PEERS, a parent-assisted manualized social skills intervention, is efficacious in improving the social competence and friendship skills of teens with ASD.   Treatment gains for hosted and invited get-togethers with friends were maintained at a three-month follow-up.  Predictors of treatment success include age and social skills ratings by parents at baseline.