Teaching Playground Staff at Schools to Improve Peer Engagement for Children with Autism Spectrum Disorders

Background: Children diagnosed with autism spectrum disorder (ASD) often experience difficulties engaging with typically developing peers in the general education setting. Well-meaning school personnel can further complicate outcomes by following intuitive intervention strategies targeting a specific child when, at times, simple strategies providing all children with a better social environment at recess could produce better results for children with ASD.

Objectives: As part of a continuing effort to help improve peer engagement for children with ASD, we aimed to test a new, more naturalistic approach to intervention within the general education setting.  We wanted to show that school personnel could easily improve key elements of the social lives of children with ASD during recess.

Methods: Using a randomized, wait-list-controlled design, a psychosocial intervention was taught to 35 school personnel and delivered to children with ASD during recess at four elementary schools.  The immediate treatment (IT) group consisted of 13 (two female) elementary school students with ASD and the wait-list (WL) group contained 11 (four female) students with ASD. The intervention aimed to 1) increase the ability of playground aides to facilitate peer engagement, and 2) increase joint engagement with peers for children with ASD.

Results:  Responsive behavior scores by the adults showed significant improvement over time for the IT group compared to the WL group, (F(1,173)=6.83, p=0.01).  The treatment effect was not maintained as the IT group did not make significant gains by follow-up compared to its entry (F(1,55)=6.76, p=0.99).

The percentage of strategic behavior use scores were not significantly different between the IT and WL groups although there was a trend for the immediate treatment group to show more of these behaviors (F(1,173)=3.09, p=0.08).  The percentage of strategic behaviors usedby group were also not significantly different by follow-up compared to entry (F(1,55)=0.1, p=0.75)).

Some of the adults were one-on-one aides (n=5), and some were assigned to supervise the entire group of children on their respective playgrounds (n= 30).  At entry one-on-one aides displayed significantly lower responsive and strategic behaviors than regular playground staff; responsive behaviors (F(1,173)=3.55; p=0.06) and strategic behaviors (F(1,173)=9.51; p=0.002).  Both one-on-one and playground staff increased strategic behaviors from entry to exit (F(1,173)=15.44; p=0.001).  The rate of improvements were significant across groups (F(1,173)=5.46; p=0.02) with one-on-one staff having better improvements from entry to exit.  Also both one-on-one and playground staff increased responsive behaviors from entry to exit (F(1,173)=8.9; p=0.003).  The rate of improvements were not significant across groups for responsive behaviors (F(1,173)=1,09; p=0.30).

There was a significant treatment by time effect with children in the IT groups rated as more engaged with peers on the playground than children in the WL groups from entry to end of treatment (F(1,108)=10.68, p=0.002). In a separate analysis, the effect of treatment for the IT groups was maintained at the follow up (F(1,35)=6.76, p=0.014).

Conclusions:  Results suggest that a low dose, brief intervention delivered by playground staff can be beneficial in jumpstarting peer engagement for children with autism in inclusive settings, but ongoing support of playground staff is likely needed.