Changes in Parental Involvement and Behavior during a Parent-Mediated Intervention for Toddlers with Autism

Background: An important focus of behavioral interventions for young children with ASD is involving parents in the treatment process.  Even in high intensity treatments, children with ASD are only exposed to the treatment a small portion of the overall day. By training parents in specific intervention techniques, we can increase the dose of early intervention that children with ASD receive throughout the day. While parents are believed to play a critical role in the success of these interventions, few studies have systematically examined the relationship between level of parental involvement and change in observed parental behaviors.

Objectives: The goal of this study was to first determine if intervention approach influences the parent’s level of involvement, as well as their implementation of specific behavioral strategies in interactions with their toddlers (in this case based on JASPER, an experimental treatment that has evidence of positive child outcomes including increases in joint engagement, number of child initiated play routines, and play skills (Kasari, Gulsrud, & Hellemann, in prep)). Additionally we assessed moderating effects of parental involvement on the implementation of these strategies.

Methods: Eighty-six parent-child dyads participated in a randomized controlled early intervention study. Participants were randomized to either a parent-mediated JASPER condition or an individualized psychoeducational intervention (PEI) targeting parental stress-reduction (Brereton and Tongue, 2006).  Families met one-on-one weekly with a trained therapist for 10 weeks. Children had an average chronological age of 31.5 months and mothers were on average 35.9 years old. The sample consisted of mostly Caucasian males with an average IQ of 68.

Results: This study highlights the efficacy of the parent-mediated JASPER intervention approach for parents of toddlers with ASD.  Increases in therapist rated parental involvement (CIS) were larger in the JASPER group (F(1,69)=9.85, p<.01, Pre:14.6, Post:17.8)) than in the psychosocial intervention group (Pre: 15.3, Post: 15.9). Similarly, videotaped dyadic interactions, rated by coders blind to timepoint (pre or post) and randomization status, showed differences for the observed  usage of environmental strategies (F(1,81)=18.3, p<.01, JASPER: .71 to .89, PEI: .70 to .73), prompting (F(1,81)=12.1, p<.01, JASPER: .48 to .69, PEI: .47 to .50), communication (F(1,81)=14.4, p<.01, JASPER:.49 to .72, PEI: .47 to 51) and imitation (F(1,81)=10.6, p<.01, JASPER: .32 to .67;  PEI (.30 to .39). While there were no significant relationships between strategy use and involvement in the PEI group (all correlations <.15, all p-levels >.20), there was evidence suggesting that parental involvement in the JASPER group was specifically associated with the use of communication strategies (r=.41, p=.01) and marginally associated with the use of environment strategies (r=.29, p=.08).  Methods for conducting these analyses will be reviewed.

Conclusions: These data suggest that a brief parent-mediated intervention can improve therapist-rated parental involvement and independently coded use of specific JASPER intervention strategies. In addition, this study suggests that maintaining and creating parental involvement is an important facet of treatment design, suggesting that child-centered interventions may also need to pay increased attention to parent involvement.