Understanding Treatment Response in a Pivotal Response Treatments-Based Program for Children with Autistic Spectrum Disorder

Background: Comprehensive treatments that start early and are based on the principles of Applied Behaviour Analysis (ABA) substantially improve outcomes in just under 50% of children with autistic spectrum disorders (ASD).  The remaining 50% of children display varying levels of response to treatment, ranging from very little or no change to modest gains.  Understanding this variable response to treatment in ASD requires examining the impact of a variety of important factors (e.g., child, family and environmental variables) on treatment outcomes.

Objectives: The overall objective of the current study is to examine specific child factors that may help to explain the variable response to treatment in a group of children with ASD enrolled in the early intensive behavioural intervention (EIBI) program in Nova Scotia, Canada. The program is based on Pivotal Response Treatment (PRT), and is supplemented by Positive Behavioural Support (PBS) practices. Key program components include parent training, one-to-one therapist-led intervention for up to 15 hours per week, and monitoring of treatment fidelity for both parents and therapists.

Methods: Data were collected for 29 children with ASD. The mean age for the total sample of children was 51.21 months (SD = 10.09) at the start of intervention.  Initial cognitive ability, as measured by the Merrill-Palmer-Revised Scales of Development (M-P-R) Cognitive scale standard score was 51.97 (SD = 27.81). Baseline video data were collected of the children interacting with trained therapists in play situations.  These videos were coded for the presence of six child variables that are empirically and/or theoretically linked to outcomes in PRT-based intervention (i.e., toy contact, avoidance, approach, stereotyped and repetitive vocalizations and non-verbal behaviours and affect). The degree to which children displayed these behaviours was examined in relation to language outcomes after 6 months of therapist- and parent-delivered PRT.

Results: Preliminary analysis (n = 6) examining child factors predictive of response to PRT indicated that children’s appropriate toy contact at baseline was significantly positively correlated with expressive language at 6 mo (r = .97, p < .006), controlling for initial language level. In addition, avoidance behaviour was negatively correlated with language, though not significantly in this small preliminary sample (r = -.41, ns). Additional data will be presented for the total sample (n = 29).

Conclusions: Pre-treatment levels of toy contact, avoidance, approach, stereotyped and repetitive vocalizations and non-verbal behaviours, and child affect may predict the variable response to PRT-based intervention in children with ASD. Ultimately, these variables may point to important initial intervention targets for children who do not display optimal response to PRT.