Development of the Observation Schedule for Children with Autism Spectrum Disorders– Anxiety and Behaviour (OSCA-AB): A New Measure of Parent and Child Behaviour for Use with Children with ASD

Background: Most young children diagnosed with Autism Spectrum Disorders (ASD) display concerning and potentially problematic behaviours and emotions. Emerging evidence suggests that parent training interventions may reduce these difficulties in young children with ASD (Postorino et al., 2017). Trials of parent training frequently use parent-report measures as their primary outcome; however, as parents are the recipients of the intervention they are not blind to treatment status, which may lead to bias is estimating treatment effects (Sonuga-Barke et al., 2013).

Objectives: We aimed to develop an objective outcome measure to allow blinded evaluations of interventions for parents of young children with ASD.

Methods: The Observation Schedule for Children with Autism Spectrum Disorders – Anxiety and Behaviour (OSCA-AB) is a measure of parent and child behaviour for use with children with ASD. It draws on existing well-validated observational measures of parent-child interaction. A range of researcher-led and parent-led tasks are completed. Tasks were modified to simulate everyday challenges that children with ASD may face and find difficult. They aim to elicit concerning behaviour and emotions by tapping into uncertainty and novelty, transition, turn taking, sensory processing and compliance. Observed frequencies of parent behaviour (positive and negative comments, clear commands, physical handling, scaffolding) and concerning child behaviours (non-compliance, destructive behaviour, aggression to self and others, avoidance) are coded. In addition, global codes are scored on a Likert scale. Intraclass correlation coefficients (ICCs) were used to establish inter-rater reliability (IRR) using a two way mixed effects model. A more modest ICC of >.60 was selected given that the measure is in development. Pearson correlation coefficients were calculated to investigate the associations between OSCA-AB scores and parent-reports of parent (Parenting Scale - PS) and child behaviour (Aberrant Behaviour Checklist - ABC, Home Situations Questionnaire-ASD - HSQ-ASD).

Results: To date, the OSCA-AB shows adequate IRR for scores on most child variables (ICCs >.60 for 10/15, Mdn=.69; n=18) and some parent variables (ICCs >.60 for 4/10, Mdn=.57). High correlations between the OSCA-AB scores and other validated measures of child behaviour are also evident (r=.64 and .61 for the ABC-Hyperactivity subscale and the HSQ-ASD, respectively). However, correlations between observed parent behaviours and parent-reports of their own parenting on the PS are low (r=.01-.07). On average, a total of 26 concerning child behaviours are displayed within the 23 minute observation. By May 2018 data from 50 families will be presented following further refinement of the measure and exploration of sensitivity to change over the course of the intervention.

Conclusions: Preliminary results indicate that the OSCA-AB may be a useful objective measure for eliciting concerning behaviour and emotions and parent-child interaction in ASD. Refinement is required to increase IRR for some child and parent variables. Further research is needed to establish whether the OSCA-AB will be a valid measure of parent and child behaviour among samples of young children with ASD and is sensitive to change following intervention.