Clinical utility of the standardized observation tool Autism Behavior Coding System for early intervention research in autism spectrum disorder.

The Autism Behavior Coding System (ABCS) was developed to help evaluating the effectiveness of early intensive interventions in children with autism spectrum disorder (ASD). The video-based ABCS assesses eight core autistic behavioral variables during therapist-child interaction using standardized quantitative criteria, four behaviors according to their frequency of occurrence, four according to their duration. The present study focuses (1) on the correspondence of ABCS scores with scores on two standard clinical instruments (the ADOS-2 and an ASD-adaptation of the Children's Global Assessment Scale, DD-CGAS), (2) on the sensitivity to change of ABCS scores by the end of an intensive 18 days intervention period (EIP) and (c) on the predictability of short- and longer-term changes in social and repetitive behaviors from ABCS scores at baseline and EIP. Data from 51 children (42 M, 9 F; median age 45 months) followed over 1 year were available. There were significant correlations at baseline between several ABCS scores and ADOS-2 as well as DD-CGAS scores. Correlations at EIP between some ABCS and DD-CGAS scores were highly significant. Four ABCS scores reflected significant changes from baseline to EIP. Several baseline ABCS scores were predictive of DD-CGAS and ADOS-2 scores at EIP and Year 1. However, associations between ABCS score changes from baseline to EIP and the clinical scale changes by Year 1 were not significant. It is concluded that several ABCS scores have adequate clinical validity and sensitivity to change. The short-term changes in ABCS scores and their relationship to longer-term clinical changes need further study.