Child Behavior Check List 1½-5 as a tool to identify toddlers with autism spectrum disorders: a case-control study.

ABSTRACT

Tools to identify toddlers with autism in clinical settings have been recently developed. This study evaluated the sensitivity and specificity of the Child Behavior Check List 1½-5 (CBCL 1½-5) in the detection of toddlers subsequently diagnosed with an Autism Spectrum Disorder (ASD), ages 18-36 months. The CBCL of 47 children with ASD were compared to the CBCL of 47 toddlers with Other Psychiatric Disorders (OPD) as well as the CBCL of 47 toddlers with Typical Development (TD) in a case control study. One-way analysis of variance (ANOVA) and logistic regression with odds ratio (OR) analyses were performed. In order to establish the optimal threshold able to discriminate children with ASD from children with OPD and TD, Receiver Operating Characteristic (ROC) analyses were performed. One-way ANOVA revealed significant differences between the three groups. Logistic regression analysis showed that the Withdrawn and the Pervasive Developmental Problems (PDP) subscales can recognize toddlers subsequently identified as ASD from both children with TD (p<0.001) and OPD (p<0.001). ROC analyses showed very high sensitivity and specificity for the PDP (0.98 and 0.91) and Withdrawn (0.92 and 0.97) subscales when ASD was compared to TD. Sensitivity and specificity of Withdrawn (0.90 and 0.83) and PDP (0.85 and 0.83) remained high when comparing ASD versus OPD. In conclusion, the CBCL 1½-5 seemed to be able to identify toddlers subsequently diagnosed with ASD from children with TD and OPD. Its high sensitivity and specificity, coupled with its efficiency in terms of time and cost, suggest this broadband tool should be tested in a pilot screening survey of toddlers in the general population.