Objective: To evaluate the properties of clinical scales of the Child Behavior Checklist in discriminating referred children with autism spectrum disorders (ASDs) (autistic disorder, Asperger's disorder, and pervasive developmental disorder not otherwise specified) from psychiatrically referred children without ASDs.
Method: Comparisons were made between children with ASDs (n = 65) with intelligence quotient >70 and children without ASDs (N = 83) on the clinical scales of the Child Behavior Checklist. Stepwise logistic regression was used to identify those scales that best predicted ASDs when compared with the non-ASD comparison group. Receiver operating characteristic curves examined the ability of the significant predictor T-scores to identify ASDs versus the non-ASD subjects.
Results: Withdrawn, Social Problems, and Thought Problems T-scores were the best independent predictors of ASD status. The Withdrawn + Social + Thought Problems T-scores yielded an area under the curve of 0.86, indicating an 86% chance that a randomly selected sample of ASD subject will have abnormal scores on these scales than a randomly selected sample of non-ASD subjects.
Conclusion: These findings suggest that a new Child Behavior Checklist-ASD profile consisting of the Child Behavior Checklist-Withdrawn, Social, and Thought Problems scales could serve as a rapid and cost-effective screening instrument to help identify cases likely to meet clinical criteria for ASDs in the clinical setting.