Diagnosing autism: comparison of the childhood autism rating scale (CARS) and the autism diagnostic observation schedule (ADOS)

Abstract

This project sought to determine the utility in using either, or both, of two instruments to diagnose autism, the CARS (Childhood Autism Rating Scale) and the ADOS (Autism Diagnostic Observation Schedule), Modules 1 or 2. Children (n=320) who were seen in the autism diagnostic clinics at the Developmental Disabilities Center of the Kansas University Medical Center, who were under the age of 72 months (6 years), and who had been evaluated with both instruments were chosen as participants in this study. Those children who received the diagnosis of autism after being evaluated numbered 220; 100 received another or no diagnosis. Three levels of data analysis were conducted in this study. The first level included developing item-item correlation matrices for each instrument that was then compared to those in the original, normed study for internal consistency. Results indicated good internal consistency. At the second level, a factor analysis was conducted on each instrument that resulted in weighted factor scores and a correlation matrix of factors for each instrument. Factor analyses resulted in three factors identified for the CARS, two factors for ADOS, Module 1, and three factors for ADOS, Module 2. These factors are consistent with the criteria currently used for the diagnosis of autism. The third level of analysis utilized chi square and stepwise discriminant analysis to predict group membership (autism or no autism) with each instrument. From these results, it was concluded that both instruments are similar in their ability to diagnose autism, although they may be measuring somewhat different factors. Closer examination revealed that communication difficulty is the factor that most closely distinguished autism for this group of children.