Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental disorder whose symptoms may involve deficits across three domains: communication, socialization and atypical behaviors or interests. With a high prevalence across populations and a tendency to impact males more than females, early and accurate diagnosis appears critical. The most current literature on ASD provides a myriad of difficulties associated with diagnosis under the age of 3 years. The purpose of this study was to determine if a predictive relationship exists between a child's individual developmental domain standard deviation (SD) subscale scores (motor, language, cognitive, social-emotional and adaptive skills) on the Bayley-III assessment instrument and whether a diagnosis of ASD was applied. A retrospective file review of 151 children participating in Kentucky's early intervention program -- First Steps -- was completed. The children ranged in age from 18 to 35 months. Individual binary logistic regressions were used to assess the association between the Bayley-III subscale scores and whether or not an ASD diagnosis was applied following multidisciplinary evaluation. The results indicated that individual lower subscale scores in the cognitive, language, adaptive, and social-emotional domains on the Bayley-III were predictive of an autism diagnosis.