Utility of the Taiwan version of the screening tool for autism in two-year-olds to detect autism in children aged three years

Abstract

Background: The Taiwan version of the Screening Tool for Autism in Two-Year-Olds (T-STAT) had been developed for identifying autism disorder (AD) in children aged 24–35 months. Two studies were performed to elucidate its utility for AD in children aged 36–48 months. Method: In Study 1, the receiver operating characteristics were utilised to select a cutoff score for the T-STAT with 34 children with AD and with developmental delay (DD). In study 2, 147 children with AD, DD, and pervasive developmental disorder-not otherwise specified (PDD-NOS) were participated for validation. Results: In study 1, the results indicated that a cutoff of 1.25 would yield high sensitivity and specificity. In study 2, the concurrent agreements between T-STAT risk and clinical diagnosis were high for children with AD and DD, but not PDD-NOS. Conclusions: The results showed that T-STAT was a promising Level 2 screening tool for AD in children aged three years.