Development of the Screening Tool for Autism in Two-Year-Olds-Brief Form: Accuracy in Detecting Autism in Toddlers.

Abstract

Early diagnosis of autism is crucial for timely intervention, yet existing screening tools are often limited by administration time in clinical settings. This study developed and evaluated the Screening Tool for Autism in Two-Year-Olds-Brief Form (STAT-BF) for detecting autism in toddlers aged less than 36 months. In Study 1, two independent samples of 40 toddlers each (20 autistic and 20 with developmental delays [DDs]) were analyzed using chi-square tests to identify discriminative items. Five items (Turn-taking, Bubbles, Snack, Balloon, and Bag of Toys) were retained, with a cutoff score of 3 yielding optimal sensitivity and specificity. Study 2's validation, based on a sample of 251 toddlers (81 autism, 55 mild- autism, 115 DDs), yielded high sensitivity for autism (0.91; 74/81) and mild-autism (0.91; 50/55), together with good specificity (0.81; 93/115). Administration time was reduced to 5 to 10 min while maintaining screening accuracy. These findings suggest that the STAT-BF may serve as a feasible and efficient Level 2 screening tool for early autism detection among toddlers referred for developmental concerns, particularly in resource-limited clinical settings. Further validation in diverse populations and longitudinal follow-up is warranted to establish generalizability.Lay AbstractEarly support can make a big difference in the outcomes of autistic children, but they must first be identified as early as possible. Although reliable autism screening tools exist, many are too time-consuming for high-volume clinical settings. In this study, we developed and validated a shorter version of the Screening Tool for Autism in Two-Year-Olds (STAT) that takes 5 to 10 min. This brief tool focuses on five play-based activities that assess social communication skills in toddlers aged 18 to 36 months. We conducted a two-phase study in Taiwan. The first phase identified the most discriminating items, and the validation phase included 251 toddlers with autism, mild-autism, and developmental delays (DDs). Results showed that the brief form correctly identified over 90% of autistic children, including those with milder symptoms, and correctly ruled out autism in about 80% of children with DDs. This efficient screening method could help more health care providers detect autism early, especially in high-volume clinical settings with limited resources. Because the brief form uses structured play instead of parent questionnaires, it offers direct behavioral observation and increases cultural accessibility for families from a broad range of backgrounds. This brief screening tool could help clinicians more efficiently identify autism among toddlers already referred for developmental evaluation, facilitating timely referral for formal diagnostic confirmation and subsequent access to intervention services.