A systematic review on age-dependent screening scales for autism spectrum disorders

Early screening of autism spectrum disorder (ASD) is critical, but the psychometric validation of existing screening tools is inconsistent and often incomplete. This systematic review aimed to identify ASD screening tools by age group, and to synthesize the evidence on their psychometric properties. Following PRISMA guidelines, a comprehensive search was conducted in PubMed, Scopus, and Web of Science up to June 2024. In addition, Google scholar was checked. Studies psychometrically validating ASD screening tools were included. The methodological quality of each study was rigorously assessed using the COSMIN checklist, and data were extracted systematically. Out of 6146 initial records, 12 distinct screening tools were identified and categorized into three groups based on the target age for diagnosis: (1) up to 24 months (e.g., CHAT-23), (2) 24–30 months (e.g., M-CHAT-R/F), and (3) over 30 months (e.g., BISCUIT). The COSMIN-based appraisal revealed a skewed focus in the literature. Most studies emphasized content validity (reported in 87.5% of instruments) and hypothesis testing (62.5%), while critical psychometric properties were frequently overlooked or inadequately reported: measurement error (assessed in only 6.25% of instruments), test–retest reliability (12.5%), cross-cultural validity (18.75%), internal consistency (18.75%), and interpretability and responsiveness (0%). None of the included studies employed more comprehensive approaches, such as Item Response Theory (IRT). Current evidence reveals a predominant focus on content validity in ASD tool validation, with critical gaps in assessing reliability, measurement error, and cross-cultural applicability. These limitations constrain evidence-based tool selection and generalizability. Future research must adopt comprehensive, standardized validation frameworks, such as COSMIN, to ensure robust and equitable diagnostic practices across diverse populations.