The Psychometric Properties of Autism Mental State Examination (AMSE) in Turkish Sample

Abstract

Autism Spectrum Disorder (ASD) is a prevalent neurodevelopmental disorder, and early diagnosis plays a pivotal role in prognosis and management. This study aims to examine the psychometric properties of the Autism Mental Status Exam (AMSE), a tool that shows great promise in terms of clinical utility, within the Turkish population. This study conducted in a cohort of 307 Turkish children aged 17 to 120 months with suspected ASD. Participants underwent a multidisciplinary assessment based on DSM-5 criteria for diagnosis and were categorized into ASD and non-ASD groups. Subsequently, the research team conducted blinded administrations of the AMSE and Childhood Autism Rating Scale (CARS). Additionally, a subset of 61 children underwent retesting for AMSE and CARS after three weeks for temporal stability. The results revealed an optimal cut-off score of 4 for AMSE, yielding sensitivity and specificity rates of 84% and 97%, respectively. Internal consistency, indicated by a Cronbach's alpha of 0.80, was very good. The test-retest reliability, assessed using the Intraclass Correlation Coefficient (ICC), was excellent (ICC = 0.959). The inter-rater reliability also showed excellent agreement (ICC = 0.997). Furthermore, a significant correlation was observed between the AMSE and CARS scores (r = 0.94, p < 0.001). Notably, the AMSE scores were significantly different between the ASD and non-ASD groups (p < 0.001) with a large effect size (Cohen's d = 1.40). The findings of this study underscore the utility of AMSE as a valid and reliable tool for Turkish children with robust psychometric properties.