Validation of the Autism Mental Status Exam in Spanish: sensitivity and specificity in an Argentine clinical sample

Abstract

Introduction: The Autism Mental State Exam (AMSE) is a brief clinician-completed instrument that structures the observation and documentation of the social, communicative, and behavioral signs and characteristics of autism. In its original English version, it demonstrated high accuracy in identifying autism spectrum disorder (ASD). This study explores the sensitivity and specificity of a Spanish version of AMSE in an Argentine sample of 313 subjects, against DSM-5 clinical consensus diagnosis as the primary outcome. Materials and methods: The cutoff values were calculated using the receiver operating characteristic (ROC) curve, identifying the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The internal homogeneity of the items was determined using Cronbach's alpha, and the Cohen Kappa coefficient was calculated for inter-rater reliability. Results: Findings indicate an optimized sensitivity of 90.71% and a specificity of 92.17% for this sample of 313 patients. AMSE exhibited fair internal consistency (alpha coefficient of .75) and high inter-rater reliability (.97) in this sample of subjects evaluated for ASD in Argentina. Discussion & conclusions: AMSE is emerging as a promising tool for diagnostic assessment of ASD in children, adolescents and adults at risk, standing out for its high clinical utility. Its application is particularly relevant in regions like Latin America, where access to the Autism Diagnostic Observation Schedule training, materials and use is limited. These results support the use of AMSE as a brief, reliable, and culturally adapted diagnostic tool to support the clinical diagnosis of autism in Spanish-speaking contexts with limited resources.