Abstract

Major neurocognitive disorder (MND) is a significant public health concern and a leading cause of disability and dependency among older adults. It is characterized by a chronic and acquired decline in cognitive functioning. For early detection, screening tools such as the Mini-Addenbrooke's Cognitive Examination (M-ACE)-a brief version of the Addenbrooke's Cognitive Examination-III (ACE-III)-have been proposed. This study analyzed data from 1164 adult participants from Antofagasta, Chile, including both the general population and individuals diagnosed with MND. Analyses included reliability estimation, factorial structure assessment using Confirmatory Factor Analysis (CFA), Item Response Theory (IRT) modeling, and diagnostic utility evaluation. Results indicate excellent reliability (McDonald's Omega = 0.926), a robust unidimensional factor structure, and good model fit under IRT. Differential item functioning (DIF) analysis revealed that performance on the clock-drawing item varied significantly according to educational level. Regarding diagnostic accuracy (AUC = 0.924), a cutoff score of <= 25/30 was proposed for the full sample, yielding 98% sensitivity. Consequently, alternative cutoffs stratified by educational level were also established to optimize performance. Findings support the M-ACE as a valid and reliable tool for cognitive screening and initial assessment of cognitive impairment in adults. The relevance of the clock-drawing task for assessing visuospatial abilities is also discussed.