Background: The revalidation of the Mini Examen Cognoscitivo (MEC), first Spanish version (1978) of the Mini-Mental Status Examination (MMSE) and documentation of "population-based norms" should clarify the potential confusion induced by later versions of MMSE.
Context: The Zaragoza Study on the prevalence of dementia and depression in a representative sample of the elderly community (N = 1,080).
Instruments: MEC-35 and MEC-30 points, and validated, Spanish versions of Geriatric Mental State (GMS), History and Aetiology Schedule (HAS) and Social Status Schedule (SSS).
Procedure: a) validation of MEC (standardized lay interviewers) against the gold standard of psychiatric diagnosis (DSM-III-R), two months later; b) "population-based norms" in the "healthy" population, and c) comparison with other MMSE versions.
Results: The instrument fulfills criteria of "feasibility", "content", "procedural" and "construct validity". Test-retest reliability: weighted kappa = 0.637. MEC-30 (cut-off point 23/24), sensitivity = 89.8%, specificity = 75.1% (80.8% with the cut-off at 22/23), and ROC curve, AUC = 0.920. The coefficients of individual items were satisfactory and the specificity increases in MEC-35 (83.9%). Other MMSE Spanish versions have not improved these coefficients. "Population-based norms" confirm the hypothesized influence of age and education level. MEC-30 is the version with most comparable results with the MMSE in USA.
Conclusions: The validity of MEC is confirmed in the elderly population, with the same cut-off points recommended in the original standardization. MEC-30 is the best version for international comparisons.