RESUMEN
BACKGROUND AND PURPOSE: Mild cognitive impairment (MCI) is a heterogenous syndrome considered as a prodromal state of dementia with clinical importance in the early detection of Alzheimer's Disease. We are currently developing an MCI screening instrument, the Early Mental Test (EMT) suitable to the needs of primary care physicians. The present study describes the validation process of the 6.2 version of the test. METHODS: Only subjects (n = 132, female 95, male 37) over the age of 55 (mean age 69.2 years (SD = 6.59)) scoring at least 20 points on Mini-Mental State Examination (MMSE), mean education 11.17 years (SD = 3.86) were included in the study. The psychometric evaluation consisted of Alzheimer's Disease Assessment Scale Cognitive subscale (ADAS-Cog) and the 6.2 version of EMT. The statistical analyses were carried out using the 17.00 version of SPSS statistical package. RESULTS: The optimalised cut-off point was found to be 3.45 points with corresponding 69% sensitivity, 69% specificity and 69% accuracy measures. The Cronbach-alpha, that describes the internal consistence of the test was 0.667, which is higher as compared with the same category in the case of the ADAS-Cog (0.446). A weak negative rank correlation was found between the total score of EMT 6.2 and the age of probands (rs = -0.25, p = 0.003). Similarly, only a weak correlation was found between the education levels and the total score of EMT 6.2 (rs = 0.31, p < 0.001). Two of the subtests, the repeated delayed short-time memory and the letter fluency test with a motorical distraction task had significantly better power to separate MCI and control groups than the other subtests of the EMT. CONCLUSION: The 6.2 version of EMT is a fast and simple detector of MCI with a similar sensitivity-specificity profile to the MMSE, but this version of the test definitely needs further development.