VALIDITY AND RELIABILITY OF THE 10/66 DEMENTIA RESEARCH GROUP COGNITIVE TEST BATTERY FOR EARLY IDENTIFICATION OF COGNITIVE DECLINE IN MEXICO

Abstract

Several neuropsychological tests have been used to assess cognitive decline across countries, nevertheless, few countries have had mind about external validity and reliability of those assessment outcomes. It is well known that there are cross-cultural differences that might be affect those results, regarding on measuring irrelevant variance. In that sense, we proposed to identify psychometrical and statistical properties of 10/66 Dementia Research Group cognitive test battery, in order to empirically confirm the utility of the instrument for the early identification of cognitive decline in the elderly population in Mexico. Secondary analysis of the basal evaluation of the Dementia Research Study 10/66 in Mexico (n=2003). The 10/66 Dementia Research Group cognitive test battery included the administration of the Community Screening Instrument for Dementia (CSI ‘D’), the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) verbal fluency (VF), word list memory (WLM, immediate recall) and recall (WLR, delayed recall) tests. The psychometric analysis was implemented to estimate the 2-parameter model for the dichotomous and polytomous items response sets (Baker, 2001) difficulty and discrimination parameters. In addition we performed a Confirmatory Factor Analysis (Abell, Springer y Kamata, 2009) to confirm factor estability and validity. It was observed that the average value of discrimination of the instrument was 1.0057, which is considered as a high power of discrimination. On the other hand, the estimated average difficulty index was -2.1288, which suggests that the neuropsychological battery can effectively differentiate between subjects with very low and average levels. The instrument had a Cronbach alpha of 0.708 which suggests the consistency between the measurements obtained. When the construct validity was performed, it was observed that the items converge positively on the factor with values greater than 0.30 factorial load. What indicates that the instrument has validity. The results showed that the items are able to discriminate effectively between subjects with low neuropsychological levels and in average values. Likewise, the factorial structure was validated, favoring the internal and external validity of the construct. These evidences could help to identify subjects with early cognitive decline.