Development of simple cognitive function measures in a community dwelling population of elderly in Spain

Objective. To develop and assess the consistency and validity of simple cog nitive function measures for an elderly population with low levels of forma l education for use in a longitudinal study of dementia. Methods. Data were from the population longitudinal study 'Growing old in L eganes' (Spain). Tn 1993, a random sample of 1540 people over 65 was drawn from the City Roll of Leganes from which 1284 (83%) were successfully inter viewed. Measures of memory and orientation were based on the SPMSQ (Short P ortable Mental Status Questionnaire), the Barcelona test and the short stor y from EPESE (Established populations Epidemiologic Studies of the Elderly) . Non-response to a test item was coded as an error. internal consistency w as assessed by factor analysis and Cronbach's alpha. Construct validity was examined with multiple linear regressions of the proposed measurements on variables chosen from the existing literature on cognitive function. Results. Two factors, memory and orientation, emerged from the factor analy sis. Internal consistency of the proposed indexes for memory and orientatio n was acceptable. Memory and orientation scores were summed into one summar y index of cognition. Associations between covariates and both cognitive in dexes were in the expected direction. Among those highly functional, orient ation was influenced by illiteracy due to higher error rates in the time or ientation items based on dates: however, memory and summary scores were not significantly different by literacy status. A large proportion of the vari ance in IADL was explained by the memory and orientation measures. Conclusion. The memory and orientation indexes are valid and reliable measu rements of cognitive function for use in a population of community dwelling elderly with low levels of formal education and high rates of illiteracy, Copyright (C) 2000 John Wiley & Sons, Ltd.