P4-194: Variables associated with cognitive impairment-no dementia in a low-educated cohort aged 75+ years: The Pietà study

Data on cognitive impairment-no dementia (CIND) from developing countries, especially among low-educated and more elderly subjects, are still scarce. The goal of the study was to investigate demographic, personal health and clinical variables associated with the diagnosis of CIND in a population-based cohort of elderly (75+ years) with low education. We conducted a population-based study of cognitive and neuropsychiatric disorders in Caeté, in southeastern Brazil. A total of 639 participants (64% women, aged 81.4 ± 5.2 years and with 2.7 ± 2.6 years of schooling; 27.6% illiterates) were submitted to a thorough personal and health questionnaire, clinical, cognitive, psychiatric and functional evaluations. Individuals with suspected cognitive impairment and a subset of cognitively healthy individuals were also submitted to a comprehensive neuropsychological and functional examination. The sample was randomly selected and represented 51.1% of total population aged 75+ years living in the town. Diagnosis of CIND was ascertained through an expert consensus panel, based on published criteria, taking into account age and education. Univariate and multivariate statistical analysis explored associations between demographic variables, personal history, clinical data and CIND diagnosis. CIND was diagnosed in 161 individuals (crude prevalence = 25.2%). Ages 80–84 (OR = 2.36; 95% CI 1.49- 3.77) and 85–89 (OR = 1.98; 95% CI 1.07 - 3.63), low socioeconomic level (OR = 1.55; 95% CI 1.02 - 2.38), depression (OR = 1.65; 95% CI 1.00–2.73) and thyroid dysfunction (OR = 2.00; 95% CI 1.03 - 3.92) were associated with CIND diagnosis. In this low-educated population-based elderly cohort, advanced age, low socioeconomic level, depression and thyroid dysfunction were all significantly associated with CIND diagnosis.