Objective: To determine which factors, especially those related to nutrition, are associated with cognitive function in the oldest old, here considered those at least 80 years of age.
Design: A cross-sectional, population-based study.
Setting: Veranópolis, Rio Grande do Sul, Brazil, and surrounding rural areas.
Participants: Individuals aged 80 years and older.
Measurements: The Mini Nutritional Assessment, anthropometric measurements, and serum levels of albumin and vitamin B12 were associated with cognitive function according to the Mini Mental State Examination (MMSE) and the Clock Drawing Test (CDT). Socio-demographic data were obtained through a structured questionnaire. Multivariate analysis was used to determine the associations.
Results: According to the MMSE and the CDT, the prevalence of cognitive impairment was 47.7% [95%CI 39.7-55.7] and 58.2% [95%CI 50.3-66.1], respectively. In the adjusted analysis, the only positive linear association with MMSE scores indicating cognitive impairment was age. However, CDT scores indicating cognitive impairment were five times higher among those with low serum vitamin B12 concentrations. For the other variables, there was a positive association between age, being widowed, a low educational level and central nervous system drugs. Being single, living with children and living alone were protective factors for cognitive impairment.
Conclusions: Although cognitive impairment was positively associated with old age, being widowed and low educational level in this population, the only nutritional variable positively associated with cognitive impairment was a low vitamin B12 concentration.
Keywords: Cognitive aging; nutritional status; older adults; vitamin B12.