Background & aims: Longevity also carries its dark side of age-related chronic diseases, dementia being one of the worst and the most prevalent. Since dementia lacks effective treatments, preventing or delaying it is highly desirable. Dietary habits and nutrition have been found to be important modifiable risk factors for many chronic diseases, but evidence on the role of diet on the risk of dementia is still limited, particularly among the very old. Aim of the present work is to study the association of the Mediterranean diet and its components with prevalent and incident dementia in the oldest-old. Methods: We analyzed data from the Monzino 80-plus study, a population-based study in subjects 80 years or older in the Varese province, Italy. A validated food frequency questionnaire was used to collect information on 23 different foods consumed in the previous year. A Mediterranean diet score was calculated and its components were classified into tertiles. Multivariable models for dementia prevalence and incidence were adjusted for demographic and clinical characteristics.
Results: Information on nutrition was available for 1390 subjects in the cross-sectional study and 512 subjects in the longitudinal study, mean respective ages 93 and 92. Greater adherence to Mediterranean diet, greater consumption of eggs, fruits and vegetables, carbohydrates, and greater food intake were associated with a lower prevalence of dementia. Increasing number of portions per week and con- sumption of legumes significantly decreased the incidence of dementia during the 3.6 year mean follow- up: corresponding hazard ratios of highest vs. lowest tertiles (95% confidence intervals) were 0.66 (0.46 e0.95) and 0.68 (0.47e0.97), respectively.
Conclusion: Oldest-old eating less and having diets with less variety and nutrient density were more frequent among subjects with dementia. The longitudinal analysis confirmed oldest-old subjects who eat more portions, as well as those who have a higher intake of legumes, are at decreased risk of developing dementia even though reverse causality cannot be completely ruled out.
