Abstract
Whether dietary and nutrition and dietary patterns are associated with the development of dementia is an interesting research question. Participants of a longitudinal cohort study that included European adults who were middle to old aged at baseline and who had not been diagnosed with dementia at baseline (2006-2010) and had not been diagnosed with dementia or died within 5 years after baseline were followed up (until 2018) and analyzed. Associations between intake frequency of each food class measured by the food-frequency questionnaire at baseline and incident dementia 5 years after baseline were analyzed after correcting for confounding variables. A total of approximately 340,000 participants and 900 cases were included in the analysis for each food class. Cox proportional hazard models with self-reported intake level of each food category divided into four mostly equally divided categorical variables revealed a high intake of bread, moderate total meat and total fish intake and low vegetable and fruit intake were thus associated with a small but significant decrease in the onset risk of dementia, while poultry and cereal were not. These findings are mostly inconsistent with the idea that Mediterranean diet is associated with lower risk of subsequent incident dementia.