Cohort studies suggest that the Mediterranean diet is associated with better global cognition in older adults, slower cognitive decline and lower risk of dementia. However, little is known about the relative contribution of each component of the Mediterranean diet to dementia risk or whether the diet’s effects are due to one or more specific food components. We aimed to examine whether Mediterranean diet components are associated with all-cause dementia risk in the UK BioBank cohort. Participants joined the UK Biobank study from 2006 to 2010 and were followed until December 2020. 249,511 participants, who were at least 55 years old, without dementia at baseline were included. We used self-reported consumption of food groups, considered part of the Mediterranean diet including fruit, vegetables, processed meat, unprocessed red meat and unprocessed poultry, fish, cheese, wholegrains. Incident dementia was ascertained through electronic linkage to primary care records, hospital and mortality records or self-report. In this study with a total follow-up of 2,868,824 person-years (median 11.4), after adjusting for all covariates and other food groups, moderate fish consumption of between 2.0 and 3.9 times a week was associated with decreased risk of dementia (HR 0.84, 95%CI 0.71–0.98) compared to no consumption. Additionally, fruit consumption of between 1.0 and 1.9 servings a day was associated with reduced dementia risk (HR 0.85, 95%CI 0.74–0.99) compared to no consumption. No other Mediterranean diet components were associated with dementia risk suggesting that fish consumption may drive the beneficial effects seen from the Mediterranean diet. Further study of potential mechanisms and diet-based intervention trials are needed to establish this.

队列研究表明，地中海饮食与老年人更好的整体认知、更慢的认知衰退和更低的痴呆风险有关。然而，对于地中海饮食中每种成分对痴呆症风险的相对贡献，或者这种饮食的影响是否是由一种或多种特定食物成分引起的，人们知之甚少。我们的目的是检查地中海饮食成分是否与英国生物银行队列中的全因痴呆症风险相关。参与者于 2006 年至 2010 年加入了英国生物银行研究，并被随访至 2020 年 12 月。包括 249,511 名参与者，他们至少 55 岁，基线时没有痴呆症。我们使用自我报告的食物组消费量，被认为是地中海饮食的一部分，包括水果、蔬菜、加工肉类、未加工的红肉和未加工的家禽、鱼、奶酪、全麦。通过与初级保健记录、医院和死亡率记录或自我报告的电子链接来确定事件痴呆症。在这项总随访 2,868,824 人年（中位数 11.4）的研究中，在针对所有协变量和其他食物组进行调整后，每周适度食用 2.0 至 3.9 次鱼与痴呆症风险降低相关（HR 0.84 , 95%CI 0.71–0.98) 与无消费相比。此外，与不食用水果相比，每天食用 1.0 至 1.9 份水果可降低痴呆风险（HR 0.85，95%CI 0.74-0.99）。没有其他地中海饮食成分与痴呆症风险相关，这表明食用鱼类可能会带来地中海饮食带来的有益影响。