Long-term dietary flavonoid intake and risk of Alzheimer disease and related dementias in the Framingham Offspring Cohort

Background: Findings from existing prospective observational studies on the protective associations of flavonoid intake and the risk of Alzheimer disease and related dementias (ADRD) are inconsistent largely due to limitations of these studies. Objectives: To examine the prospective relation between total and 6 classes of dietary flavonoid intake and risk of ADRD and Alzheimer disease (AD) while addressing limitations of earlier observational studies. Methods: We used data from the FraminghamHeart Study Offspring Cohort exams 5 through 9. Participants were ADRD-free with a valid FFQ at baseline. Flavonoid intakes were updated at each exam to represent the cumulative average intake across the 5 exams, and were expressed as percentile categories of intake (<= 15th, >15th to 30th, >30th to 60th, >60th) to handle their nonlinear relation with ADRD and AD. Cox proportional hazards regression was used to estimate the HRs for the association between the flavonoid intakes and incidence of ADRD and AD. Results: Over an average follow-up of 19.7 y in 2801 participants (mean baseline age = 59.1 y; 52% females), there were 193 ADRD events of which 158 were AD. After multivariate and dietary adjustments, individuals with the highest (>60th percentile) intakes of flavonols, anthocyanins, and flavonoid polymers had a lower risk of ADRD relative to individuals with the lowest intakes (<= 15th percentile), with HRs (95% CI; P-trend) of 0.54 (0.32, 0.90; P = 0.003) for flavonols, 0.24 (0.15, 0.39; P < 0.001) for anthocyanins, and 0.58 (0.35, 0.94; P = 0.03) for flavonoid polymers. The same pattern of associations was seen with AD for flavonols and anthocyanins but not for flavonoid polymers. Conclusions: Our findings imply that higher long-term dietary intakes of flavonoids are associated with lower risks of ADRD and AD in US adults.