This study examined the associations between overall diet quality and the risk of dementia in a rural cohort among the oldest old. Included in this prospective cohort study were 2232 participants aged ≥ 80 years and dementia-free at the baseline according to the Geisinger Rural Aging Study (GRAS), a longitudinal cohort in rural Pennsylvania. In 2009, diet quality was assessed by a validated dietary screening tool (DST). Incident cases of dementia during 2009-2021 were identified using diagnosis codes. This approach was validated by a review of electronic health records. Associations between diet quality scores and the incidence of dementia were estimated using the Cox proportional hazards models, adjusted for potential confounders. Across a mean of 6.90 years of follow-up, we identified 408 incident cases of all-cause dementia. Having a higher diet quality was not significantly associated with a lower risk for incidents of all-cause dementia (adjusted HR for the highest compared with the lowest tertile: 1.01, 95% CI: 0.79, 1.29, P-trend = 0.95). Similarly, we did not observe a significant association between diet quality and altered risks of Alzheimer's disease and other forms of dementia. Overall, having a higher diet quality was not significantly associated with a lower risk of dementia among the oldest old during the full follow-up.
Keywords: aging; dementia; diet quality; dietary pattern; oldest old; prospective study.