Abstract
Tocopherols, strong antioxidants, may be useful in preventing dementia, but the epidemiological evidence is insufficient. We performed a community-based follow-up study of Japanese, the Circulatory Risk in Community Study, involving 3739 people aged 40-64 years at baseline (1985-1999). Incident disabling dementia was followed up from 1999 through 2020. For subtype analysis, we classified disabling dementia into that with and that without a history of stroke. Dietary intake of tocopherols (total, α, β, γ, and δ) were estimated using 24-h recall surveys. During a median follow-up of 19.7 years, 670 cases of disabling dementia developed. Total tocopherol intake was inversely associated with risk of disabling dementia with multivariable hazard ratios (95% confidence intervals) of 0.79 (0.63-1.00) for the highest versus lowest quartiles of total tocopherol intake (P for trend = 0.05). However, the association was strengthened when further adjusted for α-linolenic acid intake (Spearman correlation with total tocopherol intake = 0.93), with multivariable hazard ratios of 0.50 (0.34-0.74) (P for trend = 0.001) but was weakened and nonsignificant when further adjusted for linoleic acid intake (Spearman correlation with total tocopherol intake = 0.92), with multivariable hazard ratios of 0.69 (0.47-1.01) (P for trend = 0.05). Similar but nonsignificant inverse associations were observed for α-, γ-, and δ-tocopherols but not for β-tocopherol. These results were similar regardless of the presence of a history of stroke. Dietary tocopherol intake was inversely associated with risk of disabling dementia, but its independent effect was uncertain owing to a high intercorrelation of α-linolenic linoleic acids with total tocopherol intake. Even with such confounding, a diet high in tocopherols may help prevent the onset of dementia.