Public Health.

Background: Research on the relationship between diet and dementia among Koreans are lacking. This study investigated the association between dietary habits and dementia progression over 3 years in patients with Alzheimer's disease dementia (ADD).

Method: This study included 705 patients with mild-to-moderate ADD. Dietary habits were assessed using the Mini Dietary Assessment Index, comprising 10 questions. Outcome measures included the Clinical Dementia Rating scale-Sum of Boxes (CDR-SB), Seoul-Instrumental Activities of Daily Living, Caregiver-Administered Neuropsychiatric Inventory (CGA-NPI), and neuropsychological test battery (NTB) z-scores, which were evaluated annually over 3 years.

Result: In Q10 (eat all food evenly without being picky), the 3-year mean differences in CDR-SB (increases in scores represent worsening) compared to the "rarely" group were -1.86 (95% confidence interval [CI] = -3.64 ∼ -0.09, P = 0.039) for the "usually" group and -2.23 (95% CI = -4.40 ∼ -0.06, P = 0.044) for the "always" group. In Q7 (add salt or soy sauce to food, when eating), the 3-year mean differences in CDR-SB compared to the "always" group were -2.47 (95% CI = -4.70 ∼ -0.24, P = 0.030) for the "usually" group and -3.16 (95% CI = -5.36 ∼ -0.96, P = 0.005) for the "rarely" group. The "rarely" and "usually" groups in Q7 showed significantly less decline in NTB z-score and CGA-NPI compared to the "always" group.

Conclusion: Eating a balanced diet and reducing salt intake were associated with a slower decline in dementia severity, cognition, and behavioral alterations in patients with ADD.