Diet and Cognitive Function in Aging: An Individual Participant Data Meta-Analysis from 8 Cohorts in Europe and the United States

BACKGROUND: Cognitive decline is a major public health concern as populations age. Diet is a modifiable contributor to cognitive health, yet observational findings remain inconsistent, partly because of methodological heterogeneity.

OBJECTIVES: This study aimed to examine associations between adherence to multiple dietary patterns and cognitive outcomes using harmonized individual participant data (IPD) from longitudinal cohorts.

METHODS: We conducted an IPD meta-analysis including 17,368 participants from 8 longitudinal studies in Europe and the United States. Dietary intake was assessed using food frequency questionnaires, diet histories, or food records and harmonized to estimate adherence to 7 dietary patterns: Healthy Eating Index-2020, Alternate Healthy Eating Index-2010 (AHEI-2010), EAT-Lancet Planetary Health Diet, traditional and continuous Mediterranean Diet (MedDiet), Dietary Approaches to Stop Hypertension (DASH), and Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND). Food groups were also examined. Cognitive outcomes included global cognition measured by the Mini-Mental State Examination and executive function measured by the Trail Making Test Part B. Associations were evaluated using robust regression and mixed-effects models within cohorts, followed by random-effects meta-analysis.

RESULTS: Cross-sectionally, higher adherence to the continuous MedDiet was associated with better global cognition, and higher adherence to the continuous MedDiet and DASH was associated with better executive function. Substantial between-study heterogeneity was observed for DASH, MIND, and AHEI-2010. Greater consumption of red and processed meat was associated with poorer executive function. Longitudinal analyses showed limited and inconsistent findings. Only the MIND diet, assessed in 3 United States cohorts, was associated with more favorable executive function trajectories. No consistent longitudinal associations were observed for food groups.

CONCLUSIONS: Adherence to healthier dietary patterns was associated with modest cross-sectional differences in cognitive performance, whereas longitudinal associations were limited under standardized conditions. Findings suggest that diet-cognition associations may be modest and sensitive to measurement constraints, highlighting the need for more sensitive cognitive phenotyping and prospectively harmonized study designs.

The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) was registered at clinicaltrials.gov as NCT01041989 (https://clinicaltrials.gov/study/NCT01041989), and the Walnuts and Healthy Aging Study (WAHA) was registered at clinicaltrials.gov as NCT01634841 (https://clinicaltrials.gov/study/NCT01634841).