Date of Award

Fall 2023

Project Type


Program or Major

Nutritional Sciences

Degree Name

Master of Science

First Advisor

Sherman Bigornia

Second Advisor

Tammy Scott

Third Advisor

Maria Carlota Dao


Adherence to the MeD (Mediterranean Diet), MIND (Mediterranean-Dietary Approaches to Reduce Hypertension Intervention for Neurocognitive Delay), and other dietary patterns may benefit cognitive function among Hispanic/Latino adults. Greater regional and total brain volumetric measures are associated with an increased risk of dementia, but whether the MeD and MIND dietary patterns relate to brain volume among Hispanic/Latino adults remains unclear. The objective of the current study was to quantify the associations of the MeD, MIND, and empirically derived dietary patterns with brain volume among Hispanic/Latino adults. We used data from the longitudinal Boston Puerto Rican Health Study. Approximately 12.7 y from the baseline visit (n=1500), a subsample of 240 participants were invited to have an MRI brain scan. For this analysis, diet, assessed with a validated FFQ, was averaged across available visits (baseline, wave 2 (~2.2 y) and wave 3 (~6.2 y). The MeD score has six adequacy and three moderation components, whereas MIND has ten adequacy and five moderation ones. Component scores for each dietary score were summed to get a total dietary score, where a higher value indicates greater adherence to the pattern. Total scores range from 0-9 and 0-15. for the MeD and MIND scores, respectively. For principal components analysis, foods were collapsed into 35 categories and were expressed as a percent of total energy (g/total kcal). Outcomes were hippocampal, amygdalar, total grey, and total white matter volumes. These measures were adjusted for intracranial volume in regression models and z score transformed. Multiple linear regression models included age, sex, physical activity, education, smoking status, total energy, time between the baseline and MRI visits and intracranial volume. After exclusions for missing data, the sample size was n=227. The average age of the cohort was 55 ± 6 years and predominantly female (79%). On average, the time between the baseline and MRI visits was 12.8 ±1.1 years. A higher MeD score was significantly associated with greater amygdalar [β (95% CI), 0.08 (0.02, 0.15), P<0.05], and white matter volume [0.09 (0.03, 0.15), P <0.01]. Similarly, positive associations were observed between the MIND score and hippocampal [β (95% CI), [0.47 (0.06, 0.90), P <0.05] and amygdalar volume [ 0.08 (0.00051, 0.17), P <0.05]. A positive association was observed between a vegetables, poultry and seafood dietary pattern and hippocampal [0.16 (0.04, 0.27), P <0.01], amygdalar [0.14 (0.03, 0.25), P<0.05], grey matter [0.11 (0.01, 0.21), P<0.05] and white matter volume [0.15 (0.05, 0.3), P<0.05]. A rice, beans, and oil dietary pattern was inversely associated with hippocampal [-0.13, (-0.25, -0.01) P<0.05], amygdalar [-0.12 (-0.23, -0.01), P<0.05] and total grey matter volume [-0.12 (-0.22, -0.02), P<0.05]. The interpretation of these results remained similar in models including cardiovascular disease and diabetes mellitus type II as a model covariate. Our findings suggest that benefits of the MeD and MIND diet and dietary patterns characterized by vegetables and lean meats may, in part, be through protection of brain volume. Use of both dietary indices and empirically derived dietary patterns allow researchers to capture variation in the diets of specific populations compared with dietary guidelines established a priori. These results provide evidence that can inform tailored dietary recommendations and interventions targeting brain health among older Puerto Rican adults.