Skip to main content
NCBI home page
Alzheimer's & Dementia logoLink to Alzheimer's & Dementia
. 2023 Nov 6;20(2):1190–1200. doi: 10.1002/alz.13521

Associations of the Mediterranean‐DASH Intervention for Neurodegenerative Delay diet with brain structural markers and their changes

Hui Chen 1, Michelle M Dunk 2, Binghan Wang 1, Mengjia Zhao 1, Jie Shen 1, Geng Zong 3, Yuesong Pan 4, Lusha Tong 5, Weili Xu 2, Changzheng Yuan 1,6,✉
PMCID: PMC10917040  PMID: 37932860

Abstract

INTRODUCTION

The associations of the Mediterranean‐DASH Intervention for Neurodegenerative Delay (MIND) diet with brain structural changes are unclear.

METHODS

Among 26,466 UK Biobank participants, a 15‐point MIND score was calculated from 24‐hour diet recalls from 2009 to 2012. We assessed its associations with 17 magnetic‐resonance‐derived brain volumetric markers and their longitudinal changes and explored whether genetic factors modify the associations.

RESULTS

Higher MIND adherence was associated with larger volumes of thalamus, putamen, pallidum, hippocampus, and accumbens (beta per 3‐unit increment ranging from 0.024 to 0.033) and lower white matter hyperintensities (P‐trends < 0.05), regardless of genetic predispositions of Alzheimer's disease. MIND score was not associated with their longitudinal changes (P > 0.05) over a median of 2.2 years among participants with repeated imaging assessments (N = 2963), but was associated with slower atrophy in putamen (beta: 0.026, P‐trend = 0.044) and pallidum (beta: 0.030, P‐trend = 0.033) among APOE ε4 non‐carriers (N = 654).

DISCUSSION

The MIND diet showed beneficial associations with certain brain imaging markers, and its associations with long‐term brain structural changes warrants future investigation.

Highlights

  • Adherence to the Mediterranean‐DASH Intervention for Neurodegenerative Delay (MIND) diet was significantly associated with higher volumes and larger gray matter volumes in certain brain regions in UK adults, and the associations were not modified by genetic factors.

  • No significant associations were observed between MIND diet and longitudinal changes in the investigated brain structural markers over a median of 2.2 years.

  • Higher MIND score was significantly associated with slower atrophy in the putamen and pallidum among APOE ε4 non‐carriers.

Keywords: aging, brain structure, cohort study, diet, genetic factors, nutrition

1. BACKGROUND

Global aging has brought about heavier burdens of age‐related neurodegenerative diseases, such as dementia 1 , 2 and Parkinson's disease. 3 Dietary factors are considered crucial in brain aging, and certain nutrients, 4 , 5 , 6 , 7 , 8 food groups, 9 , 10 and dietary patterns 11 , 12 , 13 have been related to specific brain diseases.

The Mediterranean‐DASH Intervention for Neurodegenerative Delay (MIND) diet, proposed by Morris et al., 14 has a focus on slowing neurodegeneration 15 by emphasizing natural plant‐based foods and restricting intakes of unhealthy animal and high‐saturated‐fat foods. Previous studies have shown beneficial associations of the MIND diet with cognitive function and its decline, 16 all‐cause dementia, 17 and Parkinson's disease, 18 which often feature structural brain changes at early stages. 19 However, associations of the MIND diet with brain structural markers remain unclear and inconsistent, and elucidating these relationships is crucial for understanding the mechanisms linking this dietary pattern to brain aging. Among 1302 Women's Health Initiative participants, a higher MIND diet score was associated with higher total and temporal lobe white matter volumes. 20 In 175 French individuals, a higher MIND diet score was associated with lower diffusivity in the splenium of the corpus callosum. 21 In 1904 participants of the Framingham Offspring cohort, a higher MIND diet score was related to larger total brain volume (TBV), but not to other brain magnetic resonance imaging (MRI) measures (i.e., hippocampal volume, lateral ventricular volume, white matter hyperintensity volume, and silent brain infarcts). 22 While available evidence suggests the beneficial role of the MIND diet, large‐scale studies with repeated measurements of brain structure markers are warranted to elucidate whether the MIND diet is associated with longitudinal changes in brain structural markers. Furthermore, whether the associations differ by genetic factors remains unclear, which is essential for the development of targeted prevention strategies of dementia.

In the current study, we aim to explore the associations of adherence to the MIND diet with brain structural markers and their longitudinal changes in the UK Biobank, a large‐scale cohort study.

2. METHODS

2.1. Study population

The UK Biobank recruited ~0.5 million participants aged 37 to 73 years between 2006 and 2010 from 22 assessment centers in the UK, as described on its website (http://www.ukbiobank.ac.uk/resources/). Ethical approval was obtained from the National Health Service National Research Ethics Service (11/NW/0382). At recruitment, participants provided consent and completed touchscreen questionnaires, physical assessments, and blood sample collection. In 2009 to 2012, ~40% of participants completed a dietary assessment. From 2014 to 2020 (at least 2 years after dietary assessments), the UK Biobank started a multi‐modal imaging substudy, 23 with > 45,000 participants having undergone an assessment by 2020. 24 Between 2018 and 2022, participants in the imaging substudy were invited to attend a repeated imaging assessment after the first assessment (median interval = 2.2 years).

We conducted two primary analyses (Figure 1) among UK Biobank participants who finished the dietary assessments from 2009 to 2012, which was treated as the study baseline. Among participants in the imaging substudy without prevalent dementia or stroke at the time of 24‐hour dietary recalls (n = 26,466), we evaluated the association of the MIND diet with brain structural markers assessed between 2014 and 2020. We additionally examined the association of the MIND diet with changes in brain structural markers among 2963 participants who underwent a second imaging assessment between 2018 and 2022.

FIGURE 1.

FIGURE 1

Open in a new tab

Participant inclusion flow chart (A) and study timeline (B).

2.2. Mediterranean‐DASH Intevention for Neurodegenrative Delay diet score

UK Biobank participants completed WebQ, a validated web‐based questionnaire for 24‐hour diet recalls 25 , 26 , 27 on five occasions (April 2009 to September 2010, February to April 2011, June to September 2011, October to December 2011, and April to June 2012). The MIND diet score was calculated based on the intake of 10 food groups classified as healthy (green leafy vegetables, other vegetables, berries, nuts, whole grains, beans, non‐fried fish, non‐fried poultry, olive oil, and wine) and 5 food groups considered unhealthy (red meats, butter and margarine, cheese, pastries and sweets, and fried or fast foods; 14 Table S1 in supporting information). Olive oil was scored 1 if the participant reported its use as the primary oil and 0 otherwise. For other recommended food groups except for wine, the top tertile was assigned a concordance score of 1, the medium tertile was assigned 0.5, and the bottom tertile was assigned 0. For unhealthy food groups, the scores were reversed. We modified the criteria for wine intake to adapt to the 24‐hour recall, with a score of 1 for daily wine intake of 1 glass, 0.5 for 2 to 3 glasses, and 0 for 0 or ≥4 glasses. The total MIND diet score was computed by summing the 15 component scores (ranging from 0 to 15), and a higher score reflects greater adherence. Overall, the participants completed a mean of 2.5 times of repeated 24‐hour dietary recalls, and 71.7% of the study participants completed at least two dietary assessments. We calculated the average MIND diet score for individuals with multiple assessments.

2.3. Genetic factors

In the current study, we selected APOE ε4 carrier status (yes or no) and the polygenic risk score of Alzheimer's disease (AD‐PRS) as factors related to brain aging. Single nucleotide polymorphism data for rs429358 and rs7412 was used to determine APOE genotypes. AD‐PRS was provided by the UK Biobank in the data field 26206. Briefly, the AD‐PRS was generated using a Bayesian approach applied to meta‐analyzed summary statistics entirely from external genome‐wide association study data, as is described in a previous study. 28

RESEARCH IN CONTEXT

  1. Systematic review: The authors systematically reviewed the literature using multiple sources (e.g., Web of Science, PubMed, Google Scholar). The Mediterranean‐DASH Intervention for Neurodegenerative Delay (MIND) diet has been associated with lower risk of dementia and slower cognitive decline. However, evidence on the relation of the MIND diet to brain structural markers and their changes is scarce and inconsistent.

  2. Interpretation: In this large‐scale cohort study, adherence to the MIND diet was significantly associated with higher overall and gray matter volumes in certain brain regions in UK adults. The main independent contributors of the MIND diet included higher intake of whole grains and olive oil, and lower intake of fast fried foods. While no significant associations were observed between MIND diet and the longitudinal changes over a median of 2.2 years in the investigated brain structural markers, we observed potentially protective associations among apolipoprotein E ε4 non‐carriers.

  3. Future directions: Our study discovered that adherence to the MIND diet showed beneficial associations with white matter hyperintensities and certain subcortical brain region volumes among middle‐aged and older adults. Future studies are needed to confirm the associations with long‐term brain structural changes and elucidate specific mechanisms linking dietary factors to brain health.

2.4. Brain structural markers

Among the study participants, brain structural markers were measured using MRI performed with a 3 Tesla Siemens Skyra scanner  (Siemens Healthineers) with VD13 software and a 32‐channel head coil since 2014. 24 In this study, we used data from T1 scans to extract volumetric measures of the total brain and the subcortical regions, including the thalamus, caudate, putamen, pallidum, hippocampus, amygdala, and accumbens. 29 , 30 In addition, we included the volume of white matter hyperintensities from T2 scans because of its strong association with dementia and other neurodegenerative diseases. 31 The primary outcomes of interest were the volumes of the total brain, total white and gray matter (normalized to the total head), and the volumes of the subcortical regions. We also included the subcortical gray matter volumes as secondary outcomes from T1 scans because they have been associated with the progression of Alzheimer's disease (AD) and related dementia. 32 In the primary analysis, we calculated the z score of each brain structural volumetric marker as the original value minus the population mean divided by the standard deviation (SD) at the Phase 1 MRI assessment. 22 For white matter hyperintensity volume with a long‐tail distribution, we first log‐transformed the original value and then calculated the z score as mentioned above. To represent the changes in these markers, we first calculated the z score of the values from Phase 2 using the population mean and SD at the first MRI assessment and subtracted Phase 2 z scores from Phase 1 z scores. 22

2.5. Covariates

We used questionnaires completed at recruitment (from 2006 to 2010) to collect information on age, sex (male or female), education level (college degree or not), Townsend deprivation index (indicating social deprivation), physical activity levels according to the World Health Organization criteria (high, moderate, or low), smoking status (never, ever, or current), body mass index (BMI) categories (underweight or normal weight, overweight, or obese), depressive symptoms (yes or no), hypertension (yes or no), diabetes (yes or no), and cardiovascular diseases (CVD; yes or no). Total energy intake (continuous, in kcal/day) was calculated from the WebQ concurrently with the MIND diet score. Alcohol intake was not considered a confounding variable because wine is a component of the MIND diet.

Depressive symptoms were assessed using the two‐item Patient Health Questionnaire (PHQ‐2; each item scored from 0 to 3), and participants scoring ≥3/6 were defined as having depressive symptoms. 33 We used self‐reported diagnoses, medication, and linkages to Hospital Episode Statistics (HES) data to ascertain other health conditions coded according to the International Classification of Disease (ICD). We used ICD‐10 codes I10 and I15 for hypertension, E10–E14 for diabetes, and I20–I25 and I60–I69 for CVD.

2.6. Statistical analyses

We described participants’ baseline characteristics according to their average MIND diet score in tertiles. Normally distributed continuous variables were presented as mean (SD), and those with skewed distributions were presented as median (interquartile range). Categorical variables were reported as number (percentage).

In the primary analyses, we assessed the relations of the MIND diet score and its components (average intake levels across multiple assessments) to the z scores of brain structural markers using general linear models, with adjustments for baseline sociodemographic factors and health‐related behaviors (age, sex, total energy intake, education level, smoking status, and physical activity). The MIND diet score variable was treated as categorical (tertile) and continuous (per 3‐unit increment) in separate models. In the analyses for specific food groups, the estimates of each component were mutually adjusted. In the analyses examining changes in brain structural markers, we used the difference in volumes from the first to the second assessment as the response variable, and the corresponding volume in the first assessment was further adjusted for in the linear models.

We performed prespecified stratified analyses by age at recruitment (< 60 years or ≥60 years), sex (female or male), APOE ε4 carrying status (carrier or not), and AD‐PRS (high or low, defined by median). We tested potential interactions for these variables using analysis of variance (ANOVA) on the cross‐product terms of the exposure variable and the covariate of interest. Specifically for AD‐PRS, we additionally adjusted for the first four ancestry‐based principal components. We also used the original values of the brain structural markers (in mm3) instead of the z scores as the outcome variables.

We conducted several sensitivity analyses. First, we additionally adjusted the models for depressive symptoms, hypertension, diabetes, and CVD, which may lie in the pathways from diet to brain health. Second, we restricted our analyses to participants with at least two 24‐hour diet recalls to reduce measurement error in the dietary assessments. Third, we evaluated the associations of the MIND diet score with brain structural markers at the second assessment. Finally, we used linear mixed models to examine the association of the MIND diet score with longitudinal changes in brain structural markers.

We reported two‐sided P‐values throughout, and a Benjamini–Hochberg‐adjusted P‐value < 0.05 was considered an indicator of statistical significance. Statistical analyses were performed using R 3.6.0.

2.7. Patient and public involvement

No patient or participant was involved in the conceptualization, design, or implementation of the current study.

3. RESULTS

3.1. Baseline characteristics

In the primary analysis of the association between the MIND diet score and brain structural markers, the 26,466 participants had a mean (SD) age of 55.1 (7.5) years at recruitment, and 14,152 (53.5%) were female (Table 1). Of the 2963 participants included in the analysis examining changes in brain structural markers, the mean (SD) age was 52.8 (7.4) years, and 1585 (53.5%) were female (Table S2 in supporting information). Participants with higher MIND diet scores were older, better educated, more physically active, and less likely to be current smokers. We also described the volumetric measures of each brain region in Tables S3 and S4 in supporting information. Among the 2963 who underwent brain MRI assessment twice, the within‐person changes of all brain structural makers significantly differed from null. For example, z scores of total brain volume were lower in the second assessment (−0.45, 95% confidence interval [CI], −0.46 to −0.44) compared to the first assessment (Table S5 in supporting information).

TABLE 1.

Baseline characteristics of study participants.

MIND diet score
Overall Tertile 1 Tertile 2 Tertile 3
n 26,466 10,161 7958 8347
MIND diet score, median (IQR) 6.0 (5.0, 7.0) 4.8 (4.1, 5.2) 6.2 (6.0, 6.5) 7.6 (7.2, 8.4)
Total energy intake, kcal/day, mean (SD) 2077.6 (530.8) 2163.8 (554.2) 2036.6 (500.8) 2011.8 (515.2)
Female, N (%) 14152 (53.5) 4585 (45.1) 4426 (55.6) 5141 (61.6)
Age at recruitment, years, mean (SD) 55.1 (7.5) 54.4 (7.6) 55.2 (7.5) 55.8 (7.3)
TDI, mean (SD) –1.9 (2.7) –1.9 (2.7) –2.0 (2.7) –1.9 (2.7)
College degree, N (%) 13969 (52.8) 5066 (49.9) 4240 (53.3) 4663 (55.9)
Ethnicity, N (%) White 25669 (97.0) 9907 (97.5) 7718 (97.0) 8044 (96.4)
Asian 250 (0.9) 76 (0.7) 73 (0.9) 101 (1.2)
Black 151 (0.6) 55 (0.5) 41 (0.5) 55 (0.7)
Others 396 (1.5) 123 (1.2) 126 (1.6) 147 (1.8)
Smoking, N (%) Never 16275 (61.5) 6160 (60.6) 4881 (61.3) 5234 (62.7)
Former 8665 (32.7) 3228 (31.8) 2683 (33.7) 2754 (33.0)
Current 1526 (5.8) 773 (7.6) 394 (5.0) 359 (4.3)
Physical activity level, N (%) Low 4821 (18.2) 2156 (21.2) 1419 (17.8) 1246 (14.9)
Medium 11238 (42.5) 4344 (42.8) 3378 (42.4) 3516 (42.1)
High 10407 (39.3) 3661 (36.0) 3161 (39.7) 3585 (42.9)
BMI, mean (SD) 26.4 (4.2) 26.8 (4.3) 26.3 (4.1) 25.9 (4.0)
Alcohol intake, N (%) Never 604 (2.3) 248 (2.4) 168 (2.1) 188 (2.3)
Former 542 (2.0) 221 (2.2) 156 (2.0) 165 (2.0)
Current 25320 (95.7) 9692 (95.4) 7634 (95.9) 7994 (95.8)
Hypertension, N (%) 5260 (19.9) 2086 (20.5) 1542 (19.4) 1632 (19.6)
CVD, N (%) 806 (3.0) 309 (3.0) 232 (2.9) 265 (3.2)
Diabetes, N (%) 707 (2.7) 307 (3.0) 203 (2.6) 197 (2.4)
APOE ε4 carrier, N (%) 6125 (27.4) 2327 (27.3) 1846 (27.4) 1952 (27.5)
Follow‐up duration, year, median (IQR) 8.8 (7.3, 9.9) 8.7 (7.3, 9.8) 8.8 (7.3, 9.9) 8.8 (7.4, 9.9)
Open in a new tab

Note: Follow‐up duration is defined as the time interval from date of first dietary assessment to the date of the first imaging assessment. APOE genotype information was available for 22,353 participants.

Abbreviations: APOE, apolipoprotein E; BMI, body mass index; CVD, cardiovascular diseases; IQR, interquartile range; MIND, Mediterranean‐DASH Intervention for Neurodegenerative Delay; SD, standard deviation; TDI, Townsend deprivation index.

3.2. MIND diet, brain structural markers, and longitudinal changes in the markers

We did not observe a significant association between MIND diet score and total brain, white matter, or gray matter volumes, but significant associations were observed in subcortical regions (Figure 2 and Table S6 in supporting information). Greater adherence to the MIND diet was associated with larger volumes of the thalamus (beta per 3‐unit increment of MIND score: 0.029, 0.009 to 0.049), putamen (0.029, 0.009 to 0.049), pallidum (0.033, 0.011 to 0.055), hippocampus (0.025, 0.003 to 0.047), and accumbens (0.024, 0.003 to 0.045), but not with volumes of caudate or amygdala. We also observed that each 3‐unit increment of MIND score was related to −0.029 (−0.050 to −0.009) lower z score of white matter hyperintensities. Moreover, greater adherence to the MIND diet showed significant associations with higher volumes of gray matter in the hippocampus (0.042, 0.021 to 0.063) and amygdala (0.024, 0.003 to 0.044).

FIGURE 2.

FIGURE 2

Open in a new tab

Associations of the MIND diet scores (per 3‐unit increment) with the z‐scores of brain structural markers (N = 26,466) and their longitudinal changes (N = 2963). Beta coefficients and 95% CIs were calculated using general linear models adjusted for age, age2, sex, time interval from recruitment to brain imaging assessment, ethnicity, Townsend deprivation index, education level, physical activity, smoking, and total energy intake. CI, confidence interval; MIND, Mediterranean‐DASH Intervention for Neurodegenerative Delay.

Among participants who underwent two brain imaging assessments, no significant association was observed between the MIND diet and changes in brain structural markers (Figure 2 and Table S7 in supporting information). For example, the betas (95% CI) were −0.011 (−0.042 to 0.020) for total brain volume, −0.010 (−0.042 to 0.021) for white matter volume, 0.012 (−0.010 to 0.034) for gray matter volume, and 0.009 (−0.017 to 0.034) for hippocampus volume with each 3‐unit increment of MIND score. MIND diet score was also not associated with a change in z score of white matter hyperintensities (0.009, −0.050 to 0.068) or with volumes of subcortical gray matter (P > 0.05 for all).

In the analysis of food groups (Figure 3), we observed that the highest intake (top vs. bottom tertile) of whole grains was associated with higher volumes of total brain, gray matter, and all subcortical regions investigated except for the amygdala, with beta coefficients ranging from 0.034 to 0.094. Highest whole grain consumption was also related to 0.059 (0.090 to 0.027, Benjamini–Hochberg–adjusted P‐values < 0.05) lower z score of white matter hyperintensities. Similarly, using olive oil as the primary oil showed protective associations with several subcortical regions, including the thalamus, caudate, pallidum, and amygdala. Moreover, higher intake of fried fast foods was associated with lower volumes of thalamus, putamen, pallidum, and hippocampus. Unexpectedly, a higher intake of pastries and sweets was related to higher volumes of total brain, white matter, gray matter, thalamus, pallidum, hippocampus, and accumbens and lower volume of white matter hyperintensities (Benjamini–Hochberg–adjusted P‐values < 0.05 for all tests).

FIGURE 3.

FIGURE 3

Open in a new tab

Associations of the food groups in the MIND diet with the z scores of brain structural markers (N = 26,466). Beta coefficients and 95% CIs were calculated using general linear models adjusted for age, age2, sex, interval from recruitment to brain imaging assessment, ethnicity, Townsend deprivation index, education level, physical activity, smoking, total energy intake, and other components of the MIND diet. Red color indicated positive associations, blue indicated inverse associations. Asterisks indicate statistical significance after adjustment for multiple testing (Benjamini–Hochberg–adjusted P‐value < 0.05). CI, confidence interval; MIND, Mediterranean‐DASH Intervention for Neurodegenerative Delay.

3.3. Secondary and sensitivity analyses

In the prespecified stratified analyses, no significant interaction was shown between the MIND diet score and age (P‐interactions > 0.05), but its associations with several subcortical volumes differed by sex (Table S8 in supporting information). For example, the coefficient for the association of 3‐unit increment in MIND score with the white matter hyperintensities z score was −0.033 (−0.057 to −0.008) in participants aged below 60 years and −0.023 (−0.059 to 0.013) for those aged 60 years or above (P‐interaction = 0.068). Among men, a higher MIND diet score was suggestively associated with higher total brain (0.024, −0.004 to 0.053), white matter (0.009, −0.024 to 0.043), and gray matter (0.029, 0.003 to 0.055) volumes, but the associations were reversed for women, with betas of −0.037 (−0.063 to −0.011), −0.026 (−0.056 to 0.004), and −0.034 (−0.058 to −0.010), respectively (P‐interactions < 0.05). Moreover, the association between MIND score and white matter hyperintensities was significant for men (−0.046, −0.077 to −0.016) but not for women (−0.015, −0.042 to 0.011, P‐interaction = 0.024), although the direction persisted.

In the stratified analysis by genetic factors, the associations of the MIND diet with brain structure markers did not significantly differ by APOE ε4 status or AD‐PRS (P‐interactions > 0.05, Figure S1 in supporting information). Nevertheless, higher MIND score was significantly associated with slower atrophy in putamen (beta: 0.026, P‐trend = 0.044) and pallidum (beta: 0.030, P‐trend = 0.033) among 654 APOE ε4 non‐carriers (P‐interaction = 0.804 and 0.021, respectively, Figure 4 and Figure S2 in supporting information). The associations for the changes in brain structure markers were similar between the high/lower AD‐PRS groups (Figure S3 in supporting information).

FIGURE 4.

FIGURE 4

Open in a new tab

Associations of the MIND diet scores (per 3‐unit increment) with the z scores of brain structural markers (N = 22,353) and their longitudinal changes (N = 2607), stratified by APOE ε4 carrying status. Beta coefficients and 95% CIs were calculated using general linear models adjusted for age, age2, sex, time interval from recruitment to brain imaging assessment, ethnicity, Townsend deprivation index, education level, physical activity, smoking, and total energy intake. APOE, apolipoprotein E; CI, confidence interval; MIND, Mediterranean‐DASH Intervention for Neurodegenerative Delay.

In the sensitivity analyses (Table S9 in supporting information), the observed associations remained consistent and robust. When depressive symptoms, diabetes, hypertension, and cardiovascular diseases were further adjusted for, all associations remained in the same direction, and those for thalamus, putamen, pallidum, hippocampus, white matter hyperintensities, and gray matter in hippocampus and amygdala remained statistically significant. When restricted to participants who had at least two dietary assessments, all associations remained in the same direction, and significant associations were observed for volumes of the putamen, pallidum, and white matter hyperintensities, and gray matter in hippocampus. When we analyzed the association between the MIND diet score and volumetric measures in the second assessment, only the association for white matter hyperintensities persisted, potentially due to the reduced sample size. Finally, linear mixed modeling revealed no significant associations for longitudinal changes in brain structural markers, consistent with results from the primary analysis (Table S10 in supporting information).

4. DISCUSSION

4.1. Principal findings

In the current study, we explored the associations of adherence to the MIND diet with brain structural markers and their longitudinal changes. Higher adherence to the MIND diet demonstrated beneficial associations with white matter hyperintensities, certain subcortical brain region volumes (thalamus, putamen, pallidum, hippocampus, and accumbens), and larger gray matter volumes in the hippocampus and amygdala. The main components of the MIND diet that appeared to drive these associations included higher intake of whole grains and olive oil, and lower intake of fast or fried foods. However, we did not observe significant associations of overall MIND diet score or its dietary components with longitudinal changes in brain structure over a median of 2.2 years.

4.2. Comparison to other studies

The MIND diet has been linked to slower cognitive decline 16 and lower risks of all‐cause dementia 17 and Parkinson's disease, 18 but only several cross‐sectional studies have assessed its associations with brain structural markers. In 1302 participants of the Women's Health Initiative Memory Study, adherence to a MIND‐like diet was associated with higher total and temporal lobe white matter volumes. 20 In 175 older adults in France, a higher MIND diet score was associated with lower diffusivity values in the splenium of the corpus callosum. 21 In 1904 Framingham Offspring participants, a higher MIND diet score was related to larger total brain volume. 22 Our study provides a unique addition to the current literature by reporting significant associations of adherence to the MIND diet with higher total and regional gray matter volumes of several subcortical regions, including the thalamus, putamen, pallidum, hippocampus, and accumbens. A higher MIND diet score was also associated with a lower volume of white matter hyperintensities, which is closely related to AD. 31 Notably, we observed that some associations were only significant in men, potentially because women in the study tend to be healthier than men. 34 Underlying sex disparities in neurodegeneration may also play a role in our observed sex differences, and additional research is warranted to further clarify how the relations of diet to brain structure may differ by sex.

Notably, we found no significant association of the MIND diet with longitudinal changes in the investigated brain structural markers in the population who completed two imaging assessments. While this corresponded to the 3‐year randomized control trial of the MIND diet for prevention of cognitive decline, 35 several factors could have influenced our results. First, calculating the difference between the two MRI assessments could have increased measurement error and thus widened the confidence intervals. The smaller sample size (compared to the primary analyses) also reduced the statistical power. Second, only approximately one tenth of the overall study population underwent a second brain MRI. They were younger and had significantly lower prevalence of chronic conditions as well as more favorable brain structural markers at the first assessment. Furthermore, the short‐term time interval between the two assessments might also contribute to the non‐significant association. Finally, the relatively young study population (mean age = 55 years) may have less brain atrophy which typically occurs in older individuals, which may have interfered with the identifications of significant associations. As a result, this sample may be less prone to the harm of a suboptimal diet. Nevertheless, we observed that MIND diet was associated with slower decline in putamen and pallidum volumes among APOE ε4 non‐carriers, which may imply that the associations of the MIND diet may be modified genetic factors. Due to the exploratory nature of this study, future investigations are warranted to confirm associations between the MIND diet and brain structural markers, and whether they differ by other risk factors of neurodegeneration.

Because impaired brain health may encompass a wide range of adverse outcomes including dementia and functional impairment, identifying effective dietary approaches for protecting long‐term brain health is crucial in a global aging context. 36 The MIND diet emphasizes food groups intrinsically related to brain health, such as green leafy vegetables, olive oil, and berries. 37 In our analysis, whole grains and vegetables demonstrated significant protective associations, which are rich in vitamin E 38 and dietary fiber 39 and may slow brain aging through inhibition of oxidization and inflammatory pathways and deposition of amyloid beta. The MIND diet also restricts food groups that might be linked to poorer brain‐related health outcomes, such as fast fried food, and potential explanations include the harmful effects of saturated fats on inflammation and cerebral oxidation. 40 The seemingly protective association of the pastries and sweets may reflect the fact that some sweets, such as chocolate, may be rich in flavanols and methylxanthine and beneficial to brain health. 41 Additionally, volumes of different subcortical regions may be related to particular nutrients. For instance, a diet high in fat and refined sugar might harm neurogenesis in the hippocampus, which is essential for memory. 42 The integrity of the amygdala, which is involved in emotional functions, may be associated with B vitamins. 43 Moreover, saturated and trans‐ fatty acids in red and processed meats may harm neuronal plasticity, 44 while unsaturated fatty acids appear to have anti‐inflammatory properties. 45

4.3. Strengths and limitations of study

The strengths of the current study included long‐term follow‐up, a relatively large sample size, and repeated measurements of brain MR that enabled longitudinal analyses. Nevertheless, our findings should be interpreted with caution due to some limitations. First, the generalizability of our findings warrants confirmation because the UK Biobank participants might be healthier than the general population. Second, measurement errors are inevitable in dietary and MR assessments, although the sensitivity analysis among participants who completed at least two recalls showed robust associations and the standardized image processing pipeline may reduce the influence of the batch effect. Third, our findings may not necessarily reflect causal relations due to potential residual confounding and reverse causality, although the imaging assessments were taken long after dietary assessments. Randomized trials are needed to further clarify potential associations of the MIND diet with brain health. 37

4.4. Conclusion and public health implications

In this cohort study, greater adherence to the MIND diet showed beneficial associations with white matter hyperintensities and certain subcortical brain region volumes among middle‐aged and older adults. Future studies are needed to confirm the associations with long‐term brain structural changes and elucidate specific mechanisms linking dietary factors to brain health.

AUTHOR CONTRIBUTORS

Changzheng Yuan, Weili Xu, and Hui Chen designed and conceptualized the study, interpreted the findings, and drafted the manuscript. Geng Zong, Michelle M. Dunk, and Weili Xu revised the manuscript. Hui Chen performed the data analyses. Binghan Wang, Mengjia Zhao, Jie Shen, Yuesong Pan, and Lusha Tong provided assistance in data analysis and interpretation of the findings.

CONFLICT OF INTEREST STATEMENT

All authors declare no competing interests. Author disclosures are available in the supporting information.

CONSENT STATEMENT

All human subjects provided informed consent at recruitment.

Supporting information

Supporting Information

ALZ-20-1190-s001.docx (1.2MB, docx)

Supporting Information

ALZ-20-1190-s002.pdf (912.8KB, pdf)

ACKNOWLEDGMENTS

This study was conducted using the UK Biobank Resource (application No. 55005). Data can be accessed upon application to the UK Biobank (http://ukbiobank.ac.uk). The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. The authors express their genuine gratitude to staff and participants of the UK Biobank, which was supported by the Wellcome Trust, Medical Research Council, Department of Health, Scottish government, Northwest Regional Development Agency, the Welsh assembly government, and the British Heart Foundation. The current study is funded by Alzheimer's Association (AARG‐22‐928604, the Zhejiang University Global Partnership Fund and National Natural Science Foundation of China (8210120183). The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Chen H, Dunk MM, Wang B, et al. Associations of the Mediterranean‐DASH Intervention for Neurodegenerative Delay diet with brain structural markers and their changes. Alzheimer's Dement. 2024;20:1190–1200. 10.1002/alz.13521

Changzheng Yuan and Weili Xu contributed equally as co‐senior authors.

The manuscript's guarantors affirm that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.

REFERENCES

  • 1. Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet North Am Ed. 2020;396:413‐446. doi: 10.1016/S0140-6736(20)30367-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Glob Health. 2022;0. doi: 10.1016/S2468-2667(21)00249-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Bloem BR, Okun MS, Klein C. Parkinson's disease. Lancet North Am Ed. 2021;397:2284‐2303. doi: 10.1016/S0140-6736(21)00218-X [DOI] [PubMed] [Google Scholar]
  • 4. Yuan C, Chen H, Wang Y, Schneider JA, Willett WC, Morris MC. Dietary carotenoids related to risk of incident Alzheimer dementia (AD) and brain AD neuropathology: a community‐based cohort of older adults. Am J Clin Nutr. 2020. doi: 10.1093/ajcn/nqaa303 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Commenges D, Scotet V, Renaud S, Jacqmin‐Gadda H, Barberger‐Gateau P, Dartigues J‐F. Intake of flavonoids and risk of dementia. Eur J Epidemiol. 2000;16:357‐363. doi: 10.1023/A:1007614613771 [DOI] [PubMed] [Google Scholar]
  • 6. Cheng N, Bell L, Lamport DJ, Williams CM. Dietary flavonoids and human cognition: a meta‐analysis. Mol Nutr Food Res:2100976. doi: 10.1002/mnfr.202100976. n.d.;n/a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Crichton GE, Bryan J, Murphy KJ. Dietary antioxidants, cognitive function and dementia—a systematic review. Plant Foods Hum Nutr. 2013;68:279‐292. doi: 10.1007/s11130-013-0370-0 [DOI] [PubMed] [Google Scholar]
  • 8. Wu S, Ding Y, Wu F, Li R, Hou J, Mao P. Omega‐3 fatty acids intake and risks of dementia and Alzheimer's disease: a meta‐analysis. Neurosci Biobehav Rev. 2015;48:1‐9. doi: 10.1016/j.neubiorev.2014.11.008 [DOI] [PubMed] [Google Scholar]
  • 9. Bakre AT, Chen R, Khutan R, et al. Association between fish consumption and risk of dementia: a new study from China and a systematic literature review and meta‐analysis. Public Health Nutr. 2018;21:1921‐1932. doi: 10.1017/S136898001800037X [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Ran LS, Liu WH, Fang YY, et al. Alcohol, coffee and tea intake and the risk of cognitive deficits: a dose–response meta‐analysis. Epidemiol Psychiatric Sci. 2021;30. doi: 10.1017/S2045796020001183 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Cao L, Tan L, Wang H‐F, et al. Dietary patterns and risk of dementia: a systematic review and meta‐analysis of cohort studies. Mol Neurobiol. 2016;53:6144‐6154. doi: 10.1007/s12035-015-9516-4 [DOI] [PubMed] [Google Scholar]
  • 12. Liu Y‐H, Gao X, Na M, Kris‐Etherton PM, Mitchell DC, Jensen GL. Dietary pattern, diet quality, and dementia: a systematic review and meta‐analysis of prospective cohort studies. J Alzheimer's Disease. 2020;78:151‐168. doi: 10.3233/JAD-200499 [DOI] [PubMed] [Google Scholar]
  • 13. Akbaraly TN, Singh‐Manoux A, Dugravot A, Brunner EJ, Kivimäki M, Sabia S. Association of midlife diet with subsequent risk for dementia. JAMA. 2019;321:957‐968. doi: 10.1001/jama.2019.1432 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Morris MC, Tangney CC, Wang Y, et al. MIND diet slows cognitive decline with aging. Alzheimers Dement: J Alzheimers Assoc. 2015;11:1015‐1022. doi: 10.1016/j.jalz.2015.04.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Morris MC, Tangney CC, Wang Y, Sacks FM, Bennett DA, Aggarwal NT. MIND diet associated with reduced incidence of Alzheimer's disease. Alzheimers Dement: J Alzheimers Assoc. 2015;11:1007‐1014. doi: 10.1016/j.jalz.2014.11.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Huang L, Tao Y, Chen H, et al. MIND diet and cognitive function and its decline: a prospective study and meta‐analysis of prospective cohort studies. Am J Clin Nutr. 2023. doi: 10.1016/j.ajcnut.2023.04.025 [DOI] [PubMed] [Google Scholar]
  • 17. Chen H, Dhana K, Huang Y, et al. Association of the mediterranean dietary approaches to stop hypertension intervention for neurodegenerative delay (mind) diet with the risk of dementia. JAMA Psychiatry. 2023. doi: 10.1001/jamapsychiatry.2023.0800 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Agarwal P, Wang Y, Buchman AS, Holland TM, Bennett DA, Morris MC. MIND diet associated with reduced incidence and delayed progression of parkinsonism in old age. J Nutr Health Aging. 2018;22:1211‐1215. doi: 10.1007/s12603-018-1094-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Ewers M, Sperling RA, Klunk WE, Weiner MW, Hampel H. Neuroimaging markers for the prediction and early diagnosis of Alzheimer's disease dementia. Trends Neurosci. 2011;34:430‐442. doi: 10.1016/j.tins.2011.05.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Chen C, Hayden KM, Kaufman JD, et al. Adherence to a MIND‐like dietary pattern, long‐term exposure to fine particulate matter air pollution, and mri‐based measures of brain volume: the women's health initiative memory study‐MRI. Environ Health Perspect;129:127008. doi: 10.1289/EHP8036. n.d.. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Thomas A, Lefèvre‐Arbogast S, Féart C, et al. Association of a MIND diet with brain structure and dementia in a French population. J Prev Alzheimers Dis. 2022. doi: 10.14283/jpad.2022.67 [DOI] [PubMed] [Google Scholar]
  • 22. Melo Van Lent D, O'Donnell A, Beiser AS, et al. Mind diet adherence and cognitive performance in the framingham heart study. J Alzheimer's Disease. 2021;82:827‐839. doi: 10.3233/JAD-201238 [DOI] [PubMed] [Google Scholar]
  • 23. Miller KL, Alfaro‐Almagro F, Bangerter NK, et al. Multimodal population brain imaging in the UK Biobank prospective epidemiological study. Nat Neurosci. 2016;19:1523‐1536. doi: 10.1038/nn.4393 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Littlejohns TJ, Holliday J, Gibson LM, et al. The UK Biobank imaging enhancement of 100,000 participants: rationale, data collection, management and future directions. Nat Commun. 2020;11:2624. doi: 10.1038/s41467-020-15948-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Liu B, Young H, Crowe FL, et al. Development and evaluation of the Oxford WebQ, a low‐cost, web‐based method for assessment of previous 24 h dietary intakes in large‐scale prospective studies. Public Health Nutr. 2011;14:1998‐2005. doi: 10.1017/S1368980011000942 [DOI] [PubMed] [Google Scholar]
  • 26. Galante J, Adamska L, Young A, et al. The acceptability of repeat Internet‐based hybrid diet assessment of previous 24‐h dietary intake: administration of the Oxford WebQ in UK Biobank. Br J Nutr. 2016;115:681‐686. doi: 10.1017/S0007114515004821 [DOI] [PubMed] [Google Scholar]
  • 27. Greenwood DC, Hardie LJ, Frost GS, et al. Validation of the Oxford WebQ online 24‐hour dietary questionnaire using biomarkers. Am J Epidemiol. 2019;188:1858‐1867. doi: 10.1093/aje/kwz165 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Thompson DJ, Wells D, Selzam S, et al. UK Biobank release and systematic evaluation of optimised polygenic risk scores for 53 diseases and quantitative traits 2022:2022.06.16.22276246. doi: 10.1101/2022.06.16.22276246 medRxiv [DOI]
  • 29. Albert MS. Cognitive and neurobiologic markers of early Alzheimer disease. Proc Natl Acad Sci. 1996;93:13547‐13551. doi: 10.1073/pnas.93.24.13547 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Roh JH, Qiu A, Seo SW, et al. Volume reduction in subcortical regions according to severity of Alzheimer's disease. J Neurol. 2011;258:1013‐1020. doi: 10.1007/s00415-010-5872-1 [DOI] [PubMed] [Google Scholar]
  • 31. Kandel BM, Avants BB, Gee JC, et al. White matter hyperintensities are more highly associated with preclinical Alzheimer's disease than imaging and cognitive markers of neurodegeneration. Alzheimer's Dement: Alzheimer's Dement: Diagn Assess Dis Monit. 2016;4:18‐27. doi: 10.1016/j.dadm.2016.03.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Yi H‐A, Möller C, Dieleman N, et al. Relation between subcortical grey matter atrophy and conversion from mild cognitive impairment to Alzheimer's disease. J Neurol Neurosurg Psychiatry. 2016;87:425‐432. doi: 10.1136/jnnp-2014-309105 [DOI] [PubMed] [Google Scholar]
  • 33. Harshfield EL, Pennells L, Schwartz JE, et al. Association between depressive symptoms and incident cardiovascular diseases. JAMA. 2020;324:2396‐2405. doi: 10.1001/jama.2020.23068 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Fry A, Littlejohns TJ, Sudlow C, et al. Comparison of sociodemographic and health‐related characteristics of UK biobank participants with those of the general population. Am J Epidemiol. 2017;186:1026‐1034. doi: 10.1093/aje/kwx246 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Barnes LL, Dhana K, Liu X, et al. Trial of the MIND diet for prevention of cognitive decline in older persons. N Engl J Med. 2023:NEJMoa2302368. doi: 10.1056/NEJMoa2302368 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Gardener H, Wright CB, Rundek T, Sacco RL. Brain health and shared risk factors for dementia and stroke. Nat Rev Neurol. 2015;11:651‐657. doi: 10.1038/nrneurol.2015.195 [DOI] [PubMed] [Google Scholar]
  • 37. Liu X, Morris MC, Dhana K, et al. Mediterranean‐DASH Intervention for Neurodegenerative Delay (MIND) study: rationale, design and baseline characteristics of a randomized control trial of the MIND diet on cognitive decline. Contemp Clin Trials. 2021;102:106270. doi: 10.1016/j.cct.2021.106270 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. McCleery J, Abraham RP, Denton DA, et al. Vitamin and mineral supplementation for preventing dementia or delaying cognitive decline in people with mild cognitive impairment. Cochrane Database Syst Rev. 2018. doi: 10.1002/14651858.CD011905.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Martins IJ, Fernando WMADB. High fibre diets and Alzheimer's disease. Food Sci. Nutr. 2014;5:410‐424. doi: 10.4236/fns.2014.54049 [DOI] [Google Scholar]
  • 40. Muth A‐K, Park SQ. The impact of dietary macronutrient intake on cognitive function and the brain. Clin Nutr. 2021;40:3999‐4010. doi: 10.1016/j.clnu.2021.04.043 [DOI] [PubMed] [Google Scholar]
  • 41. Scholey A, Owen L. Effects of chocolate on cognitive function and mood: a systematic review. Nutr Rev. 2013;71:665‐681. doi: 10.1111/nure.12065 [DOI] [PubMed] [Google Scholar]
  • 42. Stangl D, Thuret S. Impact of diet on adult hippocampal neurogenesis. Genes Nutr. 2009;4:271‐282. doi: 10.1007/s12263-009-0134-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Berkins S, Schiöth HB, Rukh G. Depression and Vegetarians: association between dietary vitamin B6, B12 and folate intake and global and subcortical brain volumes. Nutrients. 2021;13:1790. doi: 10.3390/nu13061790 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Haast RAM, Kiliaan AJ. Impact of fatty acids on brain circulation, structure and function. Prostaglandins Leukot Essent Fatty Acids. 2015;92:3‐14. doi: 10.1016/j.plefa.2014.01.002 [DOI] [PubMed] [Google Scholar]
  • 45. Wei J, Gianattasio KZ, Bennett EE, et al. The associations of dietary copper with cognitive outcomes. Am J Epidemiol. 2022;191:1202‐1211. doi: 10.1093/aje/kwac040 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

ALZ-20-1190-s001.docx (1.2MB, docx)

Supporting Information

ALZ-20-1190-s002.pdf (912.8KB, pdf)

Articles from Alzheimer's & Dementia are provided here courtesy of Wiley

RESOURCES