Abstract
INTRODUCTION
We examined whether participant characteristics, including sociodemographics, APOE ε4 allele, lifestyle factors, and cardiovascular risk factors, modified the effect of the Mediterranean–Dietary Approach to Systolic Hypertension Intervention for Neurodegenerative Delay (MIND).
METHODS
The MIND trial (NCT02817074) randomized adults 65 to 85 years who were overweight or obese to the MIND diet or control diet conditions. A linear mixed‐effects model was used to examine whether participant characteristics modified the effect of MIND on the annual change in a global cognition score derived from a battery of 12 neurocognitive tests.
RESULTS
On average, participants (N = 604) were 70.4 years old, 65.5% female, and with a body mass index (BMI) of 33.9 kg/m2. In multivariable‐adjusted models, BMI was the only significant modifier (p = 0.009); only among individuals with BMI ≥ 35 kg/m2 (n = 213) was the MIND diet group associated with higher cognitive scores by 0.040 (standard error = 0.017, p = 0.018) standardized units per year versus control.
DISCUSSION
The MIND diet may have cognitive benefits in people with obesity, particularly BMI ≥ 35.
CLINICAL TRIAL REGISTRATION NUMBER
Highlights
This study examined potential effect modifiers of the MIND diet on cognition.
BMI emerged as a significant effect modifier.
In those with BMI ≥ 35 kg/m2, the MIND diet group had higher cognitive scores.
The MIND diet may have cognitive benefits for people with obesity.
Keywords: aging, clinical trial, cognition, diet, obesity
1. BACKGROUND
Reducing the risk of Alzheimer's disease (AD) and related dementias continues to be a global public health priority 1 that includes implementing behavioral strategies that could prevent or delay onset and reduce overall burden. Healthy dietary habits have shown promise for maintaining cognitive function over time 2 , 3 , 4 and for slowing progression of mild cognitive impairment (MCI) and dementia in later life. 5 , 6 As emphasized in the recent Lancet Commission, there remains a critical need to focus on evidence‐based prevention strategies. 7 Thus, promoting specific dietary habits may be a practical prevention approach that could be widely disseminated.
The Mediterranean–Dietary Approach to Systolic Hypertension (DASH) diet intervention for neurodegenerative delay (MIND) diet trial is a recently completed 3‐year randomized controlled trial (RCT) that evaluated the effects of the MIND diet compared to a control diet on cognitive function in older adults without existing cognitive impairment who were classified as overweight or obese. The MIND diet promotes “brain‐healthy” foods that have been previously linked in observational studies to slower rates of cognitive decline in aging and prevention of Alzheimer's disease 3 , 8 while limiting consumption of “unhealthy” foods. Across both the MIND and control diet groups, these older adult participants, on average, showed improvements in global cognition scores. This could potentially be explained by participants being relatively healthy and highly educated. 9 Thus, it is prudent to investigate whether specific subgroups benefit more from the MIND diet.
Building on these trial results for cognitive function, the next logical step is to identify whether specific subgroups showed a greater cognitive benefit from the MIND diet intervention and whether these subgroups had specific characteristics. Gaining a greater understanding of specific characteristics of individuals who may benefit can potentially inform dissemination into practice using a more precise, tailored approach. This approach, rooted in a precision medicine theoretical perspective, supports providing the most effective intervention for the individual, considering differences in sociodemographics, genetics, and lifestyle. 10 For instance, our earlier work indicated that individuals at higher risk due to a genetic predisposition to AD (i.e., apolipoprotein E ε4 allele [APOE ε4 allele] carriers) could benefit more from adherence to a healthy lifestyle, including the MIND diet. 4 , 10 In the current study, we examined whether participant characteristics at baseline, including sociodemographics (age, sex, racial background, and education), APOE ε4 allele carrier status, lifestyle factors (MIND diet score, smoking, physical activity, and cognitive activity), and cardiovascular risk factors (body mass index [BMI], systolic and diastolic blood pressure [SBP and DBP], respectively], total cholesterol, high‐density lipoprotein [HDL] cholesterol, and hemoglobin A1c [HbA1c]), modified the effect of the MIND diet intervention on cognitive function.
RESEARCH IN CONTEXT
Systematic review: Few recent trials have examined whether participant characteristics modify the effects of a diet intervention for cognition function. To our knowledge, the MIND diet trial is the first trial of this size to test the MIND diet in older adults without cognitive impairment.
Interpretation: Our findings indicated that only body mass index (BMI) significantly modified the effects of the MIND diet intervention. As such, those with BMI ≥ 35 kg/m2 and were assigned to the MIND diet intervention group showed higher scores in global cognition compared to those assigned to the control group. Other than BMI, no other sociodemographic, genetic, or cardiovascular risk factors modified intervention effects of the MIND diet intervention.
Future directions: Recommendations for future research include (a) examination of underlying mechanisms that could help explain findings, (b) testing in a more diverse population that represents those with lower education levels and of racial and ethnic minority backgrounds, and (c) investigation of associations across adult age groups, including midlife.
2. METHODS
2.1. Study design
This is a secondary analysis of data from the MIND Trial (NCT02817074), a two‐site randomized controlled trial that evaluated the effects of a 3‐year diet intervention on cognitive decline in older adults who were cognitively unimpaired but at greater risk of AD due to family history. 11 Data collection occurred from March 2017 to June 2021 in the Chicago and Boston metropolitan areas. The study protocol was previously detailed. 10
The Institutional Review Boards of Rush University Medical Center, Harvard School of Public Health, and Brigham and Women's Hospital approved the MIND study protocol. All participants provided written informed consent. Limited data are available to other investigators upon submission of an approved data request.
2.2. Study participants
The MIND trial participants included 604 community‐dwelling adults 65 to 84 years of age, living in either the Boston (n = 302) or Chicago (n = 302) metropolitan area. The parent trial enrolled older adults (65 to 84 years) who were at risk for dementia but without cognitive impairment (a score of 22 or greater on the Montreal Cognitive Assessment [MoCA]), had a family history of dementia (self‐reported a first‐degree family history of dementia), were overweight or obese (25 kg/m2 or greater), and had a suboptimal diet (defined as a score of 8 or less out of 14 on the MIND diet screener, which included frequency of eating from 10 brain‐healthy food groups and five unhealthy food groups). 3 , 8 , 12
Exclusion criteria included (a) having allergies to nuts, berries, olive oil, or fish; (b) having psychosis or bipolar disorder; (c) engaging in alcohol or substance abuse within the past 6 months; (d) having unstable or recent onset of cardiovascular disease, including a stroke; (e) receiving a diagnosis of cancer within the past 5 years; (f) having gastrointestinal conditions associated with weight change (e.g., colostomy or gastric bypass surgery); (g) having a history of brain injury; (h) having a history of liver disease or hepatitis C; and (i) taking medications for AD or Parkinson's disease.
Recruitment, screening, and participant characteristics were previously detailed. 9 , 11 Participants were recruited from January 2017 through April 2018. A total of 1929 persons completed screening, and 1325 were excluded, while an additional 109 persons did not meet eligibility during the run‐in process. A total of 301 participants were assigned to the MIND diet and 303 to the control diet. By year 3, 26 participants (8.6%) assigned to the MIND diet and 14 participants (4.6%) assigned to the control diet were lost to follow‐up. Trial operations were paused for approximately 3 months (March 19, 2020, to July 15, 2020) due to the coronavirus‐19 (COVID‐19) pandemic, and all participants continued to receive the intervention or control diet food as well as dietary counseling by telephone without in‐person data collection and other contacts.
2.3. Randomization and intervention groups
Participants were randomly assigned in a 1:1 ratio to follow the MIND diet with mild caloric restriction (250 kcal) for weight loss (intervention) or their usual diet with the same mild caloric restriction for weight loss (control). As outlined previously, randomization was performed with the use of permuted blocks determined with the use of R software version 3.3 (R Project for Statistical Computing) with stratification by site, sex, and age categories (65 to 69 years, 70 to 74 years, 75 to 80 years, and 81 to 84 years). The caloric restriction across both the intervention and control groups consisted of consuming 250 kcal less per day. Study investigators and data collectors were kept blinded to the randomization scheme throughout the study duration. Participants were informed of their intervention assignment during their individual intervention orientations.
Registered dieticians conducted telephone dietary counseling with all participants at the same frequency. The MIND diet intervention consisted of promoting nine brain‐healthy food groups (green leafy vegetables, other vegetables, nuts, berries, beans/legumes, whole grains, fish, poultry, and extra virgin olive oil) and limiting consumption of five unhealthy food groups (red and processed meats, fried foods, pastries and sweets, butter, and full‐fat cheese). Dietary counseling for participants assigned to the MIND diet intervention included instructions on incorporating foods from the MIND diet and strategies to lose weight. Participants assigned to the control diet were counseled on calorie tracking, portion control, and behavioral strategies to lose weight without changing the types of foods consumed. All participants were instructed to avoid making changes to exercise routines. At the end of the 3‐year intervention, both groups experienced, on average, clinically meaningful weight loss (∼5.2% weight loss from the baseline weight). 9
2.4. Measures
2.4.1. Cognitive function outcomes
Cognitive function was assessed using a battery of 12 neurocognitive tests. 11 The tests evaluated four cognitive domains: episodic memory – recall ability (Word List Memory, Word List Recall, Word List Recognition, East Boston Memory Test, and East Boston Delayed Recall); executive function – ability to organize thoughts and activities (Trail Making Test B, Flanker Inhibitory Control); perceptual speed – ability to quickly and accurately make comparison (Trail Making Test A, Pattern Comparison, Digit Symbol Substitution Test); and semantic memory – accumulated long‐term knowledge (Verbal Fluency, Multilingual Naming Test). Raw scores of all tests were converted to z‐scores. Composite scores for each domain were calculated by averaging z‐scores of the individual tests. A score of global cognition was calculated by averaging z‐scores for the four domains. Higher scores on the global cognition composite domain‐specific scores reflect better cognitive function. Cognitive function was assessed at baseline and 6, 12, 24, and 36 months (i.e., 0, 0.5, 1, 2, and 3 years).
2.4.2. Baseline participant characteristics
Participants’ characteristics as assessed at baseline (i.e., enrollment) included demographic, genetic (APOE ε4 allele carrier vs not), lifestyle, and cardiovascular risk factors, which were examined as potential intervention modifiers. Demographics included age, sex, racial background, and education.
The genetic risk factor examined was presence of the APOE ε4 allele. Sanger sequencing was used to detect rs429358 and rs7412 variants in exon 4 of the human apolipoprotein E gene. The sequencing results had 100% concordance with the published genotypes of the samples. An indicator variable was created based on the two single nucleotide polymorphisms for participants with one or more copies of the APOE ε4 risk allele.
Lifestyle risk factors included the MIND diet score, smoking history, physical activity, and cognitive activity. Assessment of adherence to the MIND diet was determined as regular consumption of 14 dietary components, with a MIND diet score ranging from 0 to 14, assessed by a validated food frequency questionnaire. Higher scores reflect higher intake of olive oil, fish (not fried), whole grains, berries, green leafy vegetables, other vegetables, nuts, beans, and poultry (not fried, skinless) with lower intake of butter and margarine, cheese, red meat and meat products, fast and fried foods, and pastries and sweets. 8 , 13 Smoking history was a dichotomous variable (ever smoked vs no history of smoking). Self‐reported physical activity was assessed by the Yale Physical Activity Survey validated for older adults that measures number of hours of household chores and leisure and recreation activities. 14 Self‐reported cognitive activity was self‐reported frequency on a five‐point scale of weekly participation in nine information processing activities with minimal physical or social demands, as previously validated in a population‐based study of older adults. 15 , 16
Cardiovascular risk factors included BMI, SBP, DBP, total and HDL cholesterol levels, and HbA1c. BMI was calculated based on weight and height. SBP and DBP levels were the average of two sitting measures at rest. Total cholesterol levels (mg/dL) from blood samples were measured using the Beckman Coulter AU5800 series automated chemistry analyzer with the Beckman Coulter cholesterol reagent. HDL cholesterol levels (mg/dL) were measured by the supernatants only containing HDL particles using the Beckman Coulter cholesterol reagent after precipitating all apoB‐containing lipoprotein particles using 50 kDa dextran sulfate with magnesium ions. 17 HbA1c as expressed as a percentage was assessed by high‐performance liquid chromatography (HPLC) using the Tosoh G8 automated glycohemoglobin analyzer.
2.5. Statistical analyses
Participant characteristics are summarized as mean (m) and standard deviation (SD), median and interquartile range (IQR), or as number (n), and percentages (%) of participants. Linear mixed‐effects models with random intercept and slope were used to investigate whether the baseline participant characteristics influenced the MIND intervention effects on annual changes in global cognition.
Two consecutive analyses were conducted. First, to identify which baseline characteristics modified the association between the MIND diet and global cognition, a three‐way interaction term was included in the model: intervention group (MIND diet vs control), follow‐up time (i.e., 0, 0.5, 1, 2, and 3 years), and the baseline risk factor variable of interest. These variables were included as continuous, except for sex, racial background, APOE ε4 allele, and smoking, which were added as dichotomous. Variables including BMI, total cholesterol, HDL cholesterol, and HbA1c were log‐transformed to improve distribution and to make potential outliers less influential. Second, for variables that showed a significant interaction (i.e., p < 0.05 for interaction: group × time × baseline characteristic), we conducted a stratified analysis to evaluate in which strata (i.e., groups) the effect of the MIND diet intervention on global cognition differed from the control diet. Additionally, for variables that showed a significant interaction, we conducted supplemental analyses to examine associations with domain‐specific cognition, including episodic memory, executive function, perceptual speed, and semantic memory.
3. RESULTS
Table 1 displays the baseline characteristics of participants. Participants (N = 604) were, on average, 70.4 years old, 65.5% female, and 87.7% of White racial background and had 16.9 years of education. A total of 174 participants (28.9%) were APOE ε4 allele carriers. On average, participants had a BMI of 33.9 kg/m2, with levels of SBP considered in the elevated range (132.1 mmHg) but DBP in the normal range (77.0 mmHg) and normal ranges of cholesterol (195.0 and 59.0 for total and HDL cholesterol, respectively) and HbA1c (5.5%). As noted in previously published manuscripts, baseline characteristics across the MIND diet intervention and control groups were similar, apart from participants in the intervention group having a slightly lower proportion of APOE ε4 allele carriers versus the control group (25.2% vs 32.3%). 9
TABLE 1.
Baseline characteristics of 604 older adult participants by intervention arms.
| Variable | Overall (n = 604) | Control diet group (n = 303) | MIND diet group (n = 301) |
|---|---|---|---|
| Demographic | |||
| Age, years | 70.4 (4.2) | 70.4 (4.2) | 70.4 (4.2) |
| Sex, male, n (%) | 211 (34.9) | 106 (35.0) | 105 (34.9) |
| Race, White, n (%) | 530 (87.7) | 267 (88.1) | 263 (87.4) |
| Education, years | 16.9 (2.6) | 17.0 (2.6) | 16.9 (2.7) |
| Genetic | |||
| APOE ε4 allele carrier, n (%) | 174 (28.9) | 98 (32.6) | 76 (25.2) |
| Lifestyle | |||
| MIND diet, score | 7.7 (1.8) | 7.8 (1.8) | 7.7 (1.9) |
| Ever smokers, n (%) | 325 (53.8) | 166 (54.8) | 159 (52.8) |
| Physical activity, h/week | 2.3 [0.5, 4.7] | 2.4 [0.8, 5.0] | 2.2 [0.5, 4.5] |
| Cognitive activity, score | 3.5 (0.6) | 3.5 (0.6) | 3.5 (0.7) |
| Cardiovascular | |||
| BMI, kg/m2 | 33.9 (6.0) | 34.0 (6.5) | 33.8 (5.4) |
| SBP, mmHg | 129.7 (17.5) | 128.3 (17.2) | 131.1 (17.7) |
| DBP, mmHg | 76.0 (10.1) | 75.5 (10.1) | 76.6 (10.2) |
| Total cholesterol, mg/dL | 189.0 [162.0, 216.0] | 186.0 [160.0, 212.8] | 190.0 [162.0, 218.0] |
| HDL cholesterol, mg/dL | 54.0 [44.0, 67.0] | 55.0 [44.0, 68.8] | 53.0 [44.0, 64.0] |
| Hemoglobin A1c, % | 5.5 [5.3, 5.8] | 5.5 [5.3, 5.8] | 5.6 [5.3, 5.8] |
Abbreviation: BMI, body mass index; DBP, diastolic blood pressure; HDL, high‐density lipoprotein; h, hours; MIND, Mediterranean–DASH Intervention for Neurodegenerative Delay; SBP, systolic blood pressure.
Results of the linear mixed‐effects models to examine whether the baseline characteristics modified the MIND intervention effects on annual changes in global cognition score (group × time × baseline characteristic) are displayed in Table 2. The only significant modifying effect was found for BMI (β = 0.165, 95% confidence interval [CI]: [0.041, 0.288], p = 0.009), which could be interpreted as the average annual change in global cognition per 10% increase in BMI; this differed by 0.016 standardized units between dietary groups, with people in the MIND diet group having higher cognitive scores than those in the control diet group. No other baseline characteristics emerged with significant modifying effects (i.e., intervention effects on cognitive function did not vary by other baseline characteristics; p > 0.05).
TABLE 2.
Influence of demographic, genetic, lifestyle, and cardiovascular risk factors on the effect of MIND diet intervention on global cognition.
| Variable | β | 95% CI | p |
|---|---|---|---|
| Demographic | |||
| Age | −0.001 | −0.006, 0.003 | 0.549 |
| Sex | 0.022 | −0.021, 0.064 | 0.311 |
| Race | −0.049 | −0.111, 0.012 | 0.118 |
| Education | −0.003 | −0.011, 0.005 | 0.457 |
| Genetic factors | |||
| APOE ε4 allele carrier | 0.011 | −0.034, 0.056 | 0.640 |
| Lifestyle factors | |||
| MIND score | 0.000 | −0.011, 0.011 | 0.996 |
| Ever smokers | 0.033 | −0.007, 0.073 | 0.110 |
| Physical activity | −0.007 | −0.033, 0.019 | 0.583 |
| Cognitive activity | −0.017 | −0.051, 0.017 | 0.322 |
| Cardiovascular risk factors | |||
| BMI | 0.165 | 0.041, 0.288 | 0.009 |
| Systolic blood pressure | −0.001 | −0.002, 0.001 | 0.310 |
| Diastolic blood pressure | −0.001 | −0.003, 0.001 | 0.250 |
| Total cholesterol | 0.016 | −0.08, 0.112 | 0.746 |
| HDL cholesterol | −0.008 | −0.077, 0.062 | 0.824 |
| Hemoglobin A1c | −0.020 | −0.198, 0.158 | 0.828 |
Note: Each table row represents a different model testing the three‐way interaction (intervention group x time x risk factor).
Abbreviation: BMI, body mass index; HDL, high‐density lipoprotein; h, hours; MIND, Mediterranean–DASH Intervention for Neurodegenerative Delay.
To further understand these results, specifically the role of BMI, multivariable adjusted linear mixed‐effects models with random intercept and slope were conducted to estimate the effect of the MIND diet intervention on annual changes in global cognition score according to baseline BMI categories (Table 3). BMI categories 25 to 29.9 (overweight), 30 to 34.9 (Class 1 obesity), and ≥ 35 kg/m2 (Class 2 and 3 obesity) were chosen based on World Health Organization, American Heart Association, and American College of Cardiology obesity classifications. 18 , 19 Models were adjusted by age, sex, racial background, education, APOE ε4 allele, and physical activity. Participants (n = 213) with BMI ≥ 35 kg/m2 (i.e., Class 2 and 3 obesity) assigned to the MIND diet intervention showed higher scores in global cognition by 0.040 (SE = 0.02, p = 0.018) standardized units per year versus participants assigned the control diet. Figure 1 shows predicted trajectories of global cognition for individuals with different BMI levels in the MIND diet intervention and control diet groups.
TABLE 3.
Effect of MIND diet intervention on global cognition across body mass index (BMI) categories.
| BMI 25 to 29.9 (n = 179) | BMI 30 to 34.9 (n = 212) | BMI ≥ 35 (n = 213) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Assignment | n | β | 95% CI | p | n | β | 95% CI | p | n | β | 95% CI | p |
| Control diet group | 93 | 0.000 | Reference | 100 | 0.000 | Reference | 110 | 0.00 | Reference | |||
| MIND diet group | 86 | 0.003 | −0.035, 0.041 | 0.865 | 112 | −0.003 | −0.039, 0.032 | 0.852 | 103 | 0.04 | 0.007, 0.073 | 0.018 |
Note: Models were adjusted by age, sex, racial background, education, APOE ε4 allele, and physical activity.
Abbreviation: BMI, body mass index; MIND, Mediterranean–Dietary Approach to Systolic Hypertension (DASH) diet intervention for neurodegenerative delay.
FIGURE 1.
Predicted trajectories of global cognition for individuals with different body mass index levels in the MIND diet intervention and control diet groups.
To further explore the characteristics of the study population, we conducted a supplementary analysis examining baseline characteristics across the three BMI classes. In addition, to better understand and corroborate the statistically significant effect of the MIND intervention on cognition among participants with BMI ≥ 35 kg/m2, we compared the baseline characteristics of individuals in the MIND diet group with those in the control diet group. Compared to participants with BMI 25 to 29.9 or BMI 30 to 34.9 kg/m2, respectively, participants with a BMI ≥ 35 kg/m2 were significantly younger, with fewer years of education, lower MIND diet score, fewer weekly hours of physical activity, lower cognitive activity score, lower HDL cholesterol, and higher HbA1c (Table S1). There were no significant differences in baseline characteristics between the MIND diet intervention and control diet groups in individuals with BMI ≥ 35 kg/m2 (Table S2). Finally, we repeated our primary analysis using cognitive assessments only at baseline and at year 3, corresponding to study entry and exit, to mitigate potential uncertainty associated with annual assessment intervals and non‐linear relationships. These findings supported our primary results, which showed that BMI was the only modifier across demographics, genetic, lifestyle, and cardiovascular risk factors in the association between the MIND diet intervention and cognition (Table S3). They also supported the finding that the MIND diet intervention was significantly associated with cognition in individuals with BMI ≥ 35 kg/m2 (Table S4). We further examined their respective association with domain‐specific cognition, including episodic memory, executive function, perceptual speed, and semantic memory. We found that, compared to the control, the MIND diet intervention was associated with higher scores in executive function and semantic memory (Table S5).
4. DISCUSSION
The MIND diet trial compared the effects of the MIND diet intervention with a control diet – both groups had caloric restriction – on cognitive function in cognitively unimpaired older adults. We expanded on the main trial findings that showed similar cognitive benefits after 3 years across both groups to further investigate if there were specific characteristics of participants that benefitted more from the MIND diet intervention. In the present study, there was a significant interaction effect of BMI with the intervention, meaning that the effect of MIND diet intervention on the global cognition score differed by BMI. In individuals with BMI > 35 kg/m2, those in the MIND diet intervention group showed higher scores in global cognition during the intervention period when compared to the control group.
Given that the MIND diet trial was one of the first large trials to test a “brain‐healthy” diet intervention in cognitively unimpaired older adults, studies examining the modifying effects of participant characteristics on cognitive function are still emerging. The landmark Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) trial is the first large, long‐term RCT that showed cognitive benefits of a multidomain lifestyle intervention that included dietary guidance. In a series of supplemental analyses, the FINGER intervention showed benefits to cognitive function despite participant differences in background characteristics, including demographic, cardiovascular (including BMI), and frailty risk factors. 20 , 21 Similarly, there were no significant modifying effects of demographic, cardiovascular, and frailty risk factors in the NU‐AGE diet trial (individually tailored Mediterranean‐like diet advice). 22 In meta‐analyses of nine large multidomain lifestyle interventions for the prevention of dementia and cognitive decline, there was high‐certainty evidence of a greater benefit of the intervention for those with increased genetic risk (i.e., presence of the APOE ε4 allele). 23 This was consistent with a finding in a longitudinal study of older adults that also showed APOE ε4 as a significant moderator on the association of dietary habits and cognitive function over time. 24 However, the presence of the APOE ε4 allele did not emerge as a significant effect modifier in our study of the MIND diet trial. Compared to these earlier studies, a key difference in the MIND diet trial is that clinically meaningful weight loss was prescribed in both the MIND diet intervention and control diet, which may have important implications for overall health.
Multifaceted, interconnected factors may help explain our significant findings for BMI. There is a known influence of obesity on cognitive function and risk of Alzheimer's Disease and Related Dementias (ADRD), with overlapping yet distinct mechanisms, and likely more that require additional investigation. 25 To further understand the positive effect of the MIND intervention on cognition in individuals with BMI ≥ 35, we stratified across the three BMI classes. The results indicated that, compared to people with BMI < 35, those with BMI > 35 were more likely to be younger, have lower levels of physical activity and cognitive activity, and have lower HDL and higher HbA1c levels. Most of these risk factors (i.e., physical activity, cognitive activity, and cardiovascular health) have been associated with a faster rate of cognitive decline and an increased risk of ADRD. 16 , 26 , 27 This may suggest that, cumulatively, risk due to these factors could amplify any potential cognitive benefits of the MIND diet. Other than BMI, no other sociodemographic, genetic, or cardiovascular risk factors modified intervention effects of the MIND diet intervention. Nevertheless, additional investigation is needed across the complex obesity‐related mechanistic pathways across the lifespan.
This study has limitations. First, the MIND diet trial enrolled older adult participants without cognitive impairment at baseline, and this sample was largely of White racial background with high levels of education. Thus, findings may not be generalizable to other aging populations. Second, this analysis did not include other anthropometric measurements (e.g., waist circumference) or measures of body composition that may elucidate the influence of muscle and fat mass on outcomes versus examining BMI alone, including body fat percentage or muscle mass. Third, this sample included individuals of ages 65 to 85, so it did not capture the midlife age group when obesity may have significant effects on neuropathology and overall brain health. Fourth, the cognitive benefits that emerged in our analyses were relatively modest. However, reducing the risk of future cognitive decline, including just maintaining cognition over time, has been shown to be a meaningful clinical benefit as a result of lifestyle interventions. 28 , 29 Moreover, we acknowledge the possibility that weight loss may interact with dietary interventions and exert a synergistic effect, potentially enhancing the cognitive outcomes observed with the MIND diet.
Despite these limitations, this study contributes to the current state of science of lifestyle interventions by examining the modifying effects of participant background characteristics in the MIND diet intervention on cognitive function. Our findings indicated that, in the MIND trial, older adult participants with BMI greater than 35 may have unique cognitive benefits as a result of a diet focused on “brain‐healthy” foods.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest. Author disclosures are available in the Supporting Information.
CONSENT STATEMENT
All participants provided written informed consent prior to data collection.
Supporting information
Supporting information
Supporting information
ACKNOWLEDGMENTS
The authors thank all the participants and the research staff of the MIND trial. The authors would like to honor the memory of Martha Clare Morris, ScD, whose pioneering work in nutrition and dementia science laid the foundation for this research. She was the co‐creator of the MIND diet and original principal investigator of the MIND diet trial. The authors also thank Kevin Grandfield, Publication Manager for the UIC Department of Biobehavioral Nursing Science, for editorial assistance. The MIND trial was funded by the National Institute on Aging grant R01AG52583. The funding organizations had no role in the design or conduct of the study; the collection, management, analysis, or interpretation of the data; or the writing of the report or the decision to submit it for publication.
Halloway S, Aggarwal N, Arfanakis K, Sacks FM, Barnes LL, Dhana K. Effect modifiers of the MIND diet for cognition in older adults: The MIND diet trial. Alzheimer's Dement. 2025;21:e70731. 10.1002/alz.70731
REFERENCES
- 1. 2024 Alzheimer's disease facts and figures. Alzheimer's Dement 2024;20:3708‐3821. doi: 10.1002/alz.13809 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Tussing‐Humphreys L, Lamar M, McLeod A, et al. Effect of Mediterranean diet and Mediterranean diet plus calorie restriction on cognition, lifestyle, and cardiometabolic health: a randomized clinical trial. Preventive Medicine Reports. 2022;29:101955. doi: 10.1016/j.pmedr.2022.101955 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Morris MC, Tangney CC, Wang Y, et al. MIND diet slows cognitive decline with aging. Alzheimer's Dement. 2015;11:1015‐1022. doi: 10.1016/j.jalz.2015.04.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Dhana K, Barnes LL, Liu X, et al. Genetic risk, adherence to a healthy lifestyle, and cognitive decline in African Americans and European Americans. Alzheimer's Dementia. 2022;18:572‐580. doi: 10.1002/alz.12435 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lissek V, Suchan B. Preventing dementia? interventional approaches in mild cognitive impairment. Neurosci Biobehav Rev. 2021;122:143‐164. doi: 10.1016/j.neubiorev.2020.12.022 [DOI] [PubMed] [Google Scholar]
- 6. Singh RB, Watanabe S, Li D, et al. Diet and lifestyle guidelines and desirable levels of risk factors and protective factors for prevention of dementia—a scientific statement from joint symposium of jaas and apcns. Biomed J Sci Technical Res. 2019;17:12844‐12864. [Google Scholar]
- 7. Livingston G, Huntley J, Liu KY, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. The Lancet. 2024;0:572‐628. doi: 10.1016/S0140-6736(24)01296-0 [DOI] [PubMed] [Google Scholar]
- 8. Morris MC, Tangney CC, Wang Y, Sacks FM, Bennett DA, Aggarwal NT. MIND diet associated with reduced incidence of Alzheimer's disease. Alzheimer's Dement. 2015;11:1007‐1014. doi: 10.1016/j.jalz.2014.11.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. 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;389:602‐611. doi: 10.1056/NEJMoa2302368 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Dhana K, Aggarwal NT, Rajan KB, Barnes LL, Evans DA, Morris MC. Impact of the Apolipoprotein E ε4 Allele on the relationship between healthy lifestyle and cognitive decline: a population‐based study. Am J Epidemiol. 2021;190:1225‐1233. doi: 10.1093/aje/kwab033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. 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]
- 12. Morris MC. Nutrition and risk of dementia: overview and methodological issues. Ann NY Acad Sci. 2016:1367. doi: 10.1111/nyas.13047 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Morris MC, Tangney CC, Bienias JL, Evans DA, Wilson RS. Validity and reproducibility of a food frequency questionnaire by cognition in an older biracial sample. Am J Epidemiol. 2003;158:1213‐1217. doi: 10.1093/aje/kwg290 [DOI] [PubMed] [Google Scholar]
- 14. Dipietro L, Caspersen CJ, Ostfeld AM, Nadel ER. A survey for assessing physical activity among older adults. Med Sci Sports Exerc. 1993;25:628‐642. [PubMed] [Google Scholar]
- 15. Wilson RS, Bennett DA, Bienias JL, et al. Cognitive activity and incident AD in a population‐based sample of older persons. Neurology. 2002;59:1910‐1914. [DOI] [PubMed] [Google Scholar]
- 16. Wilson RS, Bennett DA, Beckett LA, et al. Cognitive activity in older persons from a geographically defined population. J Gerontol B Psychol Sci Soc Sci. 1999;54:P155‐P160. [DOI] [PubMed] [Google Scholar]
- 17. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) . Third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel iii) final report. Circulation 2002;106:3143‐3421. [PubMed] [Google Scholar]
- 18. Gadde KM, Martin CK, Berthoud H‐R, Heymsfield SB. Obesity. JACC. 2018;71:69‐84. doi: 10.1016/j.jacc.2017.11.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Jensen MD, Ryan DH, Apovian CM, et al. 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults. JACC. 2014;63:2985‐3023. doi: 10.1016/j.jacc.2013.11.004 [DOI] [PubMed] [Google Scholar]
- 20. Rosenberg A, Ngandu T, Rusanen M, et al. Multidomain lifestyle intervention benefits a large elderly population at risk for cognitive decline and dementia regardless of baseline characteristics: the FINGER trial. Alzheimer's Dement. 2018;14:263‐270. doi: 10.1016/j.jalz.2017.09.006 [DOI] [PubMed] [Google Scholar]
- 21. Pöyhönen J, Lehtisalo J, Roitto H‐M, et al. Impact of frailty status on the effect of a multidomain lifestyle intervention on cognition. Age Ageing. 2025;54:afaf041. doi: 10.1093/ageing/afaf041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Marseglia A, Xu W, Fratiglioni L, et al. Effect of the NU‐AGE Diet on cognitive functioning in older adults: a randomized controlled trial. Front Physiol. 2018;9. doi: 10.3389/fphys.2018.00349 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Hafdi M, Hoevenaar‐Blom MP, Richard E. Multi‐domain interventions for the prevention of dementia and cognitive decline. Cochrane Database Syst Rev. 2021;2021:CD013572. doi: 10.1002/14651858.CD013572.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Keum M, Lee BC, Choe YM, et al. Protein intake and episodic memory: the moderating role of the apolipoprotein E ε4 status. Alzheimer's Research & Therapy. 2024;16:181. doi: 10.1186/s13195-024-01546-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Nguyen JCD, Killcross AS, Jenkins TA. Obesity and cognitive decline: role of inflammation and vascular changes. Front Neurosci. 2014;8:375. doi: 10.3389/fnins.2014.00375 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Halloway S, Dhana K, Desai P, et al. Free‐living standing activity as assessed by seismic accelerometers and cognitive function in community‐dwelling older adults: the MIND trial. J Gerontol A Biol Sci Med Sci. 2021;76:1981‐1987. doi: 10.1093/gerona/glab106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Dhana A, DeCarli CS, Dhana K, et al. Cardiovascular health and cognitive outcomes: findings from a biracial population‐based study in the United States. Alzheimers Dement. 2023;19:4446‐4453. doi: 10.1002/alz.13421 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Baker LD, Espeland MA, Whitmer RA, et al. Structured vs self‐guided multidomain lifestyle interventions for global cognitive function: the US POINTER randomized clinical trial. JAMA. 2025;334(8):681‐691. doi: 10.1001/jama.2025.12923 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Stephen R, Barbera M, Peters R, et al. Development of the first WHO guidelines for risk reduction of cognitive decline and dementia: lessons learned and future directions. Front Neurol. 2021;12:763573. doi: 10.3389/fneur.2021.763573 [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
Supporting information