Feasibility of the MIND+SOUL Culturally Tailored Brain Healthy Diet: A Pilot Study

Racial Disparities in AD/Impact on Black Community

It is estimated that 6.9 million Americans aged 65 and older are affected by Alzheimer’s disease (AD), a figure projected to triple by 2060 due to the aging population and rising life expectancy rates in the United States (U.S.).^1,2 However, this prevalence varies significantly among racial groups within the U.S. Specifically, Black Americans are two to three times more likely to develop Alzheimer’s disease (AD) compared to non-Hispanic Whites. ³ Health disparities, particularly prevalent within Black communities are believed to contribute significantly to the elevated dementia risk observed in this population. Structural racism can both directly and indirectly impact the risk for AD by shaping social determinants of health, including access to quality education, housing, health care, employment opportunities, and exposure to environmental toxins.^4-8 Additionally, structural racism affects healthy dietary consumption by limiting access to affordable, nutritious foods through the creation of food deserts and food swamps in predominantly Black neighborhoods. This lack of access to healthy foods contributes to poor dietary patterns, increasing the risk of conditions like hypertension, diabetes, and cardiovascular disease, all of which are associated with a higher risk of AD.^9,10 While important to note that race is a socially constructed concept lacking significant genetic or biological foundation, these systemic issues contribute to racial and ethnic disparities in health outcomes, including an increased risk of AD among Black adults.^9,10

Diet and AD

Although age and genetic makeup are nonmodifiable risk factors for dementia, 40% of cases are due to modifiable factors such as lifestyle and diet. ¹¹ Heart-healthy diets, such as the Mediterranean Diet (MeDi), Dietary Approaches to Stop Hypertension (DASH), and the Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet, emphasize consumption of fruits, vegetables, whole grains, healthy fats, poultry, beans, and nuts, while limiting sugar, red meats, and saturated fats. ¹² The MIND diet, combining elements of MeDi and DASH, focuses on neuroprotective foods like berries and leafy greens and has been linked to reduced AD risk and slower cognitive decline.^13-16 Adhering to the MIND diet has been associated in observational studies with a cognitive decline rate equivalent to being 7.5 years younger, with improvements in cognitive domains by 30%-78%. ¹⁷ However, a recent trial found no significant cognitive change over a three year period, possibly due to selection bias as participants tended to be more educated, had healthier medical histories, and had higher baseline MIND diet scores compared to those in other studies where the MIND diet showed benefits.^17,18 Despite the findings, a more recent systematic review supports the MIND diet’s protective effects on global cognition. ¹⁹

The MIND diet is rich in nutrients critical for supporting brain health. Berries and vegetables provide polyphenols and antioxidants, nuts and olive oil offer vitamin E with anti-inflammatory and vascular health benefits, and fish reduce amyloid-β formation and inflammation due to omega-3 fatty acids.^20-22 Leafy greens contain flavonoids and carotenoids that may influence cognitive function indirectly by modulating neuronal receptors. ²³ Additionally, whole grains, poultry, and leafy greens supply B vitamins essential for regulating homocysteine levels. ²⁴ Preventive dietary interventions are most effective when implemented before initial dementia symptoms, typically around age 50, when nutritional status and cognitive function can compensate for neuropathological changes. ²⁵ However, much research on AD prevention focused on diet has not adequately represented populations disproportionately affected by the disease, limiting the generalizability of findings to underrepresented racial and ethnic groups in aging research. ²⁶

Barriers to Healthy Dietary Patterns Among Black Americans

Previous research indicates that studies aimed at promoting behavioral change in diverse cohorts have, in certain studies resulted in lower levels of adherence among Black Americans. ²⁷ Diet-related disparities are intricate and influenced by various demographic (e.g., age, race/ethnicity, gender), psychosocial factors (e.g., self-efficacy and social support), and systematic social determinants of health (structural racism, education quality, economic stability).^28,29 Additionally, it is important to recognize that Black Americans are not a homogeneous group. Variations in food choices, religious practices, health beliefs, and socioeconomic conditions exist across this population, which can further influence dietary behaviors. Studies suggest that the perceived absence of social support, social determinants of health (e.g., perceived affordability of healthy foods and limited access to quality food options), significant cultural influences on food preferences and preparation, together contribute to reduced adherence and acceptability seen among Black Americans.^27,29-32 Cultural tailoring of interventions is a vital approach to bridging the health disparities gap. Specifically, previous studies have suggested that modifying traditional soul food to align better with nutritional guidelines would be more effective than advocating for the elimination of such foods from a healthy diet altogether. ³³ While several behavioral interventions tailored to Black American populations have shown improvements in reduction in cardiovascular risk factors related to AD (e.g., weight loss) and dietary behaviors, there has been limited research focused on cultural adaptation with respect to dietary AD prevention interventions within Black American communities.^34,35

Community Engaged Research Interventions for Prevention, Reducing Health Disparities, and as a Means to Address Structural Disparities and Inequalities

Community engaged research (CEnR) is increasingly recognized as a critical approach to address health disparities. This asset-based approach values the experiences and knowledge of communities facing inequalities, marking a significant paradigm shift in scientific practice. ³⁶ CEnR actively involves underrepresented communities in various forms, such as collaborating to formulate research questions and design and disseminate studies.^36-38 Community leaders and minority serving organizations play pivotal roles by leveraging their credibility and longstanding partnerships to engage communities meaningfully in research. ³⁷ CEnR combined with cultural tailoring has shown positive outcomes in dietary interventions focused on Black communities. For example, a 12-week culturally tailored plant-based intervention for rural Black adults resulted in significant reductions in weight (9.6%), total cholesterol (10.9%), low-density lipoprotein (13.9%), and C-reactive protein (25.8%). ³⁹ Similarly, a 9-month community faith-based diet intervention among Black Americans saw increased vegetable consumption in 47% of participants, improved overall diet quality, and reduced blood pressure. ⁴⁰ Despite these successes, culturally tailored dietary interventions targeting brain health in Black Americans remain limited.

A critical step in developing such interventions is to assess feasibility and acceptability prior to proceeding to a large-scale replication. ⁴¹ The primary aim of this study was to assess feasibility and acceptability of a community informed, culturally tailored brain healthy dietary intervention for older Black Americans aged 55 years and older. The secondary, which was exploratory in nature, was aimed to assess changes in cardiometabolic risk, body composition, nutritional health status, and cognitive function.

Study Design

This was a single arm study conducted to assess feasibility and acceptability of the adapted brain healthy diet intervention. The secondary aim explored changes in cognitive functions associated with the MIND + SOUL diet. Given that the primary aim of this study was feasibility and acceptability, no sample size calculations were conducted. However, it is important to note that pilot studies that have sample sizes of at least 12 yield a reliable estimate of standard deviation, with a probability of less than 20% from the power to fall below 60%. ⁴²

Formative Work

The design and development of the MIND + SOUL intervention was informed by the capability, opportunity, motivation, and behavioral (COM-B) framework, ⁴³ MIND dietary model,^18,44 extensive preliminary research focused on culturally sensitive nutrition interventions among Black adults,^45-49 and a non-interventional mixed methods study that explored knowledge and beliefs that influence dietary practices among older Midwestern Black Americans to inform a culturally tailored brain healthy diet intervention. ⁵⁰ In our former CEnR work using focus group discussions, the MIND diet emerged as the most favored and feasible dietary model due to its food variety and alignment with practices common among members of the Black community, compared to the MeDi and DASH diets. ⁵⁰ Participants expressed the need for modifications in the MIND dietary model to increase acceptability (e.g., increasing recommended butter and cheese amounts and offering plant-based alternatives). Additionally, there was a consensus among participants for a culturally adapted intervention that incorporates Christian faith components (e.g., scripture) and ongoing education with resources such as cooking classes, simple recipes, and preparation videos to better reflect the cultural practices common among members of the Black community. ⁵⁰ To develop the MIND + SOUL intervention ongoing discussions between the principal investigator, three academic research experts (neurology, community-based research, dietetics), a registered dietician, and three community leaders took place over a six-month period. All aspects of the intervention were thoroughly examined and discussed to ensure its scientific and cultural suitability, resulting in only minimal adjustments to the intervention protocol.

MIND + SOUL Intervention

The MIND + SOUL diet builds upon the foundations of the MIND diet by incorporating a holistic approach to healthy eating encompassing aspects of the mind, body, and spirit (Figure 1). This dietary approach is designed to align with the cultural dietary practices in the Black community and is made feasible by food variability. The MIND + SOUL intervention incorporates education, skill-building cooking classes, health coaching, and financial support as part of its holistic approach.OPEN IN VIEWER

Education and skill-building cooking classes were conducted in group sessions over six consecutive weeks as part of the MIND + SOUL diet intervention and held at the University of Kansas Medical Center. Each of the six sessions (Table 1) lasted 60 minutes and was facilitated by a registered dietitian, a white female with prior experience working with this population. The behavioral strategies shared during these sessions, aimed at enhancing dietary consumption and adherence to the MIND + SOUL diet, were based on the principles of the COM-B framework. Specifically, participants were provided with education and skills to enhance both psychological and physical skills to improve dietary consumption and adherence to the MIND + SOUL diet. Furthermore, participants received various educational resources during the first six weeks, including sample weekly menu plans, recipe cards, culturally tailored recipe books and handouts, grocery shopping guides, Christian faith-related materials such as culturally adapted devotional books and scripture readings focusing on healthy eating, affirmations for healthy eating, and a manual outlining the MIND + SOUL diet goals.

OPEN IN VIEWER

Table 1.

Demographics of Participants in the MIND + SOUL Pilot Study.

Baseline Characteristics	N = 29
Age	70.3 (6.69)
Years of education	21.3 (3.11)
Gender
Male	3 (10.3%)
Female	26 (89.7%)
Marital status
Single (never married)	5 (17.2%)
Married	11 (37.9%)
Widowed	6 (20.7%)
Divorced	7 (24.1%)
Separated	0 (0.0%)
Primary race
Black or African American	29 (100.0%)
Additional race
American Indian or Alaska Native	2 (6.9%)
Not reported	27 (93.1%)
Ethnicity
Hispanic	0 (0.0%)
Not Hispanic or Latino	28 (96.6%)
Not reported	1 (3.4%)

Health coaching sessions were conducted by two Black American community health coaches, one female and one male, both of whom had prior experience serving in the Black community. Coaches offered ongoing support and continuing education tailored to the MIND + SOUL diet which was guided by the Telephonic Health Coaching Intervention Toolkit; an established process that promotes healthy dietary behaviors. ⁵¹ During these sessions, participants received personalized education and recommendations based on their individual needs, preferences, and local circumstances. The health coaches assisted participants in setting dietary goals aligned with the MIND + SOUL diet, aided in developing action plans for various scenarios (e.g., holiday parties and vacations), and provided encouragement to foster the adoption of healthy eating habits. All health coaching sessions were conducted via Zoom online platform, and occurred bi-weekly for 20 minutes over 12 weeks (totaling six sessions per participant).

To enhance adherence to the MIND + SOUL intervention, each participant received weekly groceries valued at up to $50 per week (with an average total of $600) for a total of 12 weeks. These groceries were exclusively comprised of items from the specific food groups in the MIND + SOUL diet. Groceries were selected each week by the research team from a community supermarket.

Recruitment and Participants

We employed both purposive and convenience sampling techniques to recruit cognitively normal older Black adults, aged 55 years and older, with 1 or more cardiovascular risk factors from a Midwestern metropolitan area between June 2022 and August 2023.^52,53 Our recruitment strategies included utilization of existing outreach and recruitment registries, distribution of flyers by community partners, and leveraging an email list server established through previous outreach efforts by the PI. Those interested in participating contacted the research staff via phone and underwent a brief telephone screening to determine eligibility.

The study team employed a brief phone pre-screening interview guide (32-item) to assess eligibility prior to scheduling in-person baseline assessments. Study eligibility requirements included (1) self-identifying as African American or Black (2) 55 years and older in age (3) English speaking (4) 1 or more of the following cardiovascular risk factors (high blood pressure, high cholesterol, controlled type 2 diabetes overweight/obesity with a BMI of 25 or more as calculated from weight and height at baseline visit) (4) cognitively normal with an Eight-item Informant Interview to Differentiate Aging and Dementia (AD8) < 2 (normal cognition), ⁵⁴ (5) no active depression as determined by a score of below 4 on the Patient Health Questionnaire (PHQ-9). AD8 was selected due to its ability to identify early cognitive changes associated with dementia-related conditions like AD with high sensitivity (> 84%), where a score below 2 on the AD8 indicates normal cognitive function. ⁵⁵ PHQ-9 was selected due to it acceptability and ease among patients in previous studies and its modest performance in dementia diagnosis. ⁵⁶

If deemed potentially eligible based upon completion of a pre-screening interview, individuals were invited to a baseline visit in which study staff followed a guided script that explained the study’s objectives, potential risks, benefits, and addressed any outstanding questions individuals may have had. Written informed consent was obtained from each of the participants prior to the start of baseline assessments. At the baseline visit, assessments were conducted to confirm the one or more cardiovascular conditions for eligibility, defined as high blood pressure, high cholesterol, controlled type 2 diabetes overweight/obesity with a BMI of 25 as calculated from weight and height. If an individual was considered ineligible based on measured CV outcomes, they were promptly notified and withdrawn from the study. Eligible participants were subsequently invited to complete study assessments at the University of Kansas Medical Center. Individuals were excluded from participation in the study for the following reasons (1) no CVD risk factor (2) score above a 2 on AD8 assessment and (3) dietary restriction.

Data Collection

All study visits were conducted by trained study staff, including individuals who identified as Black American and White, all of whom had prior experience working with this population. The visits took place in a dedicated private room at the University of Kansas Medical Center. Specifically, all study staff were trained in clinical protocols, data collection, data entry, confidentiality, and safety procedures. Each study visit lasted ∼2.5 hours.

Blood Draw

Participants were instructed to abstain from consuming any food or beverages except water for 12 hours prior to their scheduled appointment. Confirmation of fasting status was obtained from the participant prior to the blood draw. Blood draws were administered to gather cardiometabolic measurements including lipid profile (cholesterol, low-density lipoprotein, high-density lipoprotein, triglycerides), glucose metabolism (fasting glucose, fasting insulin, HbA1c), and inflammatory biomarkers of AD (IL-6, IL-8, CRP). Approximately ∼10cc of blood was drawn into an EDTA vacutainer tube both at baseline and at 12 weeks (with a variation of ± 2 weeks). Lipid profile was evaluated through a lipid panel, ⁵⁷ while glucose metabolism was assessed via fasting glucose, fasting insulin, and hemoglobin A1c (HbA1c) using the Siemens DCA Vantage HbA1c Analyzer. Inflammation and CRP were evaluated using ELISA. ⁵⁸

Blood Pressure

Blood pressure was assessed at baseline and at 12 weeks (with a variation of ± 2 weeks) using an automated sphygmomanometer (DinaMap ProCare 100). We followed the National Health and Nutrition Examination Survey (NHANES) protocol ⁵⁹ and the recommendations of the American Heart Association ⁶⁰ for conducting all blood pressure assessments.

Body Composition

Body composition was measured using dual energy x-ray absorptiometry (DXA),^61,62 waist circumference, and body mass index (BMI) which was calculated by height and weight. DXA (GE Lunar iDXA) was used to assess fat-free mass, fat mass, and percent body fat uses very low X-ray doses. DXA used low-Xray doses (0.02mREM) to detect changes in body composition on the order of 1.6-3.8%. Weight was measured using a digital scale (± 0.1 kg; Befour Inc model #PS6600, Saukville, WI) and height was measured using a stadiometer (Model PE-WM-60-84, Perspective Enterprises, Portage, MI). To calculate BMI the study staff entered participants height (feet and inches) and weight (pounds) into REDCap database that automatically calculated BMI.^63,64 Waist circumference was measured to the nearest 0.1 cm using measuring tape, following standard protocols in which the waist was measured at the inferior margin of the last rib and the crest of the ilium. ⁶⁵ All body composition assessments were measured at baseline and at 12 weeks (with a variation of ± 2 weeks)

Dietary Assessments

Nutritional health status was assessed during the clinic visit using the NHANES Dietary Screener Questionnaire (31-item) which assessed intake frequency of food groups consumed (e.g., fruits, vegetables) in the past 30 days. Also, nutritional health status was assessed using the skin carotenoid assessment (Veggie Meter) ⁶⁶ which assessed skin carotenoids as a proxy for dietary intake of carotenoid rich foods such as fruits and vegetables. All nutritional assessments were assessed at baseline and at 12 weeks (with a variation of ± 2 weeks).

Cognitive Assessments

Psychometricians administered cognitive assessments using the NIH Toolbox to evaluate memory, executive function, language, reading, vocabulary, and processing speed. ⁶⁷ Specifically, the following tests were conducted during the study visit: NIH-TB List Sorting Working Memory Test, NIH-TB Picture Vocabulary Test, NIH-TB Oral Reading Recognition Test, and the NIH-TB Pattern Comparison Processing Speed Test, and NIH-TB Flanker Inhibitory Control and Attention at baseline and at 12 weeks (with a variation of ± 2 weeks).

Satisfaction and Acceptability Surveys

Participants completed 2 surveys at the 12 weeks (± 2 weeks) study visit. The acceptability survey (23-item) was guided by the theoretical framework of acceptability ⁶⁸ and focused on the domains of affective attitude (how a participant feels about the MIND + SOUL intervention), burden (perceived amount of effort required to participate in the MIND + SOUL intervention), ethicality (extent to which the intervention has good fit with the participants’ value system), perceived effectiveness (extent to which the intervention is perceived as likely to achieve its purpose), and self-efficacy (participants’ confidence that they can perform the behaviors required to participate in the intervention).

The second survey was a 10-item satisfaction survey, where participants rated various aspects of the MIND + SOUL diet using a visual analog scale (VAS). For each item, participants rated their satisfaction on a scale of 0 to 10 where 0 indicated “not satisfied at all” and 10 indicated “completely satisfied.” The frequency of each rating was recorded, providing a detailed distribution of participant responses.

Data Analysis

We conducted descriptive statistics for all variables which included frequencies and percentages for categorical variables and the mean (standard deviation) for continuous variables. For the acceptability survey, we analyzed responses by reporting the frequency and percentage of responses for each domain, including affective attitude, burden, ethicality, perceived effectiveness, and self-efficacy. For the satisfaction survey, which used VAS, we recorded the frequency of each satisfaction rating (from 0 to 10, where 0 indicated “not satisfied at all” and 10 indicated “completely satisfied”). These frequencies provided a detailed distribution of participant responses without calculating mean values, as satisfaction is a subjective measure. Additionally, the median satisfaction score was calculated for each item to represent central tendencies in participants’ ratings. Changes in cardiometabolic risk factors, body composition, nutritional health status, and cognitive function from baseline to 12 weeks were calculated using a linear mixed-effect model. The models were adjusted for age and years of education. Missing data were sparse and considered randomly distributed. Statistical significance was defined as two-sided alpha <0.05. Statistical significance was defined as two-sided alpha <0.05. All statistical analyses were performed using SPSS (Version 27.0. IBM Corp: Armonk, NY, USA).