Article Text
PDF
Abstract
Introduction Linking patients living with chronic, diet-related diseases and food insecurity to charitable food assistance, medically tailored groceries (MTGs) and food resource coaching may empower patients to better manage their health in a way that is economically sustainable. This protocol paper describes the implementation of a study evaluating MTGs and food resource coaching in a food pantry setting.
Methods and analysis A randomised controlled trial whereby patients of a safety-net health centre will be screened for The Emergency Food Assistance Program (TEFAP) and study eligibility. Eligible patients will be randomised to receive 4 months of usual pantry services (control), MTGs (intervention 1) or MTGs and food resource coaching (intervention 2) from a colocated food pantry. Measures collected monthly include adherence, fidelity and food selections from the pantry. Measures collected at baseline and follow-up include food and nutrition security, diet quality, household resiliency, loneliness and overall well-being. Qualitative interviews (n=60) will be conducted preintervention and postintervention. Analyses will include multiple regression models to analyse changes from baseline to follow-up as well as changes between groups over time. Deductive thematic analysis of qualitative data will be conducted.
Ethics and dissemination This study was approved by the UT Southwestern Medical Center Institutional Review Board (STU-2023-1166) and the Parkland Health Office of Research Administration. This is protocol version two, modification approved on 3 June 2024. Findings will be disseminated through academic conferences, journals and to the public.
Trial registration number NCT06242808.
- Food Insecurity
- Cardiovascular Disease
- Social Support
- Randomized Controlled Trial
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
A strength of the study includes the use of a randomised controlled trial design with participants randomised to one of three groups—usual food pantry services (control), medically tailored groceries (MTGs) (intervention 1) and MTGs coupled with food resource coaching (intervention 2) to isolate the impact of each layer of support.
Implementing Food is/as Medicine (FIM) programming by leveraging existing food assistance organisations’ inventory may be more sustainable in the long term, as participants can continue receiving services at the end of the study.
Lastly, recruitment through a local safety-net clinic in the Southern sector of Dallas, Texas, reaches African American or Black and Hispanic or Latino(a) community members with lower-income, groups that are most impacted by food insecurity and diet-sensitive health disparities39–42 and increases the generalisability and external validity of the findings.
While study arms will be masked for outcomes assessors, the team members implementing services and participants cannot be masked due to the nature of the interventions—participants know if they receive coaching or not, therefore a limitation is the inability to fully mask study groups to which participants have been assigned.
The study may be underpowered to detect small effects.
Introduction
Over 10% of households in the USA and over 17% of households with children experienced food insecurity in 2022,1 and food insecurity increases the risk of adverse health outcomes.2 3 Simultaneously, there is a rise in chronic, diet-sensitive diseases that dominate healthcare spending in the USA (eg, type 2 diabetes, high cholesterol, obesity and cardiovascular disease).4 5 The emerging Food is/as Medicine (FIM) movement6 seeks to use nourishing food in combination with other medical treatments to advance the prevention, management and treatment of diet-sensitive disease. For food-insecure communities, FIM strategies may have the potential to increase food security and improve related health conditions.7 8
FIM strategies range from population-level healthy food programmes that provide education (eg, USDA’s MyPlate) and food access (eg, Supplemental Nutrition Assistance Program (SNAP) or Women Infants and Children), to more intensive provision of medically tailored foods through prepared meals, nutrition expert-guided selection of groceries, and produce provision alongside educational interventions.8–10 The integration of these programmes varies widely and addressing health-related food needs for populations that experience food insecurity introduces additional complexity.11–14 There is growing evidence that health system partnerships with community organisations that have food assistance expertise, such as food banks and pantries, offer a pathway for the long-term sustainability of FIM and the agility to pivot strategy based on local needs and preferences.15–17
Our study will be implemented in partnership with Crossroads, which operates a high-capacity client-choice pantry and is the largest non-profit food redistributor in the North Texas region. In alignment with the types of services provided by pantries, we will implement a medically tailored groceries (MTGs) FIM strategy within the pantry context. MTGs will be identified from Crossroads’ inventory, which is selected from the bulk inventory secured by the North Texas Food Bank. We also have a small amount of funding to purchase supplemental grocery items when MTG items are not available from the food bank. We will report on the proportion of total MTG food that is secured through the normal pantry distribution systems compared with the food that is purchased from retailers.
At Crossroads, households are allocated points based on the dietary requirements of household members. Points act as a form of currency such that the amount of food each household gets at each monthly visit is enough for 21 meals for each household member up to a maximum of 500 points (five household members).18 This 500-point maximum was put into place to ensure equity in the availability of products for all shoppers seeking services, as the need for food access continues to grow and challenges were presented with supply when large households received larger quantities of food (eg, 1200 points). While the use of a point system such as the one used at Crossroads is not necessary for the implementation of MTG, the MTG approach we investigate is meant to be implemented within the context of a client-choice pantry that scales allotments with household size. Scaling food allocations by household size is the standard model employed by high-capacity, evidence-based pantries in the USA.19
MTG consists of minimally prepared foods that might be offered in a pantry setting but are carefully selected to support health goals. In some cases, MTG may require a shift in typical pantry inventories so a full mix of healthy foods can be offered. The MTG food must be prepared by the patient or caregiver at home, ideally promoting self-efficacy in health-promoting food preparation. Also called medically tailored food packages or boxes, these groceries combine ingredients across food categories to optimise a nutritionally complete dietary pattern.
While some evidence exists to support the theory and process for medically tailoring groceries,20–24 evidence is generally lacking regarding how MTG can be provided at the household level and the proper ‘dosing’ in the context of food insecure households.6 In our intervention, the quantity of MTG provided is scaled with the size of the patient’s household. Food insecurity is a household-level condition and food pantries ‘treat’ food insecurity at the household level. Thus, it seems plausible that FIM strategies that intend to address both diet quality and food security should also adopt a household-level framing.
Another strategy we are testing as a complement to MTG is food resource coaching. Since 2020, Crossroads, Ascend Dallas (a local non-profit leader in financial coaching) and coauthor TL have been developing F3: From Food to Finances, a novel food resource coaching programme that provides Crossroads shoppers with training in securing resources to promote food, nutrition and economic security. To our knowledge, the food resource coaching intervention is novel, and no peer-reviewed publications have explored the impact of food resource coaching in a no-cost food pantry or clinical setting. However, other researchers have explored using financial resource coaching in a clinic setting,25 26 which served as a model for the development of food resource coaching. Therefore, we hypothesise that patients receiving MTG and coaching in our study will have the greatest improvements in patient engagement, adherence and dietary quality over the study period.
Our study objectives include the execution of a between-group, repeated measures RCT to identify the feasibility and early-stage effectiveness of recruiting patients with lower income and at least one diet-related chronic disease (n=210) from Parkland Health, a large safety-net health centre. Participants are randomly assigned to one of three groups: (1) usual pantry services which include food for up to 21 meals for up to five people in the household, (2) pantry services with MTG and (3) pantry services with MTG and monthly one-on-one food resource coaching sessions. We also aim to collect preintervention and postintervention qualitative data from a subset of randomly selected participants (n=60) to solicit feedback on linkage to supportive services via the medical system and their experience with the programme.
In addition, we aim to describe our collaborative process for defining MTG from a pantry meeting the needs of a local community experiencing food insecurity and colocated with the safety-net health centre. First, we evaluated common evidence-based dietary patterns to identify a pattern with broad applicability to multiple chronic diseases and feasible application within an MTG intervention aligned with low-cost food assistance programmes. Next, we sought to apply the dietary pattern in a pantry setting where inventories are limited and fluctuating, and people have diverse cultural and personal food preferences. Finally, we aimed to pragmatically design and develop an MTG programme rooted in pantry inventory data. This manuscript describes both the study protocol and aspects of our protocol for implementing medically tailored groceries and food resource coaching in a food pantry.
Methods and analysis
We use Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) reporting guidelines, which are uploaded as a supplement.27 The planned study start date was 1 April 2024 but was delayed to 31 May 2024 due to delays in construction at the satellite pantry. The planned data collection end date is 31 March 2025, the planned study end date is 30 June 2025.
Sample size
The current lack of information on novel intervention effects on outcomes of interest (eg, food selections from the pantry, dietary quality) serves as a challenge to conducting a power analysis. We estimated that with a sample size of 210 people (70 per group), the study is powered to detect a small interactive group by time effect (f=0.11) on measures taken at baseline and follow-up. The sample size was selected based on what the study team, clinic and community partners estimated might be feasible to recruit from the clinic and serve in the pantry within the 18-month grant timeline, and large enough to detect a small to medium size effect.
Recruitment
Recruitment takes place via multiple routes. Providers and other healthcare staff at the Parkland C.V. Roman Health Center, a community safety-net primary care clinic, make referrals to the study via recruitment fliers posted in patient exam rooms and discussion about the study with patients. The study staff send approved recruitment emails to potentially eligible patients. These emails inform patients of the study occurring at their clinic and provide contact information to learn more. The main recruitment method employed is in-person tabling at clinic entry. The recruitment team sits at a table in the lobby on Monday, Wednesday and Friday and introduces the study to patients, screening anyone interested. The team shares fliers with a QR code, linking patients to the eligibility screener where they can submit their information on their personal device. The team also has tablets that patients can use to complete the eligibility screener if preferred. Alternatively, if an interested patient has limited time and is unable to complete the screening while onsite, they can take a recruitment flier and fill out the screener at their convenience.
Eligibility criteria
A participant must be: a patient at the C.V. Roman Health Center, over the age of 18, have reliable transportation, be eligible for The Emergency Food Assistance Program (TEFAP) (earn less than 185% of the federal poverty threshold, self-report), diagnosed with at least one of three diet-related chronic disease categories (dyslipidaemia, hypertension, and/or diabetes, verified in EHR), have no illnesses, allergies or sensitivities that severely limit what they can eat (eg, put them at risk from consuming food from the pantry), live in one of the 18 ZIP codes served by Crossroads that surround the local safety net clinic, be able to give informed consent and willing to fully participate.
Randomisation
Two randomisation tables with equal group allocation using Excel are created by a team member that is not involved with participant recruitment, consent or data collection. The randomisation tables are uploaded into RedCap, one for ‘development’ (eg, survey testing) and one for ‘production’ when data collection begins. The team members executing randomisation in RedCap are unaware of which group each participant will be randomised to until they execute the instrument to assign a group.
Study setting
Once patients are recruited, consented and complete the baseline questionnaire, they are randomised into one of the three study groups and scheduled for their first pantry appointment. This pantry is colocated within a connected one million square foot commercial retail complex, on the other side of the building from the Parkland clinic. The pantry is within a renovated storefront near the complex entrance and is approved by the City of Dallas Department of Health for storing and distributing food to community members.
Usual pantry services (control group)
All participants who use Crossroads pantry receive food for up to 21 meals for up to five people in the household. We have estimated that a family of four receives approximately US$250 worth of groceries from Crossroads each month. Appointments at Crossroads are monthly. When participants arrive at Crossroads, they complete a process called ‘intake, survey and order’. At intake, Crossroads staff collect household demographics to confirm pantry eligibility, which is set by TEFAP guidelines28 and to allocate household points for pantry shopping. At the beginning of the shopper’s appointment, their next appointment is scheduled, as authors TL and SLP have found that appointments encourage pantry attendance.29
After intake is completed, shoppers are asked to answer a brief survey module. There are four different survey modules that rotate, to maximise the amount of information obtained to improve pantry services while minimising shopper survey burden. The order process is completed using Salesforce. The Salesforce interface displays the picture and description of each item currently available in the pantry and the number of points the item costs. Crossroads pantry allocates points based on an algorithm that follows 2015–2020 dietary guidelines for recommended intake based on the age, sex and activity level of each person in the household. Participants can select any food items until they use all their allocated points. With the exception that there are some inventory restrictions in place depending on the amount of inventory available for an item each week (eg, up to three pounds of onions, up to two pounds of almonds) as availability may fluctuate. After shoppers make their grocery selections on the computer, their grocery list is printed and they select items off the shelves, which is set up like a small grocery store. The process of making food selections prior to shopping reduces congestion in the small grocery area, allowing Crossroads to serve more shoppers. After selecting all items on their list, shoppers check out with pantry staff or volunteers, have their groceries bagged and exit the pantry.
Recipes are always available to Crossroads shoppers at the main market and are available to the control group in this study as well. Recipes are in a labelled box on top of the desk that is visible to the shopper and if a shopper asks for recipes, they are provided, but they are not explicitly introduced by a team member.
Medically tailored groceries (intervention one)
A community-academic partnership team including a culinary medicine (CM)-trained registered dietitian (MS), a CM-trained physician (JA), a public health nutrition and behavioural scientist (KH-H), a health economist (TL) and pantry leadership (AF, BS), collaborated to identify an evidence-based dietary pattern and apply that pattern to groceries provided within the charitable food system. Rooted in an existing evidence-based dietary pattern and scoring system for individual consumption and household grocery purchases,24 30 31 our team developed a simple, pragmatic model to medically tailor groceries from the pantry using the Dietary Approaches to Stop Hypertension (DASH) dietary pattern recommendations. The DASH dietary pattern emphasises fruits, vegetables, whole grains, legumes, nuts and low-fat dairy while limiting added sugars, sodium and red and processed meats. DASH was selected given extensive evidence for the dietary pattern’s efficacy in improving a variety of preventable chronic diseases, including those which are a focus of this study.30 32 33 Notably, this overall model for medically tailoring pantry groceries also aligns with the American Heart Association’s recent scientific statement on popular dietary pattern alignment with health recommendations, the 2020–2025 Dietary Guidelines for Americans34 and the EAT-Lancet Commission’s report35 on optimising nourishing dietary patterns and supporting planetary health and sustainability.
The team applied the food category principles of the DASH diet to score foods available in the pantry’s inventory over the prior year on a scale of 1–5 based on alignment with the DASH diet and (a) high nutrient density; (b) balanced energy intake; (c) chronic disease prevention and management; (d) affordability and (e) sustainability. In the 5-point scoring system, a ‘5’ represents the highest recommendation for nutrient density and contribution to a dietary pattern associated with disease prevention and health promotion.36 Rather than defining a precise number of servings per category or specific windows of sodium, added sugar or other nutrients, the team developed a universally applicable strategy described in detail in the following paragraph to encourage the dietary pattern that will optimise nutrition. Specific tailoring to precise intake of certain nutrients requires ongoing engagement with a registered dietitian nutritionist, which is beyond the scope of most charitable food organisations. Our proposed approach balances practical implementation constraints and supports most people seeking to prevent or treat diet-sensitive conditions including cardiometabolic diseases, many cancers and obesity. Furthermore, as food insecurity is a household-level condition, the provision of foods that align with nutritious categories in an overall pattern supports the reality of household distribution of food.
We designed the MTG carts to include only food scoring in the higher point categories (4 or 5 points). Table 1 includes our proposed MTG scoring categories, which were developed by evaluating available pantry inventory against the DASH dietary pattern recommendations. To implement the MTG grocery carts, we use the pantry’s client-choice model whereby shoppers select food from the pantry within a given number of points that are allocated to them based on household size. Shoppers in the MTG group are shown a grocery cart with at least 50% of their point allocation prepopulated with MTG items in their cart. Shoppers can remove MTG items and add additional items if they choose. MTG categories and products provided are the same for all participants. The quantity of MTG products populated is the same for household sizes of one or two people, quantity is doubled for household sizes of three or more people. Table 2 includes a sample MTG cart based on a household size of two people. Figure 1 shows a visual depiction of how the MTGs are implemented in Crossroads Salesforce inventory system with part of a prepopulated MTG cart. This approach combines expertise in applying nutrition as part of a medical treatment or prevention plan, linking prescribed groceries to a culinary lens of utility and complementary food pairings, and a pragmatic understanding of the food preferences of our target communities based on data collected from pantry shoppers.37 38
Medically tailored grocery scoring categories
Sample medically tailored grocery (MTG) cart for a household size of 2
Visual representation of part of a Shopper’s medically tailored grocery (MTG) cart.
Participants in the MTG group have recipes handed to them directly rather than having to ask for recipes. The team running appointments is trained not to discuss recipes at length as to not contaminate this group with coaching.
MTGs + food resource coaching (intervention two)
Participants in this group receive the same MTG intervention described in the section above and food resource coaching. Food resource coaching is led by trained Crossroads and UTSW (UT Southwestern) staff. Coaches are trained in coaching methodology by experts at Ascend Dallas (40 hours of in-person training), a premier financial coaching organisation supporting community members since 1908. Ascend training is then coupled with training intended to translate coaching methodology to the context of nutrition and resource coaching for households experiencing food insecurity (80 hours of in-person training). Once trained, coaches have a period of intense mentorship whereby they conduct coaching appointments in close consultation with established resource coaches for their first 1–2 weeks of coaching. The mentorship continues after coaches are fully trained and working independently to ensure ongoing support and identification of ways in which the coaching experience can be more impactful for shoppers.
In this study, the coaching intervention takes place over 4 monthly sessions. In each session, shoppers meet with the resource coach in-person as part of their visit to the community market prior to making food selections. Shoppers self-direct the focus of their coaching based on their self-selected goals, which means sessions may focus more heavily on achieving nutritional goals or financial goals depending on the person. Examples of coaching content all participants receive include training in meal planning, developing a monthly food budget and how to access additional food and social resources (eg, SNAP enrolment, how to find discount retailers).
After coaching is completed, shoppers proceed to order groceries through the Salesforce inventory application with their coach. The coach explicitly identifies the MTG, explains how the items align with healthy dietary recommendations for people with chronic diseases and shares 12 recipes each month that incorporate MTG items. At the end of the 4-month MTG + coaching intervention, shoppers are invited to continue to more in-depth financial coaching or resource coaching programmes that are available at the flagship Crossroads location, a 7-minute drive from where the FIM intervention is located.
Criteria for discontinuing
This study is minimal risk; therefore, there are no safety-related stoppage criteria. Participants in all groups are informed in the consent form that they can choose not to answer any question that makes them feel uncomfortable and can discontinue participation at any time.
Strategies to improve retention and adherence
To improve retention and adherence, participants are provided with physical appointment cards and sent text messages with their pantry appointment date and time immediately after making the appointment and the day prior. Participants receive US$15 in the form of a gift card after completing the baseline questionnaire and another after completing the month 4 pantry appointment. Subject matter experts who reviewed the study on behalf of the funder recommended a bonus payment to encourage adherence and data collection; therefore, participants receive a US$15 bonus payment if they complete all appointments and study activities (eg, questionnaires).
For participants who no-show pantry appointments, daily attempts are made to reschedule the appointment within the week of the missed appointment. Weekly attempts are made to reschedule the participant throughout the 4-month intervention period thereafter. If participants completely miss a month, the next time they return they receive those months services. For example, if a participant misses month 2 and returns to the pantry in month 3, they only have month 3 and 4 services left and would take the follow-up questionnaire at the end of month 4 as planned. There are no limits to the number of times a participant can reschedule their appointment; however, they can only reschedule during the 4-month intervention period + 10 days.
Participants are contacted through their indicated preferred method of contact first (text, call, email). However, if they continue to be unreachable using their preferred method of contact and they have provided another method, after 2 weeks of attempt to contact, the team will attempt to use another method.
Timeline
The participant timeline of enrolment, interventions and assessments is included in table 3.
Participant timeline of enrolment, interventions and assessments
Measures
Measures include demographic characteristics, address, diagnosis; intervention feasibility, adherence and satisfaction; nutrition security, food security, perceived diet quality, diet quality, cooking and food provisioning self-efficacy and food agency, general self-efficacy, adaptive capacity, loneliness and well-being. All quantitative measures collected as part of this study and measure citations are described in detail in online supplemental material.
Supplemental material
Analytic plan
We will first use analysis of variance (ANOVA) to assess whether participants differ by group on potential covariates: age, household size, children in the household, race and ethnicity, income, obesity, self-rated health, chronic disease diagnoses, self-efficacy and number of appointments attended.
Intervention feasibility, adherence and satisfaction will be compared across intervention arms using ANOVA to compare group means on continuous variables. Participants enrolled and coaching sessions attended will be quantified by study month and calendar month to examine temporal trends. We will also examine change in rates of participation across different subpopulations characterised by patient and household characteristics (eg, household size, morbidity).
Nutrition security, food security and diet quality changes between baseline and follow-up will be compared across study arms using multiple regression models that include study participant characteristics and indicators for the study arm. The impact of study arm on average monthly points spent on MTG and nutritional quality of the market order will be estimated using multiple regression models. Models will include patient characteristics and indicators for the study arm. We will also estimate a longitudinal model that includes all four observations characterising the MTG composition and nutritional quality of each participant’s market order in each study month. This model will include a trend variable indicating the study month and the interaction between trend and study arm indicators, which will provide evidence for whether duration in each intervention arm impacted changes in the market order composition.
Qualitative component
A subset of participants (n=60) are randomly selected to participate in preintervention and postintervention qualitative interviews using a randomisation website that allows participants to see their randomisation in real time (https://pickerwheel.com/tools/yes-or-no-wheel/) until 60 people complete preintervention interviews. A semistructured qualitative interview guide and codebook was developed by CP and KH-H and critically reviewed by JA, TL, SLP and MB. Interviews are conducted by CP in English and Spanish over the phone and last about 30 min. Interviews are conducted prior to the intervention and within 30 days of completing all study appointments. The qualitative interviews are focused on identifying key programme components most impactful to support nutrition behaviour change in populations with lower income and optimal design of the linkage between the health system and community resources. Deductive thematic analysis of qualitative data will be conducted using the study codebook.
Patient and public involvement
The public is included in the design of the study in that Crossroads staff assist with protocol planning and implementation. Additionally, extensive preliminary data were collected from community members to understand nutrition intervention and cuisine preferences. We also plan to share results with patients using an infographic or other brief format once study analyses are completed.
Monitoring harms
The study is minimal risk, and serious adverse events are not anticipated. Unintended effects or any unexpected adverse events that arise will be reported to the study sponsor and the institutional review board (IRB) within 5 working days.
Ethics and dissemination
Research ethics approval
This study was approved by the UT Southwestern Medical Center Institutional Review Board (STU-2023-1166) and the Parkland Health Office of Research Administration. The study is preregistered on ClinicalTrials.gov (NCT06242808) as of 2 February 2024.
Protocol amendments
Protocol amendments will be documented via American Heart Association reporting requirements as well as on ClinicalTrials.gov.
Informed consent
Participants read the consent form to themselves or have the consent form read to them by a study team member depending on their preference. Participants consent electronically via RedCap on their personal device or a study tablet after the study team determines eligibility and prior to completing the baseline questionnaire or randomisation. Participants can also elect to receive a physical or electronic copy of the signed consent form for their records. The consent form includes consent for both the quantitative and qualitative components of the study. In addition to completing the informed consent form, prior to conducting the qualitative interview, the interviewer sets expectations for the interview (eg, length of time, types of questions asked and why questions are being asked, compensation for the interview), answers any questions the participants may have about the interview and requests verbal consent to record the interview prior to beginning the interview or recording. The consent form can be found in online supplemental file 1.
Supplemental material
Confidentiality
Study staff across sites (eg, in the clinic, pantry, qualitative coinvestigator) use first and last names when scheduling participant appointments, which participants are informed of and agree to in the consent form. However, those names are not connected to the participant’s data. All participant data (eg, questionnaire responses, health information, interview transcripts) are associated with a study code number. Identifying information and clinical information is only available to those team members who are research ethics and Health Insurance Portability and Accountability Act (HIPAA) credentialed as well as IRB and Principal Investigator (PI) approved to have access. No identifying information is reported in any academic or medical journals or meetings.
Access to data
For data collected by the study team, deidentified data will be hosted in the Texas Data Repository, an open-source platform for publishing and managing datasets created by Texas higher education institutions using Harvard’s Dataverse software. The data will be made available at the time of publication and shared with anyone who creates a free account and makes a reasonable request for data. Currently, data deposited into the Texas Data Repository is available in perpetuity.
Ancillary and post-trial care
Participants can continue to receive pantry services and food resource coaching from Crossroads at the end of the study if they remain eligible for charitable food assistance.
Dissemination policy
Findings will be disseminated through academic publications and conferences, ClinicalTrials.gov, UTSW (UT Southwestern) Communications, PI KHH’s website (nicheresearch.org) and University and MPIs’ social media. The team does not intend to use any professional writers.
Ethics statements
Patient consent for publication
Acknowledgments
The authors would like to thank the entire team at Crossroads Community Services, especially Benaye Wadkins Chambers, Taylor Hall and Adelle Trogdon who were instrumental in implementing this study. Thank you also to the team at Parkland Health and specifically the Parkland C.V. Roman Clinic, Ron Sessions and Dr. Berrie for their support of study recruitment.
References
Footnotes
Contributors JA, TL and KH-H conceptualised the study and secured funding to carry out the study. MB and SLP contributed to study protocol design and edited the funding proposal. JA, MS and KH-H developed the MTG point allocation and scoring system with feedback from AF and BS. AF and BS contributed to protocol development and implementation. MS and CH developed study recipes. CH aligned recipes to the MTG scoring system and the market’s available inventory each month. CP developed the qualitative interview guide. JA, WW, TL and KH-H led manuscript preparation. All authors critically reviewed the manuscript and provided valuable feedback and edits. KH-H is the guarantor.
Funding This study was funded by an American Heart Association (AHA) Food is Medicine Grant (24FIM1262347) awarded to authors JA, TL, and KHH. The funder had subject matter experts peer review all grantees and provide recommendations for study improvement. This peer review led to our team adding intervention group 1 (MTG only) and a $15 bonus payment for participants that complete all study activities, to encourage completing follow-up data collection. The funder had no influence on the decision to submit this manuscript or the reporting of this study.
Competing interests Unrelated to this work, KH-H is a program advisor (paid consultant) on the Special Supplemental Nutrition Program for Women, Infants, and Children Community Innovation and Outreach program (WIC CIAO) on behalf of the Center for Nutrition and Health Impact. Unrelated to this work, JA serves on the advisory board and is a committee chair for the American College of Culinary Medicine focused on the development of educational strategy and tools in graduate medical education. All other authors have no competing interest to declare.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; peer reviewed for ethical and funding approval prior to submission.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.