Food As MedicinE for Diabetes (FAME-D)

Food is Medicine: A Randomized Clinical Trial of Medically Tailored Meals For Individuals With Type 2 Diabetes Mellitus and Food Insecurity

National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), Community Servings, Massachusetts General Hospital

The investigators propose a single-site, open-label, Phase II, community-based randomized clinical explanatory trial to test the efficacy of a medically tailored meal (MTM) interventions for adults with food insecurity and T2DM (Type 2 Diabetes Mellitus). While the efficacy of MTM needs to be established, given the nature of the population served by MTM, it is important to design the study to enhance the external validity/generalizability of the findings as much as possible. The investigators will recruit 200 participants from the referral list of Community Servings, a Boston-based MTM provider, and randomly assign them to MTM (N=100) or usual care + monthly food subsidy (N=100). MTM intervention will consist of 6 months of weekly delivery of 10 ready-to-eat meals (approximately half an individual's weekly food intake), along with telephone-based lifestyle intervention that prepares participants for the post-treatment period. The usual care + food subsidy recipients will receive usual care along with 6 months of a $40/month food subsidy.

Food insecurity, "lack of access to enough food for an active, healthy life", affects over 20% of the 30 million Americans with diabetes. Food insecurity is more common in racial/ethnic minorities and those with lower socioeconomic status. Moreover, food insecurity is associated with worse diabetes control and increased complications, even after adjusting for other risk factors. Food insecurity is a major contributor to disparities in diabetes outcomes. Medically tailored meal (MTM) delivery programs are a promising intervention for individuals with diabetes and food insecurity. MTM programs deliver fully prepared food, tailored by a registered dietitian nutritionist to the specific medical needs of the individual, and provide education to help optimize disease self-management. Food insecurity is typically addressed with food subsidies-offering additional financial resources that can be spent on food. While both food subsidies and MTM can increase healthy food access, MTM can help overcome other barriers to diabetes management including lack of time, ability, knowledge, and skills needed to prepare appropriate meals. The use of these programs is often called 'food as medicine', as the purpose is to provide exactly the foods needed to help prevent diabetes complications. MTM are receiving increasing public attention, but there are, as of yet, no full-scale trials to test its effects on diabetes outcomes when compared with other food insecurity interventions. Our research team has developed a medically tailored meal intervention that combines provision of healthy food, tailored to the specific nutrition needs of the individual, with an evidence-based lifestyle intervention that uses the period of meal delivery as springboard to improve diabetes self-management, both while receiving meals and after meal delivery is completed. This study is an explanatory-focused randomized trial to assess a community-based medically tailored meals intervention (n=200). It will be conducted among diverse participants referred for medically tailored meals. Adults with type 2 diabetes, Hemoglobin A1c between 7.0% and 12.0%, and BMI ≥ 25 kg/m2 (≥ 23 kg/m2 for those with Asian ancestry) will be enrolled and randomly assigned to intervention or usual care + food subsidy. The intervention group will receive meal delivery and its attendant lifestyle intervention for 6 months, while the usual care + food subsidy group will receive a $40/month food subsidy, along with usual diabetes care.

Owing to the nature of the intervention, participants, providers, and many study staff will know to which group individual participants have been randomized. However, efforts will be made to keep the study statistician and outcome assessors blinded to randomization group during the course of the trial.

The Medically Tailored Meal (MTM) intervention consists of weekly home meal delivery; an explanation of the medical tailoring of the meals; and a 6-session telephone lifestyle intervention change program designed to complement the period of meal delivery and prepare for the period after meal delivery with behavioral and skill-building approaches to sustain the benefit of the intervention.

As a comparison group, those not randomized to receive the MTM intervention will receive usual care provided by their clinicians not associated with the study, plus a food subsidy ($40/month) for 6 months, along with healthy eating information to guide use of that subsidy.

weekly home meal delivery; an explanation of the medical tailoring of the meals; and a 6-session telephone lifestyle intervention change program designed to complement the period of meal delivery and prepare for the period after meal delivery with behavioral and skill-building approaches to sustain the benefit of the intervention.

Food Insecurity Score. Score ranges from 0-10, with higher scores indicating greater food insecurity.

Food Insecurity Score. Score ranges from 0-10, with higher scores indicating greater food insecurity.

Health-Related Quality of Life Score as assessed by PROMIS (Patient-Reported Outcomes Measurement Information System)-29 at Month 6

The PROMIS-29 includes seven health related quality of life domains (Physical Functioning, Anxiety, Depression, Fatigue, Sleep Disturbance, Social Functioning, and Pain), and the pain domain has two subdomains (interference and intensity). Raw scores, except pain intensity, are transformed using the T-score metric based on the item response theory calibrations in which scores have a mean of 50 and standard deviation of 10 for the general population in the US. A higher PROMIS T-score implies more of the concept being measured; i.e., a higher PROMIS score on physical function indicates better functioning, whereas a higher score on depression indicates more severe depressive symptoms. Will report overall score and scores for each domain. From these data we will also calculate a PROMIS-Preference (PROPr score) (PROPr scores range from -0.022 (worst) to 1.0 (best)).

The PROMIS-29 includes seven health related quality of life domains (Physical Functioning, Anxiety, Depression, Fatigue, Sleep Disturbance, Social Functioning, and Pain), and the pain domain has two subdomains (interference and intensity). Raw scores, except pain intensity, are transformed using the T-score metric based on the item response theory calibrations in which scores have a mean of 50 and standard deviation of 10 for the general population in the US. A higher PROMIS T-score implies more of the concept being measured; i.e., a higher PROMIS score on physical function indicates better functioning, whereas a higher score on depression indicates more severe depressive symptoms. Will report overall score and scores for each domain. From these data we will also calculate a PROMIS-Preference (PROPr score) (PROPr scores range from -0.022 (worst) to 1.0 (best)).

Depressive Symptoms as assessed by PROMIS SF (Short Form) v1.0 - Depression 4a. Raw scores are transformed using the T-score metric based on the item response theory calibrations in which scores have a mean of 50 and standard deviation of 10 for the general population in the US. A higher PROMIS T-score implies more of the concept being measured; i.e., a higher PROMIS score on depression indicates more severe depressive symptoms.

Depressive Symptoms as assessed by PROMIS SF v1.0 - Depression 4a. Raw scores are transformed using the T-score metric based on the item response theory calibrations in which scores have a mean of 50 and standard deviation of 10 for the general population in the US. A higher PROMIS T-score implies more of the concept being measured; i.e., a higher PROMIS score on depression indicates more severe depressive symptoms.

Diet Quality as assessed by Brief Dietary Assessment Scale. This score is comprised of 3 sections (or subscales) used to assess dietary patterns. Subscales are scored independently. The subscales are: Vegetables, Fruit, Whole Grains, and Beans subscale (range 0 - 20), Drinks, Desserts, Snacks, Eating Out, and Salt subscale (range 0- 20), and Fish, Meat, Poultry, Dairy, and Eggs subscale (range 0-10): Total score equals the sum of the 3 subscales. For subscales and total score, higher scores indicates better diet quality.

Diet Quality as assessed by Brief Dietary Assessment Scale. This score is comprised of 3 sections (or subscales) used to assess dietary patterns. Subscales are scored independently. The subscales are: Vegetables, Fruit, Whole Grains, and Beans subscale (range 0 - 20), Drinks, Desserts, Snacks, Eating Out, and Salt subscale (range 0- 20), and Fish, Meat, Poultry, Dairy, and Eggs subscale (range 0-10): Total score equals the sum of the 3 subscales. For subscales and total score, higher scores indicates better diet quality.

Medication Adherence as assessed by 7-item ARMS-D scale. Scores range from 7 to 28, with higher scores indicating more problems with adherence.

Medication Adherence as assessed by 7-item ARMS-D scale. Scores range from 7 to 28, with higher scores indicating more problems with adherence.

Medication Adherence as assessed by 1-item summary of diabetes self-care activities measure. Scores range from 0-7 with higher scores indicating better adherence.

Medication Adherence as assessed by 1-item summary of diabetes self-care activities measure. Scores range from 0-7 with higher scores indicating better adherence.

Single item-indicators of trading off medication for food or food for medication. An affirmative response indicates the presence of a trade-off.

Single item-indicators of trading off medication for food or food for medication. An affirmative response indicates the presence of a trade-off.

Diet Self-Efficacy as assessed by cardiac diet self-efficacy scale. Scores range from 16 to 80 with higher scores indicating greater self-efficacy.

Diet Self-Efficacy as assessed by cardiac diet self-efficacy scale. Scores range from 16 to 80 with higher scores indicating greater self-efficacy.

Diabetes Self-Efficacy as assessed by Perceived Diabetes Self-Management Scale. Scores range from 8 to 40 with higher scores indicating greater self-efficacy

Diabetes Self-Efficacy as assessed by Perceived Diabetes Self-Management Scale. Scores range from 8 to 40 with higher scores indicating greater self-efficacy

Stress as assessed by perceived stress scale. Scores range from 0 to 40 with higher scores indicating higher perceived stress.

Stress as assessed by perceived stress scale. Scores range from 0 to 40 with higher scores indicating higher perceived stress.

Inclusion Criteria: Diagnosis of Type 2 Diabetes Mellitus Hemoglobin A1c ≥ 7.0% and ≤ 12.0% in last 12 months Experiencing food insecurity as indicated by 2-item Hunger Vital Sign English speaking BMI ≥25 kg/m^2 (≥ 23 kg/m^2 if self-report Asian ancestry) No plans to move from the area for at least 1 year Free living to the extent that participant has control over dietary intake Willing and able to provide written informed consent and participate in all study activities Exclusion Criteria: Participant in diabetes, nutrition, or weight research intervention in last 12 months Another family member or household member is a study participant. Only one member of each household may take part in this study. Considering bariatric surgery in the next year or prior bariatric surgery in the past 2 years Lack of safe, stable residence and ability to store meals Lack of telephone Pregnancy/breastfeeding or intended pregnancy in the next year History of malignancy, other than non-melanoma skin cancer, unless surgically or medically cured > 5 years ago or in remission. Patients with localized prostate and breast cancer diagnosed during the course of routine screening will not be excluded. Advanced kidney disease (estimated creatinine clearance < 30 mL/min) Known drug or alcohol misuse in the past 2 years Known psychosis or major psychiatric illness that prevents participation with study activities Intermittent use of medications (e.g., oral or intravenous glucocorticoids) that are likely to affect blood sugar

Individual participant data that can be deidentified without risk of confidentiality breach will shared upon written request made to study investigators. In addition, all NIH policies regarding data sharing will be followed. Deidentified individual data that supports the results will be shared provided the investigator who proposes to use the data has approval from an Institutional Review Board (IRB), Independent Ethics Committee (IEC), or Research Ethics Board (REB), as applicable, and executes a data use/sharing agreement with UNC.

Unless preempted by NIH policy, data will become available 12 months after publication of main findings, and be available for 60 months.

Unless otherwise required by NIH policy, ability to deidentify data, plan for secure storage of data and agreement to protect participant privacy, relevant research question, to be assessed by study investigators.