Project MiNT: Assessing the Impact of Food & Video-Based Nutrition Education on Patients With Poorly Controlled Diabetes

Project MiNT: Assessing the Impact of Food & Video-Based Nutrition Education on Patients With Poorly Controlled Diabetes

A Pragmatic Randomized Controlled Trial Assessing the Impact of Medically Tailored Meals and Medical Nutrition Therapy Via Telehealth Among Patients With Poorly Controlled Diabetes

This pragmatic randomized controlled trial will assess the efficacy of medically tailored meals and medical nutrition therapy via telehealth on clinical outcomes for patients with poorly controlled type 2 diabetes. The goal of these interventions is to improve outcomes for patients with diabetes by impacting self-care behaviors and diet self-efficacy in the short and long term. This study leverages a unique health system and community group partnership to inform the utility of reimbursement for medically tailored meal programs and medical nutrition therapy, which would ultimately facilitate their scalability and sustainability within the healthcare system.

Despite the availability of evidence-based guidance and treatments, many patients with diabetes mellitus (DM) do not achieve optimal glycemic control. In prior work, patients with DM identified primary needs of improved access to food and nutrition education. Several innovative models for food services have emerged to address patient needs for improved healthy food access. Currently, select non-profit organizations across the U.S. deliver millions of medically-tailored meals (MTM) to patients with chronic illnesses, based on the premise that food can be as helpful as medicine in improving patient outcomes. While a handful of pilot studies have demonstrated success of MTM in reducing healthcare utilization and hemoglobin A1c (HbA1c) among patients with DM, no randomized trials have assessed the sustained impact of MTM on long-term outcomes for patients with DM. Similarly, individual studies of medical nutrition therapy (MNT), which includes individualized nutrition education and counseling, have shown benefit for improving DM outcomes. However, uptake of MNT across health systems has historically been extremely limited because of patient logistical barriers to attending the in-person visits. MNT delivered by telehealth (tele-MNT) has potential to overcome these barriers. Our goal is to evaluate the effect of MTM, with and without additional tele-MNT, on outcomes for patients with poorly controlled type 2 DM. The investigators will conduct a pragmatic randomized controlled trial with patients who have poorly controlled DM. Outcomes of interest include change in HbA1c, cost effectiveness, weight, diabetes self-efficacy, self-care, diabetes quality of life, and patient satisfaction. Investigators will assess the efficacy of 1) MTM and 2) MTM + tele-MNT compared to usual care in reducing HbA1c at 3, 6 (primary outcome) and 12 months and the cost effectiveness of each intervention compared to usual care up to 12 months. Patient perspectives regarding experiences with MTM and MTM + tele-MNT will also be explored to gain a better understanding of potential factors contributing to treatment failure and success within each group. The proposed research is innovative because it develops and tests a novel tele-MNT intervention tailored to recipients of MTM, provides a comprehensive solution to patient-identified nutrition needs, includes complete utilization data provided by the HealthShare Exchange, and leverages a unique health system and community group partnership to improve sustainability and scalability. Study results will inform 1) provider and payor decisions regarding offering and covering these services; 2) how services should be structured for scalability; and 3) integration of these services into the larger care delivery system to reduce DM disparities.

Enrolled patients with HbA1c>8% will be randomized to 1 of 3 arms: 1) MTM; 2) MTM + tele- MNT; or 3) usual care in a 1:1:1 ratio using random permuted blocks stratified by HbA1c (>8 to <10 vs. >10) and hospital discharge site (inpatient vs ED). The size of any particular block will be randomly selected and only known by the study biostatistician. This stratification approach was selected as investigators hypothesize that the interventions may have differential effects based on how poorly controlled an individual's DM is at baseline, as well as the degree of acute decompensation of the patient at time of enrollment.

A computer-generated list of random numbers will be prepared in advance by the study biostatistician and loaded into a REDCap randomization tool to ensure the research staff are blinded to assignment pre-randomization. It is not feasible to blind participants and clinic staff to assignment post enrollment; however, HbA1c is an objective outcome, and as such lack of blinding should not be a significant limitation.

Participants in this arm will receive MTM for 12 weeks. Meals will be prepared and delivered by MANNA, a non-profit organization that has provided MTM for patients with chronic illnesses in Philadelphia and Southern New Jersey since 1990. MANNA will deliver 21 complete meals to the patient's home each week, providing 45-60 grams of carbohydrates per meal for optimal glucose control based on ADA guidelines and 100% of overall nutritional requirements based on USDA guidelines. In addition, children and any senior dependents for whom the participant is the primary caregiver will receive meals for the entire 12 weeks for no additional cost, as this is standard of care of MANNA services. MANNA registered dieticians will cater the program to meet the specific needs (e.g., dietary restrictions, cultural preferences). Investigators will provide information on community resources in the area, including food resources, for all patients.

Patients in this arm will receive MTM services as well as tele-MNT over 12 months. The tele-MNT intervention will be delivered by a registered dietician within the Jefferson endocrine clinic, with assistance by other endocrine dieticians and fellows. In the first months, video visits focus on supporting individuals who are not selecting, preparing or purchasing their own meals. As the end of MTM services approaches, the intervention shifts to focus on the transition from MTM to self-directed eating. Based on Academy of Nutrition and Dietetics recommendations, each participant's MNT will include the following core features: nutrition assessment, intervention, care coordination, monitoring and evaluation. The following will also be addressed: nutrition prescriptions, nutrient intake, energy intake, glycemic index and load, alcohol consumption and physical activity. The schedule includes individual visits in the first 6 months and monthly group session in months 7-12.

Patients in this arm will receive usual services offered at Jefferson for patients with DM, which includes regular visits with a diabetes provider (primary care or endocrine), standard ADA information pamphlets and referral to 1) diabetes education classes and 2) nutrition counseling by dieticians and nurse practitioners. During routine office visits, providers reinforce messages about self-management and provide lists of local and national resources related to nutrition and diabetes self-management (e.g., diabetes.org). The standard of care at Jefferson for patients with DM is to begin with a single group MNT visit lasting from 60-90 minutes. Each participant's need for additional sessions and general time-frame for follow-up is individually determined following the group session, based on patient preference. Historically, only about 2% of the Jefferson population engages in these services, thus minimizing dilution of the effect of the tele-MNT.

MTM are meals designed by a Registered Dietitian Nutritionist to reflect appropriate dietary therapy according to evidence-based nutrition practice guidelines. They address medical diagnoses, symptoms, allergies, inability to chew or swallow, medication management, and side effects to ensure the best possible nutrition-related health outcomes.

Medical nutrition therapy (MNT) is offered as a component of usual care in many health systems, and consists of individuals receiving nutrition education tailored to their unique medical needs. Tele-MNT is the use of video technology to deliver MNT.

Change in hemoglobin A1C for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care

Change in hemoglobin A1C for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care

Change in hemoglobin A1C for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care

Intervention costs of 1) medically tailored meals, 2) medically tailored meals + telehealth-delivered medical nutrition therapy, and 3) usual care

Cost of delivering usual care and MTM will be obtained from administrative billing data. Cost of delivering tele-MNT will be obtained from data logs which capture time preparing and delivering sessions.

Intervention costs of 1) medically tailored meals, 2) medically tailored meals + telehealth-delivered medical nutrition therapy, and 3) usual care

Cost of delivering usual care and MTM will be obtained from administrative billing data. Cost of delivering tele-MNT will be obtained from data logs which capture time preparing and delivering sessions.

Total participant health care cost will be obtained from the Healthshare Exchange of Southeast Pennsylvania (HSX), the regional health information exchange, or Jefferson billing and claims data. HSX captures administrative billing information (i.e., claims) for outpatient care, emergency department care, inpatient care, and post-acute care and prescription drugs.

Total participant health care cost will be obtained from the Healthshare Exchange of Southeast Pennsylvania (HSX), the regional health information exchange, or Jefferson billing and claims data. HSX captures administrative billing information (i.e., claims) for outpatient care, emergency department care, inpatient care, and post-acute care and prescription drugs.

Health utility will be measured using the EQ-5D-5L, which measures health-related quality of life by assessing 5 dimensions on a scale from 1-5. A summary index score is derived by applying an appropriate value set to the responses. The value sets provide weights to each health dimensions according to the preferences of the general population of a country or region. These values sets have been developed and verified by the EuroQol Research Foundation.

Health utility will be measured using the EQ-5D-5L, which measures health-related quality of life by assessing 5 dimensions on a scale from 1-5. A summary index score is derived by applying an appropriate value set to the responses. The value sets provide weights to each health dimensions according to the preferences of the general population of a country or region. These values sets have been developed and verified by the EuroQol Research Foundation.

The ICER is the measure of cost-effectiveness. It is calculated as the difference in cost (health care costs + intervention costs) between two strategies divided by the difference in health care related quality of life between two strategies.

The ICER is the measure of cost-effectiveness. It is calculated as the difference in cost (health care costs + intervention costs) between two strategies divided by the difference in health care related quality of life between two strategies.

Change in weight for patients with poorly controlled type 2 diabetes mellitus who receive 1)medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care

Change in weight for patients with poorly controlled type 2 diabetes mellitus who receive 1)medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care

Change in weight for patients with poorly controlled type 2 diabetes mellitus who receive 1)medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care

Change in diabetes self-efficacy for patients with poorly controlled type 2 diabetes mellitus as measured by The Diabetes Self Efficacy Scale

Change in diabetes self-efficacy for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care. The Diabetes Self Efficacy Scale is made up of 8 questions, all scored on a 10-point Likert scale. Each questions asks about the respondent's confidence in performing an activity relative to diabetes. A response of 10 indicates total confidence, while a response of 1 indicates no confidence. The score for the scale is the mean of the eight items. Higher numbers are indicative of higher self-efficacy.

Change in diabetes self-efficacy for patients with poorly controlled type 2 diabetes mellitus as measured by The Diabetes Self Efficacy Scale

Change in diabetes self-efficacy for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care. The Diabetes Self Efficacy Scale is made up of 8 questions, all scored on a 10-point Likert scale. Each questions asks about the respondent's confidence in performing an activity relative to diabetes. A response of 10 indicates total confidence, while a response of 1 indicates no confidence. The score for the scale is the mean of the eight items. Higher numbers are indicative of higher self-efficacy.

Change in self-care for patients with poorly controlled type 2 diabetes mellitus as measured by the Summary of Diabetes Self-Care Activities (SDSCA)

Change in self-care for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care. The Summary of Diabetes Self-Care Activities (SDSCA) is a survey made up of 11 questions, all scored on a 7-point scale. Each question asks how many days in the last 7 days a diabetes related task was performed. The scale represents the number of days per week this task was performed and the respondent indicates the frequency with which they performed these tasks. The SDSCA assess five regimen areas: General Diet, Specific Diet, Exercise, Blood-Glucose Testing, Foot Care, and Smoking Status. Scores are calculated for each of the five regimen areas by finding the mean number of days each task was performed within each regimen. Higher scores indicate better management of diabetes self-care activities.

Change in self-care for patients with poorly controlled type 2 diabetes mellitus as measured by the Summary of Diabetes Self-Care Activities (SDSCA)

Change in self-care for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care. The Summary of Diabetes Self-Care Activities (SDSCA) is a survey made up of 11 questions, all scored on a 7-point scale. Each question asks how many days in the last 7 days a diabetes related task was performed. The scale represents the number of days per week this task was performed and the respondent indicates the frequency with which they performed these tasks. The SDSCA assess five regimen areas: General Diet, Specific Diet, Exercise, Blood-Glucose Testing, Foot Care, and Smoking Status. Scores are calculated for each of the five regimen areas by finding the mean number of days each task was performed within each regimen. Higher scores indicate better management of diabetes self-care activities.

Change in diabetes quality of life for patients with poorly controlled type 2 diabetes mellitus as measured by Diabetes Quality of Life (DQoL) instrument

Change in diabetes quality of life for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care. The Diabetes Quality of Life (DQoL) instrument is made up of 15 questions, all scored on a 5-point Likert scale assessing either frequency (never - all the time) or satisfaction (very satisfied - very dissatisfied). Higher scores on DQOL items indicate higher frequency of negative impact of diabetes or higher dissatisfaction with diabetes care and quality of life.

Change in diabetes quality of life for patients with poorly controlled type 2 diabetes mellitus as measured by Diabetes Quality of Life (DQoL) instrument

Change in diabetes quality of life for patients with poorly controlled type 2 diabetes mellitus who receive 1) medically tailored meals and 2) medically tailored meals + telehealth-delivered medical nutrition therapy compared to usual care. The Diabetes Quality of Life (DQoL) instrument is made up of 15 questions, all scored on a 5-point Likert scale assessing either frequency (never - all the time) or satisfaction (very satisfied - very dissatisfied). Higher scores on DQOL items indicate higher frequency of negative impact of diabetes or higher dissatisfaction with diabetes care and quality of life.

Inclusion Criteria: Have type 2 diabetes mellitus (DM) Are 18 years or older of age English speaking Are a patient in the ED or inpatient at TJUH or Methodist Hospital Have a HbA1c >8% as assessed during enrollment screening Have a PCP or Endocrine provider Have access to a device (e.g., smartphone, tablet, computer with WiFi) that can support video visits Have access to an email account (either own or family member) Are able to keep food fresh for a week after delivery (e.g., not homeless) and reheat food using a microwave or oven Live in the MANNA service area Exclusion Criteria: Have type 1 DM (to reduce confounding related to underlying differences in pathophysiology and primary treatment approaches (lifestyle change vs medication regimens) between type 1 and type 2 DM) Pregnant, planning to be pregnant, or currently breastfeeding, as these patients have unique nutritional requirements outside the MNT curriculum Does not eat by mouth (i.e. receives tube feeds or total parenteral nutrition [TPN]) History of severe gastroparesis Has a life expectancy of less than one year, per clinician Inability to provide informed consent Psychiatric co-morbidity that would preclude study participation Major communication barriers such as visual or hearing impairment that would compromise the ability to participate in a video visit Lives in or is being discharged to a facility (incarcerated, IP rehab, nursing home) Currently receiving MTM from MANNA or another provider Have life threatening food allergies Currently participating in DM-I or PREVENT study Has 3 or more ED visits for alcohol or other substance use in past month