Food is Medicine: Randomized Trial of Medically-Tailored Food Support for Diabetes Health
Trial Snapshot
- Phase
- Not Applicable
- Status
- Completed
- Enrollment
- 281
- Locations
- 1
- Primary Endpoint
- Change in Hemoglobin A1c From Baseline to Six Months by Study Arm
Study Overview
Brief Summary
This is a pragmatic randomized controlled trial (RCT) of Changing Health through Food Support for Diabetes (CHEFS-DM). This pragmatic RCT will leverage Project Open Hand's (POH) real-world programs to test the impact of a six month medically tailored food support and nutrition intervention ("CHEFS-DM") on glycemic control and other cardiometabolic outcomes, investigate the paths through which CHEFS-DM may durably improve health, and assess the economic value of the intervention to society.
Detailed Description
CHEFS-DM is a six-month intervention providing a diabetes-tailored diet meeting approximately 75% of daily energy requirements and designed based on American Diabetes Association (ADA) guidelines, together with diabetes nutritional education provided by registered dietitians. In this pragmatic RCT, investigators propose to rigorously evaluate the intervention among 440* low-income adults with Type 2 Diabetes Mellitus (T2DM) recruited from networks of safety net clinics in San Francisco and Alameda counties, California.
Participants will be randomized 1:1 to the intervention (n=220*) versus control (n=220*), using a parallel design. The study aims are as follows:
Aim 1: To determine the impact of CHEFS-DM on glycemic control and other cardiometabolic outcomes.
Aim 2: To determine the impact of CHEFS-DM on intermediate outcomes that may mediate any impact of CHEFS-DM on T2DM health.
Participant will be followed for 6 months (control arm) and 12 months (intervention arm) with structured interviews, anthropometric assessments, 24-hour dietary recalls, blood pressure measurements, blood draws, and semi structured interviews. Follow-up will be broken up into two phases. From baseline to six months, researchers will implement the CHEFS-DM intervention and follow both intervention and control arm participants (n=440*). After the end of the six-month follow-up, the intervention arm will transition to receiving standard POH services comprising 33% to 67% of daily energy requirements depending on health status, and will be followed for an additional six months, in order to assess the extent to which any health benefits are sustained at 12 months (n=220*). Hence, the control arm will be followed for six months and participate in two sets of assessments (baseline and at six months), and the intervention arm will be followed for 12 months and participate in three sets of assessments (baseline, six months, and twelve months).
Study Design
- Study Type
- Interventional
- Allocation
- Randomized
- Intervention Model
- Parallel assignment
- Primary Purpose
- Treatment
- Masking
- Single (Investigator)
Masking Description
Single (Investigator) The investigator(s) reviewing the analyses of the data will be masked to whether participants were in the intervention or control arm.
Eligibility Criteria
- Ages
- 18 Years to — (Adult, Older Adult)
- Sex
- All
- Accepts Healthy Volunteers
- No
Inclusion Criteria
- Confirmed diagnosis of T2DM confirmed by medical or laboratory records. For T2DM, any of the following criteria will be considered (from the American Diabetes Association):
- glycated hemoglobin (HbA1c) ≥6.5%, or
- fasting plasma glucose of ≥126, or
- a 2-hour plasma glucose level of 200 or higher during a 75 g oral glucose tolerance test, or
- a random plasma glucose of 200 or higher in patients with symptoms of hyperglycemia
- Age ≥18 years.
- Screening positive for food insecurity (at least one positive answer) in the previous 6 months assessed using the 6-item version of the US Household Food Security Survey Module (US Department of Agriculture), or has household income <200% of the federal poverty level.
- English or Spanish language fluency.
- Adequate cognitive and hearing capacity to consent and complete study measures.
- Reside in Alameda County or San Francisco County.
Exclusion Criteria
- Type-1 diabetes mellitus
- Individuals with disorders known to affect the accuracy of the HbA1c measure (e.g., end stage renal disease and individuals with known hemoglobinopathies).
- Inability to attend the educational workshops.
- Inability to schedule baseline assessments and/or blood draw after repeated requests.
- Pregnant individuals, or individuals planning to get pregnant within 6 months, or are lactating, or are postpartum less than 6 months.
- Current POH clients, past POH clients who stopped services less than 6 months prior, or past or present participants in other POH medically tailored meals studies.
- Does not have access to food storage, including a refrigerator and freezer to safely keep food.
- Does not have access to facilities to reheat and prepare meals using Project Open Hand food.
- Anticipates moving out of study area of Alameda and San Francisco Counties in the next 6 months.
- Receives more than 1 meal per day from a free food support resource or agency.
- Allergic to or will not eat eggs, soy, wheat, nuts, seeds or seed oils, or other foods commonly included among ingredients in POH meals.
- Allergic to dairy products, or unable to tolerate any dairy products including milk, yogurt and cheese.
- Individual does not eat any or all of POH's meat meal options and will not eat the vegetarian POH meal option.
Arms & Interventions
Standard of care
During the active trial period, control participants will receive standard of care as offered by clinical partners to all T2DM patients, meaning that they will receive the same care as if they were not enrolled in the study [no intervention]. This may include referral by their providers to nutritional counseling, T2DM support groups, local diabetes self-management programs, or locally available food support services, entirely independent of the study.
At the end of 6 month follow up (after end of active comparison), the control arm will receive similar services from POH to what the intervention arm received during the intervention, regardless POH eligibility criteria (6 months of DM-tailored food support to meet 67% of their daily requirements, video recording of the 4 CHEFS-DM education classes, and access to a POH dietitian at their request). They will only receive this after their endline assessment period has ended.
Food support and nutrition education
The intervention entails two components: 1) food support that consists of weekly medically tailored meals and healthy groceries that on average covers 75% of daily energy requirements from baseline to six months and 2) diabetes-tailored nutritional education that consists of two individual counseling sessions with a Registered dietitian and four group education sessions.
Intervention: Food support (Other)
Food support and nutrition education
The intervention entails two components: 1) food support that consists of weekly medically tailored meals and healthy groceries that on average covers 75% of daily energy requirements from baseline to six months and 2) diabetes-tailored nutritional education that consists of two individual counseling sessions with a Registered dietitian and four group education sessions.
Intervention: Nutritional Counseling and education (Other)
Planned outcome measures
Primary Outcomes
Change in Hemoglobin A1c From Baseline to Six Months by Study Arm
Time Frame: Baseline and 6-months follow up
Change HbA1c levels (%) from baseline to six months by study arms.
Change in Food Insecurity Severity From Baseline to Six Months by Study Arm
Time Frame: Baseline and 6-months follow-up
The US adult food security survey module from the US Department of Agriculture (USDA) will be used to assess the change in the food security scores from baseline to six months. The score ranges from 0 to 10. Higher score indicates higher severity of food insecurity
Change in Hemoglobin A1c from baseline to six months by study arm
Time Frame: Baseline and six months
Change HbA1c levels (%) from baseline to six months by study arms.
Change in food insecurity severity from baseline to six months by study arm
Time Frame: Baseline and six months
The US adult food security survey module from the US Department of Agriculture (USDA) will be used to assess the change in the food security scores from baseline to six months. The score ranges from 0 to 10. Higher score indicates higher severity of food insecurity
Secondary Outcomes
- Change in the Percentage of Participants With Low and Very Low Food Security(Baseline and 6-month follow-up)
- Change in Percentage of Participants With HbA1c <9% by Study Arm.(Baseline and 6-month follow-up)
- Change in Systolic Blood Pressure From Baseline to Six Months(Baseline and 6-month follow-up)
- Change in Oral Medication Adherence From Baseline to Six Months by Study Arm(Baseline and 6-month follow-up)
- Change in Percentage of Participants With Acute Health Care Utilization From Baseline to Six Month by Study Arm(Baseline and 6-month)
- Change in Percentage of Participants With a Least One Hospital Admission From Baseline to Six Months by Study Arm(Baseline to six months)
- Change in Percentage of Participants Missing at Least One Outpatient Visit From Baseline to Six Months by Study Arm(Baseline to six months)
- Change in Health-related Quality of Life Scores From Baseline to Six Months by Study Arm(Baseline and 6-month follow-up)
- Change in Body Mass Index (BMI) From Baseline to Six Months by Study Arm(Baseline and six months)
- Change in Healthy Eating Index Score 2015 (HEI-2015) From Baseline to Six Months by Study Arm.(Baseline and 6-months follow-up)
- Change in Depressive Symptoms From Baseline to Six Months by Study Arm(Baseline and six months)
- Change in Diabetes Distress Scale From Baseline to Six Months by Study Arm(Baseline and 6-month follow-up)
- Change in Diabetes Self-efficacy From Baseline to Six Months by Study Arm(Baseline and six months)
- Change in the proportion of participants with low and very low food security(Baseline and six months)
- Change in proportion of participants with glucose control by study arm.(Baseline and six months)
- Change in systolic and diastolic blood pressure from baseline to six months(Baseline and 6 months)
- Change in medication adherence from baseline to six months by study arm(Baseline and six months)
- Acute health care utilization from baseline to six month by study arm(Baseline and six months)
- Hospital admissions from baseline to six months by study arm(Baseline to six months)
- Missed outpatient visits from baseline to six months by study arm(Baseline to six months)
- Change in health-related quality of life scores from baseline to six months by study arm(Baseline and six months)
- Change in body mass index (BMI) from baseline to six months by study arm(Baseline and six months)
- Change in healthy eating index score 2015 (HEI-2015) from baseline to six months by study arm.(Baseline and six months)
- Change in depressive symptoms from baseline to six months by study arm(Baseline and six months)
- Change in diabetes distress scale from baseline to six months by study arm(Baseline and six months)
- Change in diabetes self-efficacy from baseline to six months by study arm(Baseline and six months)
Results
Results posted to ClinicalTrials.gov on 2026-09-29. View on ClinicalTrials.gov
Participant flow
281 started · 130 completed
Primary outcomes
Change in Hemoglobin A1c From Baseline to Six Months by Study Arm
Percentage of glycated hemoglobin · Standard Deviation · Time Frame: Baseline and 6-months follow up
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| 0.05 (1.74) | -0.27 (1.83) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention. For example, a participant who answered the survey at 6 months (retained) but for whom we were unable to obtain an HbA1c measurement would be excluded from the HbA1c outcomes analysis.
Slope -0.345 · 95% CI -0.770–0.080 · p = 0.111 · ANCOVA
Change in Food Insecurity Severity From Baseline to Six Months by Study Arm
score on a scale · Standard Deviation · Time Frame: Baseline and 6-months follow-up
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| -0.39 (2.27) | -1.39 (2.83) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope -0.916 · 95% CI -1.502–-0.331 · p = 0.002 · ANCOVA
Other outcomes (15)
Change in the Percentage of Participants With Low and Very Low Food Security
Percentage of participants · Time Frame: Baseline and 6-month follow-up
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| -2.83 | -21.54 |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Odds Ratio (OR) 0.558 · 95% CI 0.302–1.005 · p = 0.056 · Regression, Logistic
Change in Percentage of Participants With HbA1c <9% by Study Arm.
Percentage of participants · Time Frame: Baseline and 6-month follow-up
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| -9.43 | 3.85 |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Odds Ratio (OR) 1.832 · 95% CI 0.972–3.526 · p = 0.064 · Regression, Logistic
Change in Systolic Blood Pressure From Baseline to Six Months
mmHg · Standard Deviation · Time Frame: Baseline and 6-month follow-up
| Standard of Care (n=99) | Food Support and Nutrition Education (n=115) |
|---|---|
| 0.02 (16.79) | -1.8 (17.6) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope -1.972 · 95% CI -6.021–2.076 · p = 0.338 · ANCOVA
Change in Oral Medication Adherence From Baseline to Six Months by Study Arm
percentage adherence · Standard Deviation · Time Frame: Baseline and 6-month follow-up
| Standard of Care (n=96) | Food Support and Nutrition Education (n=119) |
|---|---|
| -2.94 (19.78) | -2.52 (21.93) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope 0.136 · 95% CI -4.902–5.173 · p = 0.958 · ANCOVA
Change in Percentage of Participants With Acute Health Care Utilization From Baseline to Six Month by Study Arm
Percentage of participants · Time Frame: Baseline and 6-month
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| -7.55 | -12.31 |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Odds Ratio (OR) 0.934 · 95% CI 0.523–1.667 · p = 0.817 · Regression, Logistic
Change in Percentage of Participants With a Least One Hospital Admission From Baseline to Six Months by Study Arm
Percentage of participants · Time Frame: Baseline to six months
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| -8.49 | -6.15 |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Odds Ratio (OR) 1.304 · 95% CI 0.580–3.041 · p = 0.525 · Regression, Logistic
Change in Percentage of Participants Missing at Least One Outpatient Visit From Baseline to Six Months by Study Arm
Percentage of participants · Time Frame: Baseline to six months
| Standard of Care (n=104) | Food Support and Nutrition Education (n=127) |
|---|---|
| -5.77 | -4.72 |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Odds Ratio (OR) 0.818 · 95% CI 0.383–1.749 · p = 0.602 · Regression, Logistic
Change in Health-related Quality of Life Scores From Baseline to Six Months by Study Arm
score on a scale · Standard Deviation · Time Frame: Baseline and 6-month follow-up
| Standard of Care (n=100) | Food Support and Nutrition Education (n=121) |
|---|---|
| 0.33 (6.38) | 2.6 (7.28) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope 2.448 · 95% CI 0.830–4.067 · p = 0.003 · ANCOVA
Change in Body Mass Index (BMI) From Baseline to Six Months by Study Arm
kg/m^2 · Standard Deviation · Time Frame: Baseline and six months
| Standard of Care (n=96) | Food Support and Nutrition Education (n=110) |
|---|---|
| -0.37 (1.54) | -0.35 (2.27) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope 0.062 · 95% CI -0.468–0.592 · p = 0.817 · ANCOVA
Change in Healthy Eating Index Score 2015 (HEI-2015) From Baseline to Six Months by Study Arm.
score on a scale · Standard Deviation · Time Frame: Baseline and 6-months follow-up
| Standard of Care (n=97) | Food Support and Nutrition Education (n=113) |
|---|---|
| -1.80 (14.29) | 0.52 (16.14) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope 0.900 · 95% CI -2.329–4.130 · p = 0.583 · ANCOVA
Change in Depressive Symptoms From Baseline to Six Months by Study Arm
score on a scale · Standard Error · Time Frame: Baseline and six months
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| -0.78 (5.23) | -1.92 (5.12) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope -1.037 · 95% CI -2.118–0.009 · p = 0.055 · ANCOVA
Change in Diabetes Distress Scale From Baseline to Six Months by Study Arm
score on a scale · Standard Deviation · Time Frame: Baseline and 6-month follow-up
| Standard of Care (n=106) | Food Support and Nutrition Education (n=129) |
|---|---|
| 0.05 (1.16) | -0.37 (1.09) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope -0.303 · 95% CI -0.544–-0.060 · p = 0.013 · ANCOVA
Change in Diabetes Self-efficacy From Baseline to Six Months by Study Arm
score on a scale · Standard Deviation · Time Frame: Baseline and six months
| Standard of Care (n=106) | Food Support and Nutrition Education (n=130) |
|---|---|
| -0.32 (7.84) | 1.76 (8.32) |
Per our pre-registered protocol, this analysis includes all participants with data at both baseline and 6 months. These numbers may differ from the Participant Flow module if there was missing data on this specific outcome measure at one or more time points, despite overall study retention.
Slope 1.248 · 95% CI -0.672–3.064 · p = 0.187 · ANCOVA
Durability of HbA1c Values From Six Months to Twelve Months in the Intervention Arm.
Percentage of glycated hemoglobin · Standard Deviation · Time Frame: Six and twelve months
| Food Support and Nutrition Education (n=110) |
|---|
| 0.126 (1.426) |
Intervention participants with HbA1c data available at both 6 and 12 months (n=110) were included in the analysis.
Mean Difference (Net) 0.126 · 95% CI -0.143–0.396 · p = 0.354 · Paired t-test
Durability in Scores of Food Security From Six to Twelve Months in the Intervention Arm.
score on a scale · Standard Deviation · Time Frame: Six and twelve months
| Food Support and Nutrition Education (n=71) |
|---|
| -0.437 (2.786) |
Intervention participants with food security data from the survey available at both 6 and 12 months (n=71) were included in the analysis.
Mean Difference (Net) -0.437 · 95% CI -1.096–0.223 · p = 0.191 · Paired t-test
Safety
| Group | Serious adverse events | Deaths |
|---|---|---|
| Standard of Care | 11 of 125 | 0 of 125 |
| Food Support and Nutrition Education | 25 of 156 | 0 of 156 |
Most frequent serious adverse events (participants)
| Event | Standard of Care | Food Support and Nutrition Education |
|---|---|---|
| Any hospital admission (self-reported) | 11 of 125 | 25 of 156 |
Values as posted by the sponsor on ClinicalTrials.gov, not recalculated. Values in brackets are the posted dispersion (e.g. 95% CI).
Publications
- BackgroundPalar K, Napoles T, Hufstedler LL, Seligman H, Hecht FM, Madsen K, Ryle M, Pitchford S, Frongillo EA, Weiser SD. Comprehensive and Medically Appropriate Food Support Is Associated with Improved HIV and Diabetes Health. J Urban Health. 2017 Feb;94(1):87-99. doi: 10.1007/s11524-016-0129-7. PubMed 28097614
- BackgroundSeligman HK, Lyles C, Marshall MB, Prendergast K, Smith MC, Headings A, Bradshaw G, Rosenmoss S, Waxman E. A Pilot Food Bank Intervention Featuring Diabetes-Appropriate Food Improved Glycemic Control Among Clients In Three States. Health Aff (Millwood). 2015 Nov;34(11):1956-63. doi: 10.1377/hlthaff.2015.0641. PubMed 26526255
- BackgroundSeligman HK, Schillinger D. Hunger and socioeconomic disparities in chronic disease. N Engl J Med. 2010 Jul 1;363(1):6-9. doi: 10.1056/NEJMp1000072. No abstract available. PubMed 20592297
- BackgroundGurvey J, Rand K, Daugherty S, Dinger C, Schmeling J, Laverty N. Examining health care costs among MANNA clients and a comparison group. J Prim Care Community Health. 2013 Oct;4(4):311-7. doi: 10.1177/2150131913490737. Epub 2013 Jun 3. PubMed 23799677
- BackgroundBerkowitz SA, Terranova J, Hill C, Ajayi T, Linsky T, Tishler LW, DeWalt DA. Meal Delivery Programs Reduce The Use Of Costly Health Care In Dually Eligible Medicare And Medicaid Beneficiaries. Health Aff (Millwood). 2018 Apr;37(4):535-542. doi: 10.1377/hlthaff.2017.0999. PubMed 29608345
- BackgroundBerkowitz SA, Terranova J, Randall L, Cranston K, Waters DB, Hsu J. Association Between Receipt of a Medically Tailored Meal Program and Health Care Use. JAMA Intern Med. 2019 Jun 1;179(6):786-793. doi: 10.1001/jamainternmed.2019.0198. PubMed 31009050
- BackgroundSeligman HK, Jacobs EA, Lopez A, Tschann J, Fernandez A. Food insecurity and glycemic control among low-income patients with type 2 diabetes. Diabetes Care. 2012 Feb;35(2):233-8. doi: 10.2337/dc11-1627. Epub 2011 Dec 30. PubMed 22210570
- BackgroundSeligman HK, Jacobs EA, Lopez A, Sarkar U, Tschann J, Fernandez A. Food insecurity and hypoglycemia among safety net patients with diabetes. Arch Intern Med. 2011 Jul 11;171(13):1204-6. doi: 10.1001/archinternmed.2011.287. No abstract available. PubMed 21747017
- BackgroundMarpadga S, Fernandez A, Leung J, Tang A, Seligman H, Murphy EJ. Challenges and Successes with Food Resource Referrals for Food-Insecure Patients with Diabetes. Perm J. 2019;23:18-097. doi: 10.7812/TPP/18-097. PubMed 30939269
Investigators
Study Sites (1)
Identifiers
- NCT ID
- NCT04959487
- Other Study IDs
- R01DK124630, R01DK124630
Dates
- First Submitted
- (5 years ago)
- First Posted
- (5 years ago)
- Primary Completion
- (last year)
- Study Completion
- (last year)
- Last Verified
- (29 days ago)
- Last updated
- (yesterday)
Regulatory & Sharing
- FDA Regulated Drug
- No
- FDA Regulated Device
- No
- IPD Sharing Plan
- No
- Has Results
- Yes