Food as Medicine: What the Research Says About Using Diet to Prevent and Treat Disease

The phrase “let food be thy medicine” is widely attributed to Hippocrates, though historians note the exact quote likely never appeared in his writings. What is not in dispute is the growing body of evidence that dietary patterns play a central role in the development, management, and in some cases reversal of chronic diseases — from type 2 diabetes to cardiovascular disease to certain cancers. In the United States, where six in ten adults live with at least one diet-related chronic disease and diet-related illness costs the health care system an estimated .1 trillion annually, an entire clinical and policy movement has emerged under the umbrella term “food as medicine.”

In 2023, the American Heart Association issued a Presidential Advisory formally defining food-as-medicine interventions and calling for their integration into the health care system. In 2025, the AHA followed up with a systematic review of all randomized controlled trials in the field. And a 2025 analysis in Health Affairs estimated that scaling medically tailored meal programs to all 50 states could prevent billions of dollars in health care spending while improving patient outcomes.

This article examines what “food as medicine” actually means, what the clinical trial evidence shows, how existing programs work, and what the limitations are.

What Is Food as Medicine?

“Food as medicine” is not a single intervention but a spectrum of programs that use food and nutrition as clinical tools for preventing, managing, or treating disease. The American Heart Association’s 2023 Presidential Advisory — the first major consensus statement from a leading medical organization on the topic — defined three principal categories:

Produce prescriptions (PRx). Health care providers write prescriptions for fruits and vegetables that patients can redeem at participating grocery stores, farmers’ markets, or food pharmacies. The “prescriptions” are typically loaded onto a debit card or voucher, with a value of 0–0 per month. The goal is to improve diet quality, increase food security, and reduce cardiometabolic risk factors — particularly among low-income patients who face both economic and geographic barriers to healthy eating.

Medically tailored meals (MTM). Fully prepared meals designed by registered dietitian nutritionists (RDNs) to meet the specific nutritional requirements of a patient’s medical diagnosis, symptoms, allergies, and medication interactions. Meals are typically delivered to the patient’s home on a weekly basis. MTM programs were originally developed for people living with HIV/AIDS and have since expanded to serve patients with cancer, heart failure, diabetes, kidney disease, and other serious illnesses. Organizations like MANNA (Metropolitan Area Neighborhood Nutrition Alliance), God’s Love We Deliver, and Community Servings are among the largest MTM providers in the United States.

Medically tailored groceries (MTG). Similar to MTM but providing unprepared ingredients rather than ready-to-eat meals, along with nutrition education, recipes, and cooking instruction. MTG programs give patients more autonomy and skill-building opportunities but require the ability to cook, store food, and follow dietary guidance.

What Does the Research Say?

The evidence base for food-as-medicine interventions has grown rapidly in recent years, though it remains uneven across the three categories. A 2025 American Heart Association Scientific Statement systematically reviewed all published randomized controlled trials (RCTs) of food-as-medicine interventions for noncommunicable diseases in the United States and found that the strongest evidence exists for medically tailored meals, with more mixed results for produce prescriptions.

Medically Tailored Meals: The Strongest Evidence

The most cited study in the field is the Community Aging in Place — Advancing Better Living for Elders (CAPABLE) trial and related work by Berkowitz and colleagues. A 2018 study published in Health Affairs analyzed insurance claims data for dually eligible Medicare and Medicaid beneficiaries receiving medically tailored meals from Community Servings in Boston. Recipients had significantly fewer hospital admissions (49% reduction), fewer emergency department visits, and substantially lower total health care costs compared to a matched control group.

A randomized trial published in JAMA Internal Medicine in 2023 (Doyle et al.) tested an intensive food-as-medicine program for food-insecure adults with diabetes (HbA1c greater than 8.0%). Participants received access to a “fresh food pharmacy” where they could obtain nutritious groceries weekly, along with dietitian counseling and diabetes self-management education. The intervention lasted up to 12 months and cost approximately ,000 per participant. The results, however, were mixed: the intervention significantly improved food security and self-reported diet quality but did not produce statistically significant reductions in HbA1c or other clinical biomarkers during the study period — a finding that generated considerable discussion about the optimal design of food-as-medicine programs.

A 2025 modeling study published in Health Affairs (Deng et al.) estimated the national impact if medically tailored meal programs were scaled across all 50 U.S. states. The analysis projected that such a scale-up could prevent significant health care utilization and reduce expenditures by an estimated 8.4 billion annually for Medicare and Medicaid combined — making it one of the most cost-effective chronic disease interventions available.

In 2026, the FAME-F (Food as Medicine for Families) factorial randomized trial provided new evidence on optimal program design. The trial, led by Berkowitz and colleagues at UNC, tested two dimensions of MTM delivery: whether meals should feed only the patient or the entire household, and whether meals should be delivered by a dedicated driver (who also provides social contact) or a commercial shipper. Results showed that feeding the family improved diet quality scores, while dedicated drivers reduced loneliness — suggesting that social connection is an underrecognized component of food-as-medicine effectiveness.

Produce Prescriptions: Promising but Nuanced

Produce prescription programs are more widely implemented but have produced more variable results in clinical trials. A landmark 2023 multisite evaluation published in Circulation: Cardiovascular Quality and Outcomes (Hager et al.) assessed 9 produce prescription programs across the United States. The study found significant improvements in fruit and vegetable consumption and food security among participants, with a dose-response relationship — higher-value prescriptions led to greater dietary improvements. However, the study was not a randomized trial, and changes in clinical biomarkers like blood pressure and HbA1c were modest and inconsistent across sites.

A separate modeling study published in Circulation in 2022 (Lee et al.) estimated that if produce prescriptions were covered by Medicare and Medicaid for all beneficiaries with diet-related conditions, the program could prevent 296,000 cardiovascular events, prevent 120,000 diabetes cases, and save 9.7 billion in health care costs over a lifetime horizon — while costing 0.8 billion to implement. The net cost was estimated at just .1 billion, making it a near-break-even investment with substantial health gains.

How to Use Food as Medicine

Applying food-as-medicine principles does not require enrollment in a formal program. Decades of nutrition research point to several dietary patterns that consistently reduce the risk of chronic disease across diverse populations:

The Mediterranean diet is the most extensively studied dietary pattern in clinical medicine. Rich in olive oil, legumes, whole grains, fish, fruits, vegetables, nuts, and moderate red wine, the Mediterranean diet has been shown in the PREDIMED trial — a landmark RCT involving 7,447 participants at high cardiovascular risk — to reduce the incidence of major cardiovascular events (heart attack, stroke, cardiovascular death) by approximately 30% compared to a control diet. The trial was originally published in the New England Journal of Medicine in 2013 and republished with corrections in 2018.

The DASH diet (Dietary Approaches to Stop Hypertension) was developed specifically to lower blood pressure. The original DASH trial, published in the New England Journal of Medicine in 1997, found that the diet lowered systolic blood pressure by an average of 5.5 mmHg and diastolic by 3.0 mmHg in just 8 weeks — effects comparable to a first-line antihypertensive medication. The DASH diet emphasizes fruits, vegetables, whole grains, lean protein, and low-fat dairy while limiting sodium, saturated fat, and added sugars.

Whole-food plant-based diets have shown particular promise for type 2 diabetes management. A 2019 meta-analysis in BMJ Open Diabetes Research found that plant-based diets improved glycemic control (lowering HbA1c), reduced body weight, and improved cholesterol levels in diabetic patients. The mechanisms likely include higher fiber intake (which slows glucose absorption), lower saturated fat (which improves insulin sensitivity), and weight reduction.

All three patterns share common features: high intake of minimally processed plant foods, limited added sugars and ultra-processed foods, and moderate or no consumption of red and processed meat. None requires expensive specialty foods or supplements.

Food as Medicine Companies and Programs

The food-as-medicine landscape in the United States includes both nonprofit organizations and a growing number of companies operating at the intersection of food, technology, and health care:

Geisinger Health Fresh Food Farmacy. Launched in 2016 in rural Pennsylvania, this program provides food-insecure patients with type 2 diabetes access to a food pharmacy stocked with nutritious groceries — free of charge — along with dietitian counseling, cooking classes, and diabetes education. Published results show that participants experienced a 2.1-point reduction in HbA1c over the program’s first two years, along with reduced emergency department visits.

Boston Medical Center Preventive Food Pantry. One of the first hospital-based food pantries in the United States, serving food-insecure patients since 2001. The pantry provides nutritious groceries and has been integrated into the hospital’s electronic health record system, allowing providers to screen for food insecurity and refer patients directly.

God’s Love We Deliver. A New York City-based nonprofit that prepares and delivers medically tailored meals to people living with serious illness. The organization delivers approximately 3 million meals annually and has been the subject of several peer-reviewed studies, including the ongoing MTM4CVH randomized trial at Columbia University testing MTM for cardiovascular health equity.

MANNA (Metropolitan Area Neighborhood Nutrition Alliance). Based in Philadelphia, MANNA has provided medically tailored meals since 1990. Research on MANNA’s program published by Berkowitz and colleagues has shown significant reductions in health care costs for recipients.

The GusNIP Program. The Gus Schumacher Nutrition Incentive Program, administered by the USDA, is the primary federal funding source for produce prescription projects. As of 2025, GusNIP funds programs in 35 states, supporting local partnerships between health care providers, farmers’ markets, and community organizations.

The Mediterranean Diet: The Most Studied Food-as-Medicine Approach

No dietary pattern has been studied more rigorously than the Mediterranean diet, and no other food-based intervention has produced such consistent cardiovascular outcomes across multiple randomized controlled trials.

The PREDIMED Trial

The PREDIMED (PREvencion con DIeta MEDiterranea) trial remains the largest and most influential randomized trial in the food-as-medicine field. Conducted across 11 centers in Spain, the study enrolled 7,447 men (55–80 years) and women (60–80 years) at high cardiovascular risk but without prior cardiovascular events. Participants were randomized to one of three groups: a Mediterranean diet supplemented with extra-virgin olive oil (free provision of 1 liter per week), a Mediterranean diet supplemented with mixed nuts (30 g per day), or a control diet (advice to reduce all dietary fat).

The trial was stopped early by the Data and Safety Monitoring Board after 4.8 years. The primary endpoint — a composite of nonfatal myocardial infarction, nonfatal stroke, and cardiovascular death — occurred 30% less often in both Mediterranean diet groups compared to the control group (hazard ratio 0.70 for olive oil, 0.70 for nuts). Stroke risk specifically dropped by 33–46% depending on the subgroup. These results, published in the New England Journal of Medicine in 2013 (and republished in 2018 after methodological corrections), marked the first time a large RCT had demonstrated that a dietary pattern alone — without calorie restriction, exercise mandates, or medication changes — could produce a magnitude of cardiovascular protection comparable to statin therapy (Estruch et al., NEJM, 2013/2018).

The CORDIOPREV Trial

The CORDIOPREV trial, published in its final cardiovascular endpoints in 2022, extended the evidence from primary to secondary prevention. This single-center Spanish study enrolled 1,002 patients aged 20–75 who had already experienced a coronary event. They were randomized to either a Mediterranean diet or a low-fat diet and followed for 7 years. The Mediterranean diet group showed an 18% reduction in major adverse cardiovascular events (P = 0.024) after multivariable adjustment. This was the first RCT to demonstrate that the Mediterranean diet reduces cardiovascular recurrence in patients who already have heart disease — not just those at risk of developing it.

Meta-Analytic Evidence as of 2024

A 2024 meta-analysis by Sebastian et al. (Current Problems in Cardiology) pooled data from all RCTs of the Mediterranean diet and cardiovascular outcomes through January 2024, including the Lyon Diet Heart Study, PREDIMED, and CORDIOPREV. The combined analysis confirmed a statistically significant reduction in major cardiovascular events. A separate 2025 umbrella review of 18 meta-analyses encompassing 238 RCTs found consistent evidence for cardiovascular benefit across primary and secondary prevention populations (PMC, 2025).

What makes the Mediterranean diet evidence unusual in nutrition science is the sheer volume and quality of randomized trial data supporting it. Most dietary recommendations are based on observational studies, which can show associations but cannot prove causation. The Mediterranean diet has been tested in the gold-standard format — large, long-term RCTs — with consistent results across different countries, populations, and study designs.

Beyond the Mediterranean Diet: Other Dietary Patterns Under Study

While the Mediterranean diet dominates the RCT literature, other dietary patterns have shown promise in specific disease contexts.

DASH Diet for Blood Pressure

The Dietary Approaches to Stop Hypertension (DASH) diet was developed specifically as a blood pressure intervention through NIH-funded research. Two landmark RCTs — the original DASH trial (1997) and the DASH-Sodium trial (2001) — demonstrated that a diet rich in fruits, vegetables, whole grains, and low-fat dairy, combined with reduced sodium, lowered systolic blood pressure by 5.5–11.4 mmHg compared to a typical American diet. This reduction is comparable to a single antihypertensive medication. The DASH diet remains a first-line recommendation from the American College of Cardiology and American Heart Association for patients with stage 1 hypertension.

Low-Carbohydrate and Ketogenic Diets for Type 2 Diabetes

A growing number of clinical trials have examined low-carbohydrate and ketogenic diets as interventions for type 2 diabetes management. A 2024 meta-analysis in Diabetes Care found that very low-carbohydrate diets (<50 g/day) produced greater reductions in HbA1c (a marker of long-term blood sugar control) at 6 months compared to control diets, though the difference narrowed at 12 months. Some patients achieved medication reduction or discontinuation. The evidence is strongest for short-term glycemic control, with ongoing debate about long-term sustainability and cardiovascular effects.

Plant-Based Diets and Cardiometabolic Risk

Prospective cohort studies consistently associate higher intake of plant-based foods with lower cardiovascular risk. The Adventist Health Study 2, which followed over 96,000 Seventh-Day Adventists, found that vegetarians and vegans had significantly lower rates of hypertension, type 2 diabetes, and metabolic syndrome compared to non-vegetarians. However, randomized trial evidence specifically testing plant-based diets against cardiovascular endpoints remains limited, and most of the data come from observational studies rather than RCTs.

Food as Medicine in Policy: Legislative and Insurance Developments

The translation of food-as-medicine research into policy has accelerated in the past three years, with developments at both the federal and state level.

Federal Initiatives

The White House Conference on Hunger, Nutrition, and Health (September 2022) — the first such conference in more than 50 years — elevated food-as-medicine to a national policy priority. The resulting National Strategy included commitments to expand medically tailored meal pilot programs through Medicare and Medicaid, fund produce prescription programs through the USDA, and integrate nutrition screening into primary care.

In 2023, the Centers for Medicare and Medicaid Services (CMS) approved several state Medicaid waiver applications that included coverage for medically tailored meals as a health-related social needs service. California, Massachusetts, Oregon, and North Carolina were among the first states to implement Medicaid-funded food support programs for patients with diet-sensitive conditions.

The 2024 Farm Bill debate included provisions for expanding the Gus Schumacher Nutrition Incentive Program (GusNIP), which funds produce prescription programs. GusNIP had already invested over 20 million in nutrition incentive and produce prescription grants since its creation in 2014.

State-Level Programs

Several states have launched their own food-as-medicine initiatives:

California’s CalAIM (California Advancing and Innovating Medi-Cal) program, which began enrolling participants in 2022, allows managed care plans to cover medically tailored meals, produce prescriptions, and other nutrition interventions for Medi-Cal enrollees with qualifying conditions. Early data from the program showed reduced emergency department visits among participants receiving medically tailored meals.

Massachusetts piloted a medically tailored meal program through its MassHealth Medicaid system, partnering with Community Servings (a nonprofit meal provider) to deliver condition-specific meals to patients with HIV/AIDS, cancer, heart failure, diabetes, and renal disease. An evaluation published in Health Affairs found that participants had fewer hospitalizations, shorter lengths of stay, and lower monthly health care costs compared to a matched control group.

Insurance Coverage

Private insurance coverage for food-as-medicine interventions remains limited. Most commercial health plans do not cover medically tailored meals or produce prescriptions as standard benefits. However, some Medicare Advantage plans have begun offering supplemental nutrition benefits, including meal delivery and grocery assistance, as part of their Special Supplemental Benefits for the Chronically Ill (SSBCI). This trend has grown substantially since CMS expanded SSBCI eligibility rules in 2020.

The economic argument for insurance coverage is straightforward but requires long time horizons: food interventions cost relatively little per patient (,000–,000 per year for medically tailored meals), while the hospitalizations and complications they may prevent cost tens of thousands of dollars per event. The challenge is demonstrating this return on investment with randomized trial data at a scale that convinces payers — a challenge that currently active trials (FAME-F, CHEFS-DM, FOOD-HF, MTM4CVH) are designed to address.

The Emerging Science: Gut Microbiome and Dietary Intervention

One of the fastest-moving frontiers in food-as-medicine research involves the gut microbiome — the trillions of microorganisms that inhabit the human digestive tract and influence everything from immune function to mental health to metabolic processing.

Dietary patterns rapidly alter the composition of the gut microbiome. A 2014 study by David et al. (Nature) demonstrated that switching between plant-based and animal-based diets produced measurable shifts in gut bacterial populations within 24 hours. More recent work has shown that specific dietary components — fiber, polyphenols, fermented foods — selectively promote populations of beneficial bacteria that produce short-chain fatty acids (SCFAs), which reduce intestinal inflammation and improve metabolic markers.

The Stanford WELL trial (2024, Cell) randomized 36 healthy adults to either a high-fiber or high-fermented-food diet for 10 weeks. The fermented food group showed increased microbial diversity and reduced inflammatory markers (including interleukin-6 and C-reactive protein), while the high-fiber group showed changes in microbial carbohydrate processing capacity. These findings suggest that fermented foods — yogurt, kimchi, sauerkraut, kefir — may offer anti-inflammatory benefits through mechanisms that involve the microbiome rather than traditional nutrient pathways.

This research is still in its early stages, and translating microbiome findings into clinical dietary recommendations remains a work in progress. But it represents a potential mechanism by which dietary interventions produce health effects that go beyond the sum of their nutrient parts.

Who Benefits Most from Food-as-Medicine Interventions?

Not all patients benefit equally from food-as-medicine programs, and the emerging trial data are beginning to clarify who gains the most.

Patients with multiple chronic conditions. The strongest evidence for medically tailored meals comes from patients managing more than one diet-sensitive condition simultaneously — for example, a patient with type 2 diabetes and heart failure who also faces food insecurity. For these patients, the barrier is not knowledge (most know they should eat differently) but the practical capacity to procure, prepare, and afford appropriate food. Delivering condition-specific meals directly removes this barrier and produces measurable improvements in both diet quality and clinical markers.

Food-insecure populations. Food insecurity — defined as limited or uncertain access to adequate food — affects roughly 13% of U.S. households. For food-insecure patients with chronic disease, the treatment paradox is stark: a physician prescribes a diabetes-appropriate diet, but the patient cannot afford the foods that diet requires. Produce prescriptions and medically tailored groceries address this directly. The GusNIP-funded produce prescription programs have shown consistent improvements in fruit and vegetable intake among food-insecure participants, with some trials reporting HbA1c reductions of 0.3–0.6% — clinically meaningful for diabetes management.

Patients in post-acute transitions. The period immediately after hospital discharge is one of the highest-risk windows for readmission, particularly for heart failure patients. Several ongoing trials (FOOD-HF at UT Southwestern, MUTTON-HF at University of Pennsylvania) are specifically testing whether medically tailored meal delivery during this transition period reduces 30- and 90-day readmission rates. Early results from a pilot at Massachusetts General Hospital suggested that patients who received medically tailored meals post-discharge had 50% fewer readmissions at 30 days compared to historical controls, though this was not a randomized comparison.

Children and families. The FAME-F trial (Food as Medicine for Families), a 2024–2025 factorial randomized trial at Tufts University, is one of the first to test whether feeding the entire household — not just the patient — improves outcomes. The hypothesis is that dietary change is more sustainable when the whole family eats the same meals, removing the practical burden of preparing separate food for the patient. Results are expected in 2026.

The Limitations and Open Questions

Despite the momentum behind food-as-medicine initiatives, important limitations and unanswered questions remain.

The evidence base is still developing. As the AHA’s 2025 systematic review noted, many food-as-medicine studies have been observational or quasi-experimental rather than randomized controlled trials. The JAMA Internal Medicine trial (Doyle et al., 2023) — one of the few large, well-designed RCTs — failed to produce significant improvements in HbA1c for its fresh food pharmacy intervention, raising questions about whether improving food access alone is sufficient to change clinical outcomes without addressing other barriers (medication adherence, health literacy, chronic stress, housing instability).

Sustainability and scale. Most MTM programs operate with philanthropic funding and are not yet integrated into insurance reimbursement systems at scale. Several states — including Massachusetts, California, and Oregon — have obtained Medicaid waivers to cover MTM for qualifying patients, but nationwide coverage remains a policy aspiration, not a reality.

Cultural appropriateness. Early food-as-medicine programs were criticized for providing foods that did not align with participants’ cultural preferences, leading to lower adherence. Newer programs, including the FAME-F trial, have explicitly incorporated cultural tailoring into meal design. Research consistently shows that culturally appropriate meals improve both adherence and satisfaction.

Food is not a substitute for medical treatment. The food-as-medicine framework is intended to complement, not replace, conventional medical care. No credible food-as-medicine program recommends that patients stop taking prescribed medications in favor of dietary changes. The evidence supports diet as an adjunct that can reduce the need for medication, improve quality of life, and prevent disease progression — but it does not support the more extreme claims sometimes made in popular media that specific foods can “cure” diseases.

For individuals managing specific conditions, working with a registered dietitian nutritionist (RDN) who can tailor recommendations to their clinical situation remains the gold standard. The Academy of Nutrition and Dietetics maintains a searchable directory of credentialed RDNs at eatright.org.

Measuring What Matters: The Challenge of Outcomes in Food-as-Medicine Research

One of the persistent challenges in food-as-medicine research is defining what counts as success. Different studies measure different things, making it difficult to compare results across programs and draw overarching conclusions.

Clinical trials of medically tailored meals typically track biomarkers (HbA1c for diabetes, blood pressure for hypertension, LDL cholesterol for cardiovascular risk) and health care utilization (hospitalizations, emergency department visits, total cost of care). Produce prescription programs more commonly measure diet quality scores (such as the Healthy Eating Index), fruit and vegetable consumption, and food security status. Some programs track patient-reported outcomes like quality of life, mental health, and social isolation.

This variation is not accidental — it reflects the fact that food-as-medicine programs serve different populations with different goals. A medically tailored meal program for post-discharge heart failure patients is fundamentally trying to prevent readmission (a health system cost outcome), while a produce prescription program for food-insecure families with prediabetes is trying to shift long-term dietary patterns (a public health outcome). Both are legitimate goals, but they operate on different timescales and require different measurement frameworks.

The Food is Medicine Coalition (FIMC), a national network of nonprofit medically tailored meal providers, has developed standardized nutrition standards and outcome measurement tools to address this fragmentation. The AHA’s Healthcare by Food initiative, launched in 2024, is also working to harmonize measurement across pilot studies, with the explicit goal of producing data that can support policy decisions about insurance coverage.

A related challenge is the duration of follow-up. Most published food-as-medicine trials have run for 6 to 12 months. The cardiovascular diet trials (PREDIMED, CORDIOPREV) are notable exceptions, with follow-up periods of 5–7 years. For chronic disease management, the question is not just whether a 12-week meal delivery program improves HbA1c during the intervention period, but whether the dietary changes persist after the meals stop arriving. Few studies have collected post-intervention follow-up data, and those that have suggest that some benefits attenuate when the food support ends — reinforcing the case for sustained programs rather than short-term pilots.

Where Food as Medicine Goes from Here

The food-as-medicine field is at an inflection point. The research base is growing rapidly, with multiple large RCTs (CHEFS-DM, FOOD-HF, MTM4CVH, FAME-F) expected to report results between 2025 and 2027. Federal and state policy is moving toward coverage, with Medicaid waivers and Medicare Advantage benefits expanding. And the infrastructure — nonprofit meal providers, produce prescription networks, clinical dietitian partnerships — is being built out in real time.

The outstanding questions are not about whether diet affects health (that is settled) or whether food-as-medicine interventions can improve clinical outcomes in controlled settings (the evidence says yes). The questions are about scale, sustainability, and cost-effectiveness at the population level. Can a medically tailored meal program that works in a 200-person pilot produce the same results when it serves 50,000 patients across a state Medicaid system? Can produce prescription programs retain participants long enough to produce lasting dietary change? And can the health care savings from prevented hospitalizations and reduced medication use offset the cost of the food interventions — or at least bring them close enough to break even that payers are willing to invest?

These are implementation questions, not efficacy questions. And they are the right questions to be asking at this stage of the field’s development.

References

1. Volpp KG, Berkowitz SA, Sharma SV, et al. Food Is Medicine: A Presidential Advisory From the American Heart Association. Circulation. 2023;148(18):1417-1439. doi:10.1161/CIR.0000000000001182
2. Seligman HK, et al. A Systematic Review of “Food Is Medicine” Randomized Controlled Trials. AHA Scientific Statement. Circulation. 2025;152(4):e32-e46. doi:10.1161/CIR.0000000000001343
3. Mozaffarian D, et al. “Food Is Medicine” Strategies for Nutrition Security and Cardiometabolic Health Equity: JACC State-of-the-Art Review. J Am Coll Cardiol. 2024;83(8):843-864. doi:10.1016/j.jacc.2023.12.023
4. Berkowitz SA et al. Meal Delivery Programs Reduce the Use of Costly Health Care. Health Affairs. 2018;37(4):535-542. doi:10.1377/hlthaff.2017.0999
5. Doyle J, Alsan M, Skelley N. Effect of an Intensive Food-as-Medicine Program on Health and Health Care Use. JAMA Intern Med. 2024;184(2):154-163. doi:10.1001/jamainternmed.2023.6670
6. Deng S, Hager K, Wang L, et al. Estimated Impact of Medically Tailored Meals on Health Care Use and Expenditures in 50 US States. Health Affairs. 2025;44(4):433-442. doi:10.1377/hlthaff.2024.01307
7. Hager K et al. Impact of Produce Prescriptions on Diet, Food Security, and Cardiometabolic Health Outcomes: A Multisite Evaluation. Circ Cardiovasc Qual Outcomes. 2023;16:e009520. doi:10.1161/CIRCOUTCOMES.122.009520
8. Lee Y et al. Cost-Effectiveness of Financial Incentives for Improving Diet and Health Through Medicare and Medicaid. Circ Cardiovasc Qual Outcomes. 2022. doi:10.1161/CIRCOUTCOMES.121.008537
9. Berkowitz SA et al. FAME-F Factorial RCT: Medically Tailored Meals and Cardiovascular Health. Circ Cardiovasc Qual Outcomes. 2026. PMID: 41919373
10. Estruch R et al. Primary Prevention of Cardiovascular Disease with a Mediterranean Diet (PREDIMED). NEJM. 2018;378:e34. doi:10.1056/NEJMoa1800389
11. CDC. Chronic Disease and Health Promotion. cdc.gov/chronicdisease
12. USDA. GusNIP Produce Prescription. usda.gov
13. Office of Disease Prevention and Health Promotion. Research on Food Is Medicine Interventions. odphp.health.gov
14. Estruch R, Ros E, Salas-Salvado J, et al. Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts. N Engl J Med. 2018;378:e34. doi:10.1056/NEJMoa1800389
15. Delgado-Lista J, Alcala-Diaz JF, Torres-Pena JD, et al. Long-Term Secondary Prevention of Cardiovascular Disease with a Mediterranean Diet and a Low-Fat Diet (CORDIOPREV). Lancet. 2022;399:1876–1885.
16. Sebastian SA, Padda I, et al. Long-Term Impact of Mediterranean Diet on Cardiovascular Disease Prevention: A Systematic Review and Meta-Analysis of RCTs. Curr Probl Cardiol. 2024;49(5):102509. PubMed
17. Mozaffarian D, Aspry KE, Garfield K, et al. “Food Is Medicine” Strategies for Nutrition Security and Cardiometabolic Health Equity: JACC State-of-the-Art Review. J Am Coll Cardiol. 2024;83:843–864.
18. Comparing Two Food Is Medicine Trials Using the RE-AIM Framework. medRxiv. 2025. PMC
19. ODPHP. Research on Food Is Medicine Interventions. odphp.health.gov

Last updated: September 26, 2026