When doctors tell patients with diabetes, heart disease or kidney disease to improve their diets, the advice can sound straightforward. But eating healthier is not always as simple as choosing different groceries. Some people cannot afford the food they need. Others have health conditions, mobility limitations or demanding circumstances that make shopping and cooking difficult.
A real-world observational study published in Nature Medicine found that Massachusetts Medicaid members who received medically tailored meals had fewer hospitalizations and emergency department visits while enrolled in the program. Their overall health care costs were also lower.
Because the study was not a randomized clinical trial, it cannot prove that the meals alone caused those improvements. Still, the findings suggest that providing appropriate food may help some high-risk patients manage serious health conditions more effectively.
The study adds to growing interest in “food is medicine” programs, which aim to make nutrition support a more integrated part of health care. Senior author Dariush Mozaffarian, director of the Food is Medicine Institute at Tufts University, has reported advisory and consulting relationships with several health, nutrition and technology companies. First author Kurt Hager has reported consulting work with the nonprofit Food is Medicine Coalition. Those relationships do not invalidate the research, but they are important context when considering the findings.
Medically tailored meals are not generic prepared foods or meal kits. They are designed for people with diet-sensitive conditions and adapted to their medical needs, dietary restrictions and preferences.
In the Massachusetts program, participants first met with a registered dietitian nutritionist. They then received five lunches and five dinners each week, along with snacks. The meals were prepared and delivered by Community Servings, a Boston-based nonprofit.
The researchers analyzed Medicaid claims data collected from 2020 to 2023 across 11 Massachusetts health systems. They compared 1,866 adults who received medically tailored meals for more than 90 days with 1,372 eligible adults who did not enroll in the program.
Participants received meals for an average of about seven months. During that time, meal recipients had 31% fewer hospitalizations and 20% fewer emergency department visits than the comparison group.
Their total health care costs were also $3,433 lower per person during enrollment. The meal program cost an average of $3,512 per participant, meaning the reduction in medical spending offset 98% of the program’s cost.
The researchers also found that primary care visits did not decline. That detail matters because it suggests participants continued receiving routine medical care even as their use of more intensive and expensive services decreased.
“These findings show that medically tailored meals can be both clinically effective and economically sustainable within Medicaid,” Hager said.
The financial impact was not identical for every participant. The largest reductions in health care costs were seen among people with more complex medical needs, including cardiovascular disease, chronic kidney disease, diabetes and depression. People with multiple health conditions also appeared to benefit more.
That distinction is important. The study does not suggest that home-delivered meals would produce the same savings for everyone. Instead, the findings indicate that medically tailored meal programs may be most useful for patients who are already at elevated risk of hospitalization and other costly health emergencies.
The meals may help in several ways. Tailoring food to a patient’s medical needs could improve nutrition. Providing meals at no cost could also reduce food insecurity and ease difficult financial tradeoffs. Someone who receives prepared meals may have more money available for medications, rent or utilities. Reliable food access may also reduce stress and make it easier to follow a treatment plan.
The study was not designed to measure how much each of those factors contributed to the results. It also did not track exactly what participants ate, so the researchers could not confirm how closely people followed the meal plans.
There are other limitations. Because participants were not randomly assigned to receive meals, people who enrolled may have differed from eligible nonparticipants in ways that Medicaid claims data could not capture. They may have been more engaged with their health care or better positioned to benefit from the program.
The findings also reflect a program run by an established nonprofit serving people with serious health conditions and food insecurity. Results could differ with other providers, patient populations or state programs.
The length of enrollment may matter, too. The main analysis focused on participants who received meals for more than 90 days. When the researchers included people with shorter enrollment periods, hospitalizations and emergency department visits still decreased, but the reduction in overall health care costs was smaller and was not statistically significant.
The study highlights a practical issue that is easy to overlook in conversations about healthy eating. Nutrition advice can only go so far when people do not have consistent access to the food they need or the ability to prepare it.
Medically tailored meals will not replace medical care. But for some patients, access to appropriate food may be one part of making that care work better.
The study was supported by the National Institutes of Health and the Massachusetts Executive Office of Health and Human Services.