A recent randomized clinical trial investigating the impact of an intensive food-as-medicine program on patients with uncontrolled type 2 diabetes and self-reported food insecurity has shown limited effectiveness in improving glycemic control compared to standard medical care.
The trial, involving 500 patients (81% White, 55% women, mean age 55 years) with type 2 diabetes and glycated hemoglobin (A1c) levels of ≥ 8%, aimed to assess the influence of the food-as-medicine program on glycemic control and healthcare utilization. Participants were selected based on their residence within the service area of the participating clinics and their reported food insecurity.
In the food-as-medicine program, participants received healthy groceries to cover 10 meals per week for the entire household. Additionally, they benefited from dietitian consultations, nurse evaluations, health coaching, and diabetes education. The randomized nature of the trial assigned patients to either participate in the program immediately (treatment group) or after a 6-month delay (control group receiving usual care).
The primary outcome of the study was the participants’ A1c levels at the 6-month mark, with secondary outcomes including healthcare utilization, self-reported diet, and healthy behaviors at both 6 and 12 months.
Results revealed that after 6 months, both the treatment and control groups experienced a substantial decline in A1c levels (1.5 and 1.3 percentage points, respectively). However, the adjusted mean difference of −0.10 (P = .57) between the groups was nonsignificant.
While patients in the treatment group exhibited higher engagement with preventive healthcare at 6 months, making more visits to the program clinic (13.00 vs 0.72) and dietitians (2.7 vs 0.6), the study found no detectable impact on total claims. Although there was a significant increase in outpatient visits in the treatment group compared to the control group at 6 months (P = .007), the study did not observe a notable effect on total claims. The insignificant reduction in inpatient or emergency department claims was offset by an insignificant increase in outpatient claims.
Acknowledging the limitations of the study, which was conducted during the COVID-19 pandemic and within a single large-scale healthcare system in Pennsylvania, the authors emphasized the need for additional research to design more effective food-as-medicine programs aimed at improving health. The study, led by first author Joseph Doyle, PhD, from MIT Sloan School of Management, Cambridge, was published online on December 26, 2023, in JAMA Internal Medicine.
It’s important to note that this study was funded by the Robert Wood Johnson Foundation, the Abdul Latif Jameel Poverty Action Lab, and the Massachusetts Institute of Technology Sloan Health Systems Initiative. Additionally, coauthor John Cawley disclosed receiving personal fees from Novo Nordisk, Inc., outside the submitted work.