Grocery delivery cut A1c in Medicaid patients with diabetes

Clinical takeaway: When A1c remains high despite medication in patients facing food insecurity, food access may be a treatable part of the problem. It's worth checking whether a patient's Medicaid plan or health system offers a medical grocery program.

Food is Medicine programs have spread through Medicaid waivers and health system pilots often faster than the evidence to support them. But new data suggest that this approach could be particularly helpful for patients whose diabetes stays poorly controlled due to economic and social barriers that keep healthy food out of reach.

Delivering groceries improved HbA1c in a group where it's often difficult to get results: adults with persistently poor glycemic control, most of them facing food insecurity. Those randomized to weekly medically tailored groceries saw A1c fall 0.66 points over six months versus 0.25 points with usual care, an added 0.40-point reduction from a mean starting A1c of 9.4. That margin is in the range some second-line glucose-lowering drugs achieve.

A less expensive grocery basket, worth $100 to $170 per month, lowered A1c as much as the costlier one at $135 to $210. The two arms were otherwise identical, with the same weekly deliveries, recipes, and counseling access. Participants receiving groceries also saw the odds of food security roughly double and the odds of nutrition security more than triple. Blood pressure and BMI did not significantly change.

The trial enrolled 460 Medicaid-insured adults with type 2 diabetes and at least two A1c readings of 7.5% or higher in the past year, recruited from 12 Kaiser Permanente Southern California medical offices between November 2021 and July 2022. Participants were randomized to usual care or one of the two grocery arms and followed for six months; 85% reported Hispanic ethnicity, and 83% of grocery recipients ate most or all of the food provided.

What patients did with the program points toward what worked: most ate the food, few used the counseling, but A1c fell anyway. That pattern suggests that the groceries themselves drove the change. It also lowers the bar for scaling, since the cheaper basket matched the pricier one and the costliest add-on, counseling, went mostly unused. Coverage for these programs varies widely from plan to plan, so understanding local programs and resource availability is key.

"This trial shows that food-based support can help patients with low incomes lower their blood sugar levels above and beyond the health care they receive," said lead author Claudia Nau, PhD, of the Kaiser Permanente Southern California Department of Research & Evaluation.

Source: Nau C, et al. (2026 Jul 21) Circulation. Effects of a "Food Is Medicine" Intervention on Glucose Control Among Medicaid-Insured Patients With Type 2 Diabetes: A Randomized Controlled Trial