Testing the Feasibility and Dietary Impact of a 'Produce Prescription' Program for Adults with Undermanaged Type 2 Diabetes and Food Insecurity in Australia.

BACKGROUND: There is growing interest in Food is Medicine programs that incorporate food-based interventions into health care for patients with diet-related conditions. OBJECTIVES: We aimed to test the feasibility of a "produce prescription" program and its impact on diet quality for people with type 2 diabetes (T2D) experiencing food insecurity in Australia. METHODS: We conducted a pre-post intervention study in n = 50 adults experiencing food insecurity with T2D and glycated hemoglobin (HbA1c) ≥8%. Once enrolled, participants received healthy food boxes weekly free of charge, with the contents sufficient to create 2 meals/d, 5 d/wk for the entire household, over 12 wk. Participants were also provided with tailored recipes and behavioral change support. The primary outcome was change in diet quality assessed by 24-h diet recalls. Secondary outcomes included differences in cardiovascular disease risk factors; blood micronutrients; and feasibility indicators. Differences in the baseline and 12-wk mean primary and secondary outcomes were assessed by paired t tests. RESULTS: Participants were older adults with mean ± SD age 63 ± 9 y (range: 40-87 y), HbA1c 9.8% ± 1.5%, and 46% were female. Overall, 92% completed the final study follow-up for the primary outcome. Compared with baseline, diet quality improved at week 12, with an increase in the mean overall diet quality (Alternate Healthy Eating Index score) of 12.9 (95% CI: 8.7, 17.1; P < 0.001), driven by significant improvements in vegetables, fruits, whole grains, red/processed meat, trans fat, sodium, and alcohol consumption. Blood lipids also improved (total:HDL cholesterol: -0.48; 95% CI: -0.72, -0.24; P < 0.001), and there was significant weight loss (-1.74 kg; 95% CI: -2.80, -0.68 kg, P = 0.002), but no changes in other clinical outcomes. Participants reported high levels of satisfaction with the program. CONCLUSIONS: These findings provide strong support for an adequately powered randomized trial to assess effe