Abstract Body (Do not enter title and authors here): Background: Patients with heart failure (HF) experience high rates of post-discharge readmission and emergency department (ED) visits. Food insecurity and malnutrition are associated with adverse outcomes, however, evidence supporting food-as-medicine interventions and their optimal delivery methods for improving post-discharge outcomes remains limited.
Hypothesis: We hypothesized that 90 days of food supplementation would reduce composite HF readmissions and ED visits compared to usual care, and that conditional delivery tied to medication pickup and clinic attendance would have greater impact than unconditional delivery.
Methods: We conducted a two-center factorial RCT with recruitment predominantly at an urban county safety-net hospital in Dallas, TX (NCT06115369; AHA#:24RPGFIM1198188). Patients hospitalized for acute HF were randomized within 2 weeks of discharge 1:1:1 to medically tailored meals, weekly fresh produce boxes, or usual care with dietary counseling only. Participants receiving food interventions also received dietitian support and underwent secondary randomization to conditional vs unconditional delivery during the 90-day intervention period. Conditional delivery was tied to medication pharmacy fills and clinic attendance, with supplements paused for not meeting criteria and reinstated once met. The primary endpoint was composite 90-day HF readmission or ED visits. Secondary endpoints included clinic attendance rate, medication adherence, quality of life, and cardiac biomarkers.
Results: We successfully enrolled our target sample size of 150 participants (median age 60 years, 39% female) equally across three randomization groups. The population was diverse with 42% Black participants and 33% Hispanic ethnicity. Participants had median left ventricular ejection fraction (LVEF) of 35% and 29% had HF with EF>50%. Median baseline NT-proBNP was 1,197 pg/mL. Hypertension (93%) and diabetes (55%) were highly prevalent, with substantial burden of food insecurity (53%), nutrition insecurity (55%), medium to low self-reported medication adherence (69%), and limited insurance coverage (41%).
Conclusions: This trial successfully enrolled a diverse post-discharge HF population with substantial baseline food insecurity. Database lock will occur in October 2025, and baseline characteristics and study outcomes will be presented. These results will provide initial evidence for the efficacy of food-as-medicine interventions in post-discharge HF care.