Obesity Protects Against Heart Failure Readmissions for Patients Living in Food Deserts

Objective: This study examined whether body mass index (BMI) and access to fresh food affects heart failure outcomes. Methods: We identified patients admitted to a university hospital between 01/01/13 and 12/31/13 with a primary diagnosis of heart failure and extracted relevant data through a chart review. Using patient zip codes in the USDA Food Access Research Atlas we identified patients who lived in areas with low access to fresh food or food deserts. Low-access areas are defined as regions where ≥ 33% of the population live > 1 mile (in urban areas) or > 10 miles (in rural areas) from a supermarket. Food deserts have the additional stipulation of areas where median income is ≤ 80% of statewide median. We used the World Health Organization definitions of obesity to classify patients as obese (BMI ≥ 30) or non-obese (BMI < 29.9) and also further stratified as underweight (BMI < 18.5), normal (BMI 18.5–24.9), overweight (BMI 25–29.9) obesity class I (BMI 30–34.9), obesity class II (BMI 35–39.9), or obesity class III (BMI ≥ 40). Using regression analysis, we compared 30-day readmission, in-hospital mortality, and length of stay for obese or non-obese heart failure patients based on their access to fresh food. Results: There were 2063 patient encounters in our study (54.5% male, 43.7% African American, 47.2% heart failure with reduced ejection fraction, 39.9% obese) of which 1039 lived in adequate fresh food access areas and 1024 combined lived in low access or food desert areas with similar demographics within each group. Obesity was associated with living in a food desert (RR 1.26, P = .036). Living in a food desert was associated with a higher 30-day readmission rate for non-obese patients compared to obese patients (RR 1.96 vs. 0.51, P = .032). When stratified, patients of normal BMI patients had higher rates of 30-day readmissions in low access areas (RR 1.50, P = .032) and food deserts (RR 2.30, P = .005). There was a trend towards fewer readmissions with increasing obesity class but this was not statistically significant. No association for underweight or overweight patients and readmissions was present. There was no difference in in-patient mortality or length of stay between obese and non-obese patients regardless of food access. Conclusion: Living in a food desert is associated with higher rates of obesity for this heart failure sample. However, there appears to be an obesity paradox for heart failure in food deserts as obese patients have fewer 30-day readmissions than non-obese patients. Specifically, heart failure patients with normal BMI have the strongest association with 30-day readmissions. Further investigation is needed in a larger cohort with a broader geography.