Evaluating predictors of food insecurity among patients with cancer in the safety-net setting.

Abstract

e18550 Background: Patients with cancer, especially minority and low-income individuals, are at increased risk of financial toxicity and food insecurity. The COVID-19 pandemic has added dire economic challenges to vulnerable populations leading to a global increase in food insecurity. We sought to evaluate the severity and predictors of food insecurity among low-income patients with cancer. Methods: We conducted a cross-sectional multi-lingual survey (i.e., English, Spanish, and Chinese) on a convenience sample of patients with cancer who receive oncologic care at a safety-net hospital. Food insecurity and financial toxicity were measured using validated scales [i.e., US Department of Agriculture Household Food Security Survey and the Comprehensive Score for Financial Toxicity (COST)]. The primary outcome was food insecurity during COVID-19 and predictors of interest included sex, financial toxicity, insurance change, and degree of acculturation. Data was summarized using descriptive statistics and we explored associations between food insecurity and predictors of interest using bivariate regression. Results: A total of 140 patients participated in the study, of whom 56% were male and 47% were 50-64 years old. The most common cancer diagnoses were breast cancer (33%), followed by prostate (13%) and lung cancer (12%). The diverse study participants self-reported being 42% Hispanic/Latinx, 33% Asian, 18% Black or African American, and 14% White. Over half (52%) reported an annual household income ≤$24,999 and 50% experienced a decrease in income during the COVID-19 pandemic. The median COST score was 24 (IQR: 19-31) with 41% experiencing financial toxicity (COST < 26). Most participants experienced food insecurity, including 42% with low food security and an additional 19% with very low food security. In bivariate analysis, increasing financial toxicity (i.e., lower COST score) was associated with a 21% increased risk of very low or low food security (95% CI: 1.11-1.32) and 13% increased risk of low food security (95% CI: 1.05-1.21). Male sex was associated with 312% increased risk of very low food insecurity when compared to female sex (95% CI: 1.02-9.55). Acculturation and changes in insurance coverage were not associated with increased risk of food insecurity. Conclusions: Food insecurity was highly prevalent in this multi-ethnic cohort of low-income patients with cancer. Interestingly, male sex was significantly associated with increased risk of very low food insecurity. Further analyses should explore this potentially at-risk population, their access to nutrition-related support, and the impact of food insecurity in cancer outcomes.