Financial toxicity screening in an urban underserved patient population with cancer.

169 Background: Cancer-related financial hardship is associated with treatment delays, non-adherence and adverse survival outcomes. Racial/ethnic minorities and low-income households are disproportionately impacted by financial toxicity. The Comprehensive Score for Financial Toxicity (COST) measure is used to measure financial toxicity in research but has not been validated in low-income, racially diverse cancer patient populations. Methods: We conducted this study at a large urban medical center from January-August 2023. We included cancer patients on systemic therapy for at least 2 months with ECOG performance status 0-2. Participants completed the COST measure, the Functional Assessment of Cancer Therapy: General (FACT-G) and the European Organization for Research and Treatment of Cancer (EORTC) QOL to assess health-related quality of life (HRQOL). Cancer diagnosis/stage, demographic data, and social work/financial counseling interventions were abstracted. Factor analyses were conducted to determine the underlying factor structure of the COST measure responses and bivariate analyses were conducted to identify relationships between COST scores and other data. Results: 100 patients participated in the study; the average age was 65 and both sexes were equally represented. Most patients reported African American (67%) race/ethnicity. 72% of patients held less than a 4-year college degree, 22 % were unemployed and 55% reported a yearly household income of less than 35, 000. Lung (46%) and gastrointestinal (38%) cancers were the most represented. 73% were enrolled in a government sponsored insurance program (Medicare, Medicaid). The average COST score among participants was 22 range 0-44 and mean FACT-G and EORTC scores were 16. 5 range 0-28 and 59. 6 range 0 – 100 respectively. Fifty-eight patients were referred to social work or financial counseling for financial interventions. Factor analysis revealed three predominant factors among the COST responses (Eigenvalues > 1) including Factor 1: items 3, 7, 8, 9, 10, 11; Factor 2: item 4; Factor 3: item 5. COST was weakly associated with FACT-G and EORTC (Pearson range 0. 27 – 0. 37). The most highly correlated value was between the COST and EORTC physical wellbeing subscale (Pearson 0. 38). The COST score was significantly associated with income, with higher incomes reporting less financial toxicity (p = 0. 004). No significant differences in COST scores were noted across gender, race, or level of education. Conclusions: Low levels of financial toxicity were detected using the COST measure in a diverse, low-income cancer patient population despite many patients being referred to social work/financial counseling. Compared to similar studies, we found a weaker correlation between the COST measure and HRQOL. These findings suggest that the COST measure may need to be further adapted for use in underserved populations and patients with government sponsored insurance.