Patient Preferences in Hepatocellular Carcinoma Surveillance: A Multi-site Conjoint Analysis.

Abstract

Several emerging modalities are available for hepatocellular carcinoma (HCC) surveillance; however, patient preferences for surveillance tests - particularly across key subgroups such as race/ethnicity, income, and education level - are poorly characterized. We conducted a choice-based conjoint survey among patients with cirrhosis and chronic hepatitis B undergoing HCC surveillance (from 2022-2024) at seven centers in the US. Participants were presented with 15 scenarios, where they chose between surveillance modalities based on test attributes. We included 649 patients, with a median age of 60 years, 51% male, 60% White, 10% Black, and 20% Hispanic. The cohort reflected substantial socioeconomic variation including 12% with Medicaid and 38% with Medicare. The highest priority attribute was surveillance benefits (42.9%; 95% CI: 40.6-45.1%), followed by financial harms (19.4%; 95% CI: 18.1-20.7%). Patients placed less priority on test logistics (12.3%; 95% CI: 11.5-13.2%), test location (9.2%; 95% CI: 8.5-10.0%), test duration (8.3%; 95% CI: 7.8-8.7%), and physical harms (7.9%; 95% CI: 7.2-8.6%). In subgroup analyses, Black participants placed significantly lower importance on surveillance benefits and higher importance on financial harms (p<0.001) and test duration (p=0.001) compared to White participants. Hispanic participants also prioritized financial harms more than non-Hispanic participants (p<0.001), a pattern consistent with cost-related concerns seen across other lower socioeconomic status (SES) subgroups. Lower SES individuals with below a high school education (p=0.01), Medicaid insurance (p=0.001), and income below $25,000 (p<0.001) assigned greater importance to financial harms and less to surveillance benefits. Patients with undergoing HCC surveillance prioritize surveillance test sensitivity; however, results vary by race, ethnicity, and SES. Improving access by reducing financial and logistical barriers through shared-decision-making may enhance surveillance, particularly when modalities present with similar effectiveness.