Practice Patterns and Attitudes of Primary Care Providers and Barriers to Surveillance of Hepatocellular Carcinoma in Patients With Cirrhosis

BACKGROUND & AIMS: Fewer than 20% of patients with cirrhosis undergo surveillance for hepatocellular carcinoma (HCC), therefore these tumors often are detected at late stages. Although primary care providers (PCPs) care for 60% of patients with cirrhosis in the United States, little is known about their practice patterns for HCC surveillance. We investigated factors associated with adherence to guidelines for HCC surveillance by PCPs. METHODS: We conducted a web-based survey of all 131 PCPs at a large urban hospital. The survey was derived from validated surveys and pretested among providers; it included questions about provider and practice characteristics, self-reported rates of surveillance, surveillance test and frequency preference, and attitudes and barriers to HCC surveillance. RESULTS: We obtained a clinic-level response rate of 100% and a provider-level response rate of 60%. Only 65% of respondents reported annual surveillance and 15% reported biannual surveillance of patients for HCC. Barriers to HCC surveillance included not being up-to-date with HCC guidelines (68% of PCPs), difficulties in communicating effectively with patients about HCC surveillance (56%), and more important issues to manage in the clinic (52%). Approximately half of PCPs (52%) reported using ultrasound or measurements of a-fetoprotein in surveillance; 96% said that this combination was effective in reducing HCC-related mortality. However, many providers incorrectly believed that clinical examination (45%) or levels of liver enzymes (59%) or a-fetoprotein alone (89%) were effective surveillance tools. CONCLUSIONS: PCPs have misconceptions about tests to detect HCC that contribute to ineffective surveillance. Reported barriers to surveillance include suboptimal knowledge about guidelines, indicating a need for interventions, including provider education, to increase HCC surveillance effectiveness.