Statewide Survey of Primary Care, Gastroenterology, and Hepatology Providers on Hepatocellular Carcinoma Risk Stratification and Surveillance Practices

Introduction: Hepatocellular carcinoma (HCC) surveillance in patients with cirrhosis is associated with increased survival. Provision of HCC surveillance is low in the US, particularly among patients seen only in primary care (PCP) settings. To evaluate current HCC risk stratification and surveillance practices, we conducted a statewide survey of family medicine (FM), internal medicine (IM), gastroenterology (GI) and hepatology providers in North Carolina (NC). Methods: We developed and pilot tested a survey on HCC risk stratification and surveillance. Using the NC Health Professionals Data System, we identified all physicians, PAs and NPs actively practicing in FM, IM, GI or hepatology in NC. We used the Tailored Design Approach to send emails (NPs) or letters (PAs, MD/DOs) inviting providers to complete an online survey with up to 5 points of contact and a drawing for an iPad as an incentive. Tabulations were weighted to represent the larger population using statistical software to adjust for sampling weights (SAS, V9.4, Cary, NC). We performed chi-square tests to compare survey responses between FM/IM and GI/hepatology. Results: 7,654 providers were contacted and 290 responded (response rate 4%). Most respondents were physicians (68%), followed by PAs (24%) and NPs (8%). The most common specialty was FM (50%) followed by IM (31%) and GI/hepatology (11%). Among FM or IM, the top methods for identifying cirrhosis were ultrasound, CT, or MRI (75%), labs (62%), physical exams (35%), and transient elastography (30%). Compared to FM/IM providers, GI/hepatology providers were more likely to report performing any HCC surveillance (98% vs 39%, P< 0.0001), ordering serum AFP (in addition to US) (58% vs 37%, P=0.028) and performing q6 mo (vs annual) surveillance (98% vs 35%, P< 0.0001). There were several reported barriers to HCC surveillance (Figure 1A). Most respondents believed there was strong data to support HCC surveillance but did not know which liver disease patients needed screening and which patients had cirrhosis. Most expressed interest in potential solutions to improve HCC risk stratification/surveillance (Figure 1B). Conclusion: In this statewide survey, most respondents reported a lack of knowledge about who needed HCC surveillance and expressed interest in strategies to improve HCC surveillance including an HCC risk dashboard and electronic risk calculators. After further validation and testing, such tools may help improve low HCC surveillance rates in patients with cirrhosis.Figure 1.: Responses to survey questions regarding potential barriers to HCC surveillance (1A) and interest in strategies to improve HCC surveillance (1B).