Patient, clinician, and system-related barriers may affect adherence to hepatocellular carcinoma (HCC) surveillance programmes. The impact of a dedicated automated recall HCC surveillance programme on retention rates in patients eligible for screening is unknown. We aimed to describe and evaluate a large HCC surveillance programme in a publicly funded healthcare system.
Data were collected from January 1, 2013, to December 31, 2022, from a retrospective cohort of subjects enrolled in a publicly funded automated recall semi-annual surveillance programme as per the American Association for the Study of Liver Disease HCC guidance in the Calgary Health Zone (~1.6 million), Canada. Patients were excluded if there was incomplete data or did not meet indications for surveillance. Cox regression was used to identify predictors of non-retention to surveillance.
A total of 7269 patients were included. The median was age 55.5 years (IQR: 45.5–63.8), 60% were male, 46% were of Asian descent, 51% had HBV infection, and 36% had cirrhosis (35% alcohol-related). Median follow-up was 4.9 years (IQR: 1.5–7.2). Overall, 52% (n = 3768) of patients were retained in the surveillance programme, while 8.3% (n = 603) left for potential medical reasons, and 40% (n = 2898) were lost in follow-up. The median time in the programme for those lost in follow-up was 0.81 years (IQR: 0.0–2.8) compared to 6.75 years if retained (IQR: 5.6–8.6; p < 0.001). In multivariable Cox regression analysis, HCV aetiology (HR 1.41; CI 1.23–1.62, p < 0.01), African ethnicity (HR 1.20, CI 1.02–1.42, p = 0.03), and cirrhosis (HR 1.16, CI 1.05–1.28, p < 0.01) increased risk of dropout. On interaction analysis, Hepatitis B amongst cirrhotic patients also increased risk of dropout (HR 1.48, CI 1.05–2.07, p = 0.02).
A dedicated automated recall HCC surveillance programme has a high retention rate in a large multi-ethnic cohort of patients while identifying certain marginalised patient populations, such as those with viral liver disease, cirrhosis, or African ethnicity, as particularly vulnerable to loss to follow-up.