Factors Associated with Primary Liver Cancer Survival in a Southern Italian Setting in a Changing Epidemiological Scenario

Abstract

A retrospective observational study utilising cancer incidence data from a population-based registry investigated determinants affecting primary liver cancer survival in a southern Italian region with high hepatitis viral infection rates and obesity prevalence. Among 2687 patients diagnosed between 2006 and 2019 (65.3% male), a flexible hazard-based regression model revealed factors influencing 5-year survival rates. High deprivation levels [HR = 1.41 (95%CI = 1.15-1.76); p < 0.001], poor access to care [HR = 1.99 (95%IC = 1.70-2.35); p < 0.0001], age between 65 and 75 [HR = 1.48 (95%IC = 1.09-2.01); p < 0.05] or >75 [HR = 2.21 (95%CI = 1.62-3.01); p < 0.0001] and residing in non-urban areas [HR = 1.35 (95%CI = 1.08-1.69); p < 0.01] were associated with poorer survival estimates. While deprivation appeared to be a risk factor for primary liver cancer patients residing within the urban area, the geographic distance from specialised treatment centres emerged as a potential determinant of lower survival estimates for residents in the non-urban areas. After balancing the groups of easy and poor access to care using a propensity score approach, poor access to care and a lower socioeconomic status resulted in potentially having a negative impact on primary liver cancer survival, particularly among urban residents. We emphasise the need to interoperate cancer registries with other data sources and to deploy innovative digital solutions to improve cancer prevention.

Keywords: access to care; cancer registries; digital prevention; epidemiology; health services; liver cancer; survival.

Figure 1

A comparison of 5-year net survival curves in HCC patients diagnosed in the periods 2006–2014 (blue line) and 2015–2018 (red line); the latter intercepts the COVID-19 pandemic. Palermo Province Cancer Registry.

Figure 2

A comparison of absolute standardised mean difference values of the covariates (DC.1, DC.2, the place of residence, gender and age group) in the two balanced groups of patients (“poor” versus “easy access to care”), before (white dots) and after (black dots) applying the nearest neighbour matching of the propensity scores. HCC cases from the Palermo Province Cancer Registry, 2006–2019.