Trends in all cause and viral liver disease-related hospitalizations in people with hepatitis B or C: a population-based linkage study
National Centre in HIV Epidemiology and Clinical Research, The University of New South Wales, Sydney, Australia
BMC Public Health 2011, 11:52 doi:10.1186/1471-2458-11-52Published: 24 January 2011
Previous studies have reported an excess burden of cancer and mortality in populations with chronic hepatitis B (HBV) or C (HCV), but there are limited data comparing hospitalization rates. In this study, we compared hospitalization rates for all causes and viral liver disease in people notified with HBV or HCV in New South Wales (NSW), Australia.
HBV and HCV notifications were linked to their hospital (July 2000-June 2006), HIV and death records. Standardized hospitalization ratios (SHRs) were calculated using rates for the NSW population. Random effects Poisson regression was used to examine temporal trends.
The SHR for all causes and non alcoholic liver disease was two-fold higher in the HCV cohort compared with the HBV cohort (SHRs 1.4 (95%CI: 1.4-1.4) v 0.6 (95%CI: 0.6-0.6) and 14.0 (95%CI: 12.7-15.4) v 5.4 (95%CI: 4.5-6.4), respectively), whilst the opposite was seen for primary liver cancer (SHRs 16.2 (95%CI: 13.8-19.1) v 29.1 (95%CI: 24.7-34.2)). HIV co-infection doubled the SHR except for primary liver cancer in the HCV/HIV cohort. In HBV and HCV mono-infected cohorts, all cause hospitalization rates declined and primary liver cancer rates increased, whilst rates for non alcoholic liver disease increased by 9% in the HCV cohort but decreased by 14% in the HBV cohort (P < 0.001).
Hospital-related morbidity overall and for non alcoholic liver disease was considerably higher for HCV than HBV. Improved treatment of advanced HBV-related liver disease may explain why HBV liver-related morbidity declined. In contrast, HCV liver-related morbidity increased and improved treatments, especially for advanced liver disease, and higher levels of treatment uptake are required to reverse this trend.