Protocol for surveillance of the fraction of cirrhosis and hepatocellular carcinoma attributable to viral hepatitis in clinical centres of excellence
The importance of mortality data
WHO proposes an approach to viral hepatitis surveillance that examines: (i) acute hepatitis, reflecting the incidence of new infections; (ii) prevalence of chronic hepatitis B virus [HBV] and hepatitis C virus [HCV] infections; and (iii) mortality from sequelae, including cirrhosis and hepatocellular carcinoma (1). Reduction in mortality due to HBV and HCV infection is one of the two criteria that the Global Health Sector Strategy (GHSS) uses to define the elimination of hepatitis as a public health threat by 2030 (2). Hence, countries require methods to measure mortality from HBV and HCV infection.
Technical approaches to estimating HBV- and HCV-associated mortality
In the early 2000s, WHO estimates of mortality from viral hepatitis considered only deaths from acute infections, as they were reported directly with their own International Classification of Disease (ICD)-10 codes. This approach missed out on the majority of mortality that was secondary to cirrhosis and hepatocellular carcinoma, which were reported with their own ICD codes. Since the ICD codes for cirrhosis and hepatocellular carcinoma did not make a reference to hepatitis, the link to hepatitis was broken. Since 2010, various iterations of the Global Burden of Disease project (GBD) have developed estimates of the worldwide mortality secondary to chronic HBV and HCV infections, which accounted for deaths from the sequelae of chronic hepatitis according to their individual viral causes (3). This key step was made possible by estimating the proportion of cirrhosis and hepatocellular carcinoma attributable to HBV and HCV infections at national, regional and global levels on the basis of published studies (4). The GBD and WHO estimates now use this approach to estimate mortality.
Need for national mortality estimates
The “mortality envelope” from cirrhosis and hepatocellular carcinoma is attributed to HBV and HCV on the basis of evidence from published studies that reported the proportion of patients with these sequelae who had HBV and HCV infection (which is referred to as “the attributable fraction”). At the national level, however, most countries lack a systematic process to generate national estimates of mortality from viral hepatitis. Thus, countries need to institutionalize methods that have been used in an ad-hoc manner in the past for published studies and turn these into a routine surveillance system. To generate national estimates of hepatitis-related mortality, it is necessary (i) to estimate mortality from chronic liver disease (including cirrhosis and hepatocellular carcinoma), and (ii) to estimate the fraction of these conditions that are attributable to various hepatitis viruses. For the first step, national mortality data are usually available. Alternatively, in countries where the quality of data from vital registration systems is not optimal, estimates regarding mortality from cirrhosis and hepatocellular carcinoma are available from WHO or the GBD. However, for the second step, most countries lack a system to estimate which proportion of the sequelae is attributable to the various hepatitis viruses and which due to other causes (e.g. alcohol, metabolic syndrome), and consolidate data from various sources in order to estimate mortality.
Estimating attributable fraction in centres providing care for cirrhosis and/or hepatocellular carcinoma
The aim of this protocol is to describe simple methods to be used in sentinel centres (e.g. hepatology or gastroenterology units) to estimate the proportion of patients with cirrhosis and hepatocellular carcinoma who have HBV and HCV infection. In countries where hepatitis D virus (HDV) infection is of public health importance, the prevalence of HBV/HDV coinfection in HBV-infected patients with cirrhosis and hepatocellular carcinoma should also be estimated. This proportion of patients with HBV, HCV or HBV/HDV infection and cirrhosis or hepatocellular carcinoma may be used to estimate the fractions of these sequelae attributable to HBV and HCV infection, which may be used to estimate national hepatitis mortality rate. Given the impact of recent advances in antiviral treatments for hepatitis B (5) and hepatitis C (6) on the prognosis of chronic liver disease, we need to estimate this proportion for:
- all patients with cirrhosis or hepatocellular carcinoma;
- patients with late-stage liver disease meeting the European Association for the Study of Liver disease [EASL] criteria) (7);
- patients dying from or who received transplants for cirrhosis or hepatocellular carcinoma.
While the main objective of the selection of patients at sentinel sites is to generate data on the attributable fraction that could be used along with national mortality data, the absolute numbers of cirrhosis and hepatocellular carcinoma cases reported in these centres may also be followed up over time to examine trends.