Detailed simulation models are needed to assess strategies for prevention and treatment of hepatitis B virus (HBV) infection, the world’s leading cause of liver disease. We sought to develop and validate a simulation model of chronic HBV that incorporates virological, serological and clinical outcomes.

We developed a novel Monte Carlo simulation model (the HEPA-B Model) detailing the natural history of chronic HBV. We parameterised the model with epidemiological data from the Western Pacific and sub-Saharan Africa. We simulated the evolution of HBV DNA, ‘e’ antigen (HBeAg) and surface antigen (HBsAg). We projected incidence of HBeAg loss, HBsAg loss, cirrhosis, hepatocellular carcinoma (HCC) and death over 10-year and lifetime horizons. We stratified outcomes by five HBV DNA categories at the time of HBeAg loss, ranging from HBV DNA10⁶ copies/mL. We tested goodness of fit using intraclass coefficients (ICC).

Model-projected incidence of HBeAg loss was 5.18% per year over lifetime (ICC, 0.969 (95% CI: 0.728 to 0.990)). For people in HBeAg-negative phases of infection, model-projected HBsAg loss ranged from 0.78% to 3.34% per year depending on HBV DNA level (ICC, 0.889 (95% CI: 0.542 to 0.959)). Model-projected incidence of cirrhosis was 0.29–2.09% per year (ICC, 0.965 (95% CI: 0.942 to 0.979)) and HCC incidence was 0.06–1.65% per year (ICC, 0.977 (95% CI: 0.962 to 0.986)). Over a lifetime simulation of HBV disease, mortality rates were higher for people with older age, higher HBV DNA level and liver-related complications, consistent with observational studies.

We simulated HBV DNA-stratified clinical outcomes with the novel HEPA-B Model and validated them to observational data. This model can be used to examine strategies of HBV prevention and management.