Statistical and machine learning models for chronic hepatitis B disease progression and treatment eligibility: A scoping review protocol

HBV affects 296 million people globally and is a leading cause of cirrhosis and liver cancer. Statistical models are increasingly used to predict disease progression and guide treatment, but no comprehensive synthesis of these modelling approaches currently exists. It is unclear which methods are used, how they differ in assumptions and data requirements, how applicable they are across populations and settings, and how well they have been validated. This is especially important given major regional differences in HBV burden and healthcare infrastructure, particularly between Asia and Africa. This scoping review will systematically map the literature on statistical and machine learning models used to study chronic hepatitis B (HBV) disease progression and treatment eligibility. It will catalogue modelling approaches, summarize outcomes and data types, describe study characteristics and geographic distribution, and identify methodological gaps. This review will produce: (1) a comprehensive map of statistical models used in HBV progression research; (2) a descriptive summary of study characteristics, including geographic distribution (with particular attention to Africa vs. Asia); (3) identification of methodological gaps (e.g., limited use of longitudinal methods, sparse external validation, poor calibration reporting); and (4) a manuscript for peer-reviewed publication.