Proactive case-finding and risk-stratification in people at risk of chronic liver disease in Greater Manchester: a cost-effectiveness analysis

Introduction: We urgently need innovative strategies to combat a growing epidemic of chronic liver disease (CLD). Integrated Diagnostics for Early Detection of Liver Disease (ID-LIVER) was a collaborative project aiming to improve detection of reversible-stage CLD in a region with high prevalence of critical risk factors. This study assesses the cost-effectiveness of different ways to identify people with significant CLD (defined as METAVIR stage F2 or higher, using liver stiffness of ≥8 kPa on transient elastography as a proxy measure). Strategies of interest include proactive case-finding in the community (supplementing a reactive pathway where hepatology referrals are passively received from primary care) and/or risk-stratification (using Fibrosis-4 (FIB-4) or ID-LIVER-Machine Learning (ML)-a novel machine-learning risk-stratification tool). Methods: We developed a state-transition decision-analytic model estimating lifetime healthcare costs (2023/2024 GBP) and quality-adjusted life-years (QALYs) associated with six alternative strategies for case-finding and risk-stratification. We simulated cohorts of people with alcohol-related liver disease and metabolic dysfunction-associated steatotic liver disease. We populated the model with data collected in ID-LIVER, supplemented by parameters from the literature and routine data sources. We estimated incremental cost-effectiveness and performed deterministic and probabilistic sensitivity analyses. Results: Any case-identification strategy costing ≤£3300 per person with significant CLD identified would meet English cost-effectiveness thresholds (£20 000/QALY). In our decision set, the cheapest strategy is to use FIB-4 in the reactive-only population; however, this misses 43.6% of people with significant CLD. ID-LIVER-ML (using a cut-off of 0.4) generates more population health at a reasonable cost (£10 498/QALY gained). Introducing proactive case-finding generates further health benefits, costing £12 952/QALY gained. Using ID-LIVER-ML in the proactive-and-reactive population has the highest probability of maximising cost-effectiveness, when valuing QALYs at £20 000. Conclusions: Smart methods of case-finding and risk-stratification identify people with significant CLD in the community and are likely to represent good value for money in England.