POSC55 Model-Based Health Economic Evaluations of Screening Using Chest Computed Tomography: A Systematic Review

Abstract

Low-dose computed tomography (LDCT) is a promising technology for population-based screening programs because it is non-invasive, relatively inexpensive and has low radiation. Chest LDCT could be used to screen for different diseases of the lungs and heart. This review aimed to provide a detailed overview of the methodological decisions including organizational aspects (e.g. target screening population) reported in model-based health economic evaluations of screening programs using chest LDCT. Scopus and PubMed were searched for publications up to 25 June 2021. Eligible publications comprised prospective studies evaluating the long-term costs and health effects of screening with chest LDCT in a high-risk population, were written in English and published in the last ten years. Reporting quality was assessed using the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist. The search yielded 795 unique papers, of which 33 were included. Most papers focused on lung cancer screening (n=31, 94%), followed by coronary calcium scoring (n=2, 6%). There were no evaluations of screening for multiple diseases. Cohort models (e.g. decision trees) were the most common modelling technique (n=21, 64%), followed by microsimulation (n=12, 36%). Studies were in the healthcare context of the United States (n=10), Canada (n=3), the United Kingdom (n=3) and 11 other countries. Only 10 papers (24%) included the effect of an imperfect attendance rate. The reported incremental cost-effectiveness of lung cancer screening compared to no-screening was US$34,000/QALY (US$10,000/QALY-US$90,000/QALY) and US$21,700/QALY (US$10,500/QALY-US$32,900/QALY) for coronary calcium scoring compared to current practice. The incremental cost-effectiveness reported in model-based health economic evaluations of screening using chest LDCT differ substantially between studies as reported for lung cancer and calcium scoring. This variation could be caused by the healthcare contexts in different countries, modelling methods, inputs and assumptions. A gap still exists in literature to evaluate the combination of screening for multiple diseases.