RÉSUMÉ
AIM: To identify the optimal strategy for gastric cancer (GC) prevention by evaluating the cost-effectiveness of esophagogastroduodenoscopy (EGD)-based preventive strategies. METHODS: We conducted a model-based cost-effectiveness analysis. Adopting a healthcare payer's perspective, Markov models simulated the clinical experience of the target population (Singaporean Chinese 50-69 years old) undergoing endoscopic screening, endoscopic surveillance and usual care of do-nothing. The screening strategy examined the cohort every alternate year whereas the surveillance strategy provided annual EGD only to people with precancerous lesions. For each strategy, discounted lifetime costs ($) and quality adjusted life years (QALY) were estimated and compared to generate incremental cost-effectiveness ratio (ICER). Deterministic and probabilistic sensitivity analysis was conducted to identify influential parameters and quantify the impact of model uncertainties. RESULTS: Annual EGD surveillance with an ICER of $34 200/QALY was deemed cost-effective for GC prevention within the Singapore healthcare system. To inform implementation, the models identified six influential factors and their respective thresholds, namely discount rate (<4.20%), age of starting surveillance (>51.6 years), proportion of program cost in delivering endoscopy (<65%), cost of follow-up EGD (<$484), utility of stage 1 GC patients (>0.72) and odds ratio of GC for high-risk subjects (>3.93). The likelihood that surveillance is the most cost-effective strategy is 69.5% accounting for model uncertainties. CONCLUSION: Endoscopic surveillance of gastric premalignancies can be a cost-effective strategy for GC prevention. Its implementation requires careful assessment on factors influencing the actual cost-effectiveness.