The optimal endoscopic screening interval for gastric cancer (GC) has not yet been established. We compared survival outcomes in patients with GC based on their history of endoscopic screening before diagnosis.

Using the Korean national insurance claims data, we identified all patients newly diagnosed with GC in 2010. Patients were categorized based on the interval between their last screening endoscopy and diagnostic endoscopy as follows: ≤1 year, ≤2 years, ≤3 years, ≤4 years, ≤5 years, >5 years, and never screened. On the basis of the results of these categories, we further classified the screened patients into 2 groups: ≤3 years and >3 years. We compared the 2 groups before and after propensity score matching.

A total of 26,199 patients with GC were included in this study. Compared with the never screened group, the hazard ratios (HRs) for GC-specific survival in the ≤1-year, ≤2-year, ≤3-year, ≤4-year, ≤5-year, and >5-year groups were 0.47, 0.34, 0.35, 0.47, 0.44, and 0.62, respectively. When divided into 2 groups, the ≤3-year group had a significantly lower HR for GC-specific mortality (HR, 0.72; 95% confidence interval [CI], 0.64-0.80) than the >3-year group. In the matched cohort, the adjusted HR was 0.71 (95% CI, 0.62-0.80) in the ≤3-year group compared with the >3-year group.

Patients who underwent screening at intervals ≤3 years had a significantly reduced risk of GC-specific mortality compared with those screened at intervals >3 years. A 3-year screening interval may be an effective strategy for high-risk regions.