BACKGROUND & AIMS: Single-center studies have estimated that 4.6% to 25.8% of gastric cancers are missed at endoscopy. We performed a population-based study to make a more precise estimate of factors associated with missed lesions in England. METHODS: We performed a retrospective population-based observational cohort study of 2727 patients diagnosed with gastric cancer from April 2011 through March 2012 in England, using linked records from 3 national data sets. The primary outcome was the proportion of patients who had undergone endoscopy in the 3 to 36 months before a diagnosis of gastric cancer. We determined this proportion for the entire cohort and for subgroups. RESULTS: Of the 2727 patients in the cohort, 8.3% (95% confidence interval, 7.2%-9.3%) underwent endoscopic evaluation in the 3 to 36 months before their diagnosis of gastric cancer. An endoscopy within 3 to 36 months of diagnosis was associated with a diagnosis of early stage cancer (stages 0 or 1, 11.5%; stage 2, 7.9%; stages 3 or 4, 6.9%; P = .01 for stage 0 or 1 vs stage 2 or greater), younger age at diagnosis (<55 y, 13.3% vs ≥55 y, 7.8%; P = .03), and female sex (10% of women vs 7.3% of men; P = .01). Gastric ulcers were detected in 15% of endoscopies performed at any time in the 3 years before cancer diagnosis, and in 64% of endoscopies performed 3 to 6 months before a diagnosis of gastric cancer. CONCLUSIONS: Based on a retrospective analysis of medical records in England, in 8.3% of patients with gastric cancer, their cancer was missed at endoscopy within the 3 previous years. A previous endoscopy detected benign gastric ulcers more frequently than any other lesion in patients who later were diagnosed with gastric cancer.