Mass screening for gastric cancer originated in Miyagi Prefecture, Japan, in 1960. This review summarizes studies assessing the efficacy of screening using data from a population-based cancer registry in the prefecture that was started in 1959. Sensitivity and specificity of screening photofluorography (indirect radiography) is 81.1% and 88.8%, respectively. Ten-year survival rates are 30-40% better in screen-detected cases than in symptom-diagnosed cases. No randomized controlled trials have reported the efficacy of the screening in the country. In a 18-year follow-up study of 7,008 residents in the prefecture, incidence from gastric cancer did not differ between screened and unscreened subjects, but mortality decreased significantly in screened subjects compared with unscreened subjects for men (61.9 vs. 137.2 per 100,000 person-years, P<0.005) and women (28.1 vs. 53.8 per 100,00 person-years, P<0.01). In a population-based case-control study with 198 cases and 577 controls, odds ratio (95% confidence interval) of gastric cancer mortality was 0.41 (0.28-0.61) for ever vs. never screened within 5 years. Studies conducted in other regions have provided similar findings. Substantial evidence indicates that the Japanese screening program is effective in reducing the mortality from gastric cancer. Population-based cancer registries play a crucial role in assessing the efficacy of cancer screening programs.