A population-based case-control study of stomach cancer was conducted among the teachers and staff of primary and middle schools in Shanghai. A total of 84 cases of stomach cancer were identified in all schools and 2,676 controls were drawn from the teachers and staff of a randomly selected sample of 40 primary and 15 middle schools. Data were analyzed by a multivariate logistic regression model. The analysis indicated that a positive family history of stomach cancer, cigarette smoking, low consumption of fruits, and low consumption of strong tea were significantly associated with stomach cancer incidence.