A retrospective cohort study of esophageal (including gastric cardia) cancer was conducted to examine dietary and other potential risk factors in Linxian, a high-risk area in P.R. China. Study subjects were identified based on participation in a cytology examination conducted in 1974. They were interviewed in 1989 to obtain information on esophageal cancer risk-factors and identify new cases and deaths. A total of 1,162 subjects from the analytic cohort of 12,693 were determined to have developed esophageal cancer over the 15-year follow-up period. Results indicate that increased age, male gender, a positive family history, low education level, surface-water use, and pork consumption were the strongest risk factors for esophageal cancer identified in this cohort, while use of corn as a primary staple and infrequent consumption of fresh vegetables also were possible risk factors. Traditional or suspected risk factors for esophageal cancer in this and other populations--smoking and alcohol use, and pickled vegetable and moldy food consumption--were not risk factors in this study. Some variation in risk was seen based on the subject's cytology result from 1974. We conclude that dietary factors appear to play a role in the etiology of esophageal cancer in this high-risk population, but are less important than other constitutional factors such as age, gender, and family history.