STUDY OBJECTIVE--The aim was to analyse the relationship between partial gastrectomy and gastric cancer risk. DESIGN--This was a case control study, using a structured questionnaire to obtain problem orientated medical history and sociodemographic data. SETTING--The study was conducted in a network of hospitals in the Greater Milan area between January 1985 and February 1990. SUBJECTS--Subjects were 563 incident cases of histologically confirmed gastric carcinoma (347 males, 216 females, median age 60 years, range 28 to 74) and 1501 controls (885 males, 626 females, median age 58 years, range 23 to 74) in hospital for acute, non-neoplastic, non-digestive-tract disorders. Less than 3% of cases or controls refused to be interviewed. MEASUREMENTS AND MAIN RESULTS--Relative risks (RR) and the corresponding 95% confidence intervals (CI) were determined, adjusted for age and sex plus area of residence, education, and smoking when specified. Within 20 years after gastrectomy, the relative risk of gastric cancer was not significantly raised (RR = 1.2, 95% CI 0.5-2.8), but a positive association emerged after longer time intervals. The RR was 1.6 (95% CI 0.7-4.1) after 20 to 29 years, and 3.5 (95% CI 1.3-10.0) after 30 years or more. These results were consistent in the two sexes and in the subsequent age groups, and not materially influenced by allowance for a number of identified potential confounding factors using multivariate analysis. CONCLUSIONS--The risk of gastric cancer is increased in the long term (20 years or more) after gastrectomy. This is explainable in terms of increased intragastric carcinogen formation following gastrectomy, and/or potential similarities in aetiological correlates of gastric ulcer and carcinoma of the stomach.