To assess the influence of the number of surgical procedures performed at a hospital on cancer survival, we used data on the surgical patients with stomach cancer (n=31,545) and lung cancer (n=5,534) diagnosed between 1986-95 in the Osaka Cancer Registry. We calculated the average annual number of surgical operations for each hospital to obtain the surgical volume index. In the logistic regression model, odds ratio for survival for every 30th day after surgery up to the 1,830th day (5-year period) was calculated. The independent variables that were studied were the volume index (continuous variable), patient' sex and age, clinical stage, additional treatment such as chemotherapy and/ or radiation, and hospital category. We selected the complete data set of 18,582 stomach cancer and 2,819 non-small cell lung cancer patients. For both diseases, the surgical volume index was significantly positively associated with the patient survival at every point during 5 years after surgery. For stomach cancer, the volume effect was highest on the 60th day (OR=0.990) after surgery between 1986-95. For lung cancer, the volume effect was highest on the 90th day (OR=0.974) after surgery between 1986-95. Our data indicates that a surgical volume influence on stomach and lung cancer survival over 5 years and appears more prominently between 60th-120th days after surgery.