[Time trends in cancer incidence in Osaka]

Changes in environments and life styles in Japan have caused the recent changes in the time trends of cancer incidence for various sites. Using the data from the Osaka Cancer Registry, time trends during 1966-80 were analysed for cancer incidence of the leading 5 sites: stomach, lung, liver, uterus and breast. Age-adjusted incidence rates of cancers of the stomach (both sexes) and uterus (invasive cancer) decreased to 75% and 64% respectively between the two periods of 1966-68 and 1978-80 while cancers of the lung, liver, and breast reached 163%, 140%, and 143 % during the same period. Decrease of stomach cancer incidence was observed in all age-groups under 79, however, not in the 25-44 age-groups among females. Analyzing the histological data in the registry, it was noticed that estimated incidence of the intestinal type of stomach carcinoma had decreased more rapidly than the diffuse type. In the 30-49 age-groups among females, no decrease was observed of the diffuse type of carcinoma. Concerning lung cancer, a marked increase was observed over 60 years of age. The age-specific incidence curves by birth cohort showed no or very small cohort effects for the population born 1920-29. Among males, percentages of adenocarcinoma and undifferentiated carcinoma have increased and that of epidermoid carcinoma decreased. The change was more marked in the age-groups younger than 59. Liver cancer showed the 3rd highest incidence rate among males and 6th among females. A rising trend in recent years was noticeable over 45 years of age among males. For the invasive uterine carcinoma, the incidence rate has been decreasing in all ages. Comparing these figures with those of whites in Connecticut or of Japanese in Hawaii, the former was higher than the latter and the difference was larger in age-groups over 40. The recent age incidence curve of breast cancer in Osaka came to be close to that in Iceland in 1930-49 when the curve had kept a constant level for age-groups after menopause. Birth cohort effect was observed for in these age classes.