Search for a command to run...
During recent decades, breast cancer incidence has been increasing in Japan. According to the latest reports from several cancer registries in Japan, the breast has become the leading cancer site in female cancer incidence. To analyze the trend of breast cancer incidence in detail, we summarized female breast cancer incidence in Miyagi Prefecture, Japan during 1959-1997, and evaluated the period and cohort effect on breast cancer incidence using the age-period-cohort model. Age-specific and age-standardized rates have increased over successive calendar periods. Around 1980, an accelerated increase in these incidence rates took place. A full model including age, period and cohort was best fitted to the trend of incidence. In the model, the effects of period and cohort were statistically significant. The nonlinear effect for cohort indicates an increasing trend, beginning with the cohort in 1888-1897, and the nonlinear effect for period showed a clear increase in risk with calendar period. Furthermore, the full model including a linear component showed a steadily upward trend in the cohort effect. Based on our own epidemiologic studies previously conducted in Miyagi Prefecture, and other published reports, the cohort effect is likely to be related to the change in prevalence of women with risk factors such as low parity and insufficient breastfeeding. We believe that the emergence of the cohort effect is an important finding, although the period effect may also persist. The significant cohort effect may give a caution for continuous increase of breast cancer incidence in Japan.