Abstract
BACKGROUND: Evidence on historical trends extracted embedded in recent data can advance our understanding of the epidemiology of breast cancer for Chinese women. China is a country with significant political, socioeconomic, and cultural events since the 1900s; however, no such studies are reported in the literature. METHODS: Age-specific mortality rates of breast cancer during 1990-2015 in China were analyzed using APC modeling (age-period-cohort modeling) method. Net effect from birth cohort was derived to measure cancer mortality risk during 1906-1990 when no mortality data were collected, and net effect from time period was derived to measure cancer mortality risk during 1990-2015 when data were collected. Model parameters were estimated using intrinsic estimator, a novel method to handle collinearity. The estimated effects were numerical differentiated to enhance presentations of time/age trend. RESULTS: Breast cancer mortality rate per 100,000 women increased from 6.83 in 1990 to 12.07 in 2015. After controlling for age and period, the risk of breast cancer mortality declined from 0.626 in 1906-10 to - 1.752 in 1991-95 (RR = 0.09). The decline consisted of 3 phases, a gradual phase during 1906-1940, a moderate phase with some fluctuations during 1941-1970, and a rapid phase with large fluctuations during 1971-1995. After controlling for age and cohort, the risk of breast cancer mortality increased from - 0.141 in 1990 to 0.258 in 2015 (RR = 1.49) with an acceleration after 2005. The time trends revealed by both the cohort effect and the period effect were in consistency with the significant political and socioeconomic events in China since the 1900s. CONCLUSIONS: With recent mortality data in 1990-2015, we detected the risk of breast cancer mortality for Chinese women over a long period from 1906 to 2015. The risk declined more than 90% from the highest level in 1906-10 to the lowest in 1990-95, followed by an increase of 49% from 1990 to 2015. Findings of this study connected historical evidence with recent data, supporting further research to exam the relationship between development and risk of breast cancer for medical and health decision-making at the population level and prevention and treatment at the individual level.