| ObjectsAccording to the breast cancer mortality rates of Chinese urban and rural female aged 20-84 years old during 1988-2012, we studied the variation trend of breast cancer mortality, the effect of age, period and birth cohort on breast cancer mortality among urban and rural female, and explored the risk factors which affect breast cancer mortality further. It is hope that this thesis could provide some scientific reference for future breast cancer prevention and control work, the establishment of relevant public health interventions and health policy.MethodsThe breast cancer motality rates were collected from "China Health Statistics Yearbook" and the World Health Organization Cancer Mortality Database (WHO Cancer Mortality Database). The breast cancer standardar mortality rates were calculated by direct standardized method, and the Joinpoint Regression model was used to analyze secular trends in breast cancer mortality among urban and rural women during 1998-2012, and calculate Annual Percent Change and Average Annual Percent Change. In order to study the influence of age, period and cohort on breast cancer mortality trend, age effect, period effect and cohort effect of urban and rural female breast cancer mortality were estimated by Age-Period-Cohort Model and Intrinsic Estimator operator. Fitting variance, AIC and BIC were introduced to compare the goodness of fit of Intrinsic Estimator and conventional two factors model. The variation velocity of age effect, period effect and cohort effect were computed digital differential method. Finally, we reviewed the methods of parameters estimation and analyzed their advantages and disadvantages.Results1. There was no significant secular trend in breast cancer standard mortality of urban female aged 20-84 years old during 1988-2012. For rural female, the Annual Percent Change of rural female breast cancer mortality was 3.67% (P< 0.001) during 1988-2001, and there was no significant secular trend for the rest of years.2. The Average Annual Percent Change of breast cancer age-specific mortality for urban female 25-29 years old, 30-34 years old, and 70-74 years old during 1988-2012 was 1.5%, 2.1% and 0.7% (P< 0.001). However, the mortality of subjects 55-59 years old increased at a 1% (P< 0.001) Average Annual Percent Change. For rural female, the Average Annual Percent Change of mortality for subjects 40-44 years old, 50-54 years old, 55-59 years old, 60-64 years old and 64-65 years old was 1.7%, 2.9%, 3.0%, 2.1%, 3.0%, and P<0.001.3. The age effect on Chinese urban and rural female breast cancer mortality increased and showed a reversed "J" shape with age in general, the mortality risk in 55-59 years old was higher than other age groups for urban female except 80-84 yeas old, and the highest for rural female. Period effect for urban and rural female increased with time gradually. The mortality risk increased by 59.3% and 106% from 1988-2012 respectively. After controlling for age and period effect, the mortality risk of urban and rural female decreased with birth cohort. Compared to the oldest birth cohort, the mortality risk decreased 7.6 times and 8.5 times respectively.4. According to the fitting variance, AIC and BIC, Intrinsic Estimator performd better goodness of fit than conventional two factor models..5. The mortality risk of breast cancer for urban and rural female experienced three "deteriorating period" and a "better period" together. Conclusions1. Breast cancer mortality of Chinese urban female is higher than rural female’s during 1988-2012.ban and rural female.2. The breast cancer age-specific mortality of urban and rural female aged 55-59 years old increased with time. The average annual percent change of rural female is bigger than urban female.3. Intrinsic Estimator overcame the imperfection of traditional estimation methids and solved the "Un-identification Problem" well.4. The variation velocity of mortality risk with different birth cohort for urban and rural female illustrated that the, war, disaster and the unrest of medical and health services industry and social would brought threats for life and health. |
PDF Full Text Request