Analysis Of Trends In Mortality Rate Of Cervical Cancer In Henan Province With APC Model, 1985～2004

Object: Data from fifteen cancer registry regions in Henan province for the causes of death since 1984 were collected which covered one tenth of total population in Henan province. The classification of diseases followed the international standard from ICD-9 to ICD-10. The records were inputted into computer every year.But the most current analysis of the data is the calculation of the crude rate, the adjusted rate, life expectancy. They can describe the distribution of the diseases adequately. But it is not enough to give the clues of the cause of the diseases. APC models (age-period-cohort models) can make good use of the information of the data; give some advice to the government and more valuable information on the cause of the diseases. The APC models analysis is used in many foreign cancer registries but seldom in domestic cancer registries. So we hope more registries can understand the model analysis.Methods: 1.The cervical cancer death data and the corresponding population were drawn from the database. The quality of the data was evaluated. 2. The death rate of the cervical cancer by sex, age, year and cancer registries was evaluated with the software SAS. 3. The geographic map of the adjusted death rate of the cervix cancer was drawn with the software MAPINFOR. 4. Analysis of the cervix cancer data between 1985 and 2005 in Henan province was performed with the JOINPOINT model. 5. We demonstrated the parameters of the APC model and evaluated the effect of the period and the cohort.Results: 1. The number of person died from the cervical cancer during 1984-2005 in the cancer registries of Henan province was 4017 and the total population is 86 million. The average crude mortality rate was 4.67 per 100000. The age adjusted mortality rate decreased from 6.97 per 100000 in 1984 to 2.10 per 100000 in 2005. The decline trend is clearly. 2. The JOINPOINT model showed a clearly decline pattern with no statistical significance point. The estimated EAPC was -4.23 and the 95% confidence interval was (-5.27,-3.17). 3. The age-period-cohort analysis showed a cohort effect only. The trend of the relative risk is decline before 1940 and after 1950.Conclusions: 1.The age-period-cohort analysis can make full use of the information in the data of the cancer registries. Furthermore it can suggest the clues of the cause of the diseases. 2. The age adjusted mortality rate is declined continuously. And the ranking among all female malignant tumors is declined to the 6th. 3. The decline trend of the cervical cancer mortality rate is mainly caused by the cohort effect which is the improvement of the environmental sanitation and the private healthful habit. 4. The focus of the prevention and control for the cervical cancer should put on education of the public health to make more people know to prevent themselves. And secondly we should focus on the screening of the women aged from 35 to 65 in the more risk area.