上海市糖尿病死亡趋势APC模型分析

Analysis of Temporal Trends in Shanghai Diabetes Mortality:an Application of APC Model

[目的]通过定量描述糖尿病死亡率的年龄、时期和出生队列特征，预测上海市糖尿病死亡趋势，为预防控制糖尿病提供依据。[方法]死亡数据来自上海市疾病预防控制中心死亡登记系统。部分年份的年龄别人口数采用内插法推算。1992年至2003年间登记的疾病编码标准前后有2种（ICD-9和ICD-10），选择占全人群糖尿病提及死亡98％的30～89岁的糖尿病提及死亡个案作为研究对象。采用年龄-时期-队列模型（Age-Period-Cohort Model，APC模型）的分析方法，根据模型拟合结果进行预测。[结果]模型拟合结果以同时加入年龄、时期、队列效应的模型为最佳。分性别糖尿病死亡率的年龄、时期和队列效应均表现出序列趋势。应用模型拟合预测2013～2015年上海市分性别的30～89岁年龄别糖尿病死亡率，各年龄组死亡率相对于2001～2003年均大幅上升。[结论]糖尿病将持续成为上海市居民的主要死因之一，开展以代谢综合症（MS）为先导的社区综合防治应成为糖尿病预防控制的优先策略。

[ Objective ] To predict diabetes mortality trend in Shanghai through the study on the features of diabetes mortality including age, period and cohort. [ Methods ] Residents＇ death data from Shanghai death registry system was employed. ICD-9 and ICD-10 were the standard for disease classification. Age 30-89 years death records of diabetes during 1992-2003 were selected to construct model according to the method of age-period-cohort（ APC ） model. [ Results ] The gender-specific model of age＋period＋cohort was best fitted. In gender specific models, effects of age, period and cohort on diabetes mortality present all trend sequence. Using the best-fit model, gender-and age-specific diabetes mortality rates for age 30-89 years in 2013-2015 were predicted. Compared with those in 2001-2003, gender-and age-specific diabetes mortality rates in 2013-2015 would rise sharply in Shanghai. [ Conclusion ] Diabetes will be still one of the primary death causes for Shanghai residents. It suggests that prevention and control of metabolic syndrome （ MS ） should be the priority strategy for diabetes control.