Ⅱ型糖尿病危险因素的多状态cox模型分析  

Analysis Risk Factors of NIDDM Using Multistates Cox's Proportional Hazard Regression Model

作　　者：陈冠民[1,2] 倪宗瓒[1,2] 张菊英[1,2] 王建琼[1,2] 潘晓平[1,2] 

机构地区：[1]湖北医科大学卫生统计教研室 [2]华西医科大学卫生统计教研室

出　　处：《数理医药学杂志》1998年第2期97-99,共3页Journal of Mathematical Medicine

基　　金：国家教委博士点基金

摘　　要：采用多状态ｃｏｘ比例风险回归模型对ＮＩＤＤＭ的危险因素进行了分析，结果表明：年龄、教育程度、平均血压、体力活动、饮食因素对两阶段生存时间都有影响。年龄大、平均血压高是加快糖尿病自然病程的因子；教育程度高、体力活动量大、良好的饮食习惯是阻止糖尿病自然病程的因子。体重指数仅影响第二阶段生存时间，精神心理因素仅影响第一阶段生存时间：体重指数高、精神心理评分高是加快糖尿病自然病程的因子。在多因素多状态ｃｏｘ比例风险回归模型分析中，除年龄、平均血压、教育程度未进入模型外。Multistates cox's proportional hazard regression model was used to analyze the risk factors affecting the course of NIDDM(non-insulin- dependent diabetes mellitus). the results illustrate the age, education level, mean blood pressure, quantity of physical activity, dietary factors affected patients survival time of two stages. Old, mean of blood pressure higher were found to work as increasing the transformation of NIDDM status, Higher education level, quantity of physical activity and better dietary habits worked as delaying the transformation of NIDDM status. i. e., they prolonged transformation of IGT to NIDDM and NIDDM to NIDDM with complication. Meanwhile, body mass index only affect the survival time second stage in the way of shorten the transformation time. The results also showed that the score of psychological factors affect the survival time of first stage with the same way as body mass index. In the multivariate multistates cox's regression model, All significant risk factors in the univariate analysis were entered the model except the age, mean of blood pressure and education level. Meanwhile application of multistates cox's proportional hazard regression model was discussed in this papers.

关 键 词：糖尿病 危险因素 COX模型 流行病学 预后 

分 类 号：R587.107[医药卫生—内分泌]