新疆地区健康体检人群糖尿病风险评估方法探讨

摘要

Objective To establish the risk assessment method of diabetes for the health population in Xin-jiang.Methods On the basis of epidemiological investigation of diabetes consulting literature and analysis of diabetes risk factors.According to the latest 2013 China guidelines on prevention of diabetes and combi-ning with the life characteristics in Xinjiang,the main risk factors of diabetes in Xinjiang adults initially i-dentified.Tthe study population was divided into two teams according to the prospective research methods. In the cohort I,The diabetes prevalence rate trend was tested by Chi-square test.Logistic regression analy-sis was performed for predicting the category,parallel inspection line and the main effect analysis.The risk factors of diabetes were screened,and ten times of the regression coefficient value was as the value of the integral for different variables.The statistical methods were used to establish the risk factors for diabetes scoring.The practicability and feasibility of the method were verified by the data of cohort Ⅱ.Results When the scores of male and female were 80 and 70,the Youden′s index was maximal.Therefore,the 80 and 70 were as discriminating different gender diabetes risk threshold,respectively,and the sensitivity of the screening were 90.4% and 83.3% and specificity were 89.3% and 97.9%;the positive predictive value was 57.8% and 80.1%;negative predictive value was 98.4% and 98.3%.The area under the curve was 0.946 and 0.878,respectively.The risk factor scoring method for screening of diabetes,and the difference was statistically significant (P =0.000 ).Conclusion The data of the evaluation method came from physical examination of the population in Xinjiang area,so it is targeted to the Xinjiang population and eas-y to operate which could be used for individuals with diabetes and community groups to measure the degree of the risk or used for health education.This method is expected to become an effective method for screening high-risk diabetes in health examination center in Xinj iang area populations and for further intervention.%目的：建立适合新疆地区健康体检人群糖尿病发病风险的评估办法。方法在糖尿病流行病学调查基础上，查阅文献资料并分析糖尿病发病风险因素，依据最新2013中国糖尿病防治指南，并结合新疆地区不同民族生活特点，初步确定新疆地区成年人糖尿病发病的主要风险因素，采用前瞻性研究的方法将研究人群分为两队列，队列1对糖尿病患病率趋势检验做检验，运用 Logistic回归分析做预测类别、平行线检验及主效应分析。筛选糖尿病患病危险因素，并将回归系数数值的10倍作为不同变量的积分值。运用统计学方法建立糖尿病危险因素评分法。并以队列2的资料验证该方法的实用性、可行性。结果男性与女性积分值分别为80分与70分时，约登指数最大。分别以80分与70分作为判别不同性别发生糖尿病风险的阈值，其筛检的灵敏度分别为90．4％与83．3％，特异度分别为89．3％与97．9％，阳性预测值分别为57．8％与80．1％，阴性预测值为98．4％与98．3％，曲线下面积分别为0．946与0．878。危险因素评分法用于筛查糖尿病，差异有统计学意义（P ＝0．000）。结论本评估方法资料来源于新疆地区体检人群，针对性强，操作简易，可用于糖尿病个体和社区群体的患病危险度测定及健康教育，有望成为新疆地区健康体检中心筛查糖尿病高危人群并做进一步干预的有效方法。