中美指南推荐的糖尿病风险评估量表在中国人群中筛查糖尿病的效能评估

Evaluation of recommend risk scores by Chinese Diabetes Society and American Diabetes Association for screening diabetes in Chinese population

目的评估2013年中国糖尿病防治指南推荐的风险评分表（CDSDRS）与2017年美国糖尿病协会糖尿病指南推荐的风险评分表（ADADRS）在中国人群中筛查糖尿病的效能。方法选取2016年3月～2017年1月于我科门诊行75g口服葡萄糖耐量试验（OGTT）筛查糖尿病的人群454例作为研究对象进行问卷调查，分别构建两种风险评分量表的受试者工作特征（ROC）曲线，比较不同评分量表筛查糖尿病及糖尿病前期人群的敏感度、特异度、约登指数及ROC曲线下面积（AUC）。结果CDSDRS和ADADRS筛查糖尿病的效能无明显差异（AUC分别为0．646和0．612，P=0．059）；筛查糖尿病前期及糖尿病时CDSDRS的效能更佳（AUC分别为0．626和0．583，P=0．005）。ADADRS在中国人群中筛查糖尿病的最佳切点为≥3分，此时的敏感度为87．4％，特异度为29．0％。在女性人群中，相比于ADADRS，CDSDRS最佳切点筛查糖尿病的特异度更高（50．9％比39．2％），但敏感度稍低（84．8％比87．9％）。结论相比于ADADRS，CDSDRS在筛查糖尿病的效能上无明显差异，但筛查糖尿病前期及糖尿病人群时更有优势，ADADRS在中国人群中筛查糖尿病的最佳切点与原指南推荐切点有所差异，但仍需大样本人群进一步求证。

Objective To compare the effects of Chinese Diabetes Society Diabetes Risk Scores （CDSDRS） and American Diabetes Association Diabetes Risk Scores（ADADRS） for screening diabetes in Chinese population. Methods A total of 454 participants who conducted 75 g oral glucose tolerance test（OGTF） in our department from March 2016 to January 2017 were included in this study. Two questionnaires CDSDRS and ADADRS were performed to evaluate the risk of diabetes in all subjects. Sensitivity, specificity, Youden index and the area under the receiver operating characteristic （ROC） curve （AUC） for both questionnaires were calculated and compared. Results The AUCs for CDSDRS and ADADRS were similar when predicting diabetes in our population （AUC：0. 646 vs. 0. 612, P = 0. 059 ）, but CDSDRS had lager AUC when predicting pre-diabetes as compared to ADADRS （ AUC：0. 626 vs. 0. 583 ,P = 0.005 ）. At the cut-off value of 3 by ADADRS,the sensitivity was 87.4% and specificity was 29.0% for predicting diabetes in Chinese population. When compared with ADADRS, CDSDRS had better specificity （ 5 0.9 % vs. 3 9.2 % ） , but lower sensitivity （ 8 4.8 % vs. 8 7.9 % ） in females. Conclusion There is no difference of ADADRS and CDSDRS for predicting diabetes in our population, but CDSDRS predicts pre-diabetes better. The cut-off for ADADRS when predicting diabetes in Chinese population may be differ from ADA criteria, which will required to be further checked by a large sample.