高尿酸血症患病风险预测模型的构建  

Construction of a Predictive Model for the Risk of Hyperuricemia

作　　者：宋小玲 蒋芝月[1,2] 葛雯 汪俊华[1,2] 王子云[1,2] SONG Xiao-ling;JIANG Zhi-yue;GE Wen;WANG Jun-hua;WANG Zi-yun(Prevention Medical Grade 2017 of School of public health and health,Guizhou Medical University,Guiyang,Guizhou,550025,China)

机构地区：[1]贵州医科大学公共卫生与健康学院2017级预防医学专业,贵州贵阳550025 [2]贵州医科大学公共卫生与健康学院环境污染与疾病监控教育部重点实验室,贵州贵阳550025 [3]贵阳市疾病预防控制中心,贵州贵阳550018

出　　处：《中国初级卫生保健》2022年第10期51-54,共4页Chinese Primary Health Care

基　　金：贵州省卫生健康委2021年科学技术基金项目(gzwkj2021-529)。

摘　　要：目的:利用某乡镇国家基本公共卫生服务项目中老年人群的体检数据构建风险预测模型,为高尿酸血症的早期防控提供参考。方法:收集某乡镇2021年≥65岁老年人群的健康体检资料1821份,将资料随机分为85.00%的训练集和15.00%的验证集;在训练集中采用二元非条件logistic回归分析构建模型,以受试者工作特征曲线下面积评估模型预测效果。结果:高尿酸血症检出率为78.53%,对高尿酸血症有预测价值的指标有性别、年龄、血肌酐、空腹血糖、谷丙转氨酶、高密度脂蛋白和低密度脂蛋白;训练集的工作特征曲线下面积为0.688(95%CI:0.657~0.719),验证集的工作特征曲线下面积为0.679(95%CI:0.594~0.764),模型总正确率为83.15%,预测患病准确率为99.56%,预测正常者的准确率为4.26%。结论:利用某乡镇老年人群的健康体检数据构建的高尿酸血症患病风险预测模型,在该老年体检人群中具有一定的预测能力。OBJECTIVE To construct a risk prediction model using the data from the national basic public health service physical examination of the elderly in a township and provide a reference for early prevention and control of hyperuricemia.METHODS 1821physical examination data of the elderly aged 65 and above in a township in 2021 were collected,and randomly divided into 85%training set and 15%verification set.In the training set,the model was established by binary logistic regression,using the area under the curve(AUC)of the receiver operator characteristic curve to evaluate the prediction performance of the model.RESULTS The detection rate of hyperuricemia was 78.53%,and indicators that were predictive of hyperuricemia include gender,age,serum creatinine,fasting blood glucose,glutamic pyruvic transaminase,high density lipoprotein and low density lipoprotein.The AUC of the training set model was 0.688(95%CI:0.657,0.719),and the AUC of the validation set was 0.679(95%CI:0.594,0.764),the overall accuracy of this model was 83.15%,the model predicted the prevalence accuracy rate was 99.56%,and the accuracy of normal subjects was 4.26%.CONCLUSION The risk prediction model based on the physical examination data of the elderly has certain predictive ability.

关 键 词：高尿酸血症 风险预测模型 老年人群 

分 类 号：R181[医药卫生—流行病学]