非酒精性脂肪肝病发生风险列线图预测模型的建立与应用  

Establishment and application of a column chart prediction model for the risk of non-alcoholic fatty liver disease

作　　者：罗发燕 陈铁霞 罗平平[1] 诸伟红[1] LUO Fayan;CHEN Tiexia;LUO Pingping;ZHU Weihong(Hangzhou First People's Hospital,Hangzhou,Zhejiang 310006,China)

机构地区：[1]杭州市第一人民医院,浙江杭州310006

出　　处：《中国公共卫生管理》2024年第6期911-914,共4页Chinese Journal of Public Health Management

基　　金：浙江省自然科学基金项目(LQ18H070006)。

摘　　要：目的了解非酒精性脂肪肝病(Non-alcoholic Fatty Liver Disease,NAFLD)患病的危险因素,并构建相关风险预测模型。方法采用便利抽样法选取2023年1—12月到杭州市第一人民医院健康体检的人群作为研究对象,将检出的218例NAFLD患者纳入NAFLD组,按照1∶1的比例另外选取同期到该院体检的218名健康人群作为对照纳入非NAFLD组。采用二元logistic回归模型分析NAFLD患病的危险因素,利用R 4.3.2软件建立预测NAFLD发生的列线图预测模型,并绘制校准曲线图。结果男性(OR=3.012)、身体质量指数>24 kg/m^(2)(OR=1.095)、甘油三酯升高(OR=2.601)、空腹血糖升高(OR=1.178)、丙氨酸氨基转移酶升高(OR=1.310)、高敏C反应蛋白升高(OR=2.085)是发生NAFLD的危险因素(P<0.05)。构建NAFLD风险预测模型曲线下面积为0.873,95%CI为(0.825,0.910),特异性为79.6%,敏感性为86.1%。结论预测模型具有良好的区分度和准确度,可科学、直观、简便地识别NAFLD风险人群。Objective To understand the risk factors of non-alcoholic fatty liver disease(NAFLD)and construct a risk prediction model.Methods Convenience sampling was used to select the population who underwent health examinations at Hangzhou First People's Hospital from January to December 2023 as the research subjects.A total of 218 NAFLD patients detected were included in the NAFLD group,and an additional 218 healthy individuals who underwent physical examinations at the hospital during the same period were selected in a 1∶1 ratio as controls and included in the non NAFLD group.Using a binary logistic regression model to analyze the risk factors for NAFLD,a column chart prediction model for predicting NAFLD occurrence was established using R 4.3.2 software,and a calibration curve was plotted.Methods Male(OR=3.012),body mass index>24 kg/m^(2)(OR=1.095),elevated triglycerides(OR=2.601),elevated fasting blood glucose(OR=1.178),elevated alanine aminotransferase(OR=1.310),elevated high-sensitivity C-reactive protein(OR=2.085)were identified as risk factors for NAFLD(P<0.05).The area under the curve for constructing a NAFLD risk prediction model was 0.873,with a 95%confidence interval of(0.825,0.910),specificity of 79.6%,and sensitivity of 86.1%.Conclusion The prediction model has good discrimination and accuracy,and can scientifically,intuitively,and simply identify NAFLD risk populations.

关 键 词：非酒精性脂肪肝 危险因素 列线图 预测模型 

分 类 号：R575.5[医药卫生—消化系统]