摘要
目的:分析影响儿童咳嗽变异性哮喘发生的危险因素,构建基于不同中医体质联合肺功能参数的小儿咳嗽变异性哮喘的临床预测模型。方法:回顾性搜集2017年6月至2020年1月期间宝安区中医院咳嗽变异性哮喘患儿80例和体检健康儿童120例,分为哮喘组(80例)和正常组(120例),记录患者一般资料、过敏原、家族哮喘史、肺功能指标及中医体质等因素,对相关因子进行单因素分析,并对有统计学意义的因子进行多因素logistic回归分析,建立风险预测模型。应用模型进行内部验证,同时将模型应用于临床实际,对2020年2月至2021年2月的90例患者进行评估,结果采用H-L拟合优度检验评估实际值和预测值的拟合程度,用受试者工作特征(receiver operating characteristic,ROC)曲线下面积(area under curve,AUC)进行统计分析,评价该临床预测模型的鉴别效度和预测能力。结果:哮喘组和正常组相比,在饲养宠物、家族哮喘史、FEV1、FVC、FEV1/FVC、PEF和中医体质这7项危险因素上具有统计学意义(P<0.05),对这些因素使用logistic回归分析构建基于不同中医体质联合肺功能参数的小儿咳嗽变异性哮喘的临床预测模型。经H-L拟合优度检验,显示实际值与预测值具有较好的拟合(P=0.292)。使用模型对2020年2月至2021年2月的90例患者进行评估显示模型总正确率为82.22%,通过ROC曲线评价该模型的鉴别效度,结果显示:AUC=0.818,敏感度为77.5%,特异度为86.0%。结论:小儿咳嗽变异性哮喘的发生受多种因素影响,以此建立的临床风险预测模型对小儿发生咳嗽变异性哮喘预测具有良好的判别能力,值得临床引用与进一步研究。
Objective:To analyze the risk factors of cough variant asthma(CVA)in children,and to establish the risk prediction model of CVA in children based on different TCM constitution combined with lung function parameters.Methods:A total of 80 children with CVA and 120 healthy children in the First Affiliated Hospital of Guangzhou University of Traditional Chinese Medicine from June 2017 to January 2020 were retrospectively collected.They were divided into a asthma group(80 cases)and a normal group(120 cases).The general information,allergens,family history of asthma,lung function index and TCM Constitution of the patients were recorded,and the related factors were analyzed by univariate analysis.The risk prediction model was established by multivariate logistic regression analysis of statistically significant factors.Then the model was tested by internal validation and 90 patients from February 2020 to February 2021 were used to evaluated.Results:H-L goodness of fit test was used to evaluate the fitting degree of actual value and predicted value.Area under curve(AUC)of receiver operating characteristic(ROC)was used for statistical analysis and evaluating the differential validity and predictive ability of the clinical prediction model.Results:Compared with the normal group,the asthma group had significant differences in pet raising,family history of asthma,FEV1,FVC,FEV1/FVC,PEF and TCM constitution(P<0.05).Logistic regression analysis was used to construct the clinical prediction model of CVA in children based on different TCM constitution combined with lung function parameters.The H-L goodness of fit test showed that the actual value had a good fit with the predicted value(P=0.292).The model was used to evaluate 90 patients from February 2020 to February 2021.The total accuracy of the model was 82.22%.The ROC curve was used to evaluate the differential validity of the model.The results showed that AUC=0.818,sensitivity was 77.5%,and specificity was 86.0%.Conclusion:The occurrence of CVA in children is affected by many factors,and the clinical risk prediction model has good discriminant ability for the prediction of CVA in children,which is worthy of clinical reference and further research.
作者
黄少鹏
姚衍竹
HUANG Shaopeng;YAO Yanzhu(Bao’an Hospital of Traditional Chinese Medicine(Group)Affiliated to Guangzhou University of Traditional Chinese Medicine\the 7^(th)Clinical School of Medical College of Guangzhou University of Traditional Chinese Medicine,Shenzhen Guangdong 518133,China;The First Clinical Medical School of Guangzhou University of Traditonal Chinese Medicine,Guangzhou Guangdong 510499,China)
出处
《四川中医》
2023年第2期86-90,共5页
Journal of Sichuan of Traditional Chinese Medicine
