儿童矮小症的影响因素及列线图模型的构建

Influencing factors of dwarfism in children and construction of the nomogram model

目的探讨儿童矮小症的影响因素,并构建列线图模型。方法抽取2020年6月—2022年12月1500例儿童为研究对象,实际有效调查1422例。根据矮小症发生情况分为正常组1351例和矮小症组71例。采用单因素及多因素Logistic回归分析探讨儿童矮小症发生的影响因素;采用R软件构建预测儿童矮小症发生的列线图模型,并采用受试者工作特征(ROC)曲线及校准曲线评估列线图模型的区分度和一致性。结果1422例儿童中矮小症71例,发生率为4.99%。多因素Logistic回归分析结果显示,出生体质量、矮小症家族史、牛奶摄入情况、体育锻炼均是儿童矮小症发生的影响因素(P<0.05)。列线图模型预测儿童矮小症发生的ROC曲线的曲线下面积为0.897(95%CI:0.856~0.938),区分度较好;列线图模型预测儿童矮小症发生的校准曲线斜率趋近1,且Hosmer-Lemeshow拟合优度检验=5.020,P=0.740,一致性较好。结论基于出生体质量、矮小症家族史、牛奶摄入情况、体育锻炼4项影响因素构建的预测儿童矮小症发生的列线图模型的区分度及一致性较好,能够为临床拟定个体化的干预措施提供参考。

Objective To explore the influencing factors of dwarfism in children and construct a nomogram model.Methods From June 2020 to December 2022,1,500 children were selected as the research objects,and 1,422 cases were effectively investigated.According to incidence of dwarfism,the children were divided into normal group(n=1,351)and dwarfism group(n=71).Univariate and multivariate Logistic regression analyses were used to explore the influencing factors of dwarfism in children;the R software was used to construct a nomogram model for prediction of the occurrence of dwarfism in children,and the receiver operating characteristic(ROC)curve and calibration curve were used to evaluate the discrimination and consistency of the nomogram model.Results Among the 1422 children,71 cases had dwarfism,with an incidence rate of 4.99%.Multivariate Logistic regression analysis showed that birth weight,family history of dwarfism,milk intake and physical exercise were the influencing factors for the occurrence of dwarfism in children(P<0.05).The area under the curve of the ROC curve predicted by the nomogram model for the occurrence of dwarfism in children was 0.897(95%CI,0.856 to 0.938),with good discrimination;the calibration curve slope of the nomogram model for predicting the occurrence of dwarfism in children approached 1,and the Hosmer-Lemeshow goodness of fit test showed that the was 5.020 and P was 0.740,indicating good consistency.Conclusion The nomogram model for predicting the occurrence of dwarfism in children based on four influencing factors of birth weight,family history of dwarfism,milk intake and physical exercise has good discrimination and consistency,which can provide reference for the development of personalized intervention measures in clinical practice.