正畸患者牙釉质脱矿风险预测模型的构建与验证

Construction and Validation of a Risk Prediction Model for Enamel Demineralization in Orthodontic Patients

目的:探讨影响正畸患者牙釉质脱矿的因素,并构建Logistic回归预测模型。方法:选取2019年12月~2022年12月南京市3所三甲医院1934例正畸患者为研究对象,将脱矿组(n=226)和非脱矿组(n=1708)的各项危险因子进行对比,运用Logistic回归分析建立风险预测模型,采用ROC曲线评价模型的预测效果。选取2023年1月~6月符合标准的828例患者进行模型预测效果验证。结果:本研究共纳入龋坏、菌斑、牙面酸蚀情况、喜爱软食物、刷牙频率、含氟牙膏、正畸疗程7个独立危险因素构建风险预测模型。本模型的ROC曲线下面积为0.849,灵敏度为0.771,特异度为0.723,Youden指数为0.494。模型验证结果:灵敏度为0.813,特异度为0.733,准确率为74.09%。结论:本模型预测效果良好,可有效预测牙釉质脱矿发生风险,为医护人员及时采取预防管理措施提供参考。

Objective:To investigate the factors affecting enamel demineralization in orthodontic patients and to establish a logistic regression prediction model.Methods:A total of 1934 orthodontic patients were selected from three hospitals in Nanjing from December 2019 to December 2022 as the research subjects.The risk factors between the demineralization group(n=226)and the non-demineralization group(n=1708)were compared.Logistic regression analysis was used to establish a risk prediction model,and the prediction effect of the model was evaluated by ROC curve.Additionally,828 patients who met the criteria from January to June 2023 were selected for model prediction effect validation.Results:Seven independent risk factors including caries,dental plaque,tooth surface erosion,preference for soft food,brushing frequency,fluoride toothpaste and orthodontic treatment duration were included to construct the risk prediction model.The area under the ROC curve of this model was 0.849,with the sensitivity,the specificity and the Youden index were 0.771,0.723,and 0.494,respectively.The model validation results showed the sensitivity of 0.813,the specificity of 0.733,and the accuracy of 74.09%.Conclusions:The model is effective in predicting the risk of enamel demineralization,and can provide reference for healthcare professionals to implement timely preventive management measures.