无症状老年女性骨质疏松症列线图临床预测模型的构建及效果

Construction and effect of a nomogram clinical prediction model for predicting osteoporosis in asymptomatic elderly women

目的构建列线图预测模型,专门用于预测无症状的绝经后老年女性骨质疏松症的确切概率。方法将无症状的绝经后老年女性招募到训练组(n=319)和验证组(n=104)。获取并分析他们的临床特征和骨骼矿物质密度结果。通过单变量和多变量Logistic回归分析,筛选构建预测模型的因素。通过受试者工作特征(receiver operating characteristic,ROC)曲线、校准曲线、决策曲线分析和临床影响曲线对构建的预测模型进行统计学评价。根据模型的临界值对验证集进行实际预测,评价模型的预测效果。结果多因素Logistic回归分析显示,受教育程度较低和体质量较轻是独立的危险因素(P<0.05)。基于年龄、受教育程度和体质量,构建了列线图临床预测模型,具有中等预测值[曲线下面积(area under the curve,AUC)>0.7]、良好的校准性、临床获益和临床影响。构建的在线动态列线图(https://shibaolu.shinyapps.io/DynamicNomogram/)具有交互性且易于推广。以训练集所得的临界值=0.452为预测无症状绝经后老年女性骨质疏松症的标准,对验证集实际预测结果显示,列线图预测模型的预测效果与训练集比较接近(灵敏度=0.82,特异度=0.63),并且预测结果与实际结果具有中高度的一致性(Kappa值),表明预测模型具有一定的临床应用价值。结论该列线图临床预测模型具有良好的实际应用价值和良好的可推广性,有助于实现骨质疏松症的早预测、早诊断和早治疗,从而为无症状绝经后老年女性的骨骼健康作出贡献,促进公共卫生事业的发展。

Objective To construct and validate a nomogram clinical prediction model dedicated to predicting the exact probability of osteoporosis in asymptomatic postmenopausal elderly women.Methods Asymptomatic postmenopausal elderly women were recruited into the training(n=319)and validation(n=104)groups.Their clinical characteristics and bone mineral density(BMD)results were collected and analyzed.Univariate and multivariate Logistic regression analysis were performed.Construction of general and dynamic nomogram clinical prediction models.Validate the model through receiver operating characteristic(ROC)curves,calibration curves,decision curve analysis(DCA)curves,and clinical impact curves.Results Lower education level and lower body weight were independent risk factors.Based on age,education level,and weight,a nomogram clinical prediction model was constructed,which had moderate predictive value[area under the curve(AUC)>0.7],good calibration,clinical benefit,and clinical impact.The constructed online dynamic nomogram(https://shibaolu.shinyapps.io/DynamicNomogram/)was interactive and easy to generalize Taking the critical value 0.452 obtained from the training group as the standard for predicting osteoporosis in asymptomatic postmenopausal elderly women,the actual prediction results of the validation group showed that the prediction effect of the nomogram prediction model was relatively close to that of the training group(sensitivity=0.82,specificity=0.63),and the predicted results had a medium to high degree of consistency(Kappa value)with the actual results,indicating that the predictive model has certain clinical application value.Conclusions This nomogram clinical prediction model has good practical application value and good generalizability,which could help achieve early prediction,early diagnosis and early treatment of osteoporosis,thus contributing to the bone health of asymptomatic postmenopausal elderly women and promoting the development of public health.