肺结核患者服药依从性预测模型的建立

Establishment of prediction model of medication compliance in patients with pulmonary tuberculosis

目的:探究肺结核患者服药依从性预测模型的建立。方法:选取2017年1月至2021年6月于我院结核科检查治疗的初治肺结核患者为研究对象,收集患者临床资料,分别使用单因素和多因素回归分析服药依从性差的影响因素,并建立相关列线图模型。结果:年龄≥60岁、文化程度低、自费支付、社会支持评定量表(SSRS)评分低、无人监督用药、用药种类多和有药物不良反应是初治肺结核患者服药依从性差的独立危险因素(P<0.05)。据此建立预测初治肺结核患者服药依从性差风险的列线图模型,模型验证结果显示C-index为0.819,校正曲线趋近于理想曲线,受试者工作特征(ROC)曲线的曲线下面积(AUC)为0.839(95%CI为0.815~0.862),在1%~78%范围内,列线图预测的净获益值较高,表明该模型具有良好的预测能力。结论:导致初治肺结核患者服药依从性差的危险因素较多,基于危险因素构建的列线图模型能够有效预测初治肺结核患者服药依从性差的风险。

Objective:To establish a drug compliance prediction model for patients with pulmonary tuberculosis.Methods:The pulmonary tuberculosis patients who were examined and first treated in the tuberculosis department of our hospital from January 2017 to June 2021 were selected as our research subjects,and their clinical data were collected.Single factor and multivariate regression analyses were used to analyze the influencing factors of poor medication compliance to establish a related nomogram model.Results:Age≥60 years,low education level,self-payment,low score of SSRS,unsupervised medication,multiple types of medications,and adverse drug reactions were independent risk factors for poor medication compliance in newly treated pulmonary tuberculosis patients(P<0.05).Based on this,a nomogram model was established to predict the risk of poor medication compliance.The model verification results showed that the C-index was 0.819.And the calibration curve was close to the ideal curve.The area under the receiver operating characteristic(ROC)curve(AUC)was 0.839(95%CI:0.815-0.862).In the range of 1%to 78%,the net benefit value predicted by the nomogram was higher,indicating that the model had good predictive efficiency.Conclusion:There are many risk factors that lead to poor medication compliance in newly-treated pulmonary tuberculosis patients.Establishment of nomogram model based on risk factors can effectively predict the risk of poor medication compliance in these patients.