摘要
目的:探究慢性阻塞性肺疾病(COPD)患者呼吸功能锻炼行为影响因素,并构建列线图预测模型.方法:选取2021年3月至2023年3月我院COPD患者112例,根据呼吸功能锻炼行为情况分为低依从性组(n=68)和高依从性组(n=44).采用单因素分析、多因素Logistic回归分析COPD患者呼吸功能锻炼行为低依从性的影响因素,构建预测模型,并采用受试者工作特征(ROC)分析模型预测价值.结果:年龄≥ 60岁、在职、无家庭支持、活动后呼吸困难、疾病不确定感高、运动自我效能低为COPD患者呼吸功能锻炼行为低依从性的独立危险因素(P＜0.05).年龄、在职、家庭支持、活动后呼吸困难、疾病不确定感、运动自我效能及列线图预测模型的曲线下面积(AUC)分别为0.645、0.645、0.717、0.600、0.660、0.672、0.908,预测模型对COPD患者呼吸功能锻炼低依从性的预测价值更高,当取截断值(cut-off)值为0.498时,其灵敏度为0.912,特异度为0.795.Bootstrap法(B=1000)内部验证显示,修正偏差后的预测曲线与理想线基本重合,一致性指数(C-index)为0.816,表明该模型的预测能力较好.决策曲线显示,阈值概率范围为0.01～0.92,净收益率＞0.结论:年龄≥ 60岁、在职、无家庭支持、活动后呼吸困难、疾病不确定感高、运动自我效能低为COPD患者呼吸功能锻炼行为低依从性的独立危险因素.预测模型有助于评估锻炼依从性并制定干预措施.
Abstract
Objective:To explore the influencing factors of respiratory function exercise behavior in patients with chronic obstructive pulmonary disease(COPD),and to construct a nomogram prediction model.Methods:112 COPD patients in our hospital from March 2021 to March 2023 were selected,and patients were divided into low compliance group(n=68)and high compliance group(n=44)according to their respiratory function exercise behavior.The influencing factors of low compliance of respiratory function exercise behavior in COPD patients were analyzed by univariate analysis and multivariate logistic regression analysis,and a prediction model was constructed,the predictive value of the model was analyzed by receiver operating characteristic(ROC).Results:Age≥60 years old,on the job,no family support,dyspnea after activity,high disease uncertainty,and low exercise self-efficacy were independent risk factors for low compliance of respiratory function exercise in COPD patients(P＜0.05).The area under the curve(AUC)of age,on the job,family support,dyspnea after activity,disease uncertainty,exercise self-efficacy and nomogram prediction model were 0.645,0.645,0.717,0.600,0.660,0.672 and 0.908 respectively,the prediction model had higher predictive value for low compliance of respiratory function exercise in COPD patients,when the cut-off value was 0.498,the sensitivity was 0.912 and the specificity was 0.795.The internal validation of the Bootstrap method(B=1000)showed that the predicted curve after the correction of the deviation was basically coincident with the ideal line,and the consistency index(C-index)was 0.816,indicating that the prediction ability of the model was better.The decision curve shows that the threshold probability range was 0.01～0.92,and the net return rate was＞0.Conclusion:Age≥ 60 years old,on the job,no family support,dyspnea after activity,high uncertainty of disease,and low exercise self-efficacy are.independent risk factors for low compliance of respiratory function exercise in COPD patients.The predictive model can help to assess exercise compliance and develop interventions.
NETL
NSTL
万方数据