老年下肢骨关节炎患者衰弱风险评估模型的构建及验证

Construction and validation of a risk assessment model for frailty in elderly patients with lower extremity osteoarthritis

目的分析老年下肢骨关节炎患者衰弱的影响因素,构建并验证风险评估模型。方法采用方便抽样法,于2022年1月—9月选取杭州市某三级甲等医院和某社区卫生服务中心的535例老年下肢骨关节炎患者为调查对象,其中建模组357例(66.73%),验证组178例(33.27%)。采用单因素分析和多因素Logistic回归分析确定衰弱危险因素,构建风险评估模型并绘制列线图。通过受试者操作特征曲线下面积和Hosmer-Lemeshow检验评价模型的区分度和校准度。采用Bootstrap法进行内部验证,采用时间验证法进行外部验证。结果模型变量包括受累关节数、共病指数、疼痛、营养状况、久坐时间、日常生活活动能力、骨关节炎指数、下肢肌力、社会支持评定量表得分。Hosmer-Lemeshow检验P=0.202,受试者操作特征曲线下面积为0.942,最佳临界值为0.392,灵敏度为0.914,特异度为0.893,准确率为0.902。内外部验证显示,C-统计量分别为0.935、0.919,校准曲线显示拟合良好。结论该研究构建的老年下肢骨关节炎患者衰弱的风险评估模型具有较好的区分度和校准度,能直观、简便地筛查衰弱高风险的下肢骨关节炎老年患者,为早期监测、识别和防控提供参考。

Objective To analyze the influencing factors of frailty in elderly patients with lower extremity osteoarthritis,and to construct and validate the risk assessment model.Methods Convenient sampling method was used to select 535 elderly patients with lower extremity osteoarthritis from tertiary hospitals and community health service centers in Hangzhou from January to September 2022 as the survey subjects including 357 in the modeling group and 178 in the validation group.Univariate and multivariate logistic regression analysis were used to determine the risk factors of frailty,construct a risk assessment model and draw a nomogram.The discrimination and calibration of the model were evaluated by the area under the receiver operating characteristic curve and the Hosmer-Lemeshow test.The Bootstrap method was used for internal validation of the model,and the time verification method was used for external validation.Results The model variables included the number of affected joints,age-adjusted Charlson comorbidity index,pain,nutritional status,sedentary time,activity of daily living,osteoarthritis index,lower limb muscle strength,and Social Support Rating Scale score.The Hosmer-Lemeshow test results of the model showed that P=0.202,the area under the receiver operating characteristic curve was 0.942,the optimal critical value was 0.392,the sensitivity was 0.914,the specificity was 0.893,and the accuracy rate was 0.902.The internal and external validation showed that the C-statistics were 0.935 and 0.919,respectively,and the calibration curve showed good fitting.Conclusion The risk assessment model has a good degree of discrimination and calibration,which can more intuitively and easily screen elderly patients with lower extremity osteoarthritis at high risk of frailty,and provide references for early monitoring,identification,prevention and control.