Xu Yun, Ou Shan, Yuan Xiu, Yao Jie, Yang Tianwen
Department of Gastroenterology, The First Hospital Affiliated of Chongqing Medical and Pharmaceutical College, Chongqing, 400060, People's Republic of China.
Ther Clin Risk Manag. 2025 Jun 11;21:885-894. doi: 10.2147/TCRM.S503389. eCollection 2025.
To establish a prediction model for the risk of inadequate bowel preparation for colonoscopy in the elderly, and to validate the model to test its prediction effect.
A total 927 patients who underwent colonoscopy in the First Affiliated Hospital to Chongqing Medical and Pharmaceutical College from January to December 2023 were selected as the modeling group, and were divided into the success group (839 patients) and failure group (88 patients) according to whether or not inadequate bowel preparation occurred, and then a logistic regression prediction model was established and transformed into a risk scoring system. The area under the ROC curve and the Hosmer-Lemeshow test were used for model differentiation and calibration, respectively, and the 10-fold cross-validation method was used for internal validation of the model. A total of 548 patients who underwent colonoscopy in our hospital from January to June 2024 were selected for external validation of the model.
Logistic regression analysis showed that age ≥65 years, diabetes mellitus, exercise habits, Bristol stool characterization scale, frailty, and ambulatory functional status were independent factors influencing the inadequate bowel preparation for colonoscopy in the elderly. The area under the ROC curve for the modeling group was 0.806, and the Hosmer-Lemeshow test showed p=0.870. The area under the ROC curve of the validation group was (0.785±0.090). The area under the ROC curve of the validation group was 0.824, and the Hosmer-Lemeshow test showed =0.907, with a sensitivity of 73.13%, a specificity of 80.70%, and a Jordon's index of 0.538.
The predictive model of the risk of inadequate bowel preparation for colonoscopy in elderly people can predict the risk of inadequate bowel preparation in elderly people, and can be used as a tool for clinical caregivers to screen the high-risk group of inadequate bowel preparation in the elderly.
建立老年患者结肠镜检查肠道准备不充分风险的预测模型,并对该模型进行验证以检验其预测效果。
选取2023年1月至12月在重庆医药高等专科学校附属第一医院接受结肠镜检查的927例患者作为建模组,根据肠道准备是否不充分分为成功组(839例)和失败组(88例),然后建立逻辑回归预测模型并转化为风险评分系统。分别采用ROC曲线下面积和Hosmer-Lemeshow检验进行模型区分度和校准,采用10倍交叉验证法对模型进行内部验证。选取2024年1月至6月在我院接受结肠镜检查的548例患者进行模型外部验证。
逻辑回归分析显示,年龄≥65岁、糖尿病、运动习惯、布里斯托大便性状量表、衰弱及活动功能状态是影响老年患者结肠镜检查肠道准备不充分的独立因素。建模组ROC曲线下面积为0.806,Hosmer-Lemeshow检验显示p=0.870。验证组ROC曲线下面积为(0.785±0.090)。验证组ROC曲线下面积为0.824,Hosmer-Lemeshow检验显示=0.907,灵敏度为73.13%,特异度为80.70%,约登指数为0.538。
老年患者结肠镜检查肠道准备不充分风险预测模型可预测老年患者肠道准备不充分的风险,可作为临床医护人员筛查老年患者肠道准备不充分高危人群的工具。