新疆地区肠道准备不充分预测模型的外部验证  

External validation of a prediction model of inadequate bowel preparation in Xinjiang

作　　者：热孜亚·阿帕儿 徐甜 饶薇 高峰[1,2] 黄晓玲 REZIYA·Apaer;XU Tian;RAO Wei;GAO Feng;HUANG Xiaoling(Department of Gastroenterology,People′s Hospital of Xinjiang Uygur Autonomous Region,Urumqi 830001;Xinjiang Clinical Medical Research Center for Digestive Diseases,China)

机构地区：[1]新疆维吾尔自治区人民医院消化内科,新疆乌鲁木齐830001 [2]新疆消化系统疾病临床医学研究中心

出　　处：《胃肠病学和肝病学杂志》2024年第4期385-389,共5页Chinese Journal of Gastroenterology and Hepatology

摘　　要：目的对已建立的新疆地区肠道准备不充分的预测模型进行外部验证,明确该预测模型是否能够用于临床实践。方法收集新疆维吾尔自治区人民医院消化内科2022年1月至2022年6月住院治疗的1221例患者。根据波士顿评分标准(Boston bowel preparation scale,BBPS),≥6分为准备充分,≤5分或任意一段肠道评分<2分为准备不充分,其中肠道准备充分者990例,肠道准备不充分者231例,采用已建立的预测模型,通过患者性别、年龄、民族、肠道准备不充分史、便秘、饮酒等指标对患者进行肠道准备不充分的风险预测,将预测结果和实际观测结果的不同程度进行比较。预测模型的区分度和校准度分别采用ROC曲线和GiViTI校准曲线带验证。结果预测模型外部验证的ROC曲线下AUC为0.900(95%CI:0.762~0.902),可以更好地区分肠道准备不充分的患者。GiViTI校准曲线带的95%CI区域未穿过45°对角平分线(P=0.461),预测模型的预测概率接近实际观测到的概率,具有良好的校准性,预测模型以列线图的形式呈现。结论模型预测肠道准备不充分风险概率的准确性高,对提高此类高危患者的早期识别和筛选有很大的帮助。Objective To externally validate the established predictive model for inadequate bowel preparation in Xinjiang,and to ascertain whether the predictive model can be utilized in clinical practice.Methods A total of 1221 cases admitted to the Department of Gastroenterology at the People′s Hospital of Xinjiang Uygur Autonomous Region from Jan.2022 to Jun.2022 were collected.Cases were categorized based on the Boston Bowel Preparation Scale(BBPS)criteria:BBPS scores≥6 were considered well-prepared,while BBPS scores≤5 or any section of the bowel rated<2 were considered under-prepared.Among them,990 cases were classified as well-prepared bowel and 231 cases as under-prepared bowel.The established prediction model was applied to predict the risk of under-prepared bowel based on patient characteristics such as gender,age,ethnicity,history of colonoscopy,constipation,alcohol consumption,and other indicators.Differences between predicted and observed results were compared.The discrimination and calibration of the prediction models were assessed using ROC curves and GiViTI calibration curve bands,respectively.Results The external validation of the prediction model demonstrated an area under the ROC curve of 0.900(95%CI:0.762-0.902),indicating its ability to differentiate patients with different bowel preparation outcomes effectively.The 95%CI region of the GiViTI calibration curve band did not intersect the 45°diagonal parity(P=0.461),suggesting good calibration where the predicted probability closely matched the observed probability.The prediction model was visually presented using a column line plot.Conclusion The high accuracy of the model in predicting the probability of inadequate bowel preparation risk contributes to enhancing the early identification and screening of high-risk patients.

关 键 词：结肠镜检查 肠道准备 预测模型 外部验证 

分 类 号：R574[医药卫生—消化系统]