儿童结肠镜检查前肠道准备不合格影响因素分析与模型构建

Analysis of influencing factors and model construction of unqualified bowel preparation before colonoscopy in children

目的:分析儿童结肠镜检查前肠道准备不合格的影响因素并构建其列线图模型。方法:收集医院消化科2021年1月—2021年6月180例结肠镜检查患儿的临床相关资料。使用Boston肠道准备量表(BBPS)将患儿分为肠道准备合格组(BBPS≥6分)与肠道准备不合格组(BBPS<6分)。采用二元Logistic回归模型分析儿童结肠镜检查前肠道准备不合格的独立影响因素,使用R软件构建预测儿童结肠镜检查肠道准备不合格影响因素的列线图模型,并进行验证。结果:二元Logistic回归分析显示,性别(OR=2.181)、便秘史(OR=0.400)、是否订肠道准备营养餐(OR=2.893)、肠道准备好后等待时间(OR=0.638)是儿童结肠镜检查前肠道准备不合格的独立影响因素(P<0.05)。据此构建列线图模型并进行验证,ROC曲线显示该模型预测儿童结肠镜检查前肠道准备不合格风险曲线下面积为0.759,校准曲线为斜率接近于1的直线,Hosmer-Lemeshow拟合优度检验(χ^(2)=9.893,P=0.273)显示该模型能较为准确地预测儿童结肠镜检查前肠道准备不合格的发生风险。结论:构建的儿童结肠镜检查前肠道准备不合格影响因素列线图模型具有良好的精准度与区分度,可个性化较为准确地预测儿童结肠镜检查前肠道准备不合格的发生风险,提高临床护理质量。

Objective:To analyze the influencing factors of unqualified intestinal preparation before colonoscopy in children and construct its nomogram model.Methods:Clinical data of 180 children undergoing colonoscopy from January 2021 to June 2021 were collected.The children were divided into qualified group(BBPS≥6 points)and unqualified group(BBPS<6 points)using the Boston Intestinal Readiness Scale(BBPS).Binary Logistic regression model was used to analyze the independent influencing factors of unqualified intestinal preparation before colonoscopy in children,and R software was used to build a line graph model predicting the influencing factors of unqualified intestinal preparation before colonoscopy in children,and then verified.Results:Binary Logistic regression analysis showed that gender(OR=2.181),bowel secret history(OR=0.400),whether to order nutritional meals for intestinal preparation(OR=2.893),and waiting time after intestinal preparation(OR=0.638)were independent influencing factors for substandard intestinal preparation before colonoscopy in children(P<0.05).The ROC curve showed that the area under the curve of risk of unqualified intestinal preparation before colonoscopy predicted by this model was 0.759,and the calibration curve was a line with slope close to 1.Hosmer-Lemeshow goodness-of-fit test(χ~2 squared=9.893,P=0.273)showed that the model can more accurately predict children′s risk of bowel preparation before colonoscopy is unqualified.Conclusion:The established model of influencing factors of unqualified intestinal preparation before colonoscopy has good accuracy and differentiation,which can predict the risk of unqualified intestinal preparation before colonoscopy more accurately and personalized,and improve the quality of clinical care.