Abstract:

Objective To analyze the predictive value of fecal form for the quality of bowel preparation in patients scheduled for colonoscopy，identify risk factors for bowel preparation failure，and develop and validate a risk prediction model. Methods This was a prospective cohort study using convenience sampling. Patients scheduled for colonoscopy in the Digestive Department of a tertiary A hospital in Jiangsu Province from June to December 2022 were included in the modeling cohort. General information sheet and the Bristol Stool Form Scale（BSFS） were used for data collection. Patients were categorized into a successful bowel preparation group and a bowel preparation failure group based on the quality of bowel preparation. The optimal cutoff value for BSFS was determined using the best cutoff value method. Logistic regression analysis was employed to identify risk factors for bowel preparation failure，and a nomogram risk prediction model was constructed. Patients undergoing colonoscopy in the same hospital from January to February 2023 were served as the validation cohort. Results The modeling cohort included 569 patients，and the validation cohort included 212 patients，with bowel preparation failure rates of 19.0% and 19.8%，respectively. The risk prediction model formula derived from logistic regression analysis was P=-2.209+0.619 × hospitalized patients+0.635 × age≥65 years-0.710 × previous colonoscopy history+2.031 × BSFS type 1~2. The area under the receiver operating characteristic curve for the model was 0.751，with a sensitivity of 54.6%，a specificity of 85.9%，and the optimal cutoff value was 0.225，corresponding to a risk score of 80 for bowel preparation failure. The Hosmer-Lemeshow test showed χ²=4.429，P=0.351. External validation demonstrated an area under the receiver operating characteristic curve of 0.775，indicating good model fit and high predictive value. Conclusion The incidence of bowel preparation failure in colonoscopy patients is relatively high. Patients aged≥65 years，hospitalized patients，those with no history of colonoscopy，and those with BSFS type 1~2 are more likely to experience bowel preparation failure. The risk prediction model developed in this study has good predictive performance and can provide a basis for clinical nurses to quickly assess the risk of bowel preparation failure in patients.

Key words: Colonoscopy, Stool Form, Risk Factors, Nomogram, Prediction Model, Nursing Care

摘要：

目的 分析大便性状分型对结肠镜检查患者肠道准备失败的预测价值及肠道准备失败的危险因素，构建并验证风险预测模型。方法 该研究为前瞻性队列研究，采用便利抽样法，选取2022年6月—12月在江苏省某三级甲等医院消化内科就诊及住院的结肠镜检查患者作为建模队列，采用一般情况调查表、Bristol大便性状量表（Bristol Stool Form Scale，BSFS）收集相关资料，并根据肠道准备质量将患者分为肠道准备成功组和肠道准备失败组。采用最佳截断值法确定BSFS的最佳截断值，通过单因素和Logistic回归分析肠道准备失败的危险因素，并构建列线图风险预测模型。将2023年1月—2月在同一所医院行结肠镜检查的患者作为验模队列，进行模型的外部验证。结果 最终建模队列纳入569例患者，验模队列纳入212例，肠道准备失败发生率分别为19.0%和19.8%。经Logistic回归分析后构建的风险预测模型公式为P=-2.209+0.619 × 住院患者+0.635 × 年龄≥65岁-0.710 × 有结肠镜检查史+2.031 × BSFS分型为1~2型。该模型受试者操作特征曲线下面积为0.751，灵敏度为54.6%，特异度为85.9%，最佳截断值为0.225，肠道准备失败的风险值为80分。Hosmer-Lemeshow检验结果显示，χ²=4.429，P=0.351。模型外部验证中，受试者操作特征曲线下面积为0.775，提示模型具有较好的拟合效果和较高的预测价值。结论 结肠镜检查患者肠道准备失败发生率较高，年龄≥65岁、住院、无结肠镜检查史以及BSFS分型为1~2型的患者更易发生肠道准备失败。该研究构建的风险预测模型具有良好的预测性能，可为护理人员快速判别患者肠道准备失败风险提供依据。

关键词: 结肠镜检查, 大便性状, 危险因素, 列线图, 预测模型, 护理