Abstract:

Objective To externally validate the Classification and Regression Tree and the Nomogram model for risk prediction in patients receiving nasogastric feeding constructed in the earlier stage of this study,and to evaluate the clinical practicability and effectiveness of the 2 models. Methods Convenience sampling method was applied to collect the information of nasogastric feeding patients from July 2019 to July 2020 in a tertiary hospital of Chongqing. The patients were divided into an aspiration group and a non-aspiration group according to the incidence of aspiration. The ROC curves were drawn to evaluate the performance of the 2 models. Results The area under the ROC curves of the Nomogram and Classification and Regression Tree was 0.91（95%CI:0.89~0.98）and 0.92（95%CI:0.91~0.95）,respectively. The sensitivity,specificity and accuracy in 2 groups were 89.2% and 64.9%,89.7% and 96.6%,89.6% and 89.0%,respectively. Conclusion Both the Nomogram and the Classification and Regression Tree can predict the aspiration risk of nasogastric feeding patients effectively. The prediction performance of the Classification and Regression Tree is slightly better than that of the Nomogram.

Key words: Enteral Nutrition, Aspiration, Risk Management, Prediction Model, Nursing Care

摘要：

目的 对该研究团队前期构建的鼻饲患者误吸风险预测列线图和分类回归树开展外部验证,评估模型的临床适用性及有效性。 方法 采用便利抽样法,选取重庆市某三级甲等医院2019年7月—2020年7月收治的鼻饲患者作为研究对象,根据是否发生误吸分为误吸组和非误吸组,绘制受试者工作特征曲线评价两种模型的性能。 结果 鼻饲患者误吸风险预测列线图和分类回归树的受试者工作特征曲线下面积分别为0.91（95%CI为0.89~0.98）和0.92（95%CI为0.91~0.95）,灵敏度分别为89.2%和64.9%,特异度分别为89.7%和96.6%,预测准确率分别为89.6%和89.0%。 结论 该研究团队前期构建的鼻饲患者误吸风险预测列线图和分类回归树均能有效预测鼻饲患者的误吸风险,分类回归树的预测性能略优于列线图。

关键词: 肠道营养, 误吸, 风险管理, 预测模型, 护理