Construction and validation of a risk prediction model for unplanned extubation in postoperative patients
10.3760/cma.j.cn115682-20211008-04541
- VernacularTitle:外科术后患者非计划性拔管风险预测模型的构建与验证
- Author:
Qing ZHAO
1
;
Wujia JIANG
;
Liping ZHOU
;
Chaoli XU
;
Yi CHEN
;
Xin YU
Author Information
1. 南京医科大学附属无锡人民医院肝胆外科，无锡 214023
- Keywords:
Operative;
Postoperative period;
Unplanned extubation;
Risk management;
Predictive model;
Catheter nursing
- From:
Chinese Journal of Modern Nursing
2022;28(23):3090-3095
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct a risk prediction model for unplanned extubation (UEX) in postoperative patients, and to verify the predictive performance of the model.Methods:From January 2019 to April 2020, 314 patients who underwent surgy in Wuxi People's Hospital Affiliated to Nanjing Medical University and had postoperative indwelling catheters were selected by convenience sampling as the research object. The patients were divided into UEX group ( n=25) and non-UEX group ( n=289) according to whether UEX occurred postoperatively. The gender, age, admission to ICU, tracheal intubation, physical restraint, catheter number, analgesia, sedation, taking sleeping pills and postoperative fever of the two groups of patients were counted. Binomial Logistic regression was used to analyze the risk factors of postoperative UEX, and a risk prediction model was established. The area under the curve ( AUC) of receiver operating characteristic curve was used to test the predictive performance of the model. According to the same standard, the clinical data of 134 patients who underwent surgy from May to December 2020 were selected to validate the model. Results:The risk prediction model for UEX in postoperative patients finally included 3 risk factors, physical restraint, tracheal intubation, and age, and assigned 0-2, 0-4, and 0-5 respectively. The scores of the 3 indicators were summed up as a total score, with a total score of 0 to 11. The AUC was 0.887 [95% CI (0.807, 0.967) ] , the maximum Youden index was 0.591, the sensitivity was 0.640, the specificity was 0.951, and the predictive critical value was 5.5. The model validation showed that the sensitivity was 66.7%, the specificity was 92.6%, and the correct rate was 90.3%. Conclusions:The risk prediction model for UEX in postoperative patients has a good predictive performance, which can help clinical medical and nursing staff to quickly identify patients with high risk of UEX after surgery, and provide a reference for formulating targeted interventions.
