Abstract:Objective To establish a risk prediction model for enteral nutrition-associated diarrhea in ICU patients. Methods Data of 335 ICU patients who received enteral nutrition therapy were retrospectively collected. The independent risk factors for enteral nutrition-associated diarrhea were analyzed using univariate tests and multivariate logistic regression model to construct a risk prediction model, which was displayed with a nomogram. Bootstrap method was used for internal verification, the receiver operating characteristic (ROC) curve was used to evaluate the discriminability of the model, the calibration curve was used to evaluate the accuracy of the model, and the clinical decision curve (DCA) was used to evaluate the effectiveness of the model. Results The risk prediction model for enteral nutrition-associated diarrhea included such 5 factors as APACHE-Ⅱ score, days of oral potassium preparations, use of gastrointestinal motility drug, fasting days, and albumin level. The area under the ROC curve of the prediction model was 0.940, the sensitivity was 81.82%, the specificity was 91.18%, and the accuracy was 86.57%. The maximum Youden index was 0.73, which pointed to a cutoff value of 122 points. The calibration curve was a straight line with a slope close to 1, meaning good accuracy. The clinical decision curve showed that the net benefit (NB) value of the model was high. Conclusion This model can effectively predict the occurrence of enteral nutrition-associated diarrhea in ICU patients, and the indexes can be obtained when enteral nutrition is started, which can provide reference for ICU medical staff to take preventive nursing intervention in time.