Risk factors for intensive care unit patients with multidrug-resistant organism infection and construction of a decision tree model
ZHAO Jing
WU Min
GUO Yunyun
Abstract：Objective To establish a risk prediction model for intensive care unit(ICU)patients with multidrug-resistant organism infection(MDRO)based on decision tree,to provide references for clinical implementation of MDRO prevention and treatment strategies.Methods Using a retrospective analysis method,clinical data of a total of 210 nosocomial infection patients from January 2022 to January 2024 were collected.They were divided into MDRO group(61 cases)and non-MDRO group(149 cases)according to the onset of MDRO.Multivariate logistic regression analysis and decision tree method were implemented to establish prediction models for MDRO infection in ICU patients,and the predictive performance of two models was compared.Results Logistic regression analysis showed that age(OR=1.036,95%CI:1.002-1.072),length of hospital stay(OR=1.289,95%CI:1.072-1.550),hypoproteinemia(OR=2.604,95%CI:1.186-2.719),and antibacterial drugs>3 types(OR=3.250,95%CI:1.566-3.791)were independent risk factors for predicting MDRO in ICU(all P<0.05).Decision tree model demonstrated that the length of hospital stay was the most critical predictor of MDRO in ICU.The area under the curve(AUC)of the Logistic regression model for predicting MDRO was 0.719(95%CI:0.653-0.778),while that of the decision tree model was 0.813(95%CI:0.753-0.863).The prediction performance of the decision tree model was better than the Logistic regression model(Z=2.666,P=0.008).Conclusion The decision tree model of MDRO in ICU patients has high predictive performance and can be used as an effective tool for screening potential MDRO.
Keywords：intensive care unitmultidrug-resistant organismprediction modeldecision tree
Publication Date：2026-01-26
Online Publishing Date：2026-03-17(First online date of this platform, not the publication date of the document)
Pages：4( 87-90 )