Development and validation of a novel prediction model for Carbapenem-resistant organism infection in a large-scale hospitalized patients.

ABSTRACT

Carbapenem-resistant organism (CRO) are defined as gram-negative bacteria. The lack of safe and effective antibiotics has led to an increase in incidence rate. The purpose of this study is to establish and determine a risk nomogram to predict CRO infection in hospitalized patients. Hospitalized patients' information were collected from the electronic medical record system of hospital between January 2019 and December 2022. Based on the inclusion and exclusion criteria, we identified 131390 inpatients who met the criteria for this study. For the training cohort, the area under the curves (AUC) for predicting the CRO infection was 0.935. For the validation cohort, the AUC for predicting the CRO infection was 0.937. We have developed the first novel nomogram to predict CRO infection in hospitalized patients, which is reliable and high-performance. The nomogram performs well among hospitalized patients and has good predictive ability.