Abstract
BACKGROUND: Community-acquired bloodstream infections (CABSIs) are common in the emergency departments, and some progress to sepsis and even lead to death. However, limited information is available regarding the prediction of patients with high risk of death. METHODS: The Emergency Bloodstream Infection Score (EBS) for CABSIs was developed to visualize the output of a logistic regression model and was validated by the area under the curve (AUC). The Mortality in Emergency Department Sepsis (MEDS), Pitt Bacteremia Score (PBS), Sequential Organ Failure Assessment (SOFA), quick Sequential Organ Failure Assessment (qSOFA), Charlson Comorbidity Index (CCI), and McCabe-Jackson Comorbid Classification (MJCC) for patients with CABSIs were computed to compare them with EBS in terms of the AUC and decision curve analysis (DCA). The net reclassification improvement (NRI) index and integrated discrimination improvement (IDI) index were compared between the SOFA and EBS. RESULTS: A total of 547 patients with CABSIs were included. The AUC (0.853) of the EBS was larger than those of the MEDS, PBS, SOFA, and qSOFA (all P<0.001). The NRI index of EBS in predicting the in-hospital mortality of CABSIs patients was 0.368 (P=0.04), and the IDI index was 0.079 (P=0.03). DCA showed that when the threshold probability was < 0.1, the net benefit of the EBS model was higher than those of the other models. CONCLUSION: The EBS prognostic models were better than the SOFA, qSOFA, MEDS, and PBS models in predicting the in-hospital mortality of patients with CABSIs.