Zhi D, Zhang M, Lin J, et al. - This research sought to construct a model to prognosticate in-hospital death of sepsis patients based on the Medical Information Mart for Intensive Care-Ⅲ database, and use clinical data to externally validate the model. Researchers used a total of 1,839 sepsis patients for model development, and 125 clinical cases were used for external validation. They ascertained discriminatory ability of the model by estimating the area under the curve with 95% confidence intervals. It was shown that the random forest model had a better predictive ability and a good applicability to external populations with a SOFA score of 13-15.