The length of stay (LOS) in the intensive care unit (ICU) is one of the most commonly used metrics for quality of care. Despite its potential limitations, ICU LOS is easy to measure, reproducible and can be used as a proxy for resource use, costs, and efficiency.(1) Moreover, it is a patientcentered outcome; therefore, it is of interest to multiple stakeholders, such as patients and families, managers, payors, and ICU personnel. However, in most circumstances, the ICU LOS is used retrospectively to assess ICU efficiency or to understand patients with a long LOS and, thus, elevated costs of care. Therefore, patient LOS prediction at ICU admission could help coordinate care, implement preventive measures, and better communicate with managers, payors, and families, setting realistic expectations.