Description of variables, algorithms, and outcomes as well as pre-processing steps for the overall dataset (panel A). Retrospective validation of the model derived on the European cohort based on cross-validation as well as external validation on France as the country with highest ICU admissions (panel B), while the results for the rest of the European countries are shown in Tables A-C in S6 Text. Prospective validation of the model derived on the European cohort of patients admitted to ICUs before the cut-off date (December 1st, 2020) and validated on the European cohort of patients admitted to ICUs after the cut-off date, during the subsequent pandemic wave (panel C). External validation of the model derived on the overall European cohort and validated on the cohort of patients admitted to Asian, African and American ICUs (panel D). Note, EU is an abbreviation of Europe.