TY - JOUR TI - Development and evaluation of a machine learning-based in-hospital COVID-19 disease outcome predictor (CODOP): A multicontinental retrospective study AU - Klén, Riku AU - Purohit, Disha AU - Gómez-Huelgas, Ricardo AU - Casas-Rojo, José Manuel AU - Antón-Santos, Juan Miguel AU - Núñez-Cortés, Jesús Millán AU - Lumbreras, Carlos AU - Ramos-Rincón, José Manuel AU - García Barrio, Noelia AU - Pedrera-Jiménez, Miguel AU - Lalueza Blanco, Antonio AU - Martin-Escalante, María Dolores AU - Rivas-Ruiz, Francisco AU - Onieva-García, Maria Ángeles AU - Young, Pablo AU - Ramirez, Juan Ignacio AU - Titto Omonte, Estela Edith AU - Gross Artega, Rosmery AU - Canales Beltrán, Magdy Teresa AU - Valdez, Pascual Ruben AU - Pugliese, Florencia AU - Castagna, Rosa AU - Huespe, Ivan A AU - Boietti, Bruno AU - Pollan, Javier A AU - Funke, Nico AU - Leiding, Benjamin AU - Gómez-Varela, David A2 - Giamarellos-Bourboulis, Evangelos J A2 - van der Meer, Jos W A2 - Giamarellos-Bourboulis, Evangelos J A2 - Bermejo-Martin, Jesus VL - 11 PY - 2022 DA - 2022/05/17 SP - e75985 C1 - eLife 2022;11:e75985 DO - 10.7554/eLife.75985 UR - https://doi.org/10.7554/eLife.75985 AB - New SARS-CoV-2 variants, breakthrough infections, waning immunity, and sub-optimal vaccination rates account for surges of hospitalizations and deaths. There is an urgent need for clinically valuable and generalizable triage tools assisting the allocation of hospital resources, particularly in resource-limited countries. We developed and validate CODOP, a machine learning-based tool for predicting the clinical outcome of hospitalized COVID-19 patients. CODOP was trained, tested and validated with six cohorts encompassing 29223 COVID-19 patients from more than 150 hospitals in Spain, the USA and Latin America during 2020–22. CODOP uses 12 clinical parameters commonly measured at hospital admission for reaching high discriminative ability up to 9 days before clinical resolution (AUROC: 0·90–0·96), it is well calibrated, and it enables an effective dynamic risk stratification during hospitalization. Furthermore, CODOP maintains its predictive ability independently of the virus variant and the vaccination status. To reckon with the fluctuating pressure levels in hospitals during the pandemic, we offer two online CODOP calculators, suited for undertriage or overtriage scenarios, validated with a cohort of patients from 42 hospitals in three Latin American countries (78–100% sensitivity and 89–97% specificity). The performance of CODOP in heterogeneous and geographically disperse patient cohorts and the easiness of use strongly suggest its clinical utility, particularly in resource-limited countries. KW - COVID-19 KW - machine-learning KW - prediction KW - triage JF - eLife SN - 2050-084X PB - eLife Sciences Publications, Ltd ER -