Disclaimer: This is a prototype tool that calculates the 30-day mortality risk after emergency admission for suspected or confirmed SARS-CoV-2 infection in patients over 50 years. Use permitted for research purposes and tool validation only.

This tool consists on a Random Forest predictive model trained on 10399 COVID-19 cases of patients over 50 years from two spanish hospitals: Hospital 12 de Octubre de Madrid and Hospital Clínico Universitario de Valencia. Supported by comprehensive development and independent validation processes, the final model performs with a Cross-Validated Area Under Receiver Operating Characteristic (AUC-ROC) of .82, and at a selected decision threshold with a sensitivity of .80 and specificity of .70.

For more information about the model development please see our publication:

Publication pending. In process of submission.

SUBCOVERWD-19 project

Developed in the context of the project 'Severity Subgroup Discovery and Classification on COVID-19 Real World Data through Machine Learning and Data Quality assessment (SUBCOVERWD-19)' funded by Fondo Supera COVID-19 by CRUE (Conferncia de Rectores de las Universidades Españolas) - Santander Universidades (Santander Bank), from Oct 2020 to Dec 2021.

Research team

Universitat Politècnica de València (Biomedical Data Science lab - ITACA Institute): Carlos Sáez Silvestre (Principal Investigator), Inés Llorente Lipe, Pablo Ferri Borredá, Lexin Zhou, Alejandro Ibáñez Peña, Francisco Tomás García Ruiz, Jeanne Sudour, Aleksandar Rodríguez Dujic, Sabina Asensio Cuesta, Juan M García Gómez.

Instituto de Investigación IMAS12 - Hospital 12 de Octubre, Madrid : Agustín Gómez de la Cámara, Maria Teresa García Morales, Pilar Cancelas Navia, David Lora Pablos.

Instituto de Investigación Sanitaria INCLIVA - Hospital Clínico Universitario de Valencia: Rafael Badenes Quiles, Nekane Romero García.

If you are interested in evaluating this tool or adapting it to a new problem feel free to contact us.