COVID-19 Clinical footprint to infer about mortality

TL;DR

This study analyzes a large dataset of COVID-19 patients in Mexico to understand how comorbidities, symptoms, and hospitalizations relate to mortality, using a probabilistic model to predict death outcomes.

Contribution

It introduces a comprehensive clinical footprint model based on multivariate Bernoulli distribution for robust Bayesian inference of COVID-19 mortality risk.

Findings

Identified key factors associated with COVID-19 mortality.

Developed a predictive model for death outcomes.

Provided insights into the interplay of symptoms and comorbidities.

Abstract

Information of 1.6 million patients identified as SARS-CoV-2 positive in Mexico is used to understand the relationship between comorbidities, symptoms, hospitalizations and deaths due to the COVID-19 disease. Using the presence or absence of these latter variables a clinical footprint for each patient is created. The risk, expected mortality and the prediction of death outcomes, among other relevant quantities, are obtained and analyzed by means of a multivariate Bernoulli distribution. The proposal considers all possible footprint combinations resulting in a robust model suitable for Bayesian inference.