A Bayesian Model to Predict COVID-19 Severity in Children.

Area of Science:

• Pediatric Infectious Diseases
• Critical Care Medicine
• Epidemiology

Background:

• COVID-19 poses risks to hospitalized children.
• Identifying predictors of critical illness is essential for timely intervention.
• Understanding risk factors aids in resource allocation and patient management.

Purpose of the Study:

• To identify risk factors for critical illness in pediatric COVID-19 patients.
• To develop a predictive model for critical care needs in children with SARS-CoV-2.
• To stratify risk based on clinical syndromes and patient characteristics.

Main Methods:

• Multicenter, prospective study across 52 Spanish hospitals.
• Inclusion of 350 pediatric patients with SARS-CoV-2 infection.
• Application of a multivariable Bayesian model to estimate critical care probability.

Main Results:

• 24.2% of children with relevant COVID-19 required critical care.
• Identified risk factors include high C-reactive protein, creatinine, lymphopenia, low platelets, anemia, tachycardia, age, neutrophilia, leukocytosis, and low oxygen saturation.
• Risk factors' impact on critical disease severity varied by clinical syndrome (MIS-C, bronchopulmonary, gastrointestinal, mild).

Conclusions:

• Key risk factors for severe pediatric COVID-19 include inflammation, cytopenia, age, comorbidities, and organ dysfunction.
• The predictive power of risk factors is amplified in more severe clinical syndromes.
• A Bayesian model effectively predicts the risk of severe COVID-19 in children, with an online tool developed for practical application.