Risk Prediction Score for Pediatric Patients with Suspected Ebola Virus Disease

Area of Science:

• Pediatric infectious diseases
• Epidemiology
• Clinical diagnostics

Background:

• Rapid diagnostic tools are crucial for timely isolation and treatment of children with Ebola virus disease (EVD).
• Existing diagnostic criteria may not be optimal for pediatric populations.

Purpose of the Study:

• To evaluate a predictive diagnostic tool for EVD in children.
• To develop and validate a simplified risk score for pediatric EVD diagnosis.

Main Methods:

• Retrospective analysis of data from Ebola Treatment Centers (2014-2015) in West Africa.
• Development of a 7-point Pediatric Ebola Risk Score using statistically derived predictors.
• External validation using data from 2018-2019 in the Democratic Republic of the Congo.

Main Results:

• The Pediatric Ebola Risk Score demonstrated strong model discrimination with an area under the curve (AUC) of 0.87.
• Predictors associated with EVD included evidence of bleeding and known/unknown Ebola contacts.
• External validation yielded an AUC of 0.70, outperforming World Health Organization criteria (AUC 0.56) in initial analysis and proving simpler to use.

Conclusions:

• The Pediatric Ebola Risk Score is a promising tool for improving EVD diagnosis in children.
• The score offers a simpler and more effective alternative to current diagnostic criteria for pediatric EVD.
• Further implementation and validation in diverse settings are warranted.