Narrative review of clinical prediction models for paediatric community acquired pneumonia.

Abstract

Community acquired pneumonia (CAP) is a disease experienced by children the world over, though CAP-related morbidity and mortality differ markedly between low- and middle-income countries (LMICs) and high-income countries (HICs). Thus, setting-specific clinical prediction models are needed to identify children at risk for CAP-related morbidity and mortality. Here, we outline published clinical prediction models from LMICs and HICs for pediatric CAP-related outcomes. To date, there have been four clinical prediction models to predict treatment failure, two to predict a composite outcome of poor outcomes, and eight models for mortality prediction for CAP in LMICs. No prediction models developed in LMICs had publications that described their impact on clinical care through implementation. In HICs, to date there are three published clinical prediction models evaluating disease severity and one examining the need for major medical interventions. While clinical prediction models described in this review provide a strong foundation for risk stratification for children with CAP in HICs, there is a need for widespread external validation and implementation of optimally performing models.