Performance of Three Asthma Predictive Tools in a Cohort of Infants Hospitalized With Severe Bronchiolitis

Area of Science:

• Pediatric Pulmonology
• Respiratory Medicine
• Clinical Epidemiology

Background:

• Childhood asthma develops in 30-40% of infants with severe bronchiolitis.
• Accurate prediction of asthma post-bronchiolitis remains a clinical challenge.
• Existing predictive tools require validation in diverse, high-risk populations.

Purpose of the Study:

• To evaluate the predictive performance of the Asthma Predictive Index (API), modified Asthma Predictive Index (mAPI), and Pediatric Asthma Risk Score (PARS).
• To assess these tools for predicting asthma at age 5 years in infants hospitalized with severe bronchiolitis.
• To compare performance in a real-world, multi-center cohort against original validation studies.

Main Methods:

• A prospective cohort study (MARC-35) of infants hospitalized with severe bronchiolitis (2011-2014).
• Application of API, mAPI, and PARS using data from the first 3 years of life (parent interviews, chart review, IgE testing).
• Asthma at age 5 defined by parent-reported clinician diagnosis; performance assessed by AUC and likelihood ratios.

Main Results:

• Among 875 children, 34% developed asthma by age 5.
• The Area Under the Curve (AUC) for prediction tools ranged from 0.57 to 0.68, indicating sub-optimal performance.
• Positive likelihood ratios were lower than previously reported, suggesting reduced clinical utility in this cohort.

Conclusions:

• The Asthma Predictive Index (API), modified Asthma Predictive Index (mAPI), and Pediatric Asthma Risk Score (PARS) demonstrated sub-optimal predictive performance in infants hospitalized with severe bronchiolitis.
• Current asthma prediction tools do not meet the desired accuracy (AUC >0.8) for this high-risk population.
• Development of novel, highly accurate prediction tools is needed for early asthma identification in infants post-severe bronchiolitis.