The bronchiolitis severity score: An assessment of face validity, construct validity, and interobserver reliability.

Area of Science:

• Pediatric Medicine
• Clinical Assessment Tools
• Respiratory Illness

Background:

• Acute bronchiolitis is a common pediatric respiratory infection requiring hospitalization.
• Assessing disease severity and predicting escalation of care are crucial for effective management.
• The Bronchiolitis Severity Score (BSS) was developed to standardize bronchiolitis severity assessment.

Purpose of the Study:

• To evaluate the face validity, interobserver reliability, and predictive ability of the Bronchiolitis Severity Score (BSS).
• To determine if the BSS can discriminate children requiring escalations of care within 24 hours of admission (late rescues).

Main Methods:

• A mixed-methods study combining retrospective and prospective data collection.
• Face validity was assessed via questionnaires from 39 respondents.
• Interobserver reliability was evaluated using kappa calculations on 72 blinded BSS assessments.
• Construct validity for predicting late rescues was analyzed using ROC curve analysis.

Main Results:

• Clinicians reported positive perceptions of BSS face and content validity.
• Interobserver reliability was weak (kappa = 0.22).
• The BSS showed limited ability to discriminate late rescues (AUC = 0.61), with a threshold of ≥6 yielding 56% sensitivity and 69% specificity.

Conclusions:

• While clinicians perceive the BSS as useful, its psychometric properties require enhancement.
• Weak interobserver reliability and limited predictive accuracy for care escalations were observed.
• Further refinement and validation of the BSS are recommended prior to its routine clinical implementation.