Modified Brighton Pediatric Early Warning Score predicts clinical deterioration in hospitalized children with cancer

Pediatric cancer inpatients are vulnerable to rapid physiological deterioration, often preceded by subtle clinical changes that may be overlooked in busy oncology wards, particularly in resource-limited settings. Pediatric Early Warning Scores (PEWS) are designed to facilitate early recognition of clinical instability; however, prospective evidence evaluating their utility in pediatric oncology populations from low- and middle-income countries remains scarce. This study assessed the predictive performance of modified Brighton PEWS and its dynamic change (Delta PEWS) in identifying adverse clinical outcomes among hospitalized pediatric cancer patients. This prospective observational study was conducted over 18 months in a tertiary pediatric oncology unit in India, in children aged 1–12 years with confirmed malignancy. Modified Brighton PEWS was recorded within two hours of admission and subsequently monitored every 24 h. Diagnostic performance was evaluated using receiver operating characteristic analysis and multivariable logistic regression. Among 126 enrolled children, 17 (13.5%) developed adverse clinical events, and 9 (7.1%) died during hospitalization. Admission PEWS demonstrated excellent discrimination for predicting adverse events (AUROC 0.953), with optimal performance at a threshold ≥ 3. Delta PEWS showed strong predictive ability for deterioration (AUROC 0.854) and near-perfect discrimination for mortality prediction (AUROC 0.988). Both admission PEWS and Delta PEWS independently predicted adverse outcomes. Sepsis and chemotherapy-related complications were identified as major clinical risk factors for deterioration. Modified Brighton PEWS and serial score trends are practical bedside tools for early detection of clinical deterioration in pediatric oncology patients, enabling timely escalation of care, particularly in resource-limited settings.