Vol. 2, Issue 2, Part A (2025)

Background: Rapid and accurate clinical decision-making in paediatric emergency nursing is critical to patient outcomes. Artificial intelligence (AI)-assisted decision support tools offer the potential to enhance triage precision, accelerate interventions, and optimize clinical workflows. This study evaluated the impact of AI-assisted decision support on nursing performance and patient care in a paediatric emergency setting.
Methods: A prospective quasi-experimental study was conducted in the paediatric emergency department of a tertiary care hospital over six months. A total of 120 patients were enrolled and randomly assigned to either an AI-assisted triage and management group or a standard care group. The AI system provided real-time clinical risk stratification based on vital signs and patient characteristics. Primary outcomes included triage accuracy and time to first critical intervention. Secondary outcomes included early deterioration prediction performance, ICU transfer rates, and nurse perceptions of usability. Data were analysed using chi-square tests, t-tests, Mann-Whitney U tests, and logistic regression, with p < 0.05 considered statistically significant.
Results: AI-assisted decision support significantly improved triage accuracy (88.3% vs 72.0%; p=0.010) and reduced median time to critical intervention (18 vs 27 minutes; p<0.001). The AUROC for early deterioration prediction increased from 0.74 to 0.86 (p=0.004). Although ICU transfer and 24-hour return rates were lower in the AI group, differences were not statistically significant. Nurses reported higher levels of decision confidence, trust, and system usability in the AI-assisted group (mean SUS score: 78.2 vs 68.5; p<0.001). Adjusted analyses confirmed independent associations between AI assistance and improved triage accuracy and intervention timeliness.
Conclusion: AI-assisted decision support significantly enhances paediatric emergency nursing performance by improving triage precision and accelerating critical interventions, while also supporting nurse confidence and workflow efficiency. These findings underscore the value of integrating AI tools into clinical practice, supported by appropriate training, governance, and continuous performance monitoring to ensure safe and effective implementation.