Optimising antibiotic switching via forecasting of patient physiology

2026-03-09

AI Summary

This study addresses the clinical challenge that a substantial number of eligible patients fail to transition promptly from intravenous to oral antibiotics, leading to prolonged hospital stays, elevated infection risks, and increased healthcare costs. To tackle this issue, the authors propose a method based on Neural Processes that performs probabilistic modeling of multivariate time-series vital signs. Rather than mimicking historical clinician decisions, the approach dynamically assesses patient eligibility for route conversion by integrating explicit clinical guideline rules and generates an interpretable priority ranking for physician review. The framework preserves clinician autonomy in final decision-making, supports seamless updates to guidelines without retraining, and demonstrates 2.2–3.2 times higher efficiency than random selection in identifying convertible patients across two real-world datasets—MIMIC-IV and UCLH.

Technology Category

• Natural Language Processing: Safety and Robustness
• Planning, Routing, and Scheduling: Planning with Language Models
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Data transparency and provenance

Abstract

Timely transition from intravenous (IV) to oral antibiotic therapy shortens hospital stays, reduces catheter-related infections, and lowers healthcare costs, yet one in five patients in England remain on IV antibiotics despite meeting switching criteria. Clinical decision support systems can improve switching rates, but approaches that learn from historical decisions reproduce the delays and inconsistencies of routine practice. We propose using neural processes to model vital sign trajectories probabilistically, predicting switch-readiness by comparing forecasts against clinical guidelines rather than learning from past actions, and ranking patients to prioritise clinical review. The design yields interpretable outputs, adapts to updated guidelines without retraining, and preserves clinical judgement. Validated on MIMIC-IV (US intensive care, 6,333 encounters) and UCLH (a large urban academic UK hospital group, 10,584 encounters), the system selects 2.2-3.2$\times$ more relevant patients than random. Our results demonstrate that forecasting patient physiology offers a principled foundation for decision support in antibiotic stewardship.

Problem

• antibiotic switching
• clinical decision support
• patient physiology forecasting
• antibiotic stewardship
• IV-to-oral transition

Innovation

• neural processes
• antibiotic stewardship
• physiological forecasting
• clinical decision support
• IV-to-oral switch