Dynamic contagion potential framework for optimizing infection control in healthcare

Alexandra Fedrigo¹Mohamad Nassar²Jennifer Bail¹Antonia Bates-Ford¹Satyaki Roy^1*

• ¹University of Alabama in Huntsville, Huntsville, Alabama, United States
• ²University of New Haven, West Haven, Connecticut, United States

Hospital-acquired infections (HAIs) caused by bacterial and viral pathogens continue to affect millions of individuals annually, posing a challenge to healthcare systems. Traditional infection control strategies often fall short due to their inability to dynamically assess real-time spatial and movement data within healthcare facilities. To address this gap, this study leverages contagion potential (CP), a metric that quantifies infection risk based on individual characteristics and behavior, to propose a framework for minimizing the incidence of HAIs. CP accounts for the infection susceptibility and transmissibility of individuals, incorporating their movement patterns and interactions across units within a healthcare facility. Concentrating on mobility-driven and contact-mediated HAIs, the framework integrates approximate location data, modeling the infection risk landscape without requiring precise tracking. Through continuous learning, the CP parameters are inferred and refined over time, enabling accurate assessments of infection risk at both the individual and unit levels. This framework also includes CP-based optimization, which represents an early effort for patient-to-unit assignments by jointly optimizing contagion minimization and meeting the clinical and logistical demands. The efficacy of the framework is validated through experiments, both on individual modules and in an integrated evaluation, where mobility trends are generated to mimic homogeneous and heterogeneous mixing and social contact, while the infection rates follow realistic trends observed within hospitals. The results show that leveraging CP enhances infection control efforts, optimizes healthcare resource allocation, and improves patient safety. Overall, this dynamic, data-driven approach offers a strategy to combat HAIs, contributing to improved healthcare environments and patient outcomes.