Predicting hospital-onset Clostridium difficile using patient mobility data: A network approach

Area of Science:

• Infectious Disease Epidemiology
• Healthcare Network Analysis
• Hospital Infection Prevention

Background:

• Clostridium difficile infection (CDI) remains a significant healthcare-associated infection.
• Understanding transmission dynamics within hospitals is crucial for effective prevention strategies.
• Patient mobility between units is a potential driver of CDI spread.

Purpose of the Study:

• To investigate the link between unit-level CDI susceptibility and inpatient mobility patterns.
• To develop and validate a novel predictive measure for CDI transmission, termed contagion centrality (CC).

Main Methods:

• A retrospective cohort study analyzed 2 years of electronic health record data from a large hospital (72,636 admissions).
• A mobility network was built to represent patient transfers between hospital units.
• Network centrality measures and daily unit-wide CDI susceptibility scores were calculated and compared.

Main Results:

• Closeness centrality, reflecting incoming patient mobility, was significantly associated with unit CDI susceptibility (P < .05).
• Contagion centrality (CC), incorporating transfer rates, unit susceptibility, and current infections, significantly predicted hospital-onset CDI (P < .05).
• CC effectively captures transmission risks associated with patient movement.

Conclusions:

• Inpatient mobility, particularly incoming transfers, plays a key role in unit-level CDI susceptibility.
• Contagion centrality offers a valuable tool for real-time risk assessment and prediction of hospital-onset CDI.
• Visualizing mobility and infection data can proactively identify at-risk units, enabling targeted prevention efforts.