An application of agent-based simulation to the management of hospital-acquired infection

Abstract

Hospital patients who are colonised with methicillin-resistant Staphylococcus aureus (MRSA), may transmit the bacteria to other patients. An agent-based simulation is designed to determine how the problem might be managed and the risk of transmission reduced. Most MRSA modelling studies have applied mathematical compartmental models or Monte Carlo simulations. In the agent-based model, each patient is identified on admission as being colonised or not, has a projected length of stay and may be more or less susceptible to colonisation. Patient states represent colonisation, detection, treatment, and location within the ward. MRSA transmission takes place between pairs of individuals in successive time slices. Various interventions designed to reduce MRSA transmission are embedded in the model including: admission and repeat screening tests, shorter test turnaround time, isolation, and decolonisation treatment. These interventions can be systematically evaluated by model experimentation.

Keywords:

• agent-based simulation
• epidemiological modelling
• hospital-acquired infections
• MRSA

Acknowledgements

The authors thank the following individuals who have made a significant contribution to the project and to the collection and/or analysis of the data. These are: Ala Szczepura, Charlotte Price and Nigel Stallard of Warwick Medical School, University of Warwick, Andrew Bradbury and Savita Gossain of Heartlands Hospital, Heart of England NHS Foundation Trust, Birmingham. The research has been supported by the Department of Health. The views expressed are those of the authors and do not necessarily reflect the views of the Department of Health. The research project has also been supported by Beckton, Dickinson and Company.