Principal Investigator: Dr Carlos Lamas-Fernandez, Associate Professor in Business Analytics/ Management Science in Southampton Business School / Faculty of Social Sciences, University of Southampton.

Team:

Professor Christine Currie, School of Mathematics, Faculty of Social Sciences, University of Southampton.

Dr Dan Burns, Innovation Centre, Electronics & Computer Science, University of Southampton.

Dr Chris Duckworth, Innovation Centre, Electronics & Computer Science, University of Southampton.

Professor Michael Boniface, Innovation Centre, Electronics & Computer Science, University of Southampton. Professor Peter Griffiths, School of Health Sciences, University of Southampton.

Dr Mark Wright, University Hospital Southampton NHS Foundation Trust.

Starts: 1 April, 2024

Ends: 30 September 2024

Summary

Hospitals in the UK are in crisis with high levels of occupancy. The percentage of occupancy in England during July-September was 88%, and in UHS it reached 92.2% . These levels exceed the safety threshold for hospital occupancy, which sits at around 85%. Together with difficulties to ensure a smooth patient flow across the hospital, this results in adverse effects for patients: elongated hospital stays, increasing the backlog of elective procedures, increasing delays in ambulance handovers and increased mortality.

In practice, hospitals try and control high occupancy levels by certain interventions, such as dedicated discharge teams, re-scheduling or cancelling elective procedures or repurposing hospital wards. These measures, however, are reactive, i.e. when the occupancy is already reaching unsafe levels, rather than proactive, that is, when anticipating a high occupancy in the near future. Further, it is not clear whether occupancy levels have an effect on treatment and discharge times, but from frontline clinicians at UHS, there is the hypothesis that higher occupancy could make them longer (as clinicians are busier prioritising the sick over the well patients who could go home), compounding the occupancy issues. Higher occupancy also decreases likelihood of patients being in the optimal location. A related research project (PROCED) has shown early evidence that frequent ward/team changes increase delays in patient discharge.

The aim of this project is to investigate the feasibility and build the foundations of a simulation model that can predict accurately future, short-term, hospital bed occupancy to inform interventions. The project will have a special focus on investigating the feasibility of a model to be tailored to use in practice as a “Digital Twin” (DT), which can anticipate hospital occupancy under different scenarios, some of which can reflect proposed interventions.