Guy’s and St Thomas Hospital lies opposite the Houses of Parliament has one of the busiest Accident and Emergency departments in London, used by hundreds of patients 24 hours a day.
The 4 hour A&E maximum wait target was a challenge. It’s mathematically impossible to give each patient as much time as they need and respond to a varying demand with fixed resources and hit a waiting time target so there was no clear idea how to solve the problem. There was also no clarity about when and why the problem occurred. The Footman Walker software system used was great for recording patient details but not so good at giving the bigger picture.
Exactly the sort of challenge we love.
We analysed tens of thousands of records and came up with the interesting result that the main pressure was not late nights at weekends. That was just people’s perception. The biggest problems were mornings, especially Mondays and one of the causes was a reactive staffing rota in which clinicians present was dictated more by numbers of patients waiting than by patients arriving. Analysis and charting dispelled a lot of the myths, misconceptions and interpretation bias which anyone too close to a problem is unavoidably subject to.
We created a mathematical simulation of the demand and the throughput which showed that re-scheduling of when doctors and nurses worked solved the problem and reduced stress in the system.
It’s possible to create highly complex stochastic dynamic simulations of Accident and Emergency patients flows but that’s time-consuming and tends to create distractions as exact flows are identified and all possible routes are considered. You tend to end up with something worthy of a PhD thesis which only an expert can apply properly. We used the Pareto principle, arguing that if you can solve 80% of the problem with 20% of the work and time, then you have a good chance of hitting the target and the remaining small problems become more apparent and readily understood and are easier to tackle later.
The model itself consisted of a dashboard containing a historic patient arrival patterns which the user could use as a basis and modify to simulate any condition plus the facility to change any of a range of clinician grades, their respective capacities.
“David Robinson of Richardson Research gave us the engineer’s perspective and helped us to see the real problems then made us an easy to use software tool to help us achieve the 4 hour wait target. All A&E departments should be using this”
Dr Chris Lacy,
Consultant in Emergency Medicine and Major Trauma,
King’s College Hospital