This is our world. Welcome to our website

Improving Patient Flows Through A&E Departments

We have made important breakthroughs in understanding how to improve the processing of patients through A&E departments. Our mapping of patient flows in Hillingdon hospital in north London revealed significant hidden patterns of discharge/admission. That allowed us to create a model that correctly predicts the effect on patient throughput of changing certain practices. The model also shows how and why a strategy that works well for one hospital may not be successful elsewhere.

As a result of this study, we helped Hillingdon hospital to rethink its strategies. It has now switched to a system which focuses more on drawing patients from A&E, rather than on pushing them out from that department into the hospital.

More broadly, this work has helped us to design and seek funding for a proposal to develop a generic A&E simulation model adaptable for particular hospitals.

A simulation Model of A and E

Simulation model of the 'minors' area in the A&E department

The Project

Hillingdon hospital wanted to improve its processing of patients through A&E, mindful of the - since abandoned - 4-hour government target for admission or discharge. Clinicians wanted to know whether use of an electronic whiteboard, effective elsewhere, would help.

When we mapped patient flows, we noticed a sharp rise in the number of patients discharged or admitted from the A&E department just before the 4-hour deadline. This happens in many other UK hospitals. It is partly because of re-prioritising of patients close to the deadline, a practice not captured in previous research studies.

So, for the first time - as far as we know - MATCH was able to build a simulation model that included the re-prioritisation strategy and accurately predicts its outcomes. Using this model, we were able to show that a fast-tracking service, successfully adopted in some other hospitals, would not improve service times at Hillingdon.

Unfortunately, the model took longer than expected to build. By that time, Hillingdon had already decided to try using an Emergency Nurse Practitioner (ENP) system to fast-track minor cases. They found that the system did not improve patient throughput, producing very similar results to those predicted by our model.

This experience showed clearly how our model can accurately predict when a strategy, though successful in one hospital, may not work in another. Such modelling could be very useful in predicting the appropriateness of a general roll-out of practice that may have worked well in particular instances. It would prevent the disruption and expense caused by implementing an ineffective strategy.

As a result of this work, we have helped Hillingdon reflect on their strategies, which are now more focussed on the bed management system "pulling" patients through A & E.

Next Steps

We are seeking follow-on funding with University Hospital Coventry and Warwickshire to develop and generalise our modelling so it can be applied to other A&E departments in the UK.

Further information

Dr Julie Eatock

Julie.Eatock@brunel.ac.uk 01895 265408