Reducing A&E four hour breaches in Cornwall

South West PeninsulaGeneric Health Relevance, Injuries and Accidents
Start Date: 18 Feb 2016


Royal Cornwall Hospital (RCHT) A&E Department has seen worsening performance against the four hour Emergency Department Standard and a significant growth in recorded monthly attendances since October 2015, compared to the same months in 2014. The causes of poor performance are multi-factorial with demand, front door flow, internal hospital processes and back-door flow from the hospital all playing a role.

In the majority of cases, the official reason for an ED breach is lack of beds, but the truth is often more complex, with increased demand from various A&E streams reducing capacity to deal with other streams.

The Cornwall Systems Resilience Group partners RCHT and NHS Kernow would like to identify an optimum model of care that makes best use of existing resources at the front door of Treliske Hospital, including resources in the following areas:

  • Acute GP Service
  • Emergency Department (ED)
  • Clinical Decision Unit (CDU)
  • Ambulatory Care
  • Resus
  • Medical Admissions Unit

Project aims and activity

The project sought to determine the optimum categorisation of patient streams and what capacity (equipment, people, processes) is required to manage the variable demand of each stream to sustainably deliver the 95% Emergency Department standard.

We worked with key contacts at the A&E department to develop a process map of the A&E system. This will help us to understand the system, and move us towards a design for the simulation model.

We built an “infinite capacity” Discrete Event Simulation model of an idealised A&E at RCHT. Such models assume that there is infinite capacity within the system, and can then be tested with various numbers of resources to inform trade-off decisions. In this case, varying numbers of cubicles available for each stream were tested to predict the probability of four hour breaches if this configuration were implemented.


The optimum configuration of resources allocated to A & E streams is :

  • Between midnight and 9am to be able to triage one patient at a time, see six Majors patients simultaneously, see seven Minors patients simultaneously, and see one patient a time in the Paeds, Resus and “Other” streams
  • Between 9am and midnight to be able to triage two patients simultaneously, see ten Majors patients simultaneously, see ten Minors patients simultanouesly, see two Paeds patients simultaneously and see one patient at a time in the Resus and “Other” streams

The model predicts that the optimum configuration described above would lead to 100% of patients being triaged within 15 minutes, 96% of patients being seen by a doctor within one hour, and 88% of patients leaving within four hours.  The average time in the department per patient is predicted to be 126 minutes.

No amount of resource reconfiguration amongst triage and the streams (front and middle of A & E) will result in the four hour target being achieved.  This result emerges because there is a significant bottleneck in getting patients out of A & E once a final decision has been made.  For example, the average time from a final decision being made for a Majors patient is 113 minutes.

The results of this study were presented to the collaborators in Autumn 2016, from there we will get a better sense of specifically how these results will inform the reconfiguration of A & E services and report this in due course.

Dr Daniel Chalk