Challenge 4: Reducing Emergency Department wait times

Challenge Statement: How might we reduce patient wait times and streamline processes in WA’s emergency departments?

Supporting Mentors

Sylvia Young – Department of Health

Kai Zhen – Department of Health

Wendy Wen – North Metropolitan Health Service

Background

 

  • For this challenge, we will be focusing on a specific patient cohort: those who have been hospitalised due to a stroke. The problems faced, and therefore the solution, will be applicable (and scalable) to other cohorts such as cancer, heart disease, lung disease, etc.
  • A stroke is a sudden blockage of blood to the brain or a bleed to the brain. This can lead to loss of movement of arms or legs, loss of speech, loss of vision, or loss of ability to comprehend information.
  • Most patients with stroke will be hospitalised for about a month, which includes time spent in a rehabilitation unit.
  • Rehabilitation is a process of improving a person’s ability through a multidisciplinary team of doctors, nurses, social workers, physiotherapist, occupational therapist, speech therapist, pharmacist, dietician and psychologist.

 

Potential Focus Questions 

This is a complex Challenge, and you may find it useful to consider the following questions when building out your prototype: 

  • Could we predict the waiting time (and/or the duration of episode) in Emergency Department (ED) using the synthetic data elements? Consider: building a machine learning model to determine the weights of independent variables such as triage, age, arrival time, etc., by using wait time (and/or Length of Stay) as the dependent variable.  
  • How could we validate the prediction accuracy of the machine learning model? 
  • Could we manage frequent Emergency Department users differently when they present at the Hospital?    
  • Could we develop an AI application to simulate the ED procedure using synthetic data?  
  • Could we visualize patient journey subject to the triage category and/or major diagnosis? 

Critical Concepts 

Emergency Departments and Hospitals are incredibly complex organisations. You may want to talk to your Challenge Mentors about the following ideas as you developing you prototype:   

  • Basic Emergency Department Triage Process.    
  • Current approach and theory around Patient Journey.    
  • Predictive machine learning models for comparison.    
  • Model validation methodology.  
  • Outcomes of the machine learning model to be potentially applied in clinical care.   
  • Structure of the WA Health Network – HSPs, Hospital locations and types, Private vs Public 

Supporting Data Sets 

The Emergency Department Data Collection (EDDC) has data elements for triage category, age group, ethnicity, and four datetime variables. This dataset has been synthetically generated by the Department of Health for the WA Health Hackathon 2024.   

The EDDC includes information about:    

  • Arrival time    
  • Clinical care commencement time    
  • Bed request time    
  • Discharge time      
  • Waiting time to clinical care to commence – (To be calculated by the Hackathon participants) 
  • Length of stay at ED   — (To be calculated by the Hackathon participants) 
  • Age group  
  • Ethnicity  
  • Gender  
  • Flags indicating whether a hospital is located in metro vs regional area 
  • Flags indicating whether an ED presentation is mental health related  
  • Flags indicating whether an ED presentation is an avoidable GP-type visit 

Potential Solution Pathways 

You are free to resolve this Challenge by developing your prototype in whatever means you may like. Our mentors, partners and organising team have thought of the following techniques as being viable methods to resolve the Challenge:   

  • Predictive models provide a forecast as to an ED episode.    
  • Simulation models of patient journey through an ED.    
  • Bayesian networks to identify factors affecting length of stay in an ED.    
  • Statistical models identifying categories of patients, and locations which cause congestion in ED.    
  • Visualisation of patient journey through an ED