Emergency department (ED) waiting times are a significant predictor of the patient experience.
Simple prediction methods, such as rolling average, are used by hospitals in Australia to predict waiting time for patients. Although this approach is inexpensive to implement, the forecasts have limited accuracy and consequently most Australian hospital EDs do not report expected waiting times to the public.
A solution that is capable of sourcing data from ED information systems and feed it into prediction models to generate waiting time forecasts would bring practical benefits for staff and patients. There is also potential to assist clinicians and nurses to estimate demand for care and calibrate workflow.
For patients, the knowledge may reduce anxiety associated with uncertainty about the waiting time and reduce the number of patients who leave before treatment.
This project aims to use advanced statistical models and machine-learning algorithms to capture dynamic fluctuations in waiting time, to implement and validate the prediction performance of these models. The project will also build ED research capacity by educating staff on forecast modelling and data management techniques.
The research team has made a significant breakthrough by developing a digital health solution that can accurately predict Emergency Department (ED) waiting times for low acuity patients. This solution is based on a state-of-the-art machine-learning algorithm that utilises a large sample of ED presentations to the Princess Alexandra Hospital over the last five years. The team analysed more than 320,000 presentations to train and test the algorithm, resulting in a highly accurate predictor of ED waiting times.
Compared to the commonly used rolling average estimator in Australian hospitals, this algorithm has shown significant improvements in accuracy. Furthermore, there is a potential to seamlessly integrate with existing hospital data systems to provide accurate waiting time information to patients, allowing them to better plan their visit and reducing their overall anxiety.
The research team is now seeking further funding and collaborations to develop a patient interface, which could be in the form of an app, website, or onsite ED information screen, to communicate waiting time information to patients. The team is also engaging with consumers to gather feedback and insights to inform the development of future patient communication initiatives.
Overall, this digital health solution has the potential to significantly improve the patient experience in the ED and increase the efficiency of ED operations. The research team's work marks an important milestone in the development of predictive healthcare technologies, and their findings could have far-reaching implications for healthcare systems.
- UQ HaBS Early Career Researcher Accelerator Award, 2022: $10,000
- Trinh, K., Staib, A., Pak, A., 2023. Forecasting emergency department waiting time using a state space representation. Statistics in Medicine. https://doi.org/10.1002/sim.9870
- Pak, A., Should I stay or should I go? How can Emergency Department waiting time forecasts benefit patients? (poster presentation). Health Consumers Queensland Annual Forum 2022, Brisbane. 13-14th October 2022.
- Staib, A., ED Waiting Time Predictions (oral presentation). Australian College of Emergency Medicine, 17th Annual Queensland Autumn Symposium, Brisbane. 26th May 2022.
- 4BC (plus 7 regional Queensland stations), 5/11/20, Patients with minor injuries could soon see which hospital would be able to treat them... Audience 25,000
- WIN News, 5/11/20, 90sec interview with CI Pak
- ABC North West Radio, 5/11/20, interview with CI Pak
- Metro South Health website story, 18/8/2022, ED secures funding support from Emergency Medicine Foundation in competitive research grant round
Dr Andrew Staib
Dr Anton Pak
Dr Kelly Trinh
Prof Navonil Mustafee
Dr Rob Eley
Prof Brenda Gannon
Dr Mohammad Jahanbakht