Optimising hospital flow

Clockwise from back left: Professor Jerzy Filar; Mr Paul Hakendorf; Professor David Ben-Tovim; and Dr Shaowen Qin
Clockwise from back left: Professor Jerzy Filar; Mr Paul Hakendorf; Professor David Ben-Tovim; and Dr Shaowen Qin

Researchers from Flinders University are using information technology, software engineering and mathematics in response to the challenge of relieving patient congestion at Flinders Medical Centre (FMC).

Using patient journey data from 2003-2011, the team of academics – together with FMC doctors and  senior staff – are applying computer science, engineering and maths-based concepts to find inpatient flow patterns from the time of admission to the time of discharge.

The study aims to establish correlations between quality of care and patient types, ward types, waiting times and length of stay, thereby identifying bottlenecks in hospital flow that can lead to overcrowding.

Flinders senior lecturer in software engineering, Dr Shaowen Qin, said the research would apply “process modelling, simulation and optimisation” techniques to show patient flow trends across time scales, days, weeks and months under current and speculative conditions, with the wider aim of streamlining patient flow and reducing bottlenecks.

“In the context of this study, process mining will be used to extract information from the hospital’s electronic tracking database to learn more about the various paths taken by inpatients moving from the emergency department to other wards, identifying the most frequent paths taken,” Dr Qin said.

“The study will also examine the impact of inliers and outliers – patients who spend time in a ward that is not the speciality ward for their condition – on the total length of stay,” she said.

“Due to the nature of the hospital, the capacity is quite limited so from time to time a patient with a cardiac condition, for example, might be allocated to the respiratory ward which means they’d be an outlier.

“Clinicians generally have the perception that outlier patients end up staying longer in hospital compared to inliers because they’re being treated in a ward that’s not the best ward for their condition but this remains a hypothesis that may yet be shown to be too simplistic.”

Dr Qin said the study also aimed to establish models for predicting ward capacity for particular days of the week, with the findings used to help inform hospital staff on how to best manage patient flow and cope during peak periods.

“There are times when congestion is inevitable but we want to find out how we can recover from that as soon as possible,” she said.

“The advantage of this research is that it’s an integrated effort between medical experts at FMC and mathematicians and process specialists from the University who are working together to manage a problem that occurs in hospitals worldwide.”

Posted in
College of Science and Engineering Corporate Engage News Research School of Computer Science, Engineering and Mathematics Uncategorized

Leave a Reply

Your email address will not be published. Required fields are marked *