Cloud helps improve occupancy planning

MISSISSAUGA, ONT. – Since the start of the pandemic, it’s not uncommon for Ontario hospitals to operate at occupancy rates above 100-percent. That’s why finding a way to match bed supply with patient demand is key to improving overall patient care, safety, and operational costs. The solution to this challenge may lie with ‘bed occupancy forecasting’, a process that involves analysis of admission and discharge data – often done manually.

At Mississauga-based Trillium Health Partners (THP), Canada’s largest community-based hospital system, their research and innovation arm, the Institute for Better Health (IBH), is using cloud technology to improve the hospital’s two-to-seven-day capacity forecasting by analyzing bed-use data to identify patterns and anticipate capacity requirements more accurately.

“Our analytics lab developed a prediction algorithm to help improve hospital capacity forecasting and better match bed supply with patient demand,” said Jonathan Ranisau, project lead, Data Discovery, The Institute for Better Health (IBH).

THP created a data lake – a solution that imports data from multiple sources into a centralized repository, then automatically extracts the data and identifies its schema to create a data catalog. From there, the data team can identify any gaps in the data. Centralizing the data in the data lake helps simplify collaboration between researchers, data scientists and any other members of the team.

THP is one of the first healthcare organizations in Canada to apply cloud computing and analytics to develop a predictive occupancy planning model. The model will help hospital administrators predict demand for hospital beds and plan for potential instances where surge capacity is required.

The model was developed in collaboration with the University of British Columbia (UBC) Community Health and Wellbeing Cloud Innovation Centre (CIC), a public-private partnership between UBC and Amazon Web Services (AWS).

The UBC CIC, powered by AWS, supports student teams tackling real-world issues and challenges, where application of novel and emerging technology or computing approaches are essential.

“This is the type of project that the UBC CIC set out to do when we launched in 2020 and it’s always exciting to see our students and staff build solutions that have societal impact,” said Coral Kennett, Head of Education & Digital Innovation Lead, AWS Canada. “While this solution is purpose-built for THP, it can be adopted by other organizations to solve similar challenges.”

Funded by a donation through the Trillium Health Partners Foundation, the project represents a major step forward for THP’s technological and data capabilities. Moving forward, the goal is to enable the algorithm to accurately predict bed occupancy ‘daily’ rather than the traditional 2-7-day cycle. Once refined and fully rolled out, THP will make the model available to other hospitals, to continue to help increase operational efficiency and patient care across Ontario’s healthcare system.

Launched in January 2020, UBC CIC’s mission is to solve real-world challenges that benefit British Columbia, Canada and communities beyond, by engaging and collaborating with stakeholders in community health and wellbeing innovation challenges.

The UBC CIC is a first-in-Canada public-private partnership between UBC and Amazon Web Services (AWS), and one of only twelve currently in operation in the world. For more about UBC CIC, visit cic.ubc.ca.