Submitted by Carrie Stefanson, senior consultant, Public Affairs

To provide the right care at the right time, we need the right number of staff and medical staff to meet patient demand.

Fraser Health has partnered with Deloitte Canada to develop and launch scheduling tools that leverage artificial intelligence (AI) and machine-learning solutions to accurately forecast the waves in demand we see in the health system.

Thanks to a $1.5 million Scale AI award, Fraser Health and the Deloitte project team are developing three artificial intelligence models. Two will leverage patient analytics from various electronic medical records to forecast volume demands coming to the emergency and medicine departments and assist leadership in pre-determining broader staffing requirements. A third will take the forecasted patient data and match it with business rules, such as patient-physician ratios, to create a baseline schedule with advanced analytic capabilities to generate multiple scheduling scenarios.

Emergency department long-term schedule optimization tool – Eagle Ridge Hospital and Burnaby Hospital

A new tool to forecast trends in patients requiring emergency care is expected to launch in the coming months at Eagle Ridge Hospital and Burnaby Hospital. By better understanding when patients arrive at emergency departments, physician scheduling can be adjusted accordingly.

While current scheduling methods continue to be used, the tool will help physicians predict patient volumes and trends when preparing their manual schedules. It will also help leaders to run different scheduling scenarios and digitally test out what the patient flow could look like.

The tool was piloted in July and the response was positive.

“Eventually, we will see the tool becoming more sophisticated and able to integrate factors such as respiratory and influenza season, outbreaks, air quality levels, and other local and regional events like concerts and festivals to more accurately predict how many patients will need care on any given day and match that demand with physician coverage,” says Sheazin Premji, executive director, Centre for Advanced Analytics, Data Science and Innovation, Fraser Health.

Hospitalist department short-term daily forecast tool – Burnaby Hospital

Hospitalists in Burnaby are helping to inform an AI-driven surge prediction model to enable proactive decision-making about calling in additional staff based on forecasted physician workload.

The tool helps predict the number of patients requiring hospitalist care over a seven-day period. It then proactively adjusts staffing levels in the event of a surge. During validation testing, the model predicted surges with 91 per cent accuracy for the next day and 81 per cent accuracy up to seven days in advance.

“This innovative project is an example of how clinicians can use AI in a practical way to benefit themselves and their patients,” says Dr. Neil Barclay, emergency physician, Eagle Ridge Hospital. “Being part of this project from the ground up will help us better meet our patients’ needs and provide us with a more sustainable health care system in the long run.”

The Fraser Health/Deloitte collaboration ‘AI-Driven Physician Scheduling Solution and Workflow Optimization’ is one of nine Canadian health projects to receive $21 million in funding to pioneer the deployment of AI solutions. The initiative promotes collaboration between hospitals and AI product and solutions providers to innovate further and accelerate the deployment of AI in Canadian health care.

“Enhancing the patient experience and improving access to care is at the heart of digital transformation in Fraser Health,” says Jennifer MacGregor, vice president of Digital Patient and Provider Experience. “We are excited to be co-developing these AI tools. These partnerships are essential to drive innovation in our health system to solve some of our most challenging problems.”


comments powered by Disqus
X

Cookies help us improve your website experience.
By using our website, you agree to our use of cookies.

Confirm
X

Cookies help us improve your website experience.
By using our website, you agree to our use of cookies.

Confirm