Humber River Hospital won’t open the doors to its new, state-of-the-art acute care facility in northwest Toronto until October of this year, but the surgery department already knows the operating rooms will be running smoothly.
That’s because it has pre-emptively identified possible issues with the OR block schedule, and solved them by redesigning the schedule using GE Healthcare’s Hospital of the Future simulation suite for optimizing daily operations. Essentially, the hospital turned a difficult paper exercise into an automated scenario that accurately predicts patient flow. “What we have is a model that looks at the OR block schedule and predicts what our bed utilization is going to be, and what our OR utilization is going to be,” said Dr. John Hagen, Chief of Surgery at Humber River Hospital. “It sounds pretty straightforward, but it’s actually quite complex.”
With the launch of its highly anticipated digital hospital on Wilson Ave., Humber River Hospital will close two existing acute care locations, and will convert a third into an Ambulatory Care Centre that will also have an Urgent Care Centre.
This restructuring will require the hospital to consolidate two fully operational surgery programs into one program at the new facility.
Over several months, a steering committee of surgeons, anaesthetists and administrators led by Scott Jarrett, executive vice-president, patient services, worked through many iterations of a draft OR schedule and arrived at one that seemed reasonable on paper. They then engaged GE Healthcare’s simulation expertise to test it.
Originally developed to assist in hospital planning for new builds, the Hospital of the Future simulation suite uses different categories of data inputs to generate precise computer models of how a hospital’s operations will perform as a system. It can then test the impact of changes to those inputs, based on probabilities and with a high degree of accuracy.
In the case of Humber River Hospital, the simulated model was built to look at operational capacity (operating hours and number of beds) as well as inpatient and outpatient flow, including when and where patients arrive, how they move through the system, whether they go to surgical day care following the post anaesthetic care unit (PACU), or whether they’re admitted, and their expected length of stay at each stage of their encounter.
GE Healthcare also developed profiles for each surgeon, based on their OR time, and number and type of procedures performed.
“We built the model based on their historical data, validated it within the context of the schedule they had built and then merged the two surgical programs,” explained Tamas Fixler, a senior consultant at GE Healthcare Partners. “We said, ‘If you go with the schedule as designed, this is what you’re going to see.’”
Some of the results were positive. For example, GE Healthcare was able to show with certainty that the planned 37-bed surgical day care unit and 33-bed PACU are more than adequate to handle current patient volumes with room for growth. But the model also indicated the proposed OR schedule would soon encounter issues on Wednesdays and Thursdays, leading to potential inpatient bed shortages and some cancelled surgeries.
“What we wanted to do was identify those surgeons who were really driving that peak in patient census,” explained Fixler, adding that the idea was to perform a series of “strategic swaps” between surgeon blocks in order to flatten the curve.
GE Healthcare’s simulation model identified that the bottleneck was largely due to inpatient procedures. By detecting a handful of surgeons scheduled to operate on Tuesdays and Wednesdays who had a heavy load of inpatients occupying beds, GE Healthcare was able to propose a new schedule. These surgeons and their procedures were moved earlier in the week by switching their OR time with surgeons who perform largely outpatient procedures. That eliminated the flow disruption completely by substantially reducing census variability.
Multiple scenarios were tested, and those surgeons affected had to agree to the change, but in the end their support was unanimous. When the final iteration was presented to the entire group, everyone was listening intently, Jarrett said.
“We wanted to be able to come forward to the various surgical divisions and say, ‘These x number of surgeons really have the highest impact on downstream bed utilization,’” said Jarrett. “Some of the decisions we could have made with a bit of a gut feel, but then you open yourself up to criticism and you can get into some real pointed discussions. Whereas here, the data really speaks for itself.”
From his vantage point as Chief of Surgery, Dr. Hagen was interested to see that elective surgeries were responsible for the crunch in the schedule. Intuitively, he always felt emergency admissions were responsible for stretching bed capacity. However, the GE Healthcare model clearly demonstrated that emergency room admissions were not the cause.
“The ironic thing is that the number of admissions through emergency rooms to inpatient beds is relatively constant,” said Dr. Hagen. “What is unpredictable, and the thing that we are able to control, is the elective inpatient surgery. This modelling allows us to do that.”
Humber River Hospital plans to subscribe to GE Healthcare’s simulation suite as an ongoing service. The intent is to refresh the model every six months, uploading new data to reflect changes that have taken place, such as the addition of new surgeons or changes in existing surgeons’ practice patterns, or shifts in certain inpatient surgeries to an outpatient basis.
Hospital staff will also be trained on how to run their own experiments using the model (e.g. to understand the system level impact of new surgeon recruitment or various operational changes) so that the OR schedule can continue to align to the hospital’s needs.
“If we had just combined our present schedules, there would have been problems – no question,” added Dr. Hagen. “All it meant was looking at it, analyzing it and moving a few high impact surgeons to a different day, and the whole thing smoothed out. Predictably, it will be better.”