Perhaps no other event in modern history has tested the nation's healthcare system like the COVID-19 pandemic.
With advanced predictive analytics and intelligence tools powered by artificial intelligence, Aurora-based University of Colorado Health has optimized capacity demands on beds, supplies and staff.
During a virtual session sponsored by LeanTaaS at Becker's 6th Annual Health IT + Revenue Cycle Conference, Steve Hess, CIO of UCHealth, and Mohan Giridharadas, founder and CEO of LeanTaaS, explained how LeanTaaS helped UCHealth optimize capacity.
Four learnings:
1. EHRs aren't designed for predicting tomorrow's needs. Enterprise EHR systems collect and show data that is important to making capacity optimization decisions, but EHRs do not have predictive capabilities, nor were they designed for them. The data in EHRs needs to be intepreted by tools that do have predictive capabilities. "The key to our response was … grabbing data in real time and analyzing it to predict what's going to happen tomorrow, next week, next month," Mr. Hess said. These data-driven insights helped UCHealth guide utilization practices for beds, staffing, supplies and equipment across its 12 hospitals.
2. While capacity optimization hasn't occurred in healthcare, it has in other industries. In healthcare, scheduling often involves using a calendar to schedule a patient for an open time, which does not optimize the capacity of time, rooms, or staff.. While actual capacity optimization is a tough mathematical problem, it is not a new concept; it has been used for decades in other industries. Airlines, services like UPS and ride-hailing services use advanced mathematics, data science, and predictive and prescriptive analytics to manage dynamic logistics to maximize service and business efficiency.
3. Matching and linking are pillars of capacity optimization. Mr. Giridharadas explained the two key concepts behind capacity optimization: matching supply and demand, and linking together various independent services to perform a larger service. For example, an operating room requires both correct staffing and several smaller services to perform a case.
Mr. Giridharadas said airlines have solved both of these problems, in a way that is instructive to perioperative facilities. First, an airplane seat, like an hour in an OR, is a perishable asset: when it goes unused, it is gone forever. Airlines often overbook, then solicit volunteers to take later flights, thereby matching supply and demand. Then through a well-designed choreography of linked services and actions that happen quickly when a plane lands — unloading the passengers, crew and baggage; servicing, cleaning and inspecting the plane and reloading a new crew, passengers and baggage — that plane is back in the air as fast as possible.
Similar advanced matching and linking is needed in healthcare. "Imagine if a hospital could have five times the number of patients flowing in and out of the hospital every day, with low wait times and high quality of care," Mr. Giridharadas said. "That's the magic of solving the matching problem and the linking problem correctly in a detailed, rigorous, mathematical manner."
4. Predictive analytics to optimize capacity of critical healthcare assets yields an immediate return on investment. UCHealth first integrated LeanTaaS’ predictive analytics capabilities in one of its infusion centers to help manage a surge of midday activity. The surge created a bottleneck that led to long patient wait times and nurse overtime. UCHealth sent aggregate EHR data to LeanTaaS. The resulting predictive analytics developed by LeanTaaS allowed UCHealth — using the same assets — to increase patient volume by 7 percent within 90 days and realize a 13-fold ROI. A subsequent project, to optimize OR time, resulted in a 4-fold ROI, Mr. Hess said.
To learn more about the event, click here.