Hospital administration has historically relied on highly manual processes, leading to delays in decision-making and patient care. But the rising use of analytics, machine learning, and AI is changing all that, a trend exemplified by healthcare consortium Kaiser Permanente, which set about overhauling the data operations of its 39 hospitals and more than 700 medical offices in the U.S in 2015.
“Waiting for an available bed during an emergency room visit or hospital admission, or waiting for discharge, is frustrating for patients and inefficient for hospital operations,” says Dick Daniels, executive vice president and CIO at Kaiser Permanente. “Many data points are necessary to make these processes efficient, including hospital census, bed demand, room cleaning and availability, pending discharges, and many others.”
That data came from multiple sources manually compiled into reports, a resource-intensive process that often resulted in information that was outdated and not actionable. Daniels says it also involved extensive communication via phone or text messages.
“KP’s clinical and operational leaders identified these challenges as key opportunities to improve care delivery and patient satisfaction while improving efficiency in daily operations,” Daniels says.
The result was the creation of the “Insight Driven” program, which has earned Kaiser Permanente a CIO 100 Award in IT Excellence.
The value of human-centered design
When the Insight Driven program’s focus group of clinical and operational leaders first came together, nurse managers retention and the need for a more holistic view of patient status and flow were among the top initial concerns.
As a result of those discussions, IT launched Operations Watch List (OWL) in 2018, a mobile app that provides a comprehensive, near real-time view of key hospital quality, safety, and throughput metrics (including hospital census, bed demand and availability, and patient discharges).
“The mobile app synthesizes the information to direct hospital leaders’ attention and action on issues that can cause bottlenecks in workflows and longer patient wait times,” Daniels says. “The app ensures that the best care delivery and patient experience are delivered in a seamless manner.”
The Insight Driven team used a human-centered design approach to identify opportunities to transform the information generated by Kaiser Permanente’s integrated model and electronic health record system into insight that could be leveraged for care delivery decisions. It also sought opportunities to introduce new capabilities such as predictive analytics to support decision-making at the front lines of care and improve patient flow management.
“One commonly identified use case was to use the tool for daily huddles at the hospital and unit levels,” Daniels says. “The tool now provides current and projected hospital census, bed demand, room cleaning and availability, pending discharges, and other throughput measures to help hospital leadership plan for admissions and discharges, identify opportunities to improve patient throughput, and anticipate and resolve potential bottlenecks, improving the member and staff experience.”
The team used the Scaled Agile Framework (SAFe) methodology to develop the app and for continuous improvement.
“This approach allows hospital operations leaders and front-line managers the ability to provide timely guidance and input on the needs and expectations of a solution on short intervals of development,” Daniels says.
The operational users, who gave the team guidance on initial requirements for OWL, continue to provide feedback today.
“By regularly incorporating feedback and often, the development team can continuously refine solutions to deliver the most important, relevant information and functionality to support ease of use and maximize value delivered,” Daniels says. “This tight-knight partnership between our clinical and operational leaders, the business, and our IT teams allow the Insight Driven program to continuously and more efficiently deliver value to our care delivery leaders on a regular basis.”
The SAFe approach
Kaiser Permanente’s early adoption of a cloud environment and the electronic health record was a key part of being able to develop OWL, Daniels says. Engagement and alignment with stakeholder groups and teams was a big challenge that led to the creation of a cross-functional team consisting of representatives of Kaiser Foundation Hospital Operations, Patient Care Services, Finance, The Permanente Medical Group Hospital Operations, Kaiser Permanente HealthConnect, Care Delivery Technology Services, Enterprise Architecture, and Digital Corporate Services and Solutions.
Success required changing the team’s culture from one of top-down leadership to a team-led approach, Daniels says.
“We also changed the mix of support roles over time while getting everyone trained on a new way of working with SAFe and knowledgeable on working within the SAFe practices,” Daniels says. “At the beginning of the journey, there was a lot of discovery work that took place as we were building the product from scratch. During this discovery phase, we began to identify and address gaps and pain-points within the process and within the team.”
One of those gaps was business data subject matter experts that could help the team translate business requirements into technical development. To fill the need, the team tapped into existing technology groups to socialize its project and to solicit input from data experts who had a keen understanding of operational needs.
Daniels says it was also important to maintain focus on a key SAFe concept: empowering decision-making within the team.
“As we shifted decision-making from the team leads and leadership to the team itself, we had to continually reinforce this way of working and our longstanding principle of speaking up,” Daniels says. “We increased our efforts to make sure the team felt empowered to speak up, make decisions, and step up to fill in gaps beyond what they considered their normal role if it would help move the project forward.”
OWL has now gone live across all of Kaiser Permanent’s 21 Northern California hospitals, and the plan is to expand it broadly across all Kaiser Permanente’s regions and hospitals. Pilot sites have reduced patient wait time for admission to the emergency department by an average of 27 minutes per patient. Surveys also show hospital managers have reduced the amount of time they spend manually preparing data for operational activities by an average of 323 minutes per month. Hospital managers also report an average of 114 fewer calls and messages to report out data per month.
Predictive analytics capabilities are next on the agenda. Daniels says the team is leveraging AI to build predictive census features into the app, helping hospital leaders allocate resources based on anticipated demand.