In an increasingly data-driven healthcare world, healthcare delivery organizations grapple with data complexity and often find it challenging to merge different data sources to drive real-time decision-making. To tackle these challenges, Blairsville, Ga.-based Union General Hospital recently developed and implemented a business intelligence strategy with the goal of putting data into the hands of case managers and clinicians to improve efficiency and quality of care and establish best practices across many departments.
The Union General health system includes the 45-bed hospital Union General Hospital, located in rural north Georgia, as well as Chatuge Regional Hospital, a 25-bed critical access hospital, two nursing homes and several outlying medical clinics.
According to Karen Reff, manager of decision support at Union General, hospital leadership recognized the need to enhance the organization’s analytical and reporting system for healthcare operations and care management as well as finance operations.
“The electronic health record (EHR) system was woefully lacking in its reporting capability and the initiative to start looking for another solution really originated with our chief financial officer when he finally crossed the pain threshold. He was doing a lot of Excel spreadsheets and pivot tables, which was extremely limiting and time-consuming,” she says. “That’s how the initiative originated, but we’ve seen tremendous benefits in numerous areas.”
As part of its business intelligence strategy, Union General worked with Sisense, a business intelligence software vendor, to develop and implement a data analytics technology platform that enables clinicians and case managers to pull together and analyze disparate data sources. The platform also provides a browser-based HTML data visualization tool that enables Reff and her team to build dashboards for various clinical and business analytics. The first dashboard got off the ground in August and there are now 10 departments across the system using dashboards.
The use of analytics in the healthcare delivery setting has increased rapidly in the past few years specifically in areas such as disease management, case management and performance monitoring. Larger hospitals and health systems have invested significant resources into analytics initiatives, as they move forward, and move more deeply into, the challenging work of setting up the data warehouses and business intelligence capabilities needed to support analytics.
Smaller healthcare delivery organizations, however, are often challenged with limited IT resources when tackling analytics projects. The data analytics market seems to be quickly expanding to provide healthcare delivery organizations tools that are agile and easy to use without a major IT investment.
When it came to vetting the vendors, Reff says they chose a company that could meet all of Union General’s core requirements, and “the time to deployment and the effort required was very compelling as well,” she says.
“One of our core requirements when we were looking for a BI (business intelligence) solution is that we couldn’t be dependent on or further burden our IT department to support it,” Reff says. “And that’s one area where the Sisense business intelligence platform excelled was the fact that I, not being an IT professional, could manage and build these dashboards and generate reports that have saved immense amounts of time and given us insights that we didn’t have previously.”
Sisense provided an end-to-end technology solution, referred to as self-service data analytics technology, which does not require significant IT resources or the creation of data warehouses and allows users to “drag-and-drop” data to combine large data sets. And, Reff says she can build interactive dashboards and business intelligence reports and the web-based dashboards can be shared so multiple users can access, monitor and interact with the same report.
For one initiative, Union General’s case management team is using the business intelligence platform and data analytics to examine variables in 30-day patient readmissions with the aim of driving readmission reductions and improve patient outcomes.
“When looking at 30-day readmissions and how we could reduce 30-day readmissions, there are a lot of data elements that go into assessing the high-risk readmissions and it would take me days to pull that data out of the EHR system. And, we had to do it individually, so I knew who the patients were and I had to log in and manually extract several data elements. I wanted to know, for example, whether the patient went home with a home health agency and which home health agency did they go home with? Who was their doctor? What was their diagnosis? Did they have a primary care provider when they came in? And there was no one report for this data.”
Knight continues, “With this business intelligence software, I’ve been able to reduce the time it takes me to gather that data and actually spend time analyzing the patients’ charts for more clinical things that I could drill down on that impacts that next potential readmission. So, when a patient goes home with a home health agency, I am able to drill down and compare that home health agency A has a reduced readmission rate versus home health agency B and that helps us when considering contracts and who we might potentially use as a home health agency.”
This business intelligence strategy has enabled a data-driven approach to patient care and real-time decision-making. Using data analytics to look at 30-day readmissions rates via a dashboard, the case management team found that patients referred to home health services have a reduced readmission rate. The case management team then established a best practice that every patient who qualifies is referred to home health services. Case managers also have analyzed data related to inpatient length of stay and readmission trends, Knight says.
According to Reff, Union General has seen a drop in readmission rates since implementing the analytics platform yet concrete figures are not available at this time. “We are fairly new to this platform, so I think a lot of the data is going to be qualitative rather than quantitative, but there definitely are improvements,” adds Julia Barnett, Union General chief nursing officer.
The analytics platform enables more efficient reporting and business intelligence functions and improves workflow, Knight says.
“To get the information that we can get in one dashboard, people would have to run multiple reports and merge them manually out of our system,” Reff says. “And, another good example of how dashboards are providing us data in real time is with something as easy as census. In the past, to get up to the minute census stats, we would go into our system and manually count the number of patients in each department and each unit. Now, we pull up a dashboard and it tells us how many patients are in each unit and how long they’ve been here and the admitting service. So it’s gone from 15 minutes potentially to five seconds or less.”
In the emergency department, the use of the new dashboards has resulted in significant improvements in efficiency and workflow. “Karen has built me about six dashboards that I use every single day and the three main dashboards that I use have really cut hours off of my day,” says Connie Rhodes, manager of the emergency department at Union General Hospital.
“When we send cultures out of the ED, it takes three to five days to get those results back and the ER has to follow up. Some of those were getting missed and were harder to find because you had to manually go in and look at every single chart in order to find the results. With the dashboards, what used to take me literally from two to three hours a day now takes about two to three minutes to get the results of the entire day.”
According to Sisense, the technology platform’s in-chip analytics engine enables increased speed in analytics processing. At Union General, the technology platform and data visualization tools are part of a business intelligence initiative that’s improving reporting capabilities, and these reports, such as documentation improvement, touch on a range of hospital processes from quality improvement to billing to clinical outcomes.
For Cheryl Curtis, manager of risk management, and her staff the dashboards have proven to be a valuable tool for monitoring physician quality reporting.
“We’re getting a lot of good data from a management standpoint as well, so staffing and management quality. Even things like managing overtime and watching turnover rates, we have dashboards for that,” Reff says.
She adds, “Everything we do eventually impacts patient care and if we make ourselves more efficient that has an impact as well.” Reff mentions, for example, how the dashboards impact caring for behavioral health patients. “We are able to assess, in real-time, how long, on average, we’re keeping those patients in the ER before we find a bed for them.”
According to Jeremy Sokolic, vice president of product at Sisense, while the healthcare analytics market is growing, an ongoing barrier for many organizations to deploy analytics programs is one that is more cultural than technical. “One of the biggest barriers is having the right focus and intention of what you’re trying to do, and with data analytics it takes intention from the top down. So you want a senior executive that recognizes the value and is willing to make the investment, so a champion in the project, and you need people in the field at the care delivery level that are responding to it and are willing to change their behavior,” he says.