Medical Center Health System in Odessa, Texas, created its own analytics and decision support department to facilitate analysis that helps clinicians and administrators. As a home-grown department, it’s enabled the hospital to focus business intelligence and analytic capabilities, while establishing an organizational approach to analytics structure and governance.

Its data analytics department is the one area that can link multiple data sources from across the organization, enabling the connection of links and correlations of data that previously was not available. 

Over time, it’s helping administrators and clinicians learn how to analyze data, and get excited about it, said Alan Snider, senior decision support data analyst in the department. “They don’t know how to drill down into the ‘why’ question,” he said. “Our goal is to supply them with the tools so they can answer those questions themselves.”

The approach used at the 402-bed regional medical center is a model that could be employed at other healthcare organizations that want to make the transition from merely gathering data to deriving value from it.

 

MCH Center for Women and InfantsMore hospitals are interested in developing analytic capabilities, said Gary Barnes, who is senior vice president and CIO at Medical Center Health System. “Typically, these departments are under the CIO,” he said. In those instances, results can vary depending on where the CIO is in the organization.

Medical Center Health System took a different approach when, nearly three years ago, it put decision support and analytics into one unique department. Leadership in the organization was prompted to act because it believed that simply implementing an electronic medical record and leaving analytic work at the department level would have little impact on outcomes and wouldn’t enable the organization to tie process metrics to outcomes, Barnes said.

As healthcare evolves and changes, healthcare organizations will need to conduct enterprise-wide analytics with a multi-disciplinary team – that will provide more opportunities to improve outcomes, reduce costs and facilitate planning across the continuum of care, Barnes said.

An initial assessment of the organization’s analytics capabilities revealed that Medical Center Health System had a long way to go – capabilities were spread throughout the organization, data wasn’t always trusted, and analytics skills were lacking. Creating a well-supported analytics and decision support department was intended to address these concerns and also enhance results by installing an optimized technical infrastructure specifically to build up analytics capabilities, Barnes said.

To establish the department, the organization created a business intelligence steering committee, which provides direction and oversight for the initiative, and continues to work with the department to identify activities that can help achieve improvement through data analysis. The committee draws members from finance, quality, care delivery and IT.

The department also was created to build a data background for decision support for clinical areas, said Arun Mathews, MD, the CMIO for the organization.

“When we started our (computerized provider order entry, or CPOE) journey, we were interested in analytics and the ability to track the usage of individual order sets down to individual orders, so we could let that be part of the discussion about how we were doing with our implementation,” Mathews said. “CPOE was such a huge cultural change, you can be tempted to make changes based on disgruntlement or feedback. We wanted to make changes based on the usage of order sets. We wanted to be able to look at the data, and then use that to improve processes in patient care.”

Quote“The key to success is not making it a part of IT–we make that perfectly clear. It’s designed to be separate and away from IT, on a whole different floor of this building.”

The analytics department at Medical Center Health System is small, encompassing only three people with complementary specialties. However, the team enjoys executive-level support and wide-ranging respect throughout the organization.

The separation of the analytics department from the IT department is crucial, Barnes emphasized. “The key to success is not making it a part of IT–we make that perfectly clear. It’s designed to be separate and away from IT, on a whole different floor of this building.”

The analytics and decision support department staff include Michael Yandrich, a data warehouse manager with more than 20 years of experience in almost all of the IT systems; Anusha Donepudi, a decision support analyst; and Snider, the senior decision support data analyst who has more than 30 years of experience with hospital analysis, including productivity, budgets, operations, quality and benchmarking.

With executive support within the organization and a growing track record of achievement, the small department has made an impact, Yandrich said. “We’ve made great strides over the past year,” he said. “We work with physicians and decision makers, including executives and department managers.”

The analytics staff has responsibility for nuts-and-bolts work to do their jobs, such as finding relevant information to enable analysis; normalizing the information to permit comparisons; and presenting the findings in the most understandable format. Beyond those tasks, the staff has worked hard to interact with and build trust with its internal customers.

“At first, it was a little slow in getting people to understand what you’re doing,” Yandrich said. “You have to present data that they feel is accurate. We had to build the trust, and now a lot more people are trusting and asking for more. But getting trust in the quality of the data is the first step.”

Another goal of the analytics and decision support department is to raise the awareness of the potential for using data within the warehouse. Many times, that involves a visit to two gigantic whiteboards in the hallway outside the department’s office. “They’re story boards, and it shows in a visual way what we are trying to tell people; it lets us visually try to piece together all this information,” Yandrich said. “It helps identify where we have to go and what’s next.”

The department also tinkers with the visual representation of findings, using analytic tools and creativity to meet the needs of end users. “There might be four different ways to present information, and we’re trying to find a presentation format that meets their need and corresponds with their decision making style,” Donepudi said.

As the staff collaborates with internal clients, they find out what information they need on a recurring basis. “One of our goals is to be able to supply information and give people access to the information themselves, rather than just giving them a one-time presentation of data,” Snider said. “For example, we can create bookmarks for people that show the results of reports in the exact manner they want to see them. They can look at the results in the way they want it, and I’m out of the production business.

“Although we produce reports and lists when needed, our goal is to provide progress tools for decision making as close to real time as possible, he added. “We have been effective in reducing the time several areas spend pulling the information they need on a routine basis.”

In addition to monitoring hospital operations, the analytics and decision support department also is helping drive quality improvements and adherence to clinical protocols and workflow standards.

Using data on clinician practice within the organization helps the organization achieve a wide variety of goals, Barnes said. For example, it uses scorecards to track progress toward achieving Stage 2 Meaningful Use Objectives.

Beyond achieving improved IT performance, the organization is using improved data tracking to help it improve its caregiving. The organization has a philosophy of using standards for its care, and it has a program to help clinicians meet those standards.

Data was first used to help clinicians increasingly use CPOE to enter orders for medications, lab tests and radiological exams. Now, the organization is using analytics to improve quality and patient care.

“We’re starting to do some more impressive work related to clinical outcomes,” said Mathews, the CMIO. “We are developing a readmissions dashboard, and using a modified tool for predicting readmissions. We can get close to identifying the real-time risk of readmission. Because it is data driven, we can measure how effective we are and can eventually get to the things that give us the most bang for our buck in limiting readmissions.

“The goal is to get to real time, when we’re able to refresh our analytics in such a way than an admission triggers the readmission logic, so that while a patient is still in the emergency department, we can identify patients to be at high risk of readmission,” Mathews added. “Eventually, we hope to be able to do the same thing with identifying fall risks.”

Analytics capabilities will continue to impact quality initiatives at the hospital, said Sherrill Rhodes, Divisional Director of Quality and Service Excellence at Medical Center Health System. She draws heavily on the analytics and decision support department to provide the information she needs.

As data needs grew to meet reporting requirements, Rhodes began to work more closely with IT and the analytics team. “The staff members in my department were responsible for making sure that physician order sets were up to date,” she added. “We were driving a lot of change in the order sets because of regulatory requirements or evidence-based practice.”

Rhodes monitors readmission rates, and can use analytics to keep tabs on current trends. For example, she can drill down to the level of transitional care nurses and otherwise evaluate opportunities to keep readmission rates as low as possible. She similarly uses data regularly to watch trends with medication errors. The use of CPOE has reduced these types of errors, but there is always room for improvement, and analytics help to identify issues and resolutions, she said.

Availability of information, and the ability to report on it, enables executives to use it to drive improvement and make the organization more intelligent, Snider said.

“Although all data from every system is not available yet, we have significantly reduced the number of data silos in the healthcare system,” Snider added. “Having a central source of data builds consistency and confidence in the information provided. Our next big step is to increase awareness and utilization of our current resources.”

(This article appears courtesy of our sister publication, Health Data Management)

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access