Published on 11-Oct-2010
Validated on 16 Oct 2012
"Our key challenge is that every part of the UNC system—the academic medical center, the school of medicine and schools within the university—as well as outside institutions, wants the data that the data warehouse offers" - Dr, Donald Spencer, Associate Director of Medical Informatics for the University of North Carolina (UNC) Health Care System
University of North Carolina UNC
Big Data & Analytics, Smarter Planet
Leadership Series, Smarter Healthcare
The nexus of information technology and healthcare delivery is a familiar place for Dr. Donald Spencer. In leading the creation of UNC’s healthcare data warehouse, Spencer recognized that for an organization as complex and multifaceted as UNC, addressing governance issues early on is critical to long-term success.
UNC made a quantum leap in all aspects of its research capabilities. It can identify highly specific research cohorts10 times faster than it could before— in the process of redefi ning what’s possible in the realm of medical research. It’s a big reason UNC was recently awarded a $61 million research grant by the NIH. On the clinical side, UNC is now better adapted to the “pay for performance” reimbursement model increasingly demanded by payers. UNC has translated its superior ability to demonstrate the quality of its care into higher reimbursements that strengthen its operating performance.
Advance Planning With UNC creating a completely new information resource for the entire UNC system, it needed to work out how that resource should be regulated— specifi cally, how it’s used, paid for and invested in—and do so before it was designed and built. “If you have a great idea for the logical or technical design of a warehouse, and you think you’re going to establish the governance for how to use that data afterwards, you’re going to end up spending a lot more money unnecessarily.” — Dr. Don Spencer, Associate Director of Medical Informatics, UNC Health Care System
Get senior management buy-in early The complexity of UNC’s organization elevated the importance of senior management in establishing governance strategies and resolving governance issues. As such, getting them on board early was critical. “The common mistake would be to not involve senior leadership early on in a project like this.” (Spencer)
• Ten times faster identifi cation of research cohort targets, shortening the time and cost of launching new research projects • Facilitated more than $60 million in new research funding due to superior data management capabilities • Ability to negotiate higher reimbursements from payers by demonstrating adherence to high quality care practices • Improved ability to infuse clinical research into patient treatment and medical education
Dr. Donald Spencer has been straddling the worlds of medicine and information technology since the beginning of his career—and even earlier if you count the pairing of biology with computer science in his undergraduate studies. In the 1980s, when Spencer first joined a family practice in a small North Carolina town (population: 300), it was he who took on the job of setting up the practice’s computer billing system. Later, he led an effort to implement electronic medical records (EMR) across a number of rural practices. Spencer attributes his instinctive drive to meld technology with healthcare to the “fortune” of being the son of electrical engineer, a pedigree he shares with a group of physician friends who call themselves the “Children of Engineers.”
Today, Spencer is the Associate Director of Medical Informatics for the world-renowned University of North Carolina (UNC) Health Care System. After following his own path, Spencer is now leading UNC’s efforts to bring information into every aspect of its clinical and research activities. One dimension of the challenges facing Spencer and his team relates to the sheer quantity, complexity and diversity of the information that underpin UNC’s operations, with sources ranging from patient admissions data to lab results to radiology images. For this information to be useful across all of its operations, UNC needed to first synchronize these disparate sources into a “single version of the truth”—in the form of a data warehouse—that can be counted on for consistency and accuracy.
Planning for governance the key to success
The most immediate challenge UNC would face in realizing this vision was the technical complexity of integrating a large number of diverse sources, all the while conforming to requirements around data integrity, patient confidentiality and a number of other parameters. Even more integral to the project success, Spencer believed, was the need to plan for the unique governance issues that the new capability would introduce. In essence, UNC would be creating a completely new information resource that would open up unprecedented opportunities in every part of the UNC system. Put simply, it would redefine—indeed, expand—what’s doable in the realms of research, clinical practices and administrative optimization.
Spencer’s key insight was the importance of thinking through and working out the practical considerations of what would become a highly sought after resource among all of UNC’s diverse stakeholder groups, and doing it in advance of the implementation. This meant creating a framework that would govern how projects that rely on the data warehouse are approved, how they are prioritized and how their costs are allocated. “Our key challenge is that every part of the UNC system—the academic medical center, the school of medicine and schools within the university—as well as outside institutions, wants the data that the data warehouse offers,” says Spencer. “We needed to institute a means of balancing their interests while creating a sense of common purpose.”
UNC’s answer was to create a governance framework with three tiers. At the upper-level of the governance structure are executive leaders from each part of the UNC system, whose role is to address high-level strategic issues such as the funding of new infrastructure or policies around sharing data with other institutions. Below this are a series of smaller committees and ad hoc workgroups focused on more short-term, tactical and project-specific decisions. “Through this three-tiered approach, we have stable leadership with a long-term, big-picture view along with the nimbleness to make on-the-fly decisions when new issues, problems or opportunities arise,” Spencer explains. “It’s allowed us to sort through and prioritize issues in a way that we hadn’t done before.”
Eyes on the prize
The ultimate decision to move ahead fell to Dr. William Roper, whose varied titles include Dean of the School of Medicine, Vice Chancellor for Medical Affairs and Chief Executive Officer of the UNC Health Care System. To sell the project, the leadership prepared a multi-pronged business case whose main elements addressed important areas of research and healthcare delivery. On the research front, UNC’s top priority by far was to secure a major grant from the National Institutes of Health (NIH) known as a Clinical and Translational Science Award (CTSA), part of a large program designed to promote the transfer of medical research from the university into the community. Just how high were the stakes for UNC? Roper put it simply: “Our goal is to be the best public school of medicine in the country. And to be successful, we have to receive one of these grants.”
The clinical dimension of the business case emphasized the ability to drive quality improvements, an issue of direct relevance as UNC’s payers sought to tie payments more closely to demonstrated quality care. The better UNC could document key measures of quality (such as improvements in Hemoglobin A1c tests among diabetics) or adherence to high-quality clinical practices, the better chance UNC stood to increase its payer reimbursement rates. In making the case, Spencer was cognizant of the need for rigor—and came prepared. “You can’t just assert it will improve the quality of care because of your clinical judgment,” explains Spencer, who got his MBA in the mid-1990s for this very reason. “You have to translate it into the language that’s understood by business decision-makers.”
Once the project was approved, it took two years to establish the governance framework and deploy the data warehouse solution, which is known as Carolina Data Warehouse for Health (CDWH). IBM assisted in both in the development of the governance framework (which followed the IBM Healthlink Solutions Roadmap Methodology) as well as the CDWH implementation (which leveraged the IBM Health Integration Framework). The sidebar on the following page outlines the technology elements used to implement and run the solution.
Opening new doors for research
UNC’s investment in the data warehouse solution has yielded significant hard-dollar benefits. The most prominent and direct of these is the $61 million CTSA grant that UNC received from the NIH. The magnitude of the CTSA grant represents an enthusiastic endorsement of UNC’s capability—at an institutional level—to conduct research efficiently and to get that research out to the community where it can do the most good. In that sense, the data warehouse solution has opened the door to a range of new research opportunities for UNC, each of which can be viewed as a small victory that might otherwise not have happened. Dr. Shannon Carson, a pulmonary specialist on the faculty of the UNC School of Medicine, exemplifies this. His team recently got the resources to do a multimillion dollar study on Chronic Obstructive Pulmonary Disease that relies on CTSA funding as well as the wealth of patient data made accessible by the UNC data warehouse.
One important way the data warehouse aids medical research is by taking on one of its grittiest challenges—recruiting the right patients in timely fashion. In some cases, great research ideas never leave the conceptual stage because the difficulty of finding highly specific qualifications among subjects, with researchers often required to manually pore through scattered reams of medical records to create cohort groups. In other cases, grant opportunities require researchers to put studies together very quickly for them to get funding. UNC’s data warehouse addresses both. Through the solution’s simple-to-use Research Portal feature, researchers can run smart queries against the data warehouse, enabling them to assemble a research cohort at least 10 times faster than before, translating months of effort into days, or even hours.
On the clinical side, UNC’s data warehouse has become an essential tool in improving the efficiency and quality of the care it delivers, and—just as importantly—translating them into financial results. With the principle of pay-for-performance gaining traction among health insurance companies, the reimbursements that UNC and other healthcare providers negotiate with insurers have emerged as the most salient indicator of this new dynamic. Whether you view this back-and-forth as a “battle” or a “dance” (Spencer, in various contexts, calls it both), data is the key to UNC’s negotiating power. “In complex organizations [like UNC], there are conversations with payers where the right hand doesn’t know what the left hand is doing,” says Spencer. “The fact that we now have a single, comprehensive view of that information enables us to fully demonstrate our quality and maximize our reimbursements. That kind of efficiency, in turn, is critical to our ability to provide unreimbursed care to the citizens of North Carolina who can’t get it anywhere else.”
The parameters of UNC’s advanced health informatics
Instrumented: Physicians, staff, administrators and researchers feed data ranging from patient records to X-ray images and more into a multitude of information management systems.
Interconnected: A robust data warehouse, developed and deployed by IBM, links the data systems and enables single-point access and sophisticated healthcare informatics for diverse user groups.
Intelligent: Researchers, clinicians and administrators can analyze and inter-relate data in new ways, which leads to improved patient outcomes, disease management, compliance and research.
Flexibility to hit a “moving target”
But as Spencer points out, the fact that payers frequently redefine their quality standards means that providers need the flexibility to change their analytics and reporting along with them. “Not only is it a battle,” he says of reimbursement negotiations with payers, “but it’s a battle with moving targets. Once we’ve negotiated our targets, the rules will likely be a little different next year, like maybe a requirement that blood-pressure testing on diabetics be at the 60th percentile level instead of 55 like last year.” Multiply these changing quality requirements by the sheer range of disease areas, and the data management challenges of meeting these requirements gets complex fast. On the strength of the data warehouse, UNC has achieved a string of successes in increasing its reimbursement through its ability to document the quality of care. One recent event spoke volumes about the success of UNC’s data warehouse strategy. It was a visit by the head of the NIH, Dr. Francis Collins, who had traveled the 300 miles from Bethesda, Md. to Chapel Hill to see firsthand how UNC was making good use of the CTSA grant. In his tour of the facility, Collins came away with a clear belief that the data warehouse was living up to its potential of strengthening UNC’s research capability.
Spencer is already convinced. “The success of our data warehouse has already shown us how the right kind of data management practices can drive important medical research and strengthen us—not only as a research facility, but also as a healthcare provider and medical school,” says Spencer. “But it’s early, and we’ve only begun to scratch the surface of what we’re capable of achieving.”
UNC’s solution is -
- IBM Health Integration
- IBM DB2
- IBM InfoSphere
- IBM InfoSphere Information Server
- IBM WebSphere Application Server
- IBM WebSphere Portal
- IBM WebSphere Portlet Factory
- IBM z/OS
- Linux on System z
- IBM System z10
- IBM Global Business Services: Strategy and Change
For more information
Please contact your IBM sales representative or IBM Business Partner.
Or visit us at:
View the Leadership Series Web Portal for UNC
Products and services used
©Copyright IBM Corporation 2010 IBM Corporation 1 New Orchard Road Armonk, NY 10504 U.S.A. Produced in the United States of America June 2010 All Rights Reserved IBM, the IBM logo and ibm.com are trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at www.ibm.com/legal/copytrade.shtml Other company, product or service names may be trademarks or service marks of others. This case study illustrates how one IBM customer uses IBM products. There is no guarantee of comparable results. References in this publication to IBM products or services do not imply that IBM intends to make them available in all countries in which IBM operates. Please Recycle ODC00000‐USEN‐00