Beyond the RIS: Safeguarding Radiology Through Business Intelligence

Paul Chang, MD, professor and vice chair of radiology informatics and medical director of enterprise imaging for the University of Chicago, believes radiology practices must evolve to survive—and that the path forward is to become more collaborative with clinical partners, survive on fewer reimbursement dollars, improve workflow, and enhance patient care, all of which can be accomplished by through properly leveraging data.

“When we’re talking about the challenges faced by radiology, including declining reimbursements and the imperative to improve quality while reducing costs, we’re talking about an administrative challenge. We need data that go beyond a single patient encounter. We need to understand utilization as a broad component, not just for workflow optimization but also for quality.”

–Paul Chang, MD

“In order to survive and flourish in an environment of decreased reimbursement and increased expectations, we need to be able to measure what we’re doing,” Chang says. “We have to demonstrate our value, improve quality and relevance to patients, and decrease costs, and the only way to do that is to do what every other business does: put heavy emphasis on, and strong investment in, business intelligence and analytics.” While many radiology departments might believe they’ve achieved this emphasis on analytics through their RIS and PACS, Chang cautions that these systems have yet to attain their full potential and that the real answers for radiology lie elsewhere. “The challenge is that we are moving to a more mature version of image management, where it’s more than just radiology centric,” he says. “To demonstrate our value to the enterprise, we need a comprehensive understanding of that context.” Limitations of the RIS Chang observes that one of the RIS’ most obvious limitations is its inability to provide clinical context beyond the indications given in the study order. “In the beginning, the RIS was adequate for providing context; we were concentrating on departmental efficiency—but now, when we want to demonstrate that we make a valuable contribution to patient management, we need context that comes from data captured outside the radiology department,” he says. Clinically speaking, this context can be found in the electronic medical record (EMR), but extracting clinical data from outside the RIS and PACS generally requires already-busy radiologists to enter the EMR system separately. “In order for me to interpret the study correctly, I need to understand the comprehensive clinical context, which is in the EMR,” Chang says. He notes that this presents a fresh workflow issue for radiology: how can the specialty better contribute to the continuum of care without sacrificing efficiency? The need also goes beyond clinical context. Chang notes that RIS and PACS systems are not able to access the kind of enterprise-wide billing and ordering information necessary for radiology to have a full picture of a given patient’s management. “When we’re talking about the challenges faced by radiology, including declining reimbursements and the imperative to improve quality while reducing costs, we’re talking about an administrative challenge,” he says. “We need data that go beyond a single patient encounter. We need to understand utilization as a broad component, not just for workflow optimization but also for quality.” Understanding these factors will be critical to ensuring radiology’s future as a specialty, Chang says, especially as health care becomes more focused on patient outcomes through emerging models like ACOs and integrated delivery systems: “These kinds of business intelligence questions are critical to demonstrating radiology’s relevance to outcomes.” Getting Around the Architecture At the University of Chicago, the answer has been leveraging a service-oriented architecture in which information is extracted from the hospital’s various information systems and mapped to a middleware business logic layer. Chang notes, however, that this architecture is not for everyone. Although he describes the model as very powerful (and as being used by every company in a vertical market outside medicine), he says, “It’s not how we’ve historically done things in health care, and for many facilities, it would require a complete overhaul of their IT systems.” That kind of overhaul is an undertaking that cash-strapped providers would find hard justify financially. Instead, Chang advocates the use of systems such as the Amalga health intelligence platform from Microsoft Corp (Redmond, Washington), which take HL7 and DICOM data from hospitals’ information systems, aggregate them, and selectively perform semantic normalization to transform those data into information that can readily be processed by other applications. For example, he says, “Laboratory values in HL7 are hard to interpret on their own. Normalization makes them more easily consumed by an application—for instance, a dashboard or database that provides insights to clinical users.” The advantage of leveraging a platform like Amalga is that it can be adopted without completely changing a facility’s architecture, Chang notes. “You’re only restricted by what you can feed to the system. Financial and billing data can be correlated with utilization, and you can begin to ask some very interesting questions,” he says. “You’re beginning to see more and more people understand that this is the least disruptive way forward.” With integrated enterprise data made available in this manner, radiology departments not only can produce more meaningful dashboards, Chang says, but also can produce scorecards, which offer the more global perspective favored by administrators. “We need two kinds of monitoring in radiology: tactical and strategic,” he says. “The tactical tools are the dashboards. They tell us whether we’re winning the battle—but cumulative, strategic scorecards tell us whether we are winning the war. Both tools can be created through this Amalga approach.” Winning the battle—by improving workflow and providing better patient outcomes—is important, but in order to safeguard its future as a profession, radiology also needs the tools to win the war, Chang says. Without enterprise-level business intelligence and analytics, when the specialty tries to demonstrate its value, it will come up short. “Without knowing where we are, we can’t know where we’re going to go,” he says. “In order to be able to handle this increasingly disruptive environment and navigate this very narrow corridor between commodity and value proposition, we need the data.”Cat Vasko is editor of ImagingBiz.com and associate editor of Radiology Business Journal.