The Road to Interoperability Is Paved With Benefits and Challenges
Frank ClarkNo matter which alternative imaging-delivery model they opt to follow, radiology practices and hospitals must ensure the interoperability of the systems used to share data internally, as well as with provider partners and payors. The Medical University of South Carolina (MUSC) in Charleston, a high-volume teaching hospital, has progressed quite far along this continuum. The facility, which saw 40,000 inpatients and 1.1 million outpatients and generated $1.3 billion in patient revenue in 2010, describes itself as the largest nonfederal employer in the city. In terms of connectivity, its electronic medical record (EMR) rated a stage 6 designation on the HIMSS Analytics™ seven-stage EMR Adoption Modelsm scale, according to a July 2011 U.S. News & World Report ranking of the top 118 most connected hospitals in the country. Frank Clark serves as MUSC’s CIO. He states that the hospital has had a fully digital EMR system for about four or five years, although, as he says, “it will always be a refining and evolving initiative” as additions and upgrades to it continue. The center of the hospital’s clinical IT universe is a data repository that holds 15 years’ longitudinal patient data. A Java-based platform enabled the university’s IT staff to customize the look and feel of end-user interactions throughout the system across a single sign on. Clark describes the system as highly interfaced, noting that it manages data from the RIS/PACS, the all-digital MUSC hospital pharmacy, and the 30,000 daily electronic orders that flow through the repository. “From a caregiver’s perspective, it seems as if it’s one record and one database,” Clark says. He adds that this also goes for referring physicians, who likewise have Web-based remote access to the MUSC system via secure, HIPAA-compliant connections. There is, however, more than a slight degree of behind-the-scenes maintenance required to develop and interweave the various applications that constitute the system, Clark notes. As opposed to having a single-vendor solution, such maintenance requires three to four additional full-time employees, as Clark puts it, “to keep the trains running on time.” Poised for Quality Improvement Maintaining an interoperable digital environment not only positions MUSC well for adopting newer models of service delivery and of interfacing with referring physicians, but also affords the hospital other opportunities that might make it a more attractive catch for potential alternative care-delivery partners. For example, it yields a significant opportunity for quality improvement through data mining. As a $243 million translational research center, the hospital has a single source of truth: an enterprise data warehouse, from which clinical data are extracted nightly. “We have to report outcomes to CMS, the state, Medicaid, the Joint Commission, and on and on and on,” Clark says. “The data are structured in a way that makes it much easier not only to do the analytics, but also research. Things like the warehouse have really helped grow research funding.” Moreover, the data center has been a key component of a systemwide cost-saving initiative to slash $100 million (10%) from the $1 billion MUSC budget, Clark says. This includes 5% in each of fiscal years 2011 and 2012. “If you don’t know your costs, you can’t reduce them,” Clark adds. On the horizon, Clark says, MUSC’s health IT staff will move into capturing data from additional devices, from monitors to infusion pumps. He concedes, though, that gathering that information on the fly—in real time, without human intervention—is a tall order, made ever more complex as the industry progresses into care delivered in nontraditional settings. “We continue to gather up different devices,” Clark says, because telemedicine and home monitoring are growing as health-care organizations require more and more access for patients to view such data as laboratory results. “You’d like some of the applications that are running on the iPhone or Android to run natively on those devices,” he says. “We just don’t have some of the technologies that are needed.”