by: Mari Greenberger, MPPA, co-authored by Gary W. Ozanich, Ph.D.
Two important federally funded studies that take a rigorous approach to HIE research have recently been published. The first is the RAND Corporation’s Usage and Effect of Health Information Exchange: A Systematic Review published in December, and the second is AHRQ’s Guide to Evaluating Health Information Exchange Projects published in September. What both of these studies have in common is an appreciation for how narrow HIE research has been to date as well as a discussion of how these research limitations can be addressed.
The RAND article reviews studies published between January, 2003 and May, 2014 that are structured as hypothesis tests or quantitative studies of data exchange between non-affiliated entities. The timeframe represents at least 3 different HIE-funding life stages [Pre-HITECH, HITECH, and Post – (or at least preparing for Post)-HITECH]. The conclusion is not surprising, finding limited results that HIE use “probably reduces (Emergency Department) usage and costs in some cases.”
By now it is pretty clear that health data exchange is messy and complicated and still rapidly evolving. If you have seen one HIE, you’ve seen one HIE. Their characteristics are largely regional—depending on the competitive market structure and governance. Standards and interoperability, vendor challenges, truly “shared” care plans, workflow, adoption and use, data usability, and the overarching concerns of governance and data-sharing agreements are all “intervening variables” that shape the efficacy of information exchange. HIEs are nuanced and emergent. Therefore, it is difficult to assume that an HIE’s impact will be measured as a single “direct-effect” end result structured as a testable hypothesis. Interpretation requires context. Ultimately outcomes and cost savings will be the gold standard, but the HIE market is still in an early development phase.
Another factor concerns private HIEs—whether they are extended IDN networks, vendor networks, or ACO’s—that is where much of the action is. Due to the parameters of the literature review and the secondary nature of the work, the RAND study does not touch upon these private HIEs. Finally, the study does not mention the characteristics and scope of the annual assessments required by ONC under the Cooperative Agreement Grants—which greatly shaped HIE research during the last five years.
Nonetheless, the study is well aware of the limitations of the earlier research they analyze and it succinctly states “more studies are needed to identify what does and doesn’t work and in what contexts. Until better evidence is available, all HIEs should be viewed as experiments and evaluated as such.”
The second research project, the AHRQ Guide is designed to solve the short-comings of HIE research design, implementation, analysis and interpretation. This comprehensive work provides an excellent reference document for undertaking HIE research.
These resources represent two prominent, timely, and complementary research projects which continue to move the conversation and impetus for HIEs forward.
Mari Greenberger, MPPA, Director of Informatics, Healthcare Information Management Systems Society
Gary W. Ozanich, Ph.D., Center for Applied Informatics, College of Informatics, Northern Kentucky University