It has been eight years since Dr. David Brailer announced the Decade of Health Information Technology while serving as our first national coordinator at HHS. The framework, as published in 2004, identified four strategic actions:
Thus far, much of HITECH has focused on ensuring that providers are investing in and are appropriately adopting secure, functional electronic health-record systems, and most of us are scurrying about seeking to comply with meaningful-use requirements and other rules so as to receive Medicare and Medicaid incentive payments for all of our efforts. While Stage 1 requirements seem to be centered on EHR adoption, they foreshadow, by a few indicators, things to come. I will mention a couple of those items that are paramount in building the national health information network.
The use of medical vocabularies to make the interpretations of clinical information more uniform and user-friendly for clinicians is critical. As we evolve to more data-rich uses of these structured vocabularies, we also will be more easily positioned to study trends and correlations across data to include population health findings, translational informatics research outcomes and clinical utilization support.
Also along this journey on the health IT information highway is the eventual ability to provide meaningful exchanges of clinical data across the continuum, especially between providers. These exchanges, in effect, are what make regional, state or national health information networks valuable to patient care and care economics.
One example of a practical and much-needed patient information exchange is that of a primary-care physician requesting a specialist referral or a hospital admission while actively working in the patient's record within the EHR—without needing to leave the record, log on to other applications, or further expend time and energy from the heart of that current episode of care. Then, as follow-up, the findings of the specialist or the hospital discharge summary should be sent as easily from that care venue back to the referring doctor electronically. The same sort of exchange is needed to report to the primary physician what occurred during an emergency department or urgent-care center visit. And these sorts of exchange examples could be expressed for all the various encounters that occur across the care continuum. This form of data exchange is called "direct" by the ONC.
Secured data exchange should be built to use economic means to transport the information, i.e., the Internet without the added expense of virtual private networks. EHR data exchanges will require a routing service, much like how Surescripts provides routing of pharmacy prescriptions to various pharmacies and now also how it provides medication reconciliation data from its database back to providers. These transactions should be workflow-centric for the physician by being easy-to-use functions within the active screens of a patient's EHR.
Many large EHR vendors have built the capabilities to deliver these transactions as described above; however, many use proprietary approaches that share such data only among their own EHR products' customers. Some of these approaches also will require eventual retooling because they send an image file of the information, which is useful, but which will need to be represented as real data that can be merged into the EHR for computational uses and trending (not to mention auxiliary uses for population health and bioinformatics research).
Further, health information exchanges may be sharing data by other means, such as portals, that are accessed separately through secure log-ons and that require printing of information from other providers so as to allow scanning into the accessing provider's record. This approach also may be currently useful, but if built without the work-flow efficiencies that physicians require in their busy workdays, these systems will not be adopted.
Current methods of EHR-to-EHR data exchange include packaging data as it is available in various EHR products' formats. The clinical descriptions are usually unique to the provider organization because they have been "home-grown"—in other words, clinical vocabularies may be all, partially or not currently used, depending on what clinical content is being viewed—for example, problems, procedures, history, directives, alerts, medications and other types of information.
As a result, current data exchanges cause our physicians to have to work through all sorts of representations of data that are spoken with various and unstructured clinical words from many EHR sources. The clinician then is working in the world of "what you see is what you get." This is a far stretch from ideal clinician decision support.
HITECH starts making a real difference in care when physicians use EHRs, the EHRs can directly and securely exchange specific chart elements, and those elements, where appropriate, are spoken in structured clinical vocabularies to assure understanding across providers. This is what is coming in an evolutionary way in Stages 2 and 3 of HITECH.
With the expectation that our vendors will cooperate by adhering to standards, providers will adopt new EHR features and clinical vocabularies, and infrastructure will be put in place for routing data by and between various providers' EHRs. This is happening already in certain geographies among select providers with investments in the necessary health information services providers acting as routing utilities and with EHR vendors who are keen on making these investments. As these models evolve, they will scale and thus be sustainable—a problem that has existed thus far for HIEs.
The future is looking bright. I love it when a plan comes together.
George Hickman Executive vice president and CIO Albany (N.Y.) Medical Center Member, CHIME board of trustees
Send us a letter
Have an opinion about this story? Click here to submit a Letter to the Editor, and we may publish it in print.
