What kind of network should EMRs use to talk to each other?

Getting Electronic Health Record (EHR) systems to “talk” to each other has been one of the goals of the emerging national Health IT policy. Several of the Meaningful Use criteria for doctors and hospitals, as well as HHS Certification criteria for EHR vendors, are based on the ability of an EHR system in one doctor’s office to forward clinical information to someone else. We have become so accustomed to this capability in the realm of emails that we take it for granted; with Health IT, it is a bit more complex.

Health IT has traditionally been built around enterprise client/server systems, which were constructed to be all-encompassing solutions for their customers – a soup-to-nuts offering that included billing, scheduling, note-documentation, lab integration, x-rays, in-system messaging, prompting and reporting, all in one place. Each vendor created its own system in its own silo, and (often for competitive reasons) did not have an easy way to exchange information with outside systems. The result? Such systems have been huge, expensive, and stand-alone. Kaiser, for example, invested billions in implementing their EHR for hospitals and doctors within their network – yet, those outside the Kaiser system don’t have easy access to those records. A community physician requesting Kaiser records on a patient whose insurance has changed still requests and receives those records as faxed or hard-copy printouts (not a direct transfer of the data electronically).

The Office of the National Coordinator for Health IT (ONC) has been working on addressing this problem. Besides the issue of standards of EHR record output – pertinent clinical information can be extracted and placed into a standard-format document (either a CCD or a CCR) – there is also the question of how to transport that information from one doctor to another in a secure fashion. This has been the focus of work around the Nationwide Health Information Network (NHIN).

A couple of different approaches have emerged for how the NHIN should be built, and what its role should be. The initial proposal was that the NHIN would be a “network of networks” aggregating state-level and regional Health Information Exchanges (HIEs) into a massive “library” of patient data, where providers of medical data (like labs, hospitals, and office EHRs) would upload their CCDs, and “check out” information on a given patient from this library. This kind of build-out, not surprisingly, came from the massive enterprise-style EHR vendors and systems using such products. One can imagine the issues involved – different privacy and security laws for each state, big costs for building and maintaining such huge structures (with an uncertain business model around how such HIEs would remain solvent), differences in how different HIEs might structure their data, and the effective exclusion of everyone outside the large hospitals and clinics from participation in such things (if they ever are built, and work).

Parallel to this path, a different vision for the NHIN has also emerged, in the form of the NHIN Direct. This is much more of a “health internet” approach and offers good promise for connecting smaller practices in a point-to-point fashion. This view is that the NHIN Direct should be a very lightweight platform, and may only need to aggregate a list of physician/provider names and connection information (similar to a “contact list” of email addresses). An office EHR would then simply create an output document (like a CCD or CCR) from a patient’s chart, look up the destination from the “contact list” maintained by the NHIN Direct, and send the record point-to-point to the destination in a simple yet secure fashion. Such a structure would thus require that EHR vendors build input/output channels, very much like email inboxes. This kind of lightweight “transport platform” vision for the NHIN Direct, whose main job is to maintain a Master Provider Index that contains simple email/routing information, is much more likely to be adopted by small practices.

Getting EHRs to “talk” to each other is a critical goal in healthcare reform, and is a necessary piece of Meaningful Use and HHS Certification. The conversation and vision coming out of the NHIN Direct workgroups are a refreshing, and more realistic, approach, in contrast to the heavy, enterprise-focused vision of the NHIN that had dominated the conversation initially. We encourage pursuit of this vision of NHIN Direct, where point-to-point exchange of standard-format health data (CCDs or CCRs) via a simple email-like method, is what comes into play. The likelihood that this can be achieved by the time Meaningful Use bonus moneys are available in 2011 is much higher than with a heavier, traditional approach. Keep up the good work, NHIN Direct!

Robert Rowley, MD
Chief Medical Officer
Practice Fusion EMR