Make your practice more efficient. Get in touch with our Sales team today at (415) 993-4977.
Contributing Writer · Dec 9, 2010

Practice Fusion and Health Information Exchanges

An important goal in the effort to achieve widespread adoption of Electronic Health Records (EHRs) is to allow patient data to flow easily from where it is housed to where it is needed. Though it is conceptually simple, such a goal is remarkably difficult to achieve.

The legacy of EHRs, which have been around for over a decade, has been mainly to replace paper note-taking with their electronic equivalent. Legibility issues go away, and finding a patient’s record in the clinic is no longer the headache it once was. Mostly, EHRs were built to be locally installed, and be the electronic equivalent of a doctor’s chart rack (the local paper “database” for patient health records).

This was certainly an improvement over the past, but on a larger scale, health information was still contained in silos (each clinic, or each hospital) – just like paper charts. The landscape of health data distribution did not change much with this first stage of EHRs.

In the modern era, such a fragmentation of health data is not enough. Given that a patient may see as many as many different practitioners, pieces of the patient’s record can be contained in many different charts – there is no practical way to sync these up, and the risk of preventable errors is significant.

How to address this problem? There are several approaches. One approach is to recognize that places that have already invested very large sums of money into EHR systems (like hospitals, large medical groups, etc.) are unlikely to abandon those systems. Therefore, one must overlay onto these existing systems a method of connecting them together: the Health Information Exchange (HIE).

The Office of the National Coordinator for Health IT (ONC) has tried to encourage the building of HIEs, “connecting the dots” locally and regionally into a variety of local HIEs, and then tethering all these HIEs together into an envisioned Nationwide Health Information Network (NHIN).

Several significant hurdles stand in the way of this approach achieving its goals. The first is technical: health data exchange standards are not really very standardized. As a recently described attempt to link the Veterans Affairs system and the Department of Defense system together shows, getting the right message structure and specific vocabulary to mesh (so that the sender is saying things that the receiver can interpret) is difficult (it failed). Expanded onto a landscape of several hundred different EHR systems in use (not to mention home-grown or locally adapted systems), this challenge is enormous.

There is an entire niche of business for companies that build very sophisticated software to allow HIEs to function. Standardizing messages, keeping track of senders and recipients, authenticating the parties, verifying consent and specific permissions – these are all issues addressed by such HIE-supporting software. As one would imagine, such systems are quite expensive, and represent a major expense to be born by organizations that want to serve as HIEs.

The other challenge to the HIE strategy is the business model. Once initial seed funding for the creation of HIEs has run out, how HIEs will sustain themselves is an open question. Several models have been followed – but the end result is that some HIEs will work, and others will fail. And the question of cost – will a physician have to pay to get an Emergency Room report from the local hospital, through the local HIE? – remains unanswered.

Will HIEs work? Maybe. Eventually. They will likely be the main way that hospitals will connect, and large medical groups will connect. Will it be what solo and small-group practices use to connect with each other? Probably not (at least not right away). Will it help drive EHR adoption by those same small practices? Very unlikely.

What alternatives are there? The largest example of an alternative is the Practice Fusion web-based EHR (which is the common platform for over 60,000 users and nearly 6,000,000 patient records, so far, and growing by 200-300 new users every day). By design, the single-system web-based deployment of this EHR connects all users in a common platform. Though not yet built out as fully as envisioned, the ability to share clinical data between users on this platform is quite straightforward – certainly more so that by overlaying an HIE as a necessary intermediary between practices.

Given the startling adoption of this web-based EHR by small practices in particular, such an approach may be the preferred path for sharing clinical data for this end of the practice-size spectrum.

Of course, even a web-based system like this will need to talk to hospitals and other large settings that have locally-installed legacy systems. On the physician side of this connection, there is only one connection needed (for everyone). The web-based server creates, tests and deploys a single connection, and everyone (all 60,000+ users) have immediate connectivity. This has been the case for laboratory connectivity, and can be the case for hospital connectivity.

The problem, of course, is that there are thousands of hospitals. So the single web “plug” for all the physicians would still need to connect to each and every hospital. Still untenably onerous (though less so than connecting every local doctor with every hospital on-to-one). This is where HIEs come into play. An HIE, using HIE-supporting software, would do the work of connecting the hospitals (and other health data sources), and then create a single connection to the Practice Fusion web servers.

This is likely to be the way that small practices connect to outside data sources. In-system connection (other Practice Fusion users) can take place inside the product (as these features are built) – it is clean, easy and fast. Out-of-system connection will be via a server-side arrangement with HIEs as these options mature. It will likely be an uneven maturation, with early connections to “beacon communities” that have well-functioning HIEs occurring first, and other areas occurring later, as their local HIEs start to emerge.

Robert Rowley, MD
Chief Medical Officer
Practice Fusion EMR