I came across an interesting article in E-health Insider. The article reports on comments made by Dr. David Brailer at the recent HIMSS meeting in San Diego regarding the state of US RHIO (Regional Health Information Organization) development. According to the article, Brailer said that the idea of creating a virtual network that keeps health data in its original store and shares information via a ‘Google-type’ search is not on the horizon at the moment. He went on to say, "I have not seen the epiphany… I hope people keep experimenting but I don’t see a breakthrough that’s going to take us all the way without something in the middle".
It's too bad Dr. Brailer didn't drop by our booth at HIMSS. If he had, he would have seen some interesting work done by one of our partners in the RHIO space. The partner, HealthUnity, is a small company based in Bellevue, Washington, that was showing their "RHIO in a Box" solution built entirely on Microsoft .Net.
What HealthUnity did was to first approach their RHIO offering from the business requirements perspective (see prior Blog entry on this issue), as defined by Clinicians, Hospitals and patient advocates. They combined that with their technical expertise (the founders are former Microsoft employees) in areas such as fault-tolerance, availability, scalability etc. Here is how they scoped some of the RHIO requirements.
1. Search for clinical data across a regional network - This search takes up one of the following models:
Real-time (query end systems or query caching servers at end points)
Pre-fetched data i.e. anticipating a future query
Locally caching results (address potential network failure scenario)
2. Patients want an accounting (HIPAA right) of who accessed their records and from which organization. This data is captured today in the HealthUnity solution. In the future, when patient authentication issues are solved, they'll be able to let patients see this data (patient self service).
3. Ability to exclude patients completely from the network (i.e. patient opts out entirely; let's say a VIP patient). Knowing the ability to opt-out exists is more important than actually pulling the trigger (from a patient's perspective).
4. Ability to exclude parts of a patient's chart from automated access, yet make it available for case by case requests (e.g. an AIDS test result). Driven by patient privacy and state laws...
HealthUnity founder, Prem Urali, told me "To some extent Dr. Brailer has not seen the epiphany because he hasn't seen our solution" (he added a smiley face to the end of that comment). "We are not resting on our laurels by any means" he said. "We are continually refining our approach by listening to customers and patient-rights advocates and experts like yourself".
While many others are still spinning up meetings on how to begin the road toward forming a RHIO, the team at HealthUnity has already deployed their solution in the greater Eastside areas of Bellevue-Redmond, WA, and also in Baltimore, MD. Their pipeline is growing. Of course, these guys don't have the only RHIO solution on the market or in development, and I won't make predictions on the company's long-term prospects. But they have proved something. It is possible to build a robust, distributed model for sharing patient demographic and clinical information across disparate systems, and to do so without building a centralized data base other than a record locator service.
Perhaps this is the epiphany we've been searching for.
Let us know what you think.
Bill Crounse, MD Healthcare Industry Director Microsoft Healthcare and Life Sciences