In the Healthcare Interoperability Glossary, interoperability is defined as “The ability of two or more systems or components to exchange information and to use the information that has been exchanged.” Even with the industry’s current focus on health information exchange and electronic health records, a truly interoperable healthcare system can at times seem as likely as a pot of gold at the end of the rainbow.
Despite the monumental task at hand, there are many health organizations making significant strides toward interoperability and the ultimate goal – providing patients with quality, accountable care. I asked my good friends at Perficient to provide their insight on the five following interoperability questions. (You can follow Perficient on Twitter @Perficient_HC.)
Special thanks to Michael Planchart, Enterprise Architecture and Interoperability Consultant in Health IT at Perficient, who was kind enough to participate in the interview.
1. When you first begin working with a client, are there common “first steps” you take to make the interoperability process easier?
As a solution architect, I perform several actions when I engage with the client.
First, I take a clear assessment of their current state and available resources, their integration platform and their integration engines.
I perform a very comprehensive “data discovery” of what source systems are providing HL7 messages and other types of discrete data, such as X12 and DICOM objects. This data discovery is carefully documented using specialized templates.
Second, I perform a very thorough discovery of what source systems provide unstructured data, such as clinical notes, discharge summaries, scanned documents, etc.
Third, I perform a gap analysis to determine what type of information they are missing based on common practices in the provider space.
Fourth, I assess what they want to do once they do become interoperable, or what they hope to accomplish, buy conducting a requirements gathering exercise. For example, do they want to perform predictive analytics for clinical use cases? Do they want real time point of care applications? If so, I backtrack to see if they have the required data available from the source systems. I then find out what their capabilities are and where their gaps may be. This is what I call a top-down approach to understanding how the use case will align with their current data entities.
The information I obtain from these steps is used to complete architectural diagrams and other mapping artifacts that provide a comprehensive overview of interfaces running through the integration engine, including the types of standards being used with each connection, and how information is being captured and how the data is being transmitted from one system to another. It also helps to create a conceptual data model of the future state.
2. Most providers are either already connected to a Health Information Exchange or are preparing to connect to an HIE. After personally working with several different HIEs, what do you think they need to do for long-term success?
Providers have to be willing to provide the data, this is the big challenge with HIEs. I’ve worked on several HIE projects, and it’s very complicated to align the provider if the HIE is external, or public. Many providers consider health data as “Intellectual Property” or IP and they create many roadblocks that prevent the release of the data to the HIE. I’ve worked on projects where it has taken approximately two years before a hospital is willing to release their data.
I think HIE members need to be willing to share information and stop looking at their data like IP. The patient data belongs to the patient and the provider needs to be willing to not only use the data but also provide the data to help coordinate care, move towards population health management and improve research and evidence-based medicine.
We know from a technology perspective that healthcare can become interoperable, but providers have to be willing to use and trust the data that already exists. For many providers, there is too much risk in trusting health data from another provider and this can be quite understandable considering the potentially high litigious risks for the providers due to malpractice.
3. Which HIE type has more momentum: private HIEs or public HIEs? Which is more important to quality patient care?
Private HIEs have the most momentum, largely due to the factors I mentioned above. Private HIEs own the data, so they trust the data, especially in the large healthcare organizations with matching applications.
I think we’re still a long way off from a large, public-based exchange of data and I think more government involvement is needed to push it further into reality. I think data for research, as opposed to patient care, will be exchanged first, but I am hopeful data for use in patient care will follow soon after.
It’s going to take a very long time, probably five to 10 years, before we see a nationwide HIE. Health IT technology evolves at the pace that it can evolve – changes to integrated technology with radiology and other modalities cannot be changed overnight and hospitals have huge investments in these areas.
4. Since HHS designated Consolidated CDA as the format for the clinical exchange of health information, how are you helping clients exchange data in this format?
It’s still very much in the “talking phase.” I am helping customers begin using CCD, especially in the transfer of care, and I work with other customers on the implementation of other CDA documents.
I will tell you this: I’ve gone into hospitals from coast to coast, and I stand in front of the entire healthcare organization and ask, “Does anybody know what a CCR is?” And rarely do I see one hand go up. When I begin to explain CCR, they tell me, “Speak English.”
Despite the lack of awareness, I still promote CCD and CDA on a daily basis everywhere I go because we have to get people up to speed with changes that need to be made. Providers seem to trust the technology, but they just don’t yet know what it means.
5. Are IHE-based profiles beginning to catch on for externally communicating data, or are VPNs still the predominate method?
IHE profiles have been slow to be implemented; in fact, I have yet to see an implementation that is comprehensive. VPNs are by far the most used communication method.
You can find some implementations of XDS (a Patient Care Coordination profile) in the following Google map that was created and is maintained by Keith Boone, who first published it in his post, “Where in the World is XDS?”
I’ve worked with healthcare organizations that don’t know what they need to do for Meaningful Use Stage 1. It’s technologists who know what’s going on and I see a few select health IT professionals who are up to speed on IHE profiles.
VPNs are very predominant because organizations are very sensitive and protective of their healthcare data. Plus, the word “cloud” really makes many in healthcare panic.