Clinical decision support (CDS) can improve care by delivering the right clinical guidance at the right time to the right care provider given the context of care.

Moreover, evidence-based portable CDS knowledge artifacts empower health care organizations to more rapidly and efficiently share CDS knowledge. To be truly effective, however, one must consider both the representation of the clinical content and the capabilities of the systems into which such artifacts are deployed.

Integrating CDS content into electronic health record (EHR) systems has historically been an onerous and highly manual task. Once available in the EHR system, the cost and effort involved in updating the clinical guidance results in CDS content eventually becoming stale and unusable as organizations forego maintenance over budget concerns and competing demands for resources. Ultimately, the barriers to importing and maintaining CDS artifacts cause clinical care to lag behind advances in evidence-based medicine.

Further complicating matters, CDS content has been near impossible to share because standard, expressive, computable models for it did not exist; this lack of content portability hinders greater EHR interoperability. The recent introduction of standards-based and computable knowledge artifact specifications is beginning to address this limitation.

Recently, Cognitive Medical Systems and Motive Medical Intelligence demonstrated to the Department of Veterans Affairs (VA) and the Office of the National Coordinator (ONC) for Health Information Technology the most comprehensive DSTU HL7 standard to date allowing for sharing and portability of knowledge artifacts between proprietary systems.

Developed in response to an ONC Standards & Interoperability initiative, the Health eDecisions (HeD) Knowledge Artifact specification supports an XML message format and serialization representation for order sets, simple event-condition-action rules, and structured documentation templates (e.g., forms) that ensure content semantics, behavior, and expressiveness can be retained in transmission.

“We demonstrated how content can be brought into a system via standard CDS structures, such as a documentation template,” said Claude Nanjo, Cognitive Medical Systems’ Chief Scientist. “We then rendered the document and translated its behavior so a clinician could fill it in and submit it thus demonstrating how a portable knowledge artifact can be made actionable at the point of care.”

“Our demonstration project showed how clinical content intended to support a real-world clinical scenario, which in this case was suicide risk assessment and management, could be created by a content authoring organization and deployed accurately in a proprietary clinical system,” added Julie Scherer, Chief Data Scientist at Motive Medical Intelligence.

The demo validates that the HeD specification is a sufficiently expressive standard to facilitate structured knowledge capture and content integration. This, in turn, provides more robust knowledge interoperability among health care systems by enabling the import and maintenance of clinical knowledge artifacts, ultimately lowering the traditionally steep barriers to CDS adoption.

In the future, if EHRs were built to support it, the HeD knowledge artifact specification could pave the way for automatically integrating CDS content into clinical workflows and sharing content between systems. This means content vendors could write CDS content once and interpret and deploy that content across EHRs. It also means that health care organizations could purchase content once and customize it for varying internal applications and local needs.

For healthcare providers the implications include lower cost to institute and maintain CDS, faster delivery of the latest clinical guidance through automatic content updates, and more sharable content with greater integrity.

Nanjo and Scherer hope the ONC and Centers for Medicare and Medicaid Services will make HeD and the knowledge artifact specification part of Meaningful Use to provide a standard interface for EHR systems to exchange CDS knowledge artifacts.