Federal Terminology Distribution Framework Discussion

August 9, 2007 Meeting Summary

Thepurpose of this meetingwasto bring together representatives from Federal agencies (VHA, EPA, DoD, NCI, and CDC) who provide a suite of standard terminologies (or subsets of these terminologies) for use in their agencies health, environmental and related data systems.

The meeting was held at the National Library of Medicine on August 9, 2007 with the following participants:

Veteran’s Health Administration (VHA):Steve Wagner, Michael Lincoln (by phone)

Environmental Protection Agency (EPA):William Sonntag, Michael Pendleton

Department of Defense (DoD):Nancy Orvis (by phone)

National Cancer Institute (NCI):Denise Warzel, Margaret Haber (by phone)

Centers for Disease Control and Prevention (CDC):Steven Steindel, Tim Morris

National Library of Medicine (NLM):Vivian Auld

Each group presented an overview of their current terminology distribution systems including plans for the near future (presentations will be posted separately). This was followed by a discussion identifying where the various agencies needs overlap and what steps can be taken in the near future to develop a common solution for moving forward. This summary document presents the results of this discussion.

Identified Need / Agency Indication
VHA / EPA / DoD / NCI / CDC
Distributed Workflow / X / X / X
Subset Collaboration / X / X / X / X / X
Shared Common Terminology Services / X / X / X / X / X
SNOMED US Realm Editors / X / X
Extend CHI like process (Phase II) / X / X / X / X / X
Standard new term submission process / X / X / X / X
Terminology Discovery / X / X / X / X
International Cooperation / X / X / X / X / X
CENDI Interaction / X / X
Common solution to versioning / X / X / X / X
Shared Information Models / X / X / X / X
Basic Terminology Standards/Practices / X / X / X / X

Descriptions:

Distributed Workflow: includes collaboration on quality assurance for terminology distribution

Subset Collaboration: includes code set management at agency and partner levels, federated systems, common authoring tools

Shared Common Terminology Services: includes access and API services around terminology repositories, software service standards (identifier system for services, ontology services) and evolution to a federated system and tie-in to the semantic web

SNOMED US Realm Editors: includes subsetting and intelligent use of SNOMED, training people to participate in SNOMED editing and maintenance, common tool development

Extend CHI like process (Phase II): includes domain extensions including review of current use of ICD-9-CM versus SNOMED, possible domain packaging extensions similar to medications and allergies in Phase I, review of Phase I recommendations against present realities

Standard new term submission process: includes common authoring tools for use by subject matter experts (similar to specific SNOMED request above, but more general) and maybe considered part of subset collaboration

Terminology Discovery: includes knowledge of health, environmental and related terminologies across agencies, open programmatic access to terminologies

International Cooperation: need to make sure what we use fits into evolving international picture

CENDI Interaction: Refers to the Federal STI Managers Group ( and possible need to engage them

Common solution to versioning: express the need to have common methods to address versions and maybe included in Shared terminology services.

Shared Information Models: includes use of standardized common terminology identifiers to describe the information model (metadata), class, attributes and data descriptions, change models (versioning tie-in) developed to allow flexibility cross agency and time

Basic Terminology Standards/Practices: guiding principles and compatibility guidelines need to be developed or adopted based upon existing works such as Dr. Cimino’s “Desiderata for Controlled Medical Vocabularies in the Twenty-First Century”.

It was noted that a goal was a single public access method or point to all federal health, environmental and related terminologies. This access method or point should be available to all federal agencies, include the entire list of terminologies used by federal agencies, and allow the ability to at least download current versions of all terminologies and access other services as feasible.The structure of this system should allow others to build portals to the information regardless of the physical location.

Conclusion:

The large degree of commonality between the participating federal agencies became quite clear through the presentations and subsequent discussion. The group identified the need for threerelated approaches: 1) Metadata for each terminology structure should be assembled using standard concepts; 2) a federal Metadata Registry should be assembled to allow access to the metadata for each terminologystructure and 3) web based terminology services for distributing the terminologies needed by federal agencies. By developing these items agencies would be able to extractand use unique subsets of terminologies needed for their work. Federal agreement on the key issues for this process willenable the agencies to collaborate on terminology distribution and reduce redundancy in the community at large.

Next Steps:

The participants identified the next step as the formation of a task specific workgroup of federal representatives within an existing body such as FHA or CENDI to develop anapproach for distributing terminologies throughout the federal sector. The workgroup should be comprised of federal representatives from agencies that currently provide a suite of standard terminologies (or subsets of these terminologies) for use in their agencies health, environmental and related data systems. Their approach should be the development of a draft specification or standard which will then be shared outside the federal sector to ensure compatibility and collaboration where appropriate. Steps should be taken by the workgroup to reach out to agencies that were not present at the August 9th meeting to ensure the needs of all federal agencies will be met by this effort.

  • Ideally the HHS Office of the National Coordinator for Health Information Technology (ONC) would form this workgroup within the Federal Health Architecture (FHA) or another appropriate venue. The formation of this workgroup and the production of their draft standard should take place in as short a time as reasonable in order to take advantage of the parallel efforts taking place in the private sector.

This report will be widely distributed. Recipients will include:

  • HHS Office of the National Coordinator for Health Information Technology (ONC)
  • Federal Health Architecture (FHA)
  • Healthcare Information Technology Standards Panel (HITSP) for internal distribution
  • HL7 groups including the Vocabulary Technical Committee and the Government Project Special Interest Group.

Additionally participants in this meeting will share the document internally and with partners.

V0.1-DRAFT: For internal review only-