June 25, 2003

The Honorable Tommy Thompson
Department of Health and Human Services
200 Independence Avenue SW
Washington, D.C. 20201

Dear Secretary Thompson:

The National Committee on Vital and Health Statistics (NCVHS) commends you for your commitment to move toward uniform government adoption of clinical data standards that you announced on March 21, 2003. NCVHS recognizes and appreciates that there is new momentum to adopt clinical data standards that is driven by you and the Consolidated Healthcare Informatics Initiative (CHI). Consequently, NCVHS is now working closely with CHI to study, select and recommend patient medical record information (PMRI) standards. Given your strong interest in adopting PMRI standards, we thought you should be aware of the progress that NCVHS is making in this area and of some observations that are likely to influence our next set of NCVHS clinical data standards recommendations.

NCVHS began the process of studying, selecting, and recommending patient medical record information standards under the directive of the Health Insurance Portability and Accountability Act of 1996. Accordingly, the Committee completed the NCVHS Report on PMRI Standards in August 2000. This report set forth a framework and guiding principles for the selection of these standards. NCVHS then employed these guiding principles to select and recommend PMRI Message Format Standards to you. The Department adopted them as part of the first set of clinical data standards in your announcement of March 21, 2003. At the current time, NCVHS is evaluating PMRI terminologies, which we plan to recommend as clinical data standards for your adoption.

Here is a brief review of the status and direction of NCVHS activities that will lead to recommendations for PMRI terminology standards:

  • In August 2002, NCVHS convened 1-1/2 days of hearings to solicit guidance from healthcare industry experts regarding the need, the scope, the priorities, and the criteria for selection of PMRI terminology standards. Testifiers recommended that the committee focus on identifying those PMRI terminologies that can serve as a stable and consistent core and then identify domain-specific terminologies that can be referenced or mapped to the core.
  • NCVHS then prepared a comprehensive questionnaire to solicit information from all terminology developers that wished to have their terminologies considered for selection as PMRI terminologies. This questionnaire reflected the guidance we received from the testimony in August and NCVHS criteria for selection of PMRI terminologies. We received more than forty responses to this comprehensive questionnaire. The responses were analyzed to determine which terminologies were technically qualified to be recognized as part of the core PMRI terminologies. This initial set of candidate terminologies was included in a preliminary report that was reviewed by NCVHS on March 25-26, 2003.
  • NCVHS solicited additional information from terminology developers and then updated the preliminary analysis, generating a revised list of candidates in a second draft report that was distributed to the Standards and Security Subcommittee on April 17, 2003. The revised list identifies 12 terminologies that meet NCVHS technical criteria for selection as a core PMRI terminology standard.
  • NCVHS distributed this second draft analysis on May 2, 2003, to all terminology developers to solicit a final set of reviews, comments, and corrections.
  • NCVHS invited users of the 12 terminologies that meet the technical criteria to serve as core PMRI terminologies to testify to NCVHS on May 21-22, 2003. In total, twenty-four users, including representatives from healthcare provider organizations, healthcare information system vendors, and developers of related terminology services testified to NCVHS.

Although it is premature at this time for NCVHS to recommend which terminologies should be selected as the initial set of core PMRI terminologies, which terminologies should be recognized as valuable domain-specific terminologies, and in which context they should or should not be used, it might be valuable for us to share with you the following observations that are likely to influence our recommendations.

  • In the August 2002 and May 2003 NCVHS hearings, testifiers expressed overwhelming support for the government to take leadership by identifying a core set of non-redundant clinical terminologies for healthcare.
  • In the May 2003 NCVHS hearings, terminology user testifiers expressed broad support for the adoption of SNOMED-CT and the laboratory portions of LOINC as part of an initial set of core PMRI terminologies. The support for SNOMED CT was dependant on clarification of licensure costs.
  • Testifiers also indicated that NCVHS should recommend an appropriate set of drug and medical device terminologies as part of the core set of PMRI terminologies. However, there was not a consensus as to which drug or device terminologies should be selected. NCVHS intends to explore further these two areas in hearings planned for August.
  • After the initial set of core PMRI terminologies is selected the issues of content overlap, interfacing, integrating, and mapping, as well as maintenance and distribution, will be investigated and addressed in the September recommendations and/or subsequent reports.

NCVHS plans to present its final recommendations for the set of core PMRI terminologies to you at the end of September 2003. In the meantime, we hope the information in this progress letter will be of value to you for anticipating next steps toward building this central component of the National Health Information Infrastructure.



John R. Lumpkin, M.D., M.P.H.
Chair, National Committee on Vital and Health Statistics

Cc: HHS Data Council Co-Chair