Skip to main content
AMIA Annual Symposium Proceedings logoLink to AMIA Annual Symposium Proceedings
. 2006;2006:1057.

Implementation of RxNorm as a Terminology Mediation Standard for Exchanging Pharmacy Medication between Federal Agencies

Fola Parrish 1, Nhan Do 1, Omar Bouhaddou 2, Pradnya Warnekar 2
PMCID: PMC1839491  PMID: 17238676

Abstract

The federal government is working toward its goal of achieving interoperability between health information systems through several multi-agency efforts. While some interoperability partnerships exist between federal agencies, only a few systems are involved and these projects have proven difficult to implement. This paper describes the process of implementing an interoperable standard for exchanging computable pharmacy data between the Department of Defense (DoD) and the Department of Veterans Affairs (VA).

Background

In the past year, the DoD and the VA have made significant progress toward interoperability through several health data-sharing projects such as the Clinical Health Data Repository (CHDR) initiative. CHDR is a combined effort between the DoD and the VA to enable the exchange of computable patient data in the pharmacy, allergy and clinical laboratory domains. The first phase of CHDR developed a pharmacy prototype. The second phase of CHDR builds on the pharmacy prototype and adds the clinical laboratory domain.

The pharmacy prototype supports the transfer of coded patient data including: patient demographic data, medications, and drug allergy lists from one agency national repository to another. Once the data is transferred, it is made available for pharmacy order entry and decision support functions of each agency’s native pharmacy system.

Mediation based interoperability

Both agencies agreed to the following activities:

  1. Standardization and normalization of existing local and proprietary drug terminologies across care sites along with a central enterprise maintenance process.

  2. Selection of a Consolidated Health Informatics (CHI) terminology standard with the appropriate level of granularity and content coverage for pharmacy mediation. RxNorm Semantic Clinical Drug (SCD) names and codes were agreed upon as the mediation terminology.

  3. Complete mapping of each agency’s native drug terminology concepts to RxNorm SCD concepts. Mapping was established by the National Library of Medicine (NLM) for the VA, while the DoD’s mapping was developed in-house.

  4. Comparison of the two mapping files by independent subject matter experts. Both agencies conducted an independent review and validation of common RxNorm SCD concepts in mapping files to ensure that medication names refer to identical products across VA, DoD, and RxNorm descriptions.

Results

Review of common concepts in agency mapping files revealed a few mapping errors (less than 3%) which were discussed and corrected by VA and DoD CHDR terminologists. The interoperability success rates for medication exchange were estimated based on common concepts and frequency of these concepts in the respective agency data repository. We found that 97% of the DoD’s prescribed medications will be exchanged with and used by the VA, while 96% of all VA prescribed medications will be exchanged with and used by the DoD.

Lessons learned

  • - Domain expertise and terminology informatics skills are needed for implementation.

  • - Systems should not replace the clinician’s standard practice of obtaining an oral history or confirmation of the patient’s medication and allergy profile.

  • - When mediation of coded data fails, exchange of textual information can still provide clinicians with important information in cases where patient cannot provide an oral history.

  • - Use of RxNorm as a mediation terminology partner provides quality assurance mapping and a scalable strategy over direct mapping between agency files.

Conclusion

Mediation based interoperability is a journey. Even though 100% interoperability for medication exchange is not yet achieved, patients can still receive several benefits from computerized decision support. Incomplete computable pharmacy data exchange may be an acceptable risk as long as clinicians are made clearly aware of this limitation.


Articles from AMIA Annual Symposium Proceedings are provided here courtesy of American Medical Informatics Association

RESOURCES