Guest blog by Michael Totzke, 3M data analyst.

At the AMIA 2014 convention in Washington D.C., we showcased some of our processes for mapping and maintaining RxNorm drugs into the Healthcare Data Dictionary (HDD). Our poster and podium presentations emphasized the fact that with clinical data, accurate and consistent mapping of terminology standards over successive versions is critical. With the selection of RxNorm as the drug terminology standard required to meet Meaningful Use criteria, it has become necessary for the HDD to maintain RxNorm’s drug data from a longitudinal perspective. Our former process for maintaining RxNorm dealt solely with the mapping of the current version, with limited regard to managing changes in RxNorm’s data over time. However, it’s not just the initial mapping that is important; having a long term strategy for maintaining that terminology within a larger terminology server is crucial for ensuring data quality. Medical terminologies change over time, and there is no algorithm yet that can alone guarantee the level of accuracy required for exchange of clinical data.

Mapping in an automated or programmatic fashion increases efficiency by making the volume of concepts that need to be manually mapped by a clinical terminologist as small as possible. A variety of strategies must be used, including one aspect of mapping that the HDD is well known for: curation of clinical data by subject matter experts. Clinical data mapping requires knowledge in both the clinical content area, such as pharmacy, and also with the structure and use of the terminology standard. Lack of expertise in either of these facets introduces risk when a patient’s data is applied to their record.

As more of our customers begin to use RxNorm to exchange clinical data, they trust that it is mapped accurately and consistently, and will continue to be over time. The HDD meets their data exchange needs, particularly now that hospitals and providers are attempting to meet Meaningful Use requirements.

The presentation we gave detailed some of the lessons we learned when we updated our approach maintaining RxNorm mapping within the HDD. The discussion emphasized the importance of providing historically compatible and computable data, thus maintaining continuity of patient care.

Michael Totzke is a data analyst with 3M Health Information Systems.