Announcing Apelon’s Reformatted MedDRA Data for DTS

We just overhauled the way we represent MedDRA in DTS.

The Medical Dictionary for Regulatory Affairs (MedDRA) is produced by the MedDRA Maintenance and Support Services Organization (MSSO) on behalf of the International Council on Harmonization (ICH), and is used in the United States, Europe and Japan to report and analyze drug-related adverse events. For example, when you file a report with the US Food and Drug Administration’s Adverse Event Reporting system (AERS), the report is coded with MedDRA. MedDRA use requires a license.

Looking at MedDRA as a terminologist, I see some things I really like and some things I wish were different. MedDRA uses a fixed-depth hierarchy of five levels. While this provides a nice consistency for users, it means that you have to shoehorn everything into that structure. If one kind of disease categorization is more complex than another, too bad! Everything has five levels. MedDRA also includes the Standardized MedDRA Queries (SMQs), which are designed to help users define major and minor adverse events. SMQs are a unique feature of MedDRA, and combine straightforward value sets with simple-but-powerful algorithms to power analytic applications. Other terminologies could learn from this idea! Another point of frustration for me is the way MedDRA’s Preferred Terms, which are most likely to be used for reporting, are supplemented by a large number of additional entries known as Lowest Level Terms (LLTs). In some cases, the LLTs are true synonyms (different words that mean the same thing), but in others, the LLTs are in fact more specialized “children” of a PT, but grouped under the PT for purposes of aggregation and analysis. I wish the different types of LLTs were explicitly called out.

The new DTS representation of MedDRA (see the screenshot from DTS Editor) includes all the LLTs as Terms attached to their PTs, which are represented as Concepts. We also include a separate top-level hierarchy for the SMQs, with their members attached via DTS Associations. This make it easy to fetch SMQ members, navigate the MedDRA hierarchy with an SMQ filter, and embed MedDRA in downstream applications and analytics tools (assuming you have the correct license to do all that!). We also include a way to use the MedDRA Preferred System Organ Class, a special “skip-level” hierarchical construct, via a separate DTS Association type.

That’s a lot of acronyms and details, and getting the most out of MedDRA certainly requires an understanding of its unique features. However, I’m really proud of the content team’s new data model: it faithfully represents the MedDRA data, and it provides the flexibility for users to do more with MedDRA, and more easily. For more info, contact me via jcarter@apelon.com.