Hi all,
We are preparing a community contribution to register Japanese standard vocabularies in Athena, including:
Drug vocabulary: YJ code mapped to RxNorm and RxNorm Extension
Laboratory test vocabulary: JLAC mapped to LOINC and SNOMED CT
Before submission, we would appreciate clarification on the following points.
Duplicate names
concept_name and synonym_name may be duplicated, although concept_code is unique.
This is due to constraints in the source vocabulary and cannot be modified.
Question:
Is it acceptable to register vocabularies in Athena with duplicate names?
Drug mapping granularity (YJ β RxNorm)
Question:
Is it acceptable to standardize mappings at the Clinical Drug level where possible, or should we prefer more granular mappings (e.g., Branded Drug) when available?
Laboratory test naming detail (JLAC β LOINC/SNOMED)
JLAC codes contain structured semantic information.
Question:
Do concept_name and synonyms need to explicitly reflect LOINC/SNOMED-level detail, or is it acceptable to rely on code-level semantics and mappings?
Technically, itβs possible, unless these are standard concepts.
The best practice is to preserve all concept attributes during mapping, i.e., prefer the most granular semantic equivalent available.
Usually, we leverage concept names provided by the source without any semantic modification. If the source gives us several terms for one concept code, we use the most specific as a concept name, and the rest are placed in the concept_synonym table. Also, just in case, if a concept name is in Japanese, you should assign concept name translated into English, and the original spelling goes to synonyms.
Iβd recommend you to contact our community contribution expert Alexander @Alexdavv, who can provide detailed answers to all your questions.
Hi, @m-khitrun
Thank you very much for your reply.
Although we still need to coordinate with the code issuer, the overall direction has become clear. I also now have a much better understanding of how to use synonyms.
Additionally, thank you for introducing Alexander.
Our contributions are just beginning, but we will continue to make steady progress while learning from everyone.
We look forward to your continued support!