Back

Integrating Local Vocabulary into OMOP CDM: A Step-by-Step Tutorial

Park, Y.; Yoon, J.; Zhuk, A.; Ostropolets, A.; You, S. C.

2025-05-08 health informatics
10.1101/2025.05.07.25327200 medRxiv
Show abstract

ObjectivesThis tutorial provides healthcare informatics researchers with a practical guide for integrating local vocabularies into the Observational Health Data Sciences and Informatics (OHDSI) Standardized Vocabularies, through the example of Korean Electronic Data Interchange (EDI) codes. MethodsWe present a step-by-step guide for integrating local vocabulary. The methodology includes four sequential stages: (1) comprehensive and continuous collection system for longitudinal EDI codes from November 2000 to May 2024, (2) mapping to OHDSI Standardized Vocabularies, (3) validation and integration through the OHDSI Community Contribution Pipeline, and (4) publication into the Athena, the official OHDSI vocabulary site. ResultsBy following this tutorial, researchers can achieve integration of local vocabulary into the OHDSI Vocabularies. Of the total 620,642 EDI codes collected, we successfully integrated 437,807 codes (70.5%) into the OHDSI Vocabularies, with 376,111 (85.9%) of these integrated codes mapped to standard concepts. The integrated vocabulary passed all Community Contribution Pipeline requirements and was published in September 2024 on the official OHDSI vocabulary website. ConclusionThis tutorial provides healthcare informatics researchers with practical guidance for vocabulary standardization, enabling local healthcare data integration into the OMOP CDM. The approach can be adapted for various local vocabularies, fostering international healthcare data interoperability.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.