Feasibility of Converting EMR Data to OMOP CDM and Utilizing OHDSI Analysis Tools in Japan

The study investigates the feasibility of converting electronic medical record (EMR) data to the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM) and utilizing Observational Health Data Sciences and Informatics (OHDSI) analysis tools in Japan. Data from 8,387 breast cancer patients at the National Cancer Centre Hospital East were extracted and transformed into OMOP CDM. OHDSI software and standard vocabulary were installed, and the transformed data were displayed on ATLAS for analysis. The quality of the converted datasets was assessed using the Data Quality Dashboard developed by the OHDSI community. Mapping to OHDSI standard vocabulary was performed for 7,259 terms, with SNOMED-CT, RxNorm, and LOINC frequently used. The conversion program was applied to all cases, but there were challenges, such as race data and anticancer drug specification selection. ATLAS displayed reports on the observation period and drug use. The study demonstrated the feasibility of converting EMR data to OMOP CDM and utilizing ATLAS in Japan. However, there were challenges in software installation, vocabulary mapping, and data quality evaluation.