Building the observational medical outcomes partnership's T-MSIS Analytic File common data model

Inform Med Unlocked. 2023:39:101259. doi: 10.1016/j.imu.2023.101259. Epub 2023 May 5.


Objectives: This effort used Databricks to create an Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) for Transformed MSIS Analytic File (TAF) Medicaid records.

Materials and methods: Our process included data volume and content assessment of TAF, translation mapping of TAF concepts to OMOP concepts and the creation of Extract Transform and Load (ETL) code.

Results: The final CDM contained 119,048,562 individuals and 24,806,828,121 clinical observations from 2014 through 2018.

Discussion: The transformation of TAF into OMOP can support the generation of evidence with special attention to low-income patients on public insurance. Such patients are perhaps underrepresented in academic medical center patient populations.

Conclusion: Our effort successfully used Databricks to transform TAF records into OMOP CDM. Our CDM can be used to generate evidence for OMOP network studies.

Keywords: Centers for Medicare and Medicaid Services; Common Data Model; Medicaid; Observational Health Data Sciences and Informatics; Observational Medical Outcomes Partnership.