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

Area of Science:

• Health Informatics
• Data Science
• Public Health

Background:

• Medicaid data, specifically the Transformed MSIS Analytic File (TAF), contains valuable information on public insurance beneficiaries.
• Existing data structures may not be optimal for large-scale, multi-site research, particularly for underrepresented patient populations.

Purpose of the Study:

• To transform TAF Medicaid records into the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM).
• To create a standardized dataset suitable for generating real-world evidence, with a focus on low-income patients on public insurance.

Main Methods:

• Utilized Databricks for data transformation.
• Conducted data volume and content assessment of TAF.
• Developed translation mappings from TAF concepts to OMOP concepts.
• Created Extract Transform and Load (ETL) code for data conversion.

Main Results:

• Successfully created an OMOP CDM from TAF records.
• The final CDM encompasses 119,048,562 individuals and 24,806,828,121 clinical observations (2014-2018).

Conclusions:

• The transformed TAF data in OMOP CDM format facilitates evidence generation for OMOP network studies.
• This effort enhances research capabilities for understanding the health outcomes of low-income populations covered by public insurance.