STATUS: Active Project
Common data models (CDMs) are critical for patient-centered outcomes research (PCOR) studies that use real-world data (i.e., data generated outside the settings of clinical trials). CDMs are used to standardize the structure, format, and terminologies for data, enabling researchers to collaborate, share data, and conduct analyses across different datasets. However, CDMs are not necessarily interoperable; researchers often need to transform data standardized in one CDM format to another CDM prior to use in research. To address this challenge and facilitate interoperability of data across networks, this project builds on a previous FY2017 OS-PCORTF project that harmonized different CDMs to create Code Map Services, which will provide researchers with an automated tool to transform data collected and stored in one CDM format (i.e., HL7 FHIR [Fast Health Interoperability Resources], OMOP, Sentinel, PCORNet, i2b2/ACT, TriNetX, CDISC SDTM) into another CDM format. Harmonization of Various Common Data Models and Open Standards for Evidence Generation Code Map Services will enable translation between CDMs (similar to the “Google Translate” tool for human languages), improves interoperability and expands the data accessible to PCOR researchers.
An initiative of multiple HHS agencies, Code Map Services will establish a source of truth within a sustainable ongoing government supported infrastructure, NCI’s Cancer Data Standards Registry and Repository (caDSR). NIH/NCATS will be responsible for implementation of CDM transformation services with NIH/NCI working on the caDSR extension, metadata curation, and the new application programming (API) services to support the CDM transformation services. FDA on the other hand will do the FHIR to CDISC SDTM mapping and the implementation of this mapping to enable direct submission of clinical trials data from any CDM to the FDA.
The Code Map Service will facilitate CDM network interoperability, thereby expanding researchers’ access to clinical data that can be used for addressing new questions across a range of populations and settings. Establishing Code Map Services as a resource for investigators will also improve the quality of research, decrease duplicative efforts, and lower the effort required by data coordination centers to initiate research as well as submit results by to improving the quality and efficiency of model-to-model mapping.
Code Map Services facilitate the use of open, consensus-based standards, consistent with HHS’s recognition of the need for greater interoperability of data and sharing of resources. This project will create an automated set of code mapping services of common CDMs (OMOP, Sentinel, PCORNet, i2b2/ACT, TriNetX) and tools to support cross-network and cross-CDM data sharing for patient-centered outcomes research. Through the collaboration of FDA and NIH, this project will leverage the investments of HHS agencies, including OS-PCORTF investments, and private investments in Real-World Data (RWD) sources.
- Develop Common Data Model Interoperability Services
- Improve quality and efficiency of model-to-model mappings
- Extend the functionality of Common Data Models to allow submissions of clinical trials data
- Disseminate CDM interoperability data infrastructure tools