Common Data Models and Data Standards for Tabular Health Data: A Systematic Review
Common data models and data standards for tabular health data:
Summary
This systematic review systematically evaluates Common Data Models (CDMs) and data standards necessary to integrate diverse data sources and enable federated analysis. The study concludes that OMOP CDM and FHIR scored best across various criteria, emphasizing that achieving seamless interoperability requires enabling transformations between different representations rather than relying on a single global model.
Details
The paper establishes the necessity of Common Data Models (CDMs) and data standards for supporting knowledge-based decision-making using health data. CDMs serve as conceptual frameworks to harmonize data from various sources, facilitating federated research analysis and result aggregation. Data standards include semantic ones (e.g., SNOMED CT), which focus on term meanings, and syntactic ones. This review compared major CDMs—such as i2b2, Sentinel CDM, PCORnet CDM, and OMOP CDM—and data standards like CDA, HL7 version 2, FHIR, and openEHR. Evaluation criteria included Suitability, Popularity, Adaptability, Interoperability, and Support. The results highlighted that the OMOP CDM and FHIR scored highest overall. However, the authors conclude that because each model has unique characteristics, no single global representation can be selected. To achieve truly seamless data exchange across institutions, it is essential to enable transformation between different representations and utilize various formats within a single tool. This approach addresses the complexity of data sharing (due to varied formats, terminologies, and scopes) by promoting interoperability based on FAIR principles.
Technology Note
FHIR（Fast Healthcare Interoperability Resources）は医療データ交換の国際標準。このエントリの関連技術: FHIR
Original content copyright by respective publishers