Common data models and data standards for tabular health data: a systematic review

The use of health data supports knowledge-based decision-making in healthcare. Common Data Models (CDMs) and data standards facilitate the integration of diverse data sources and enable federated analysis by harmonizing data formats and terminologies. To determine the best approaches to harmonizing patient data, we undertook a comprehensive literature search, which allowed us to identify the most popular and established CDMs (i2b2, Sentinel CDM, PCORnet CDM, OMOP CDM) and data standards (CDA, HL7 version 2, FHIR, openEHR). We established a set of criteria across the categories of Suitability, Popularity, Adaptability, Interoperability, and Support. The CDMs and data standards are evaluated based on the defined criteria. Overall criteria the OMOP CDM and FHIR scored best. We highlight the strongest CDM and data standard for each criteria category. Given the unique characteristics, strengths, and weaknesses of each CDM and data standard, no single global representation can be selected. To promote broad adoption of CDMs and data standards, it is essential to enable transformation between different representations and utilize various formats within a single tool to facilitate their interoperability. Only then seamless data exchange and research across borders can be achieved. Not applicable.