Success and Challenges Querying OMOP-transformed EHR Data from Different Healthcare Organizations

Adoption of a common data model (CDM), such as the Observational Medical Outcomes Partnership (OMOP) CDM, is a critical component of health information exchange, but the extent to which CDMs facilitate patient-level interoperability is unclear. We sought to determine the feasibility of using a CDM for health information exchange between two healthcare organizations.

We executed a single Structured Query Language (SQL) query on OMOP-transformed data from the University of California, San Francisco (UCSF) and the San Francisco VA Healthcare System for five patients.

The SQL query was successfully executed, and complementary healthcare information was obtained for five of five patients, but interoperability was limited by (1) lack of uniform patient identifiers, (2) use of different coding vocabularies, and (3) variations in mapping of source data to the CDM.

Although transformation of EHR data to a CDM can facilitate health information exchange, our study suggests that patient-level interoperability of a CDM may require further alignment of both semantics and syntax.