ByChristian Knell

First technical paper about federated analyses on the MIRACUM data with the OMOP common data model has been published

The OMOP common data model and the OHDSI tools are used as a core element in MIRACUM for developing a common data model, harmonizing terminologies as well as facilitating the federation of research queries/analyses across institutions. Already during the BMBF MI-I conceptual phase, the Chair of Medical Informatics Erlangen around Prof. Prokosch developed an OMOP/OHDSI-based pilot implementation and supported its deployment at the consortium´s eight university hospitals.

To this end, the OMOP vocabularies and terminological mappings were analysed for their applicability on the German dataset/terminologies and adaptions were implemented were required for importing the facts into the OMOP common data model. Then, the process for importing the data from the source files was designed and distributed among eight German university hospitals in form of a preconfigured virtual machine with the OMOP database and the OHDSI tools as well as the jobs to import the data and conduct the analysis. Finally, the generated results from federated/distributed queries focusing on colorectal cancer yielded 18,000 patient cases visualized in a SunBurst plot. All in all, the data of in total 400,000 inpatients with more than 900,000 visits and 1 million conditions and procedures have been imported in those MIRACUM pilot DIC research data repositories.