Using the OMOP data model
We want to be a part of this development, and the Cancer Registry of Norway has initiated and received funding to standardise the registry's data for the data model OMOP - Observational Medical Outcomes Partnership.
OMOP has been developed through an international project that is working to standardize data to a common format. In this way, it can be more easily used for collaboration, large-scale analyses and sharing sophisticated tools and methods. Varying formats of health data make it difficult to collaborate across organizations and nations. With a common data model, the starting point for sharing becomes better. The model is based on established coding systems and maps variables and structure to the OMOP model. Such a generic data model facilitates standardised analyses and sharing of analysis scripts across research networks. Central to the concept is that the data always remain locally stored and that only statistics/aggregated data go out from the Cancer Registry.
In the long term, we hope to be able to make data available based on this data model.
The figure shows an overview of which of the Cancer Registry's data have been mapped or planned to be mapped to OMOP. All core variables are mapped to the data model, and selected variables from the quality registries will be mapped in specific projects (SACT = Systemic anti-cancer treatment, PROs = Patient reported outcomes, dx = Diagnosis, Clinical registry = quality registries)
Ongoing OMOP projects:
FLORENCE: Federated Learning on Cancer Data
EHDS2 pilot
Completed OMOP projects:
AI solution for use in cancer operations under the EU program Interreg ØKS
