Professor Evgeniy Krastev and Emanuil Markov from Technical University of Sofia discuss their journey into mapping the Bulgarian Diabetes Register onto the OMOP CDM, derived from the Bulgarian National Health Insurance Fund (NHIF). In a very comprehensive presentation, they initially review Bulgaria and its healthcare system for background, before discussing their data source harmonisation, the ETL process, and the utilisation of natural language processsing to extract clincal data from parts of their unstructured dataset (in XML documents of course in Bulgarian), and then outlining their future proposed work.
One of the main challenges which stimulated discussion is how best to utilise drug coding, in particular RxNorm and ATC coding, with a custom solution, and whether this or other options would have been optimal. Expansion of this dataset in Diabetes, and the two recent publications, are highlighted at the end. Acknowledgements also to MedicalScan Ltd, in Hungary, their chosen EHDEN-certified SME, who collaboared with the Technical University on the ETL work.