An Ontology-Based Architecture to Support Language Variants of Model-Driven Electronic Health Records

This paper describes the implementation of tools to support multiple language variants of the clinical information models that are used to define a model-driven EHR system. Beyond that, it describes how a complete EHR system can be created with multiple language variants, using the example of an EHR for clinical management of patients in a Fracture Liaison Service. A clinical information model, represented as ontology, was developed in the English language. This model was translated into Spanish and Mandarin, and the modeling tools were refined, on the basis of the experience gained. A workshop was then held, where participants used the EHR tools to create additional language variants in German, French, Portuguese, Arabic, Farsi, Urdu and Somali. The results from the workshop are presented here, with a brief summary of the lessons learned; further work will focus on improving the tools in response to those lessons.