Development of an Ontological Model of Evidence for TRANSFoRm Utilizing Transition Project Data Derek Corrigan 1 , Jean-Karl Soler 2 , Brendan Delaney 3 1 Royal College of Surgeons in Ireland, Dublin, Ireland derekcorrigan@rcsi.ie 2 Mediterranean Institute for Primary Care, Attard, Malta jksoler@synapse.net.mt 3 Kings College London, London, United Kingdom brendan.delaney@kcl.ac.uk Abstract. The development of decision support tools that assist clinicians effectively practice evidence-based-medicine in primary care is dependent on the development of formal models of clinical knowledge. These formal models are a pre-requisite for bridging the knowledge gap that exists between generation of research knowledge and its application in clinical practice. The TRANSFoRm project has developed formal ontological models to represent diagnostic clinical knowledege providing a basis for future development of diagnostic decision support tools. The conceptual validity of the developed models has been tested through representation of diagnostic clinical evidence obtained from literature sources and International Classification of Primary Care Second Edition (ICPC2) coded clinical evidence captured as part of the Transition project. The models provide a basis for future development of decision support tools as part of the on-going TRANSFoRm project. These tools can assist clinicians to formulate and quantify potential diagnoses based on diagnostic cues extracted from patient electronic health records. Keywords: Ontology, Semantic Web, Evidence-Based-Medicine, Electronic Health Record, Decision Support 1 Introduction The application of systematic and rigorous approaches to diagnosis through access to the latest available clinical research has long been advocated as one way of contrib- uting to improving patient safety in family practice. The term ‘evidence based medi- cine’ has been widely associated with such approaches [1]. The effective practice of evidence based medicine implies the existence and use of an up-to-date repository of clinical knowledge. This can be used for interpretation of the diagnostic cues associ- ated with a presenting patient (whether or not this evidence be in electronic format or written) [2]. The challenges in keeping a repository of diagnostic information up to