Abstract
The disorder chapter of Version 3 of the Read codes is a rich source of clinically derived terminology. The file structure has been designed to meet a clinical need to support both enumerated and compositional taxonomy. This requirement coupled with the maintenance of multiple classification necessitates a mechanism to identify consistent hierarchical placement and equivalence between concepts. The early work on the semantic definition of the disorder chapter, to support these requirements, is outlined and the attributes that have been found to be important are presented. We also describe different categories of completeness of definition that have been identified, and the scope of those that are likely to remain incompletely characterised.
Keywords
Controlled Medical Vocabulary - Knowledge Representation - Disease - Medical Record Coding - Taxonomy
