Considerations for the Representation of Meta-Data for the Support of Structured Data Entry

Abstract

Research and decision support require patient data to be structured. Flexible data representation, to cope with expansion and new recording demands, can be achieved through abstraction of the data model. Such a model, however, does not include explicit information for the support of structured data entry (SDE), i.e. the definition of what is relevant to say in a specific context. The objective of this paper is to provide considerations for an intuitive conceptual representation of such meta-data for domain experts. The content of meta-data for SDE is compared with that of controlled medical terminologies in the context of its purpose. Upon analysis of medical descriptions, a network and a tree both emerge as an intuitive structure to represent meta-data for the support of SDE. The pros and cons of a tree versus a network are dominated by the challenges of representing multiple contexts for description. Meta-data for SDE only partially overlap with the content of controlled medical terminologies. Integration of meta-data for SDE and controlled terminologies would benefit standardization and retrieval, but the requirements for flexible data entry are at odds with the rigidity and limited granularity of a standard. The preferred strategy seems to be a mapping between the concepts from a terminology standard and meta-data for SDE.