Terminology Binding
Introduction
The term 'binding' loosely denotes the structural relationship between what terminologists call 'the information model' (largely modelled in archetypes in openEHR) and terminology itself.[IMPORTANT NOTE TO TERMINOLOGY PEOPLE: in these pages, the openEHR where you see the word 'archetype', we mean 'information model'!].
There are three main types of relationship between the two:
- the structure of attributes in archetypes with respect to terminology (e.g., should we have the 3 attributes 'procedure', 'procedure site', 'procedure protocol', each coded or partially coded, or should it be a single attribute, whose value is a single post-coordinated code?);
- if and how the semantic names (not screen names) of single nodes, or 'data points' in an archetype, such as 'Apgar result / 1 minute / breathing value' are mapped to single codes in a terminology, including conditions based on data values, e.g. gender, age etc;
- if and how the values of certain data fields are coded by 'subsets', nowadays known as intentional reference sets, or 'ref sets' for short.
The first of these is potentially the most important in the design sense, since it influences the design orf archetypes. In these pages, we have denoted this 'IM&T equivalence' (Information Model and Terminology equivalence). In real systems, it is the use of ref-sets and their relationship with archetypes that is probably the most important to users, since ref sets are usually what appears on the screen.
See various sub-pages for the details.
Presentations
This presentation (pptx, pdf) is from the recent IHTSDO Technical Committee meeting at end of April, 2010.