is one part of the process of specifying
on the way that information is structured and represented.
Consistent representation is a prerequisite for effective retrieval and reuse of clinical record information. Requirements for reuse are many and varied, ranging from direct support for the care of the individual patient, through to aggregate analysis for research, statistics and audit. The common theme of these requirements is the need to retrieve particular items of information reliably and consistently, irrespective of the environment in which the data was entered and stored.
Since both the information model and
contribute to the processable meaning of an entry in a clinical record it is essential to manage the interdependencies between these two components.
Simple requirements can be addressed by specifying a value-set consisting of the permitted coded values that can be used in a particular field. However, effective representation of clinical records requires a rich information model and an expressive terminology.
Models such as
RIM provide the necessary structural flexibility and
provide expressivity. An inevitable side-effect of a richer approach to information representation is an increase in the interdependencies and overlaps between the information model and the terminology. In order to specify and validate consistent representation of meaningful clinical records,
must be applied to both the information model and terminology. These
must address all the facets of the model and terminology (e.g. including the use of
and the effect of modeled record structures). The
on information model and terminology components must be integrated, or bound together, in ways that ensure consistency, avoid ambiguity and minimize the number of different ways in which the same meaning may be expressed.
A terminology binding is an instance of a link between a terminology component and an information model artifact. Therefore, it is necessary to consider the representation of the required terminology components and the way these are associated with relevant information model artifacts.
The information model artifact to which a
is applied may be a field of a class in a static model or a collection of fields
of one or more related classes.
Bound components include:
- Information model artifacts:
- Coded attributes in an Version 3 model, an Archetype or in the proprietary information model of an operational application.
- Terminology components: