Requirements-Value Set Content

From HL7Wiki
Jump to navigation Jump to search

Value set content defines an enumeration or expression that resolves to a set of codes from one or more code systems.

Content

Requirement Value Set content definitions may have a number of different types of annotations
Rationale See rationales for individual annotations types
Implementation


Requirement Content definitions need to be able indicate which code system the content is being drawn from.
Rationale The code system determines the codes, properties, qualifiers and other relationships that can be used to define content.
MIF mif-vocabulary-model.xsd/ContentDefinition/drawnFromCodeSystem
OMG Mapping Tagged value of a ValueSet stereotype.


Requirement Content definitions need to be able indicate which code system supplement(s), if any, they are based on. the content is being drawn from.
Rationale Code system supplements provide additional properties, relationships and representations that may be used to define value set content.
MIF mif-vocabulary-model.xsd/ContentDefinition/usesCodeSystemSupplement
OMG Mapping Tagged value of a ValueSet stereotype.


Requirement Content definitions need to be able to identify which types of representations are allowed for the concepts selected by the content definition.
Rationale Some code systems may have multiple codes and/or multiple print names for a given concept. When defining the value set, there's often a need to constrain to only a single code or print name. For example, with ISO country codes, there might be a desire to constrain to only the 2-character codes. Or for multi-language code systems, to constrain to a particular language print name.
MIF mif-vocabulary-model.xsd/ContentDefinition/allowedRepresentation
OMG Mapping Tagged value of a ValueSet stereotype.

Filters

Requirement Valueset content must support complex structures involving combinations of other content structures such as unions, intersections and exclusions.
Rationale Many value set definitions are complex. For example "all specializations of code X or code Y, except specializations of code Z"
MIF mif-vocabulary-model.xsd/ContentDefinition/combinedContent
OMG Mapping Tag of a ValueSet stereotype that points to an expression giving the combination definition.


Requirement Content needs to be able to be defined by enumerating a set of allowed codes
Rationale For simple value sets or unsophisticated code systems, enumeration of the allowed set of codes is the best or only way to define value set content
MIF mif-vocabulary-model.xsd/ContentDefinition/codeBasedContent
OMG Mapping UML provides for the definition of enumerations.


Requirement Content needs to be able to be defined as a code and/or codes that have relationships of a particular type to that code, including such constructs as direct relations to the head code, the complete set of codes found by traversing relationships of the specified type or "leaf" code - those codes that are part of the complete traversal set that have no outbound relationships.
Rationale This is a common mechanism for value set definition in code systems that have parent-child or similar relationships.
MIF mif-vocabulary-model.xsd/ContentDefinition/codeBasedContent
OMG Mapping Tagged values of a ValueSet stereotype providing links to other code definitions. (?)


Requirement Content needs to be able to be defined based on properties held by the concepts. This might be inclusion of concepts that have specified values for certain properties or exclusion of concepts with specified values for certain properties. The values may either be strings or REGEX expressions
Rationale Examples include filtering on LOINC properties such as Class or B/R/O, or supplemental properties such as "orderable". For more complex code system properties, matches may need to be on such things as "starts with 'b'".
MIF mif-vocabulary-model.xsd/ContentDefinition/propertyBasedContent
OMG Mapping Tagged value of a ValueSet stereotype.


Requirement Content needs to be able to be defined based on the actual string value of the codes for the property.
Rationale Some code systems, such as the ICD code systems embed semantics in the codes themselves rather than using properties or relationships. As a result, the only way to define expression based content filters is with filters on the code.
Methodology Two types of expression languages are supported - regex and Terminology Query Language (TQL)
MIF mif-vocabulary-model.xsd/ContentDefinition/codeFilterContent
OMG Mapping Attached filters as constraints written in the appropriate language. (?)


Requirement Content needs to be able to be defined by referencing existing value sets
Rationale Some value sets are most efficiently composed by referencing other existing value sets. This supports re-use and simplifies maintenance.
MIF mif-vocabulary-model.xsd/ContentDefinition/codeFilterContent
OMG Mapping Tag of a ValueSet stereotype the points to other value sets. (Alternatively, perhaps UML specialization could be used, as long as the value set being defined inherits all the allowed values from the reused value set.)


Requirement In some cases, content needs to be defined using free text description.
Rationale Some content definitions cannot be defined by formal mechanisms due to limitations of the code system. In other cases the initial requirement is captured as text and is only later formally represented.
MIF mif-vocabulary-model.xsd/ContentDefinition/nonComputableContent
OMG Mapping Attach the content definition as a UML constraint specified using text.


Qualifiers

Requirement Valueset content that draws from code systems that support post-coordination needs to be able to indicate what types of qualifier relationships are allowed.
Rationale Post-coordination is complex and few systems can support (or want) all of the expressivity that post-coordination might allow. Therefore it is important to define what types of relationships can be used.
MIF mif-vocabulary-model.xsd/ValueSetVersion/allowedQualifiers
OMG Mapping [I do not understand what "post-coordination" is. -- EVS]


Requirement When identifying qualifiers, there's a need to identify what types of concepts are allowed as the target for a particular relationship.
Rationale When allowing qualifiers, it's not always sufficient to say "qualifier relationships of this type". For example, limiting which types of body sites can be communicated.
MIF mif-vocabulary-model.xsd/ValueSetVersion/allowedQualifiers/targetConcepts
OMG Mapping [I do not understand how "qualifiers" are being used in this context. -- EVS]


Requirement When defining support for qualifier relationships, the value set must be able to assert conformance and cardinality constraints.
Rationale Concepts can have multiple qualifiers of the same type. However, systems may not be designed to handle multiple repetitions or may have limits on how many repetitions they can manage. In some cases, there might be a need for a certain minimum number of repetitions to be present. Similarly, the expectation for support of a particular qualifier might be optional, required, or even not permitted. And some systems might declare that they'll tolerate the presence of a given qualifier but will ignore it if present.
MIF
  • mif-vocabulary-model.xsd/ValueSetVersion/allowedQualifiers/@conformance
  • mif-vocabulary-model.xsd/ValueSetVersion/allowedQualifiers/@minimumMultiplicity
  • mif-vocabulary-model.xsd/ValueSetVersion/allowedQualifiers/@maximumMultiplicity
OMG Mapping [I do not understand how "qualifiers" are being used in this context. -- EVS]


Requirement Valueset content that draws from code systems that support post-coordination needs to indicate whether the default assumption is that qualifiers not specifically referenced are allowed or not permitted.
Rationale It is not always convenient to explicitly list every qualifier association that is possible for a particular code system, but it is not safe to assume that non-mentioned qualifiers are allowed.
MIF mif-vocabulary-model.xsd/ValueSetVersion/@areBaseQualifiersUnlimited
OMG Mapping [I do not understand how "qualifiers" are being used in this context. -- EVS]


Requirement Valueset content that draws from code systems that support post-coordination and where concepts can be represented in both post-coordinated and pre-coordinated form need to indicate whether the preference for representing a concept is using the pre-coordinated or the post-coordinated form.
Rationale While some systems might easily support both types of representations, for other types of communications it will be easier if there is only a single way a particular concept is represented in an exchange.
MIF mif-vocabulary-model.xsd/ValueSetVersion/@postCoordinationPreference
OMG Mapping [I do not understand "pre-coordination" and "post-coordination" is. -- EVS]