This page is part of the FHIR Specification (v5.0.0-ballot: R5 Ballot - see ballot notes). The current version which supercedes this version is 5.0.0. For a full list of available versions, see the Directory of published versions . Page versions: R5 R4B R4 R3 R2
FHIR Infrastructure Work Group | Maturity Level: N | Normative (from v4.0.0) | Security Category: N/A | Compartments: Not linked to any defined compartments |
A domain resource is a resource that:
As an abstract resource, this resource is never created directly; instead, one of its descendant resources (see List of Resources) is created.
This resource extends the base Resource. All of the listed Resources except Bundle, Parameters and Binary extend this resource.
Structure
Name | Flags | Card. | Type | Description & Constraints |
---|---|---|---|---|
DomainResource | «A»N | Resource | A resource with narrative, extensions, and contained resources + Rule: If the resource is contained in another resource, it SHALL NOT contain nested Resources + Rule: If the resource is contained in another resource, it SHALL be referred to from elsewhere in the resource or SHALL refer to the containing resource + Rule: If a resource is contained in another resource, it SHALL NOT have a meta.versionId or a meta.lastUpdated + Rule: If a resource is contained in another resource, it SHALL NOT have a security label + Guideline: A resource should have narrative for robust management Elements defined in Ancestors: id, meta, implicitRules, language | |
text | 0..1 | Narrative | Text summary of the resource, for human interpretation | |
contained | 0..* | Resource | Contained, inline Resources | |
extension | 0..* | Extension | Additional content defined by implementations | |
modifierExtension | ?!Σ | 0..* | Extension | Extensions that cannot be ignored |
Documentation for this format |
UML Diagram (Legend)
Structure
Name | Flags | Card. | Type | Description & Constraints |
---|---|---|---|---|
DomainResource | «A»N | Resource | A resource with narrative, extensions, and contained resources + Rule: If the resource is contained in another resource, it SHALL NOT contain nested Resources + Rule: If the resource is contained in another resource, it SHALL be referred to from elsewhere in the resource or SHALL refer to the containing resource + Rule: If a resource is contained in another resource, it SHALL NOT have a meta.versionId or a meta.lastUpdated + Rule: If a resource is contained in another resource, it SHALL NOT have a security label + Guideline: A resource should have narrative for robust management Elements defined in Ancestors: id, meta, implicitRules, language | |
text | 0..1 | Narrative | Text summary of the resource, for human interpretation | |
contained | 0..* | Resource | Contained, inline Resources | |
extension | 0..* | Extension | Additional content defined by implementations | |
modifierExtension | ?!Σ | 0..* | Extension | Extensions that cannot be ignored |
Documentation for this format |
Additional definitions: Master Definition XML + JSON + ShEx, the spreadsheet version & the dependency analysis, Resource Representation XML + JSON + Turtle
UniqueKey | Level | Location | Description | Expression |
dom-2 | Rule | (base) | If the resource is contained in another resource, it SHALL NOT contain nested Resources | contained.contained.empty() |
dom-3 | Rule | (base) | If the resource is contained in another resource, it SHALL be referred to from elsewhere in the resource or SHALL refer to the containing resource | contained.where((('#'+id in (%resource.descendants().reference | %resource.descendants().ofType(canonical) | %resource.descendants().ofType(uri) | %resource.descendants().ofType(url))) or descendants().where(reference = '#').exists() or descendants().where(as(canonical) = '#').exists() or descendants().where(as(canonical) = '#').exists()).not()).trace('unmatched', id).empty() |
dom-4 | Rule | (base) | If a resource is contained in another resource, it SHALL NOT have a meta.versionId or a meta.lastUpdated | contained.meta.versionId.empty() and contained.meta.lastUpdated.empty() |
dom-5 | Rule | (base) | If a resource is contained in another resource, it SHALL NOT have a security label | contained.meta.security.empty() |
dom-6 | Guideline | (base) | A resource should have narrative for robust management | text.`div`.exists() This is (only) a best practice guideline because:
|
The 'contained' mechanism is not permitted as a means of reducing server calls. I.e., It is not allowed to package a set of resources as one base resource and a set of contained resources to allow multiple creates or updates to happen in a single transaction, nor is it permitted to use 'contained' as a means of retrieving multiple resources in a single 'read'.
'contained' is only for use when there is a need to reference another resource, but the target resource does not have any independence from the referencing resource. Most commonly, this occurs when not enough information is known to allow uniquely identifying the target resource. For example, if a surgeon is known only as "Dr. Smith" (no full name, no license number, etc.), then there would be no way of determining which of the candidate "Dr. Smith" practitioners was being referenced. In that situation, it would be reasonable for the Procedure to have a 'contained' reference to the Practitioner.
In other cases, there might be enough information to construct an independent resource, but system behavior does not recognize/permit an independent identity. As an example, a prescriber might create a prescription for a custom compound. However, rather than registering the compound as a new 'Medication' instance (that could be pointed to by many prescriptions/dispenses), the system treats the compound recipe as 'part' of the prescription - it can only be edited/modified in the context of that one prescription and cannot be queried or manipulated independently. If the prescription were deleted, the Medication would be deleted as well. This would be an appropriate use of 'contained'.
A key consideration with respect to 'contained' resources is that if a resource is sent to a server as 'contained', then it is expected to remain 'contained' (at least until a formal update is received that removes the 'contained' resource and perhaps replaces it with a reference to a stand-alone resource due to additional information now being available).
For implementations concerned about minimizing the number of service calls, FHIR provides a number of options:
_include
and _revinclude
to return additional resources that are relevant as part of the queryREST can be somewhat heavy in terms of the number of service calls needed to manipulate all desired resources. This is part of the cost associated with the flexibility that allows a single RESTful interface to support a wide range of use-cases.
The FHIR specification does not define SearchParameters for most of the standard extensions we publish and obviously cannot define SearchParameters for extensions developed by implementers outside the core FHIR specification. However, implementers are free to define their own SearchParameters for any extensions they make use of and servers are free to support SearchParameters above and beyond those published as part of the FHIR core specification. There are no formal naming rules for extension-based search parameters. Authors SHOULD name the search parameters in such a way that the linkage between the extension name and the search parameter is obvious.
The expression for the search parameter would typically end with extension('http://yourextensionurlhere').value
. The expression would also need to provide a path to the extension, including handling situations where an extension appears on multiple paths if necessary. Additional constraints could also be added if necessary (e.g. .exists()
, =true
, etc.)
Common search parameters defined by this resource. See Searching for more information about searching in REST, messaging, and services.
Name | Type | Description | Paths |
_text | special | Search on the narrative of the resource |