This is the Continuous Integration Build of FHIR (will be incorrect/inconsistent at times).
See the Directory of published versions
FHIR Infrastructure ![]() | Maturity Level: N | Normative (from v4.0.0) | Security Category: Anonymous | Compartments: No defined compartments |
A Capability Statement documents a set of capabilities (behaviors) of a FHIR Server or Client for a particular version of FHIR that may be used as a statement of actual server functionality or a statement of required or desired server implementation.
The capability statement is a key part of the overall conformance framework in FHIR. It is used as a statement of the features of actual software, or of a set of rules for an application to provide. This statement connects to all the detailed statements of functionality, such as StructureDefinitions and ValueSets. This composite statement of application capability may be used for system compatibility testing, code generation, or as the basis for a conformance assessment. For further information about Conformance testing, see Conformance Rules and Profiling FHIR.
Specifically, capability statements are used in one of three ways:
Instance | implementation must be present and software may be present |
Capability | implementation must be absent, software must be present |
Requirements | implementation and software must be absent |
In this scenario, the capability statement describes the capabilities of a deployed and configured solution available at a particular access point or set of access points. The statement describes exactly how to interface with that deployed solution and thus provides for a degree of self-configuration of software solutions.
This is the type of statement that FHIR restful solutions are expected to make available on invocation of the capabilities operation. It is also the type of statement that forms a basis for the testing, certification or commissioning of specific software installations.
In this scenario, the capability statement describes generic capabilities of a software application or component solution. The solution might be available for purchase or other acquisition and might be deployed and configured at any number of independent sites. Because it is not dependent on any particular implementation, the profile cannot provide specific details such as endpoint addresses. It may also need to document various configurations in which the application can be set up or describe the degree of customizability associated with the solution.
This type of statement may be used as a marketing tool by software and system developers to formally describe their capabilities. It can also be used as the basis for conformance testing of software solutions independent of a particular installation.
In this scenario, the capability statement describes the capabilities of a desired system. It might be used as part of an architectural design process to document needed system capabilities, or might be used as part of an RFP process to formally document the requirements of a requested solution and to document the criteria by which proposals will be evaluated.
These three types of profiles can be used together. A requirements statement can be compared against the solution statements proffered by respondents to an RFP. A solution statement for a software package forms the starting point for the implementation statement associated with a particular installation of that software package.
CapabilityStatements of type "requirement" describe what capabilities are potentially relevant; additional documentation or extensions (see capabilitystatement-expectation) within the CapabilityStatement are expected to make more explicit statements of degree of expectation associated with each capability.
Capability Statements provide for a degree of automatic configuration and adaptation. However, capturing absolutely every variation that could impact the interoperability of two systems, let alone keeping that detailed information up-to-date as systems evolve through maintenance and upgrades, is rarely practical. Therefore, capability statements should be seen as an interim step. They provide a degree of automation. However, they also provide a great deal of human-readable content that can minimize the need for direct communication between the operators of the systems being configured to interoperate.
Applications may implement multiple versions. If they do, then a CapabilityStatement
describes the system's support for a particular version of FHIR, and the system will have multiple
statements, one for each version it supports. For further information, see Managing
Multiple Versions, and the $versions operation.
While the core of the CapabilityStatement
resource is Normative,
many of the flags that indicate exactly how the system operates are marked as trial-use
. Roughly,
the portions of the resource that correspond to OpenAPI document elements
are normative.
Applications looking for normative stability should only use the normative parts of the
resource, and not populate or ignore the portions labelled trial-use. To assist with this, clients
can ask for the server to return a 'Normative content only' CapabilityStatement using the mode parameter
on /metadata
.
Community discussion regarding more capable, efficient and computable representations of an applications capabilities may lead to change to the trial-use parts of this resource or the creation of new resources and/or functionality in future versions of this specification.
Additional definitions: Master Definition XML + JSON, XML Schema/Schematron + JSON Schema, ShEx (for Turtle) , the spreadsheet version & the dependency analysis
Path | ValueSet | Type | Documentation |
---|---|---|---|
CapabilityStatement.versionAlgorithm[x] | VersionAlgorithm | Extensible | Indicates the mechanism used to compare versions to determine which is more current. |
CapabilityStatement.status | PublicationStatus | Required | The lifecycle status of an artifact. |
CapabilityStatement.jurisdiction | JurisdictionValueSet | Extensible | This value set defines a base set of codes for country, country subdivision and region for indicating where a resource is intended to be used. Note: The codes for countries and country subdivisions are taken from ISO 3166 |
CapabilityStatement.kind | CapabilityStatementKind | Required | How a capability statement is intended to be used. |
CapabilityStatement.fhirVersion | FHIRVersion | Required | All published FHIR Versions. |
CapabilityStatement.format | SupplementedMimeTypes | Required | This value set includes all possible codes from BCP-13 (see http://tools.ietf.org/html/bcp13), and xml, json, and ttl |
Capability Format Type | starter | ||
CapabilityStatement.patchFormat | PatchMimeTypes | Required | This value set includes the possible codes from BCP-13 |
CapabilityStatement.acceptLanguage | AllLanguages (a valid code from Tags for the Identification of Languages ![]() |
Required | This value set includes all possible codes from BCP-47 (see http://tools.ietf.org/html/bcp47) |
Common Languages | starter | ||
CapabilityStatement.rest.mode | RestfulCapabilityMode | Required | The mode of a RESTful capability statement. |
CapabilityStatement.rest.security.service | RestfulSecurityService | Extensible | Types of security services used with FHIR. |
CapabilityStatement.rest.resource.type | ExtendedResourceTypes | Required | Current and past FHIR resource types (deleted or renamed), and additional resource types. Note that the binding to additional resources is not version fixed, and will change over time |
CapabilityStatement.rest.resource.interaction.code | TypeRestfulInteraction | Required | Operations supported by REST at the type or instance level. |
CapabilityStatement.rest.resource.versioning | ResourceVersionPolicy | Required | How the system supports versioning for a resource. |
CapabilityStatement.rest.resource.conditionalRead | ConditionalReadStatus | Required | A code that indicates how the server supports conditional read. |
CapabilityStatement.rest.resource.conditionalDelete | ConditionalDeleteStatus | Required | A code that indicates how the server supports conditional delete. |
CapabilityStatement.rest.resource.referencePolicy | ReferenceHandlingPolicy | Required | A set of flags that defines how references are supported. |
CapabilityStatement.rest.resource.searchParam.type | SearchParamType | Required | Data types allowed to be used for search parameters. |
CapabilityStatement.rest.interaction.code | SystemRestfulInteraction | Required | Operations supported by REST at the system level. |
CapabilityStatement.messaging.endpoint.protocol | MessageTransport | Extensible | The protocol used for message transport. |
CapabilityStatement.messaging.supportedMessage.mode | EventCapabilityMode | Required | The mode of a message capability statement. |
CapabilityStatement.document.mode | DocumentMode | Required | Whether the application produces or consumes documents. |
UniqueKey | Level | Location | Description | Expression |
![]() | Warning | (base) | Name should be usable as an identifier for the module by machine processing applications such as code generation | name.exists() implies name.matches('^[A-Z]([A-Za-z0-9_]){1,254}$') |
![]() | Warning | CapabilityStatement.url | URL should not contain | or # - these characters make processing canonical references problematic | exists() implies matches('^[^|# ]+$') |
![]() | Rule | (base) | A Capability Statement SHALL have at least one of REST, messaging or document element. | rest.exists() or messaging.exists() or document.exists() |
![]() | Rule | (base) | A Capability Statement SHALL have at least one of description, software, or implementation element. | (description.count() + software.count() + implementation.count()) > 0 |
![]() | Rule | (base) | Messaging end-point is only permitted when a capability statement is for an implementation. | messaging.endpoint.empty() or kind = 'instance' |
![]() | Rule | (base) | There should only be one CapabilityStatement.rest per mode. | rest.mode.isDistinct() |
![]() | Rule | (base) | The set of documents must be unique by the combination of profile and mode. | document.select(profile&mode).isDistinct() |
![]() | Rule | CapabilityStatement.rest | A given resource can only be described once per RESTful mode. | resource.select(type).isDistinct() |
![]() | Rule | CapabilityStatement.rest.resource | Search parameter names must be unique in the context of a resource. | searchParam.select(name).isDistinct() |
![]() | Rule | (base) | If kind = instance, implementation must be present and software may be present | (kind != 'instance') or implementation.exists() |
![]() | Rule | (base) | If kind = capability, implementation must be absent, software must be present | (kind != 'capability') or (implementation.exists().not() and software.exists()) |
![]() | Rule | (base) | If kind = requirements, implementation and software must be absent | (kind!='requirements') or (implementation.exists().not() and software.exists().not()) |
supportedMessage
element can be used in place of the event
and the work group believes it may meet
implementer needs better, however because the new mechanism has not yet been reviewed by ballot, the older 'event' mechanism has
been retained. Implementers may use one or the other to define their capabilities. Feedback is welcome.CapabilityStatement
resource
does not attempt to describe service-based use of resources. The various service specifications will need to describe this usage in their own way.
If a server has a default timezone, it should be indicated via the timezone code extension on
CapabilityStatement.meta
.
One of the most important parts of the CapabilityStatement is indicating what search functionality a server supports or a client uses. The search can specify the following things:
Note to Implementers: It is useful to support discovery of which reverse chaining values a server supports. Clients should not assume that servers support reverse chaining everywhere they support forward chaining. This will require a new field somewhere (e.g., in CapabilityStatement or in SearchParameter). Will review with FHIR-I, re: the best place to put this.
The most important use of support for reverse (or forward!) chaining might be in establishing conformance requirements for a server, rather than in supporting "live" discovery of a server's capabilities. (The latter could simply be tested by a client, rather than relying on possibly-incomplete or possibly-incorrect discovery data.)
Open question #1 would we want a way for servers to advertise which specific pairs of (searched resource + reverse chained search params) it support. For example, a server might need to say that the following is supported:
GET [base]/Patient?_has:Observation:subject:code=1234-5... even though the following is not:
GET [base]/Group?_has:Observation:subject:code=1234-5Open question #2: concern is that the possibility space is so broad that it might not be worth capturing all of this in such detail. Where to draw the line?
Note to Implementers: There is no way for a server to communicate how it supports search at this time. FHIR-I plans to address this and other search issues for R6.
A CapabilityStatement declares two different kinds of profiles for the functionality it describes. For a discussion of the use of these two types of resources, see two uses for profiles.
CapabilityStatements may represent the use of additional resource.
When doing so, CapabilityStatement.rest.resource.type
contains the type of the resource
as defined in the HL7 registry ,
along with a reference to the definition of the resource in
CapabilityStatement.rest.resource.definition
.
The definition SHALL contain a version specific reference to the definition. The definition SHALL NOT be present
if the resource type is one defined in this specification.
See general note to balloters about Additional resources
Search parameters for this resource. See also the full list of search parameters for this resource, and check the Extensions registry for search parameters on extensions related to this resource. The common parameters also apply. See Searching for more information about searching in REST, messaging, and services.
Name | Type | Description | Expression | In Common |
context | token | A use context assigned to the capability statement | (CapabilityStatement.useContext.value.ofType(CodeableConcept)) | 30 Resources |
context-quantity | quantity | A quantity- or range-valued use context assigned to the capability statement | (CapabilityStatement.useContext.value.ofType(Quantity)) | (CapabilityStatement.useContext.value.ofType(Range)) | 30 Resources |
context-type TU | token | A type of use context assigned to the capability statement | CapabilityStatement.useContext.code | 30 Resources |
context-type-quantity TU | composite | A use context type and quantity- or range-based value assigned to the capability statement | On CapabilityStatement.useContext: context-type: code context-quantity: value.ofType(Quantity) | value.ofType(Range) |
30 Resources |
context-type-value TU | composite | A use context type and value assigned to the capability statement | On CapabilityStatement.useContext: context-type: code context: value.ofType(CodeableConcept) |
30 Resources |
date | date | The capability statement publication date | CapabilityStatement.date | 31 Resources |
description | string | The description of the capability statement | CapabilityStatement.description | 29 Resources |
fhirversion | token | The version of FHIR | CapabilityStatement.fhirVersion | |
format | token | formats supported (xml | json | ttl | mime type) | CapabilityStatement.format | |
guide | reference | Implementation guides supported | CapabilityStatement.implementationGuide (ImplementationGuide) |
|
identifier | token | External identifier for the capability statement | CapabilityStatement.identifier | 35 Resources |
jurisdiction | token | Intended jurisdiction for the capability statement | CapabilityStatement.jurisdiction | 27 Resources |
mode | token | Mode - restful (server/client) or messaging (sender/receiver) | CapabilityStatement.rest.mode | |
name | string | Computationally friendly name of the capability statement | CapabilityStatement.name | 28 Resources |
publisher | string | Name of the publisher of the capability statement | CapabilityStatement.publisher | 31 Resources |
resource | token | Name of a resource mentioned in a capability statement | CapabilityStatement.rest.resource.type | |
resource-profile | reference | A profile id invoked in a capability statement | CapabilityStatement.rest.resource.profile (StructureDefinition) |
|
security-service | token | OAuth | SMART-on-FHIR | NTLM | Basic | Kerberos | Certificates | CapabilityStatement.rest.security.service | |
software | string | Part of the name of a software application | CapabilityStatement.software.name | |
status | token | The current status of the capability statement | CapabilityStatement.status | 35 Resources |
supported-profile TU | reference | Profiles for use cases supported | CapabilityStatement.rest.resource.supportedProfile (StructureDefinition) |
|
title | string | The human-friendly name of the capability statement | CapabilityStatement.title | 28 Resources |
url | uri | The uri that identifies the capability statement | CapabilityStatement.url | 34 Resources |
version | token | The business version of the capability statement | CapabilityStatement.version | 32 Resources |